[OPENMP] Outlined function for parallel and other regions with list of captured variables.

Currently all variables used in OpenMP regions are captured into a record and passed to outlined functions in this record. It may result in some poor performance because of too complex analysis later in optimization passes. Patch makes to emit outlined functions for parallel-based regions with a list of captured variables. It reduces code for 2*n GEPs, stores and loads at least.
Codegen for task-based regions remains unchanged because runtime requires that all captured variables are passed in captured record.


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@247251 91177308-0d34-0410-b5e6-96231b3b80d8

31 files changed, 542 insertions, 616 deletions

diff --git a/lib/CodeGen/CGExpr.cpp b/lib/CodeGen/CGExpr.cpp
index 5e6c4de408..dd3c69ccf8 100644
--- a/lib/CodeGen/CGExpr.cpp
+++ b/lib/CodeGen/CGExpr.cpp
@@ -2033,7 +2033,10 @@ LValue CodeGenFunction::EmitDeclRefLValue(const DeclRefExpr *E) {
     const Expr *Init = VD->getAnyInitializer(VD);
     if (Init && !isa<ParmVarDecl>(VD) && VD->getType()->isReferenceType() &&
         VD->isUsableInConstantExpressions(getContext()) &&
-        VD->checkInitIsICE()) {
+        VD->checkInitIsICE() &&
+        // Do not emit if it is private OpenMP variable.
+        !(E->refersToEnclosingVariableOrCapture() && CapturedStmtInfo &&
+          LocalDeclMap.count(VD))) {
       llvm::Constant *Val =
         CGM.EmitConstantValue(*VD->evaluateValue(), VD->getType(), this);
       assert(Val && "failed to emit reference constant expression");
diff --git a/lib/CodeGen/CGOpenMPRuntime.cpp b/lib/CodeGen/CGOpenMPRuntime.cpp
index a0b3ee5ab0..79900778fd 100644
--- a/lib/CodeGen/CGOpenMPRuntime.cpp
+++ b/lib/CodeGen/CGOpenMPRuntime.cpp
@@ -231,14 +231,18 @@ public:
 
 } // namespace
 
+static LValue emitLoadOfPointerLValue(CodeGenFunction &CGF, Address PtrAddr,
+                                      QualType Ty) {
+  AlignmentSource Source;
+  CharUnits Align = CGF.getNaturalPointeeTypeAlignment(Ty, &Source);
+  return CGF.MakeAddrLValue(Address(CGF.Builder.CreateLoad(PtrAddr), Align),
+                            Ty->getPointeeType(), Source);
+}
+
 LValue CGOpenMPRegionInfo::getThreadIDVariableLValue(CodeGenFunction &CGF) {
-  return CGF.MakeNaturalAlignAddrLValue(
-      CGF.Builder.CreateLoad(
-          CGF.GetAddrOfLocalVar(getThreadIDVariable())),
-      getThreadIDVariable()
-          ->getType()
-          ->castAs<PointerType>()
-          ->getPointeeType());
+  return emitLoadOfPointerLValue(CGF,
+                                 CGF.GetAddrOfLocalVar(getThreadIDVariable()),
+                                 getThreadIDVariable()->getType());
 }
 
 void CGOpenMPRegionInfo::EmitBody(CodeGenFunction &CGF, const Stmt * /*S*/) {
@@ -257,9 +261,9 @@ void CGOpenMPRegionInfo::EmitBody(CodeGenFunction &CGF, const Stmt * /*S*/) {
 
 LValue CGOpenMPTaskOutlinedRegionInfo::getThreadIDVariableLValue(
     CodeGenFunction &CGF) {
-  return CGF.MakeAddrLValue(
-      CGF.GetAddrOfLocalVar(getThreadIDVariable()),
-      getThreadIDVariable()->getType());
+  return CGF.MakeAddrLValue(CGF.GetAddrOfLocalVar(getThreadIDVariable()),
+                            getThreadIDVariable()->getType(),
+                            AlignmentSource::Decl);
 }
 
 CGOpenMPRuntime::CGOpenMPRuntime(CodeGenModule &CGM)
@@ -307,7 +311,7 @@ llvm::Value *CGOpenMPRuntime::emitParallelOutlinedFunction(
   CodeGenFunction CGF(CGM, true);
   CGOpenMPOutlinedRegionInfo CGInfo(*CS, ThreadIDVar, CodeGen, InnermostKind);
   CodeGenFunction::CGCapturedStmtRAII CapInfoRAII(CGF, &CGInfo);
-  return CGF.GenerateCapturedStmtFunction(*CS);
+  return CGF.GenerateOpenMPCapturedStmtFunction(*CS);
 }
 
 llvm::Value *CGOpenMPRuntime::emitTaskOutlinedFunction(
@@ -1168,25 +1172,25 @@ static void emitOMPIfClause(CodeGenFunction &CGF, const Expr *Cond,
 
 void CGOpenMPRuntime::emitParallelCall(CodeGenFunction &CGF, SourceLocation Loc,
                                        llvm::Value *OutlinedFn,
-                                       Address CapturedStruct,
+                                       ArrayRef<llvm::Value *> CapturedVars,
                                        const Expr *IfCond) {
   auto *RTLoc = emitUpdateLocation(CGF, Loc);
-  auto &&ThenGen =
-      [this, OutlinedFn, CapturedStruct, RTLoc](CodeGenFunction &CGF) {
-        // Build call __kmpc_fork_call(loc, 1, microtask,
-        // captured_struct/*context*/)
-        llvm::Value *Args[] = {
-            RTLoc,
-            CGF.Builder.getInt32(
-                1), // Number of arguments after 'microtask' argument
-            // (there is only one additional argument - 'context')
-            CGF.Builder.CreateBitCast(OutlinedFn, getKmpc_MicroPointerTy()),
-            CGF.EmitCastToVoidPtr(CapturedStruct.getPointer())};
-        auto RTLFn = createRuntimeFunction(OMPRTL__kmpc_fork_call);
-        CGF.EmitRuntimeCall(RTLFn, Args);
-      };
-  auto &&ElseGen = [this, OutlinedFn, CapturedStruct, RTLoc, Loc](
-      CodeGenFunction &CGF) {
+  auto &&ThenGen = [this, OutlinedFn, CapturedVars,
+                    RTLoc](CodeGenFunction &CGF) {
+    // Build call __kmpc_fork_call(loc, n, microtask, var1, .., varn);
+    llvm::Value *Args[] = {
+        RTLoc,
+        CGF.Builder.getInt32(CapturedVars.size()), // Number of captured vars
+        CGF.Builder.CreateBitCast(OutlinedFn, getKmpc_MicroPointerTy())};
+    llvm::SmallVector<llvm::Value *, 16> RealArgs;
+    RealArgs.append(std::begin(Args), std::end(Args));
+    RealArgs.append(CapturedVars.begin(), CapturedVars.end());
+
+    auto RTLFn = createRuntimeFunction(OMPRTL__kmpc_fork_call);
+    CGF.EmitRuntimeCall(RTLFn, RealArgs);
+  };
+  auto &&ElseGen = [this, OutlinedFn, CapturedVars, RTLoc,
+                    Loc](CodeGenFunction &CGF) {
     auto ThreadID = getThreadID(CGF, Loc);
     // Build calls:
     // __kmpc_serialized_parallel(&Loc, GTid);
@@ -1200,11 +1204,10 @@ void CGOpenMPRuntime::emitParallelCall(CodeGenFunction &CGF, SourceLocation Loc,
       CGF.CreateTempAlloca(CGF.Int32Ty, CharUnits::fromQuantity(4),
                            /*Name*/ ".zero.addr");
     CGF.InitTempAlloca(ZeroAddr, CGF.Builder.getInt32(/*C*/ 0));
-    llvm::Value *OutlinedFnArgs[] = {
-      ThreadIDAddr.getPointer(),
-      ZeroAddr.getPointer(),
-      CapturedStruct.getPointer()
-    };
+    llvm::SmallVector<llvm::Value *, 16> OutlinedFnArgs;
+    OutlinedFnArgs.push_back(ThreadIDAddr.getPointer());
+    OutlinedFnArgs.push_back(ZeroAddr.getPointer());
+    OutlinedFnArgs.append(CapturedVars.begin(), CapturedVars.end());
     CGF.EmitCallOrInvoke(OutlinedFn, OutlinedFnArgs);
 
     // __kmpc_end_serialized_parallel(&Loc, GTid);
@@ -1946,7 +1949,8 @@ emitProxyTaskFunction(CodeGenModule &CGM, SourceLocation Loc,
   FunctionArgList Args;
   ImplicitParamDecl GtidArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr, KmpInt32Ty);
   ImplicitParamDecl TaskTypeArg(C, /*DC=*/nullptr, Loc,
-                                /*Id=*/nullptr, KmpTaskTWithPrivatesPtrQTy);
+                                /*Id=*/nullptr,
+                                KmpTaskTWithPrivatesPtrQTy.withRestrict());
   Args.push_back(&GtidArg);
   Args.push_back(&TaskTypeArg);
   FunctionType::ExtInfo Info;
@@ -1966,10 +1970,8 @@ emitProxyTaskFunction(CodeGenModule &CGM, SourceLocation Loc,
   // tt->task_data.shareds);
   auto *GtidParam = CGF.EmitLoadOfScalar(
       CGF.GetAddrOfLocalVar(&GtidArg), /*Volatile=*/false, KmpInt32Ty, Loc);
-  auto *TaskTypeArgAddr = CGF.Builder.CreateLoad(
-      CGF.GetAddrOfLocalVar(&TaskTypeArg));
-  LValue TDBase =
-      CGF.MakeNaturalAlignAddrLValue(TaskTypeArgAddr, KmpTaskTWithPrivatesQTy);
+  LValue TDBase = emitLoadOfPointerLValue(
+      CGF, CGF.GetAddrOfLocalVar(&TaskTypeArg), KmpTaskTWithPrivatesPtrQTy);
   auto *KmpTaskTWithPrivatesQTyRD =
       cast<RecordDecl>(KmpTaskTWithPrivatesQTy->getAsTagDecl());
   LValue Base =
@@ -2014,7 +2016,8 @@ static llvm::Value *emitDestructorsFunction(CodeGenModule &CGM,
   FunctionArgList Args;
   ImplicitParamDecl GtidArg(C, /*DC=*/nullptr, Loc, /*Id=*/nullptr, KmpInt32Ty);
   ImplicitParamDecl TaskTypeArg(C, /*DC=*/nullptr, Loc,
-                                /*Id=*/nullptr, KmpTaskTWithPrivatesPtrQTy);
+                                /*Id=*/nullptr,
+                                KmpTaskTWithPrivatesPtrQTy.withRestrict());
   Args.push_back(&GtidArg);
   Args.push_back(&TaskTypeArg);
   FunctionType::ExtInfo Info;
@@ -2031,10 +2034,8 @@ static llvm::Value *emitDestructorsFunction(CodeGenModule &CGM,
   CGF.StartFunction(GlobalDecl(), KmpInt32Ty, DestructorFn, DestructorFnInfo,
                     Args);
 
-  auto *TaskTypeArgAddr = CGF.Builder.CreateLoad(
-      CGF.GetAddrOfLocalVar(&TaskTypeArg));
-  LValue Base =
-      CGF.MakeNaturalAlignAddrLValue(TaskTypeArgAddr, KmpTaskTWithPrivatesQTy);
+  LValue Base = emitLoadOfPointerLValue(
+      CGF, CGF.GetAddrOfLocalVar(&TaskTypeArg), KmpTaskTWithPrivatesPtrQTy);
   auto *KmpTaskTWithPrivatesQTyRD =
       cast<RecordDecl>(KmpTaskTWithPrivatesQTy->getAsTagDecl());
   auto FI = std::next(KmpTaskTWithPrivatesQTyRD->field_begin());
@@ -2112,21 +2113,17 @@ emitTaskPrivateMappingFunction(CodeGenModule &CGM, SourceLocation Loc,
                     TaskPrivatesMapFnInfo, Args);
 
   // *privi = &.privates.privi;
-  auto *TaskPrivatesArgAddr = CGF.Builder.CreateLoad(
-      CGF.GetAddrOfLocalVar(&TaskPrivatesArg));
-  LValue Base =
-      CGF.MakeNaturalAlignAddrLValue(TaskPrivatesArgAddr, PrivatesQTy);
+  LValue Base = emitLoadOfPointerLValue(
+      CGF, CGF.GetAddrOfLocalVar(&TaskPrivatesArg), TaskPrivatesArg.getType());
   auto *PrivatesQTyRD = cast<RecordDecl>(PrivatesQTy->getAsTagDecl());
   Counter = 0;
   for (auto *Field : PrivatesQTyRD->fields()) {
     auto FieldLVal = CGF.EmitLValueForField(Base, Field);
     auto *VD = Args[PrivateVarsPos[Privates[Counter].second.Original]];
     auto RefLVal = CGF.MakeAddrLValue(CGF.GetAddrOfLocalVar(VD), VD->getType());
-    auto RefLoadRVal = CGF.EmitLoadOfLValue(RefLVal, Loc);
-    CGF.EmitStoreOfScalar(
-        FieldLVal.getPointer(),
-        CGF.MakeNaturalAlignAddrLValue(RefLoadRVal.getScalarVal(),
-                                       RefLVal.getType()->getPointeeType()));
+    auto RefLoadLVal =
+        emitLoadOfPointerLValue(CGF, RefLVal.getAddress(), RefLVal.getType());
+    CGF.EmitStoreOfScalar(FieldLVal.getPointer(), RefLoadLVal);
     ++Counter;
   }
   CGF.FinishFunction();
@@ -2271,10 +2268,13 @@ void CGOpenMPRuntime::emitTaskCall(
   // Copy shareds if there are any.
   Address KmpTaskSharedsPtr = Address::invalid();
   if (!SharedsTy->getAsStructureType()->getDecl()->field_empty()) {
-    KmpTaskSharedsPtr = Address(CGF.EmitLoadOfScalar(
-        CGF.EmitLValueForField(
-            TDBase, *std::next(KmpTaskTQTyRD->field_begin(), KmpTaskTShareds)),
-        Loc), CGF.getNaturalTypeAlignment(SharedsTy));
+    KmpTaskSharedsPtr =
+        Address(CGF.EmitLoadOfScalar(
+                    CGF.EmitLValueForField(
+                        TDBase, *std::next(KmpTaskTQTyRD->field_begin(),
+                                           KmpTaskTShareds)),
+                    Loc),
+                CGF.getNaturalTypeAlignment(SharedsTy));
     CGF.EmitAggregateCopy(KmpTaskSharedsPtr, Shareds, SharedsTy);
   }
   // Emit initial values for private copies (if any).
diff --git a/lib/CodeGen/CGOpenMPRuntime.h b/lib/CodeGen/CGOpenMPRuntime.h
index da6dc108c4..d3c7728c64 100644
--- a/lib/CodeGen/CGOpenMPRuntime.h
+++ b/lib/CodeGen/CGOpenMPRuntime.h
@@ -405,7 +405,8 @@ public:
   ///
   virtual void emitParallelCall(CodeGenFunction &CGF, SourceLocation Loc,
                                 llvm::Value *OutlinedFn,
-                                Address CapturedStruct, const Expr *IfCond);
+                                ArrayRef<llvm::Value *> CapturedVars,
+                                const Expr *IfCond);
 
   /// \brief Emits a critical region.
   /// \param CriticalName Name of the critical region.
diff --git a/lib/CodeGen/CGStmtOpenMP.cpp b/lib/CodeGen/CGStmtOpenMP.cpp
index 3ac3dfaf45..d7c411c073 100644
--- a/lib/CodeGen/CGStmtOpenMP.cpp
+++ b/lib/CodeGen/CGStmtOpenMP.cpp
@@ -20,6 +20,115 @@
 using namespace clang;
 using namespace CodeGen;
 
+void CodeGenFunction::GenerateOpenMPCapturedVars(
+    const CapturedStmt &S, SmallVectorImpl<llvm::Value *> &CapturedVars) {
+  const RecordDecl *RD = S.getCapturedRecordDecl();
+  auto CurField = RD->field_begin();
+  auto CurCap = S.captures().begin();
+  for (CapturedStmt::const_capture_init_iterator I = S.capture_init_begin(),
+                                                 E = S.capture_init_end();
+       I != E; ++I, ++CurField, ++CurCap) {
+    if (CurField->hasCapturedVLAType()) {
+      auto VAT = CurField->getCapturedVLAType();
+      CapturedVars.push_back(VLASizeMap[VAT->getSizeExpr()]);
+    } else if (CurCap->capturesThis())
+      CapturedVars.push_back(CXXThisValue);
+    else
+      CapturedVars.push_back(EmitLValue(*I).getAddress().getPointer());
+  }
+}
+
+llvm::Function *
+CodeGenFunction::GenerateOpenMPCapturedStmtFunction(const CapturedStmt &S) {
+  assert(
+      CapturedStmtInfo &&
+      "CapturedStmtInfo should be set when generating the captured function");
+  const CapturedDecl *CD = S.getCapturedDecl();
+  const RecordDecl *RD = S.getCapturedRecordDecl();
+  assert(CD->hasBody() && "missing CapturedDecl body");
+
+  // Build the argument list.
+  ASTContext &Ctx = CGM.getContext();
+  FunctionArgList Args;
+  Args.append(CD->param_begin(),
+              std::next(CD->param_begin(), CD->getContextParamPosition()));
+  auto I = S.captures().begin();
+  for (auto *FD : RD->fields()) {
+    QualType ArgType = FD->getType();
+    IdentifierInfo *II = nullptr;
+    VarDecl *CapVar = nullptr;
+    if (I->capturesVariable()) {
+      CapVar = I->getCapturedVar();
+      II = CapVar->getIdentifier();
+    } else if (I->capturesThis())
+      II = &getContext().Idents.get("this");
+    else {
+      assert(I->capturesVariableArrayType());
+      II = &getContext().Idents.get("vla");
+    }
+    if (ArgType->isVariablyModifiedType())
+      ArgType = getContext().getVariableArrayDecayedType(ArgType);
+    Args.push_back(ImplicitParamDecl::Create(getContext(), nullptr,
+                                             FD->getLocation(), II, ArgType));
+    ++I;
+  }
+  Args.append(
+      std::next(CD->param_begin(), CD->getContextParamPosition() + 1),
+      CD->param_end());
+
+  // Create the function declaration.
+  FunctionType::ExtInfo ExtInfo;
+  const CGFunctionInfo &FuncInfo =
+      CGM.getTypes().arrangeFreeFunctionDeclaration(Ctx.VoidTy, Args, ExtInfo,
+                                                    /*IsVariadic=*/false);
+  llvm::FunctionType *FuncLLVMTy = CGM.getTypes().GetFunctionType(FuncInfo);
+
+  llvm::Function *F = llvm::Function::Create(
+      FuncLLVMTy, llvm::GlobalValue::InternalLinkage,
+      CapturedStmtInfo->getHelperName(), &CGM.getModule());
+  CGM.SetInternalFunctionAttributes(CD, F, FuncInfo);
+  if (CD->isNothrow())
+    F->addFnAttr(llvm::Attribute::NoUnwind);
+
+  // Generate the function.
+  StartFunction(CD, Ctx.VoidTy, F, FuncInfo, Args, CD->getLocation(),
+                CD->getBody()->getLocStart());
+  unsigned Cnt = CD->getContextParamPosition();
+  I = S.captures().begin();
+  for (auto *FD : RD->fields()) {
+    LValue ArgLVal =
+        MakeAddrLValue(GetAddrOfLocalVar(Args[Cnt]), Args[Cnt]->getType(),
+                       AlignmentSource::Decl);
+    if (FD->hasCapturedVLAType()) {
+      auto *ExprArg =
+          EmitLoadOfLValue(ArgLVal, SourceLocation()).getScalarVal();
+      auto VAT = FD->getCapturedVLAType();
+      VLASizeMap[VAT->getSizeExpr()] = ExprArg;
+    } else if (I->capturesVariable()) {
+      auto *Var = I->getCapturedVar();
+      QualType VarTy = Var->getType();
+      Address ArgAddr = ArgLVal.getAddress();
+      if (!VarTy->isReferenceType()) {
+        ArgAddr = EmitLoadOfReference(
+            ArgAddr, ArgLVal.getType()->castAs<ReferenceType>());
+      }
+      setAddrOfLocalVar(Var, ArgAddr);
+    } else {
+      // If 'this' is captured, load it into CXXThisValue.
+      assert(I->capturesThis());
+      CXXThisValue =
+          EmitLoadOfLValue(ArgLVal, Args[Cnt]->getLocation()).getScalarVal();
+    }
+    ++Cnt, ++I;
+  }
+
+  PGO.assignRegionCounters(CD, F);
+  CapturedStmtInfo->EmitBody(*this, CD->getBody());
+  FinishFunction(CD->getBodyRBrace());
+
+  return F;
+}
+
 //===----------------------------------------------------------------------===//
 //                              OpenMP Directive Emission
 //===----------------------------------------------------------------------===//
@@ -246,6 +355,7 @@ bool CodeGenFunction::EmitOMPCopyinClause(const OMPExecutableDirective &D) {
           DeclRefExpr DRE(const_cast<VarDecl *>(VD), true, (*IRef)->getType(),
                           VK_LValue, (*IRef)->getExprLoc());
           MasterAddr = EmitLValue(&DRE).getAddress();
+          LocalDeclMap.erase(VD);
         } else {
           MasterAddr =
             Address(VD->isStaticLocal() ? CGM.getStaticLocalDeclAddress(VD)
@@ -473,7 +583,8 @@ static void emitCommonOMPParallelDirective(CodeGenFunction &CGF,
                                            OpenMPDirectiveKind InnermostKind,
                                            const RegionCodeGenTy &CodeGen) {
   auto CS = cast<CapturedStmt>(S.getAssociatedStmt());
-  auto CapturedStruct = CGF.GenerateCapturedStmtArgument(*CS);
+  llvm::SmallVector<llvm::Value *, 16> CapturedVars;
+  CGF.GenerateOpenMPCapturedVars(*CS, CapturedVars);
   auto OutlinedFn = CGF.CGM.getOpenMPRuntime().emitParallelOutlinedFunction(
       S, *CS->getCapturedDecl()->param_begin(), InnermostKind, CodeGen);
   if (const auto *NumThreadsClause = S.getSingleClause<OMPNumThreadsClause>()) {
@@ -497,7 +608,7 @@ static void emitCommonOMPParallelDirective(CodeGenFunction &CGF,
     }
   }
   CGF.CGM.getOpenMPRuntime().emitParallelCall(CGF, S.getLocStart(), OutlinedFn,
-                                              CapturedStruct, IfCond);
+                                              CapturedVars, IfCond);
 }
 
 void CodeGenFunction::EmitOMPParallelDirective(const OMPParallelDirective &S) {
diff --git a/lib/CodeGen/CodeGenFunction.h b/lib/CodeGen/CodeGenFunction.h
index 34112727a1..76296dee13 100644
--- a/lib/CodeGen/CodeGenFunction.h
+++ b/lib/CodeGen/CodeGenFunction.h
@@ -2204,10 +2204,11 @@ public:
 
   LValue InitCapturedStruct(const CapturedStmt &S);
   llvm::Function *EmitCapturedStmt(const CapturedStmt &S, CapturedRegionKind K);
-  void GenerateCapturedStmtFunctionProlog(const CapturedStmt &S);
-  llvm::Function *GenerateCapturedStmtFunctionEpilog(const CapturedStmt &S);
   llvm::Function *GenerateCapturedStmtFunction(const CapturedStmt &S);
   Address GenerateCapturedStmtArgument(const CapturedStmt &S);
+  llvm::Function *GenerateOpenMPCapturedStmtFunction(const CapturedStmt &S);
+  void GenerateOpenMPCapturedVars(const CapturedStmt &S,
+                                  SmallVectorImpl<llvm::Value *> &CapturedVars);
   /// \brief Perform element by element copying of arrays with type \a
   /// OriginalType from \a SrcAddr to \a DestAddr using copying procedure
   /// generated by \a CopyGen.
diff --git a/lib/Sema/SemaOpenMP.cpp b/lib/Sema/SemaOpenMP.cpp
index 6abdd9d8cd..1a7c7760ef 100644
--- a/lib/Sema/SemaOpenMP.cpp
+++ b/lib/Sema/SemaOpenMP.cpp
@@ -1158,7 +1158,8 @@ void Sema::ActOnOpenMPRegionStart(OpenMPDirectiveKind DKind, Scope *CurScope) {
   switch (DKind) {
   case OMPD_parallel: {
     QualType KmpInt32Ty = Context.getIntTypeForBitwidth(32, 1);
-    QualType KmpInt32PtrTy = Context.getPointerType(KmpInt32Ty);
+    QualType KmpInt32PtrTy =
+        Context.getPointerType(KmpInt32Ty).withConst().withRestrict();
     Sema::CapturedParamNameType Params[] = {
         std::make_pair(".global_tid.", KmpInt32PtrTy),
         std::make_pair(".bound_tid.", KmpInt32PtrTy),
@@ -1234,7 +1235,8 @@ void Sema::ActOnOpenMPRegionStart(OpenMPDirectiveKind DKind, Scope *CurScope) {
   }
   case OMPD_parallel_for: {
     QualType KmpInt32Ty = Context.getIntTypeForBitwidth(32, 1);
-    QualType KmpInt32PtrTy = Context.getPointerType(KmpInt32Ty);
+    QualType KmpInt32PtrTy =
+        Context.getPointerType(KmpInt32Ty).withConst().withRestrict();
     Sema::CapturedParamNameType Params[] = {
         std::make_pair(".global_tid.", KmpInt32PtrTy),
         std::make_pair(".bound_tid.", KmpInt32PtrTy),
@@ -1246,7 +1248,8 @@ void Sema::ActOnOpenMPRegionStart(OpenMPDirectiveKind DKind, Scope *CurScope) {
   }
   case OMPD_parallel_for_simd: {
     QualType KmpInt32Ty = Context.getIntTypeForBitwidth(32, 1);
-    QualType KmpInt32PtrTy = Context.getPointerType(KmpInt32Ty);
+    QualType KmpInt32PtrTy =
+        Context.getPointerType(KmpInt32Ty).withConst().withRestrict();
     Sema::CapturedParamNameType Params[] = {
         std::make_pair(".global_tid.", KmpInt32PtrTy),
         std::make_pair(".bound_tid.", KmpInt32PtrTy),
@@ -1258,7 +1261,8 @@ void Sema::ActOnOpenMPRegionStart(OpenMPDirectiveKind DKind, Scope *CurScope) {
   }
   case OMPD_parallel_sections: {
     QualType KmpInt32Ty = Context.getIntTypeForBitwidth(32, 1);
-    QualType KmpInt32PtrTy = Context.getPointerType(KmpInt32Ty);
+    QualType KmpInt32PtrTy =
+        Context.getPointerType(KmpInt32Ty).withConst().withRestrict();
     Sema::CapturedParamNameType Params[] = {
         std::make_pair(".global_tid.", KmpInt32PtrTy),
         std::make_pair(".bound_tid.", KmpInt32PtrTy),
@@ -1320,7 +1324,8 @@ void Sema::ActOnOpenMPRegionStart(OpenMPDirectiveKind DKind, Scope *CurScope) {
   }
   case OMPD_teams: {
     QualType KmpInt32Ty = Context.getIntTypeForBitwidth(32, 1);
-    QualType KmpInt32PtrTy = Context.getPointerType(KmpInt32Ty);
+    QualType KmpInt32PtrTy =
+        Context.getPointerType(KmpInt32Ty).withConst().withRestrict();
     Sema::CapturedParamNameType Params[] = {
         std::make_pair(".global_tid.", KmpInt32PtrTy),
         std::make_pair(".bound_tid.", KmpInt32PtrTy),
diff --git a/test/OpenMP/for_firstprivate_codegen.cpp b/test/OpenMP/for_firstprivate_codegen.cpp
index a8a44374d8..3ac32b6958 100644
--- a/test/OpenMP/for_firstprivate_codegen.cpp
+++ b/test/OpenMP/for_firstprivate_codegen.cpp
@@ -31,7 +31,6 @@ struct S {
 // CHECK-DAG: [[S_FLOAT_TY:%.+]] = type { float }
 // CHECK-DAG: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
 // CHECK-DAG: [[ST_TY:%.+]] = type { i{{[0-9]+}}, i{{[0-9]+}} }
-// CHECK-DAG: [[CAP_TMAIN_TY:%.+]] = type { i{{[0-9]+}}*, [2 x i{{[0-9]+}}]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]* }
 
 template <typename T>
 T tmain() {
@@ -70,11 +69,11 @@ int main() {
   // LAMBDA: call void [[OUTER_LAMBDA:@.+]](
   [&]() {
 // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-// LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+// LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp for firstprivate(g, g1)
   for (int i = 0; i < 2; ++i) {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // Skip temp vars for loop
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
@@ -113,11 +112,11 @@ int main() {
   // BLOCKS: call void {{%.+}}(i8
   ^{
 // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-// BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+// BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp for firstprivate(g, g1)
   for (int i = 0; i < 2; ++i) {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // Skip temp vars for loop
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
@@ -217,11 +216,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [2 x i32]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call {{.*}} [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}}, [2 x i32]* dereferenceable(8) %{{.+}}, [2 x [[S_INT_TY]]]* dereferenceable(8) %{{.+}}, [[S_INT_TY]]* dereferenceable(4) %{{.+}})
 // Skip temp vars for loop
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
@@ -234,22 +233,20 @@ int main() {
 // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]],
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
+// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** %
+// CHECK: [[S_ARR:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** %
+
 // firstprivate t_var(t_var)
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_PTR_REF]],
 // CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_REF]],
 // CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_PRIV]],
 
 // firstprivate vec(vec)
-// CHECK: [[VEC_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_PTR_REF:%.+]],
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
 // CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
 
 // firstprivate s_arr(s_arr)
-// CHECK: [[S_ARR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 2
-// CHECK: [[S_ARR:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** [[S_ARR_REF]],
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
 // CHECK: [[S_ARR_PRIV_END:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_PRIV_BEGIN]], i{{[0-9]+}} 2
 // CHECK: [[IS_EMPTY:%.+]] = icmp eq [[S_INT_TY]]* [[S_ARR_PRIV_BEGIN]], [[S_ARR_PRIV_END]]
@@ -261,8 +258,7 @@ int main() {
 // CHECK: br i1 {{.+}}, label %{{.+}}, label %[[S_ARR_BODY]]
 
 // firstprivate var(var)
-// CHECK: [[VAR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 3
-// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_REF_PTR]],
+// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
 // CHECK: call {{.*}} [[ST_TY_DEFAULT_CONSTR]]([[ST_TY]]* [[ST_TY_TEMP:%.+]])
 // CHECK: call {{.*}} [[S_INT_TY_COPY_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]], [[S_INT_TY]]* {{.*}} [[VAR_REF]], [[ST_TY]]* [[ST_TY_TEMP]])
 // CHECK: call {{.*}} [[ST_TY_DESTR]]([[ST_TY]]* [[ST_TY_TEMP]])
diff --git a/test/OpenMP/for_lastprivate_codegen.cpp b/test/OpenMP/for_lastprivate_codegen.cpp
index b1a49383c9..cada3a06e5 100644
--- a/test/OpenMP/for_lastprivate_codegen.cpp
+++ b/test/OpenMP/for_lastprivate_codegen.cpp
@@ -24,9 +24,7 @@ float f;
 char cnt;
 
 // CHECK: [[S_FLOAT_TY:%.+]] = type { float }
-// CHECK: [[CAP_MAIN_TY:%.+]] = type { i{{[0-9]+}}*, [2 x i{{[0-9]+}}]*, [2 x [[S_FLOAT_TY]]]*, [[S_FLOAT_TY]]* }
 // CHECK: [[S_INT_TY:%.+]] = type { i32 }
-// CHECK: [[CAP_TMAIN_TY:%.+]] = type { i{{[0-9]+}}*, [2 x i{{[0-9]+}}]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]* }
 // CHECK-DAG: [[IMPLICIT_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 66, i32 0, i32 0, i8*
 // CHECK-DAG: [[X:@.+]] = global double 0.0
 // CHECK-DAG: [[F:@.+]] = global float 0.0
@@ -61,18 +59,17 @@ int main() {
   // LAMBDA: call void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp for lastprivate(g, g1)
   for (int i = 0; i < 2; ++i) {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // LAMBDA: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     // LAMBDA: [[GTID_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %{{.+}}
     // LAMBDA: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_REF]]
     // LAMBDA: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 [[GTID]], i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
@@ -115,18 +112,17 @@ int main() {
   // BLOCKS: call void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp for lastprivate(g, g1)
   for (int i = 0; i < 2; ++i) {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // BLOCKS: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     // BLOCKS: [[GTID_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %{{.+}}
     // BLOCKS: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_REF]]
     // BLOCKS: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 [[GTID]], i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
@@ -199,16 +195,15 @@ int main() {
 // CHECK: define i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, %{{.+}}*)* [[MAIN_MICROTASK1:@.+]] to void
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, %{{.+}}*)* [[MAIN_MICROTASK2:@.+]] to void
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, %{{.+}}*)* [[MAIN_MICROTASK3:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [2 x i32]*, [2 x [[S_FLOAT_TY]]]*, [[S_FLOAT_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[MAIN_MICROTASK1:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[MAIN_MICROTASK2:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[MAIN_MICROTASK3:@.+]] to void
 // CHECK: = call {{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call void [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 
-// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK]](i32* noalias [[GTID_ADDR:%.+]], i32* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}}, [2 x i32]* dereferenceable(8) %{{.+}}, [2 x [[S_FLOAT_TY]]]* dereferenceable(8) %{{.+}}, [[S_FLOAT_TY]]* dereferenceable(4) %{{.+}})
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
@@ -220,19 +215,16 @@ int main() {
 // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
 
+// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+// CHECK: [[VEC_REF:%.+]] = load [2 x i32]*, [2 x i32]** %
+// CHECK: [[S_ARR_REF:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** %
+// CHECK: [[VAR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** %
+
 // Check for default initialization.
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_PTR_REF]],
 // CHECK-NOT: [[T_VAR_PRIV]]
-// CHECK: [[VEC_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_PTR_REF:%.+]],
 // CHECK-NOT: [[VEC_PRIV]]
-// CHECK: [[S_ARR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 2
-// CHECK: [[S_ARR_REF:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[S_ARR_REF_PTR]],
 // CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_FLOAT_TY]]*
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[S_ARR_PRIV_ITEM]])
-// CHECK: [[VAR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 3
-// CHECK: [[VAR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[VAR_REF_PTR]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
 // CHECK: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 %{{.+}}, i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
 // <Skip loop body>
@@ -277,7 +269,7 @@ int main() {
 // CHECK: ret void
 
 //
-// CHECK: define internal void [[MAIN_MICROTASK1]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK1]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: [[F_PRIV:%.+]] = alloca float,
 // CHECK-NOT: alloca float
 // CHECK: [[X_PRIV:%.+]] = alloca double,
@@ -318,7 +310,7 @@ int main() {
 // CHECK: call i32 @__kmpc_cancel_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i{{[0-9]+}} [[GTID]])
 // CHECK: ret void
 
-// CHECK: define internal void [[MAIN_MICROTASK2]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK2]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: [[F_PRIV:%.+]] = alloca float,
 // CHECK-NOT: alloca float
 
@@ -350,7 +342,7 @@ int main() {
 // CHECK: call i32 @__kmpc_cancel_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i{{[0-9]+}} [[GTID]])
 // CHECK: ret void
 
-// CHECK: define internal void [[MAIN_MICROTASK3]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK3]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: [[CNT_PRIV:%.+]] = alloca i8,
 
 // CHECK: [[GTID_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[GTID_ADDR_REF]]
@@ -400,11 +392,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [2 x i32]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call void [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}}, [2 x i32]* dereferenceable(8) %{{.+}}, [2 x [[S_INT_TY]]]* dereferenceable(8) %{{.+}}, [[S_INT_TY]]* dereferenceable(4) %{{.+}})
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
@@ -417,19 +409,16 @@ int main() {
 // CHECK: [[VAR_PRIV_REF:%.+]] = alloca [[S_INT_TY]]*,
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
 
+// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** %
+// CHECK: [[S_ARR_REF:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** %
+
 // Check for default initialization.
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_PTR_REF]],
 // CHECK-NOT: [[T_VAR_PRIV]]
-// CHECK: [[VEC_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_PTR_REF:%.+]],
 // CHECK-NOT: [[VEC_PRIV]]
-// CHECK: [[S_ARR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 2
-// CHECK: [[S_ARR_REF:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** [[S_ARR_REF_PTR]],
 // CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_INT_TY]]*
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]])
-// CHECK: [[VAR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 3
-// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_REF_PTR]],
+// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]])
 // CHECK: store [[S_INT_TY]]* [[VAR_PRIV]], [[S_INT_TY]]** [[VAR_PRIV_REF]]
 // CHECK: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 %{{.+}}, i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
diff --git a/test/OpenMP/for_linear_codegen.cpp b/test/OpenMP/for_linear_codegen.cpp
index 61122f0fc0..461be1db48 100644
--- a/test/OpenMP/for_linear_codegen.cpp
+++ b/test/OpenMP/for_linear_codegen.cpp
@@ -24,9 +24,7 @@ float f;
 char cnt;
 
 // CHECK: [[S_FLOAT_TY:%.+]] = type { float }
-// CHECK: [[CAP_MAIN_TY:%.+]] = type { float**, i64* }
 // CHECK: [[S_INT_TY:%.+]] = type { i32 }
-// CHECK: [[CAP_TMAIN_TY:%.+]] = type { i32**, i32* }
 // CHECK-DAG: [[IMPLICIT_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 66, i32 0, i32 0, i8*
 // CHECK-DAG: [[F:@.+]] = global float 0.0
 // CHECK-DAG: [[CNT:@.+]] = global i8 0
@@ -50,11 +48,11 @@ int main() {
   // LAMBDA: call void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp for linear(g, g1:5)
   for (int i = 0; i < 2; ++i) {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: [[G_START_ADDR:%.+]] = alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
@@ -64,7 +62,6 @@ int main() {
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // LAMBDA: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     // LAMBDA: [[GTID_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %{{.+}}
     // LAMBDA: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_REF]]
     // LAMBDA: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 [[GTID]], i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
@@ -102,11 +99,11 @@ int main() {
   // BLOCKS: call void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp for linear(g, g1:5)
   for (int i = 0; i < 2; ++i) {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: [[G_START_ADDR:%.+]] = alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
@@ -116,7 +113,6 @@ int main() {
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // BLOCKS: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     // BLOCKS: [[GTID_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %{{.+}}
     // BLOCKS: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_REF]]
     // BLOCKS: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 [[GTID]], i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
@@ -166,13 +162,12 @@ int main() {
 // CHECK: define i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 2, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, float**, i64*)* [[MAIN_MICROTASK:@.+]] to void
 // CHECK: = call {{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call void [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 
-// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, float** dereferenceable(8) %{{.+}}, i64* dereferenceable(8) %{{.+}})
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: [[PVAR_START:%.+]] = alloca float*,
 // CHECK: [[LVAR_START:%.+]] = alloca i64,
@@ -186,12 +181,10 @@ int main() {
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
 
 // Check for default initialization.
-// CHECK: [[PVAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[PVAR_REF:%.+]] = load float**, float*** [[PVAR_PTR_REF]],
+// CHECK: [[PVAR_REF:%.+]] = load float**, float*** %
+// CHECK: [[LVAR_REF:%.+]] = load i64*, i64** %
 // CHECK: [[PVAR_VAL:%.+]] = load float*, float** [[PVAR_REF]],
 // CHECK: store float* [[PVAR_VAL]], float** [[PVAR_START]],
-// CHECK: [[LVAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[LVAR_REF:%.+]] = load i64*, i64** [[LVAR_PTR_REF]],
 // CHECK: [[LVAR_VAL:%.+]] = load i64, i64* [[LVAR_REF]],
 // CHECK: store i64 [[LVAR_VAL]], i64* [[LVAR_START]],
 // CHECK: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 [[GTID:%.+]], i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
@@ -220,11 +213,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 2, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32**, i32*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call void [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i32** dereferenceable(8) %{{.+}}, i32* dereferenceable(4) %{{.+}})
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: [[PVAR_START:%.+]] = alloca i32*,
 // CHECK: [[LVAR_START:%.+]] = alloca i32,
@@ -239,12 +232,10 @@ int main() {
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
 
 // Check for default initialization.
-// CHECK: [[PVAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[PVAR_REF:%.+]] = load i32**, i32*** [[PVAR_PTR_REF]],
+// CHECK: [[PVAR_REF:%.+]] = load i32**, i32*** %
 // CHECK: [[PVAR_VAL:%.+]] = load i32*, i32** [[PVAR_REF]],
 // CHECK: store i32* [[PVAR_VAL]], i32** [[PVAR_START]],
-// CHECK: [[LVAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[LVAR_REF:%.+]] = load i32*, i32** [[LVAR_PTR_REF]],
+// CHECK: [[LVAR_REF:%.+]] = load i32*, i32** %
 // CHECK: [[LVAR_VAL:%.+]] = load i32, i32* [[LVAR_REF]],
 // CHECK: store i32 [[LVAR_VAL]], i32* [[LVAR_START]],
 // CHECK: store i32* [[LVAR_PRIV]], i32** [[LVAR_PRIV_REF]],
diff --git a/test/OpenMP/for_private_codegen.cpp b/test/OpenMP/for_private_codegen.cpp
index 65bfa4397e..bc2922675f 100644
--- a/test/OpenMP/for_private_codegen.cpp
+++ b/test/OpenMP/for_private_codegen.cpp
@@ -21,10 +21,7 @@ volatile double g;
 volatile double &g1 = g;
 
 // CHECK: [[S_FLOAT_TY:%.+]] = type { float }
-// CHECK: [[CAP_MAIN_TY:%.+]] = type { i8 }
-// CHECK: type { i8 }
 // CHECK: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
-// CHECK: [[CAP_TMAIN_TY:%.+]] = type { i8 }
 template <typename T>
 T tmain() {
   S<T> test;
@@ -48,13 +45,12 @@ int main() {
   // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp for private(g, g1)
   for (int i = 0; i < 2; ++i) {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double,
-    // LAMBDA: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     g = 1;
     g1 = 1;
     // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4(
@@ -82,13 +78,12 @@ int main() {
   // BLOCKS: call {{.*}}void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* {{.+}})
+  // BLOCKS: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp for private(g, g1)
   for (int i = 0; i < 2; ++i) {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca double,
-    // BLOCKS: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     g = 1;
     g1 = 1;
     // BLOCKS: call {{.*}}void @__kmpc_for_static_init_4(
@@ -135,13 +130,12 @@ int main() {
 // CHECK: define i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[MAIN_MICROTASK:@.+]] to void
 // CHECK: = call i{{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call void [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
 // CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
 // CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]],
@@ -166,11 +160,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call void [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
 // CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
 // CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]],
diff --git a/test/OpenMP/for_reduction_codegen.cpp b/test/OpenMP/for_reduction_codegen.cpp
index 3d326bda0a..8535741c33 100644
--- a/test/OpenMP/for_reduction_codegen.cpp
+++ b/test/OpenMP/for_reduction_codegen.cpp
@@ -23,8 +23,6 @@ struct S {
 
 // CHECK-DAG: [[S_FLOAT_TY:%.+]] = type { float }
 // CHECK-DAG: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
-// CHECK-DAG: [[CAP_MAIN_TY:%.+]] = type { float*, [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]*, float*, [2 x i{{[0-9]+}}]*, [2 x [[S_FLOAT_TY]]]* }
-// CHECK-DAG: [[CAP_TMAIN_TY:%.+]] = type { i{{[0-9]+}}*, [[S_INT_TY]]*, [[S_INT_TY]]*, i{{[0-9]+}}*, [2 x i{{[0-9]+}}]*, [2 x [[S_INT_TY]]]* }
 // CHECK-DAG: [[ATOMIC_REDUCE_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 18, i32 0, i32 0, i8*
 // CHECK-DAG: [[IMPLICIT_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 66, i32 0, i32 0, i8*
 // CHECK-DAG: [[REDUCTION_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 18, i32 0, i32 0, i8*
@@ -61,11 +59,11 @@ int main() {
   // LAMBDA: call void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp for reduction(+:g, g1)
     for (int i = 0; i < 2; ++i) {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* %{{.+}})
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double,
 
     // Reduction list for runtime.
@@ -122,11 +120,11 @@ int main() {
   // BLOCKS: call void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp for reduction(-:g, g1)
     for (int i = 0; i < 2; ++i)  {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* %{{.+}})
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca double,
 
     // Reduction list for runtime.
@@ -197,13 +195,12 @@ int main() {
 // CHECK: define {{.*}}i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 6, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, float*, [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]*, float*, [2 x i32]*, [2 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK:@.+]] to void
 // CHECK: = call {{.*}}i{{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call {{.*}} [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, float* dereferenceable(4) %{{.+}}, [[S_FLOAT_TY]]* dereferenceable(4) %{{.+}}, [[S_FLOAT_TY]]* dereferenceable(4) %{{.+}}, float* dereferenceable(4) %{{.+}}, [2 x i32]* dereferenceable(8) %vec, [2 x [[S_FLOAT_TY]]]* dereferenceable(8) %{{.+}})
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca float,
 // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: [[VAR1_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
@@ -214,23 +211,20 @@ int main() {
 
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[T_VAR_REF:%.+]] = load float*, float** [[T_VAR_PTR_REF]],
+// CHECK: [[T_VAR_REF:%.+]] = load float*, float** %
+// CHECK: [[VAR1_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** %
+// CHECK: [[T_VAR1_REF:%.+]] = load float*, float** %
+
 // For + reduction operation initial value of private variable is 0.
 // CHECK: store float 0.0{{.+}}, float* [[T_VAR_PRIV]],
 
-// CHECK: [[VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[VAR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[VAR_PTR_REF:%.+]],
 // For & reduction operation initial value of private variable is ones in all bits.
+// CHECK: [[VAR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** %
 // CHECK: call {{.*}} [[S_FLOAT_TY_CONSTR:@.+]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
 
-// CHECK: [[VAR1_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[VAR1_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[VAR_PTR_REF:%.+]],
 // For && reduction operation initial value of private variable is 1.0.
 // CHECK: call {{.*}} [[S_FLOAT_TY_CONSTR:@.+]]([[S_FLOAT_TY]]* [[VAR1_PRIV]])
 
-// CHECK: [[T_VAR1_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[T_VAR1_REF:%.+]] = load float*, float** [[T_VAR1_PTR_REF]],
 // For min reduction operation initial value of private variable is largest repesentable value.
 // CHECK: store float 0x47EFFFFFE0000000, float* [[T_VAR1_PRIV]],
 
@@ -467,12 +461,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_TMAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 6, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [[S_INT_TY]]*, [[S_INT_TY]]*, i32*, [2 x i32]*, [2 x [[S_INT_TY]]]*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call {{.*}} [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}}, [[S_INT_TY]]* dereferenceable(4) %{{.+}}, [[S_INT_TY]]* dereferenceable(4) %{{.+}}, i32* dereferenceable(4) %{{.+}}, [2 x i32]* dereferenceable(8) %{{.+}}, [2 x [[S_INT_TY]]]* dereferenceable(8) %{{.+}})
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
@@ -488,23 +481,20 @@ int main() {
 
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_PTR_REF]],
+// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+// CHECK: [[VAR1_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
+// CHECK: [[T_VAR1_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+
 // For + reduction operation initial value of private variable is 0.
 // CHECK: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[T_VAR_PRIV]],
 
-// CHECK: [[VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_PTR_REF:%.+]],
 // For & reduction operation initial value of private variable is ones in all bits.
+// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
 // CHECK: call {{.*}} [[S_INT_TY_CONSTR:@.+]]([[S_INT_TY]]* [[VAR_PRIV]])
 
-// CHECK: [[VAR1_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[VAR1_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_PTR_REF:%.+]],
 // For && reduction operation initial value of private variable is 1.0.
 // CHECK: call {{.*}} [[S_INT_TY_CONSTR:@.+]]([[S_INT_TY]]* [[VAR1_PRIV]])
 
-// CHECK: [[T_VAR1_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[T_VAR1_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR1_PTR_REF]],
 // For min reduction operation initial value of private variable is largest repesentable value.
 // CHECK: store i{{[0-9]+}} 2147483647, i{{[0-9]+}}* [[T_VAR1_PRIV]],
 
diff --git a/test/OpenMP/parallel_codegen.cpp b/test/OpenMP/parallel_codegen.cpp
index 37d1679357..307476a29a 100644
--- a/test/OpenMP/parallel_codegen.cpp
+++ b/test/OpenMP/parallel_codegen.cpp
@@ -6,13 +6,9 @@
 #ifndef HEADER
 #define HEADER
 // CHECK-DAG: %ident_t = type { i32, i32, i32, i32, i8* }
-// CHECK-DAG: %struct.anon = type { i32* }
-// CHECK-DAG: %struct.anon.0 = type { i8*** }
 // CHECK-DAG: [[STR:@.+]] = private unnamed_addr constant [23 x i8] c";unknown;unknown;0;0;;\00"
 // CHECK-DAG: [[DEF_LOC_2:@.+]] = private unnamed_addr constant %ident_t { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* [[STR]], i32 0, i32 0) }
 // CHECK-DEBUG-DAG: %ident_t = type { i32, i32, i32, i32, i8* }
-// CHECK-DEBUG-DAG: %struct.anon = type { i32* }
-// CHECK-DEBUG-DAG: %struct.anon.0 = type { i8*** }
 // CHECK-DEBUG-DAG: [[STR:@.+]] = private unnamed_addr constant [23 x i8] c";unknown;unknown;0;0;;\00"
 // CHECK-DEBUG-DAG: [[DEF_LOC_2:@.+]] = private unnamed_addr constant %ident_t { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* [[STR]], i32 0, i32 0) }
 // CHECK-DEBUG-DAG: [[LOC1:@.+]] = private unnamed_addr constant [{{.+}} x i8] c";{{.*}}parallel_codegen.cpp;main;[[@LINE+14]];9;;\00"
@@ -35,39 +31,30 @@ int main (int argc, char **argv) {
 }
 
 // CHECK-LABEL: define {{[a-z\_\b]*[ ]?i32}} @main({{i32[ ]?[a-z]*}} %argc, i8** %argv)
-// CHECK:       [[AGG_CAPTURED:%.+]] = alloca %struct.anon
-// CHECK:       [[ARGC_REF:%.+]] = getelementptr inbounds %struct.anon, %struct.anon* [[AGG_CAPTURED]], i32 0, i32 0
-// CHECK-NEXT:  store i32* {{%[a-z0-9.]+}}, i32** [[ARGC_REF]]
-// CHECK-NEXT:  [[BITCAST:%.+]] = bitcast %struct.anon* [[AGG_CAPTURED]] to i8*
-// CHECK-NEXT:  call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%ident_t* [[DEF_LOC_2]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %struct.anon*)* [[OMP_OUTLINED:@.+]] to void (i32*, i32*, ...)*), i8* [[BITCAST]])
+// CHECK: store i32 %argc, i32* [[ARGC_ADDR:%.+]],
+// CHECK:  call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%ident_t* [[DEF_LOC_2]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*)* [[OMP_OUTLINED:@.+]] to void (i32*, i32*, ...)*), i32* [[ARGC_ADDR]])
 // CHECK-NEXT:  [[ARGV:%.+]] = load i8**, i8*** {{%[a-z0-9.]+}}
 // CHECK-NEXT:  [[RET:%.+]] = call {{[a-z\_\b]*[ ]?i32}} [[TMAIN:@.+tmain.+]](i8** [[ARGV]])
 // CHECK-NEXT:  ret i32 [[RET]]
 // CHECK-NEXT:  }
 // CHECK-DEBUG-LABEL: define i32 @main(i32 %argc, i8** %argv)
-// CHECK-DEBUG-DAG:   [[AGG_CAPTURED:%.+]] = alloca %struct.anon
-// CHECK-DEBUG-DAG:   [[LOC_2_ADDR:%.+]] = alloca %ident_t
+// CHECK-DEBUG:       [[LOC_2_ADDR:%.+]] = alloca %ident_t
 // CHECK-DEBUG:       [[KMPC_LOC_VOIDPTR:%.+]] = bitcast %ident_t* [[LOC_2_ADDR]] to i8*
 // CHECK-DEBUG-NEXT:  [[KMPC_DEFAULT_LOC_VOIDPTR:%.+]] = bitcast %ident_t* [[DEF_LOC_2]] to i8*
 // CHECK-DEBUG-NEXT:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[KMPC_LOC_VOIDPTR]], i8* [[KMPC_DEFAULT_LOC_VOIDPTR]], i64 24, i32 8, i1 false)
-// CHECK-DEBUG:       [[ARGC_REF:%.+]] = getelementptr inbounds %struct.anon, %struct.anon* [[AGG_CAPTURED]], i32 0, i32 0
-// CHECK-DEBUG-NEXT:  store i32* {{%[a-z0-9.]+}}, i32** [[ARGC_REF]]
-// CHECK-DEBUG-NEXT:  [[KMPC_LOC_PSOURCE_REF:%.+]] = getelementptr inbounds %ident_t, %ident_t* [[LOC_2_ADDR]], i32 0, i32 4
+// CHECK-DEBUG:       store i32 %argc, i32* [[ARGC_ADDR:%.+]],
+// CHECK-DEBUG:       [[KMPC_LOC_PSOURCE_REF:%.+]] = getelementptr inbounds %ident_t, %ident_t* [[LOC_2_ADDR]], i32 0, i32 4
 // CHECK-DEBUG-NEXT:  store i8* getelementptr inbounds ([{{.+}} x i8], [{{.+}} x i8]* [[LOC1]], i32 0, i32 0), i8** [[KMPC_LOC_PSOURCE_REF]]
-// CHECK-DEBUG-NEXT:  [[BITCAST:%.+]] = bitcast %struct.anon* [[AGG_CAPTURED]] to i8*
-// CHECK-DEBUG-NEXT:  call void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%ident_t* [[LOC_2_ADDR]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %struct.anon*)* [[OMP_OUTLINED:@.+]] to void (i32*, i32*, ...)*), i8* [[BITCAST]])
+// CHECK-DEBUG:  call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%ident_t* [[LOC_2_ADDR]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*)* [[OMP_OUTLINED:@.+]] to void (i32*, i32*, ...)*), i32* [[ARGC_ADDR]])
 // CHECK-DEBUG-NEXT:  [[ARGV:%.+]] = load i8**, i8*** {{%[a-z0-9.]+}}
 // CHECK-DEBUG-NEXT:  [[RET:%.+]] = call i32 [[TMAIN:@.+tmain.+]](i8** [[ARGV]])
 // CHECK-DEBUG-NEXT:  ret i32 [[RET]]
 // CHECK-DEBUG-NEXT:  }
 
-// CHECK:       define internal {{.*}}void [[OMP_OUTLINED]](i32* %.global_tid., i32* %.bound_tid., %struct.anon* %__context)
-// CHECK:       #[[FN_ATTRS:[0-9]+]]
-// CHECK:       [[CONTEXT_ADDR:%.+]] = alloca %struct.anon*
-// CHECK:       store %struct.anon* %__context, %struct.anon** [[CONTEXT_ADDR]]
-// CHECK:       [[CONTEXT_PTR:%.+]] = load %struct.anon*, %struct.anon** [[CONTEXT_ADDR]]
-// CHECK-NEXT:  [[ARGC_PTR_REF:%.+]] = getelementptr inbounds %struct.anon, %struct.anon* [[CONTEXT_PTR]], i32 0, i32 0
-// CHECK-NEXT:  [[ARGC_REF:%.+]] = load i32*, i32** [[ARGC_PTR_REF]]
+// CHECK:       define internal {{.*}}void [[OMP_OUTLINED]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i32* dereferenceable(4) [[ARGC_ADDR:%[^)]+]])
+// CHECK-SAME:       #[[FN_ATTRS:[0-9]+]]
+// CHECK:       store i32* [[ARGC_ADDR]], i32** [[ARGC_PTR_ADDR:%.+]],
+// CHECK-NEXT:  [[ARGC_REF:%.+]] = load i32*, i32** [[ARGC_PTR_ADDR]]
 // CHECK-NEXT:  [[ARGC:%.+]] = load i32, i32* [[ARGC_REF]]
 // CHECK-NEXT:  invoke {{.*}}void [[FOO:@.+foo.+]](i32{{[ ]?[a-z]*}} [[ARGC]])
 // CHECK:       call {{.+}} @__kmpc_cancel_barrier(
@@ -75,13 +62,10 @@ int main (int argc, char **argv) {
 // CHECK:       call {{.*}}void @{{.+terminate.*|abort}}(
 // CHECK-NEXT:  unreachable
 // CHECK-NEXT:  }
-// CHECK-DEBUG:       define internal void [[OMP_OUTLINED]](i32* %.global_tid., i32* %.bound_tid., %struct.anon* %__context)
-// CHECK-DEBUG:       #[[FN_ATTRS:[0-9]+]]
-// CHECK-DEBUG:       [[CONTEXT_ADDR:%.+]] = alloca %struct.anon*
-// CHECK-DEBUG:       store %struct.anon* %__context, %struct.anon** [[CONTEXT_ADDR]]
-// CHECK-DEBUG:       [[CONTEXT_PTR:%.+]] = load %struct.anon*, %struct.anon** [[CONTEXT_ADDR]]
-// CHECK-DEBUG-NEXT:  [[ARGC_PTR_REF:%.+]] = getelementptr inbounds %struct.anon, %struct.anon* [[CONTEXT_PTR]], i32 0, i32 0
-// CHECK-DEBUG-NEXT:  [[ARGC_REF:%.+]] = load i32*, i32** [[ARGC_PTR_REF]]
+// CHECK-DEBUG:       define internal void [[OMP_OUTLINED]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i32* dereferenceable(4) [[ARGC_ADDR:%[^)]+]])
+// CHECK-DEBUG-SAME:  #[[FN_ATTRS:[0-9]+]]
+// CHECK-DEBUG:       store i32* [[ARGC_ADDR]], i32** [[ARGC_PTR_ADDR:%.+]],
+// CHECK-DEBUG:  [[ARGC_REF:%.+]] = load i32*, i32** [[ARGC_PTR_ADDR]]
 // CHECK-DEBUG-NEXT:  [[ARGC:%.+]] = load i32, i32* [[ARGC_REF]]
 // CHECK-DEBUG-NEXT:  invoke void [[FOO:@.+foo.+]](i32 [[ARGC]])
 // CHECK-DEBUG:       call {{.+}} @__kmpc_cancel_barrier(
@@ -96,34 +80,25 @@ int main (int argc, char **argv) {
 // CHECK-DEBUG-DAG: declare void @__kmpc_fork_call(%ident_t*, i32, void (i32*, i32*, ...)*, ...)
 
 // CHECK:       define linkonce_odr {{[a-z\_\b]*[ ]?i32}} [[TMAIN]](i8** %argc)
-// CHECK:       [[AGG_CAPTURED:%.+]] = alloca %struct.anon.0
-// CHECK:       [[ARGC_REF:%.+]] = getelementptr inbounds %struct.anon.0, %struct.anon.0* [[AGG_CAPTURED]], i32 0, i32 0
-// CHECK-NEXT:  store i8*** {{%[a-z0-9.]+}}, i8**** [[ARGC_REF]]
-// CHECK-NEXT:  [[BITCAST:%.+]] = bitcast %struct.anon.0* [[AGG_CAPTURED]] to i8*
-// CHECK-NEXT:  call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%ident_t* [[DEF_LOC_2]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %struct.anon.0*)* [[OMP_OUTLINED:@.+]] to void (i32*, i32*, ...)*), i8* [[BITCAST]])
+// CHECK:       store i8** %argc, i8*** [[ARGC_ADDR:%.+]],
+// CHECK-NEXT:  call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%ident_t* [[DEF_LOC_2]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i8***)* [[OMP_OUTLINED:@.+]] to void (i32*, i32*, ...)*), i8*** [[ARGC_ADDR]])
 // CHECK-NEXT:  ret i32 0
 // CHECK-NEXT:  }
 // CHECK-DEBUG:       define linkonce_odr i32 [[TMAIN]](i8** %argc)
-// CHECK-DEBUG-DAG:   [[AGG_CAPTURED:%.+]] = alloca %struct.anon.0
 // CHECK-DEBUG-DAG:   [[LOC_2_ADDR:%.+]] = alloca %ident_t
 // CHECK-DEBUG:       [[KMPC_LOC_VOIDPTR:%.+]] = bitcast %ident_t* [[LOC_2_ADDR]] to i8*
 // CHECK-DEBUG-NEXT:  [[KMPC_DEFAULT_LOC_VOIDPTR:%.+]] = bitcast %ident_t* [[DEF_LOC_2]] to i8*
 // CHECK-DEBUG-NEXT:   call void @llvm.memcpy.p0i8.p0i8.i64(i8* [[KMPC_LOC_VOIDPTR]], i8* [[KMPC_DEFAULT_LOC_VOIDPTR]], i64 24, i32 8, i1 false)
-// CHECK-DEBUG:       [[ARGC_REF:%.+]] = getelementptr inbounds %struct.anon.0, %struct.anon.0* [[AGG_CAPTURED]], i32 0, i32 0
-// CHECK-DEBUG-NEXT:  store i8*** {{%[a-z0-9.]+}}, i8**** [[ARGC_REF]]
-// CHECK-DEBUG-NEXT:  [[KMPC_LOC_PSOURCE_REF:%.+]] = getelementptr inbounds %ident_t, %ident_t* [[LOC_2_ADDR]], i32 0, i32 4
+// CHECK-DEBUG-NEXT:  store i8** %argc, i8*** [[ARGC_ADDR:%.+]],
+// CHECK-DEBUG:  [[KMPC_LOC_PSOURCE_REF:%.+]] = getelementptr inbounds %ident_t, %ident_t* [[LOC_2_ADDR]], i32 0, i32 4
 // CHECK-DEBUG-NEXT:  store i8* getelementptr inbounds ([{{.+}} x i8], [{{.+}} x i8]* [[LOC2]], i32 0, i32 0), i8** [[KMPC_LOC_PSOURCE_REF]]
-// CHECK-DEBUG-NEXT:  [[BITCAST:%.+]] = bitcast %struct.anon.0* [[AGG_CAPTURED]] to i8*
-// CHECK-DEBUG-NEXT:  call void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%ident_t* [[LOC_2_ADDR]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %struct.anon.0*)* [[OMP_OUTLINED:@.+]] to void (i32*, i32*, ...)*), i8* [[BITCAST]])
+// CHECK-DEBUG-NEXT:  call void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%ident_t* [[LOC_2_ADDR]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i8***)* [[OMP_OUTLINED:@.+]] to void (i32*, i32*, ...)*), i8*** [[ARGC_ADDR]])
 // CHECK-DEBUG-NEXT:  ret i32 0
 // CHECK-DEBUG-NEXT:  }
 
-// CHECK:       define internal {{.*}}void [[OMP_OUTLINED]](i32* %.global_tid., i32* %.bound_tid., %struct.anon.0* %__context)
-// CHECK:       [[CONTEXT_ADDR:%.+]] = alloca %struct.anon.0*
-// CHECK:       store %struct.anon.0* %__context, %struct.anon.0** [[CONTEXT_ADDR]]
-// CHECK:       [[CONTEXT_PTR:%.+]] = load %struct.anon.0*, %struct.anon.0** [[CONTEXT_ADDR]]
-// CHECK-NEXT:  [[ARGC_PTR_REF:%.+]] = getelementptr inbounds %struct.anon.0, %struct.anon.0* [[CONTEXT_PTR]], i32 0, i32 0
-// CHECK-NEXT:  [[ARGC_REF:%.+]] = load i8***, i8**** [[ARGC_PTR_REF]]
+// CHECK:       define internal {{.*}}void [[OMP_OUTLINED]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i8*** dereferenceable(8) %argc)
+// CHECK:       store i8*** %argc, i8**** [[ARGC_PTR_ADDR:%.+]],
+// CHECK-NEXT:  [[ARGC_REF:%.+]] = load i8***, i8**** [[ARGC_PTR_ADDR]]
 // CHECK-NEXT:  [[ARGC:%.+]] = load i8**, i8*** [[ARGC_REF]]
 // CHECK-NEXT:  invoke {{.*}}void [[FOO1:@.+foo.+]](i8** [[ARGC]])
 // CHECK:       call {{.+}} @__kmpc_cancel_barrier(
@@ -131,12 +106,9 @@ int main (int argc, char **argv) {
 // CHECK:       call {{.*}}void @{{.+terminate.*|abort}}(
 // CHECK-NEXT:  unreachable
 // CHECK-NEXT:  }
-// CHECK-DEBUG:       define internal void [[OMP_OUTLINED]](i32* %.global_tid., i32* %.bound_tid., %struct.anon.0* %__context)
-// CHECK-DEBUG:       [[CONTEXT_ADDR:%.+]] = alloca %struct.anon.0*
-// CHECK-DEBUG:       store %struct.anon.0* %__context, %struct.anon.0** [[CONTEXT_ADDR]]
-// CHECK-DEBUG:       [[CONTEXT_PTR:%.+]] = load %struct.anon.0*, %struct.anon.0** [[CONTEXT_ADDR]]
-// CHECK-DEBUG-NEXT:  [[ARGC_PTR_REF:%.+]] = getelementptr inbounds %struct.anon.0, %struct.anon.0* [[CONTEXT_PTR]], i32 0, i32 0
-// CHECK-DEBUG-NEXT:  [[ARGC_REF:%.+]] = load i8***, i8**** [[ARGC_PTR_REF]]
+// CHECK-DEBUG:       define internal void [[OMP_OUTLINED]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i8*** dereferenceable(8) %argc)
+// CHECK-DEBUG:       store i8*** %argc, i8**** [[ARGC_PTR_ADDR:%.+]],
+// CHECK-DEBUG:  [[ARGC_REF:%.+]] = load i8***, i8**** [[ARGC_PTR_ADDR]]
 // CHECK-DEBUG-NEXT:  [[ARGC:%.+]] = load i8**, i8*** [[ARGC_REF]]
 // CHECK-DEBUG-NEXT:  invoke void [[FOO1:@.+foo.+]](i8** [[ARGC]])
 // CHECK-DEBUG:       call {{.+}} @__kmpc_cancel_barrier(
diff --git a/test/OpenMP/parallel_copyin_codegen.cpp b/test/OpenMP/parallel_copyin_codegen.cpp
index 4aacb80957..f29fb8b641 100644
--- a/test/OpenMP/parallel_copyin_codegen.cpp
+++ b/test/OpenMP/parallel_copyin_codegen.cpp
@@ -78,19 +78,15 @@ int main() {
   // LAMBDA: [[G:@.+]] = global i{{[0-9]+}} 1212,
   // LAMBDA-LABEL: @main
   // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]](
-  // TLS-LAMBDA: [[G_CAPTURE_TY:%.+]] = type { i{{[0-9]+}}* }
   // TLS-LAMBDA: [[G:@.+]] = {{.*}}thread_local {{.*}}global i{{[0-9]+}} 1212,
   // TLS-LAMBDA-LABEL: @main
   // TLS-LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8*
+  // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 
-  // TLS-LAMBDA-DAG: [[G_CPY_ADDR:%.+]] = getelementptr inbounds [[G_CAPTURE_TY]], [[G_CAPTURE_TY]]* [[G_CAPTURE:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // TLS-LAMBDA-DAG: [[G_CPY_VAL:%.+]] = call i{{[0-9]+}}* [[G_CTOR:@.+]]()
-  // TLS-LAMBDA:     store i{{[0-9]+}}* [[G_CPY_VAL]], i{{[0-9]+}}** [[G_CPY_ADDR]]
-  // TLS-LAMBDA:     [[G_CAPTURE_CAST:%.+]] = bitcast [[G_CAPTURE_TY]]* [[G_CAPTURE]] to i8*
-  // TLS-LAMBDA:     call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* [[G_CAPTURE_CAST]]
+  // TLS-LAMBDA:     [[G_CPY_VAL:%.+]] = call i{{[0-9]+}}* [[G_CTOR:@.+]]()
+  // TLS-LAMBDA:     call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i32* [[G_CPY_VAL]])
 
   // TLS-LAMBDA:     define {{.*}}i{{[0-9]+}}* [[G_CTOR]]() {
   // TLS-LAMBDA:     ret i{{[0-9]+}}* [[G]]
@@ -98,8 +94,8 @@ int main() {
 
 #pragma omp parallel copyin(g)
   {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
-    // TLS-LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, [[G_CAPTURE_TY]]* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
+    // TLS-LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}})
 
     // threadprivate_g = g;
     // LAMBDA: call {{.*}}i8* @__kmpc_threadprivate_cached({{.+}} [[G]]
@@ -111,8 +107,7 @@ int main() {
     // LAMBDA: store volatile i{{[0-9]+}} %{{.+}}, i{{[0-9]+}}* %{{.+}},
     // LAMBDA: [[DONE]]
 
-    // TLS-LAMBDA-DAG: [[G_CAPTURE_FIELD:%.+]] = getelementptr inbounds [[G_CAPTURE_TY]], [[G_CAPTURE_TY]]* {{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-    // TLS-LAMBDA-DAG: [[G_CAPTURE_SRC:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[G_CAPTURE_FIELD]]
+    // TLS-LAMBDA-DAG: [[G_CAPTURE_SRC:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
     // TLS-LAMBDA-DAG: [[G_CAPTURE_DST:%.+]] = call i{{[0-9]+}}* [[G_CTOR]]()
     // TLS-LAMBDA-DAG: [[G_CAPTURE_SRCC:%.+]] = ptrtoint i{{[0-9]+}}* [[G_CAPTURE_SRC]] to i{{[0-9]+}}
     // TLS-LAMBDA-DAG: [[G_CAPTURE_DSTC:%.+]] = ptrtoint i{{[0-9]+}}* [[G_CAPTURE_DST]] to i{{[0-9]+}}
@@ -145,27 +140,23 @@ int main() {
   // BLOCKS-LABEL: @main
   // BLOCKS: call {{.*}}void {{%.+}}(i8
 
-  // TLS-BLOCKS: [[G_CAPTURE_TY:%.+]] = type { i{{[0-9]+}}* }
   // TLS-BLOCKS: [[G:@.+]] = {{.*}}thread_local {{.*}}global i{{[0-9]+}} 1212,
   // TLS-BLOCKS-LABEL: @main
   // TLS-BLOCKS: call {{.*}}void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8*
+  // BLOCKS: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 
-  // TLS-BLOCKS-DAG: [[G_CPY_ADDR:%.+]] = getelementptr inbounds [[G_CAPTURE_TY]], [[G_CAPTURE_TY]]* [[G_CAPTURE:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // TLS-BLOCKS-DAG: [[G_CPY_VAL:%.+]] = call i{{[0-9]+}}* [[G_CTOR:@.+]]()
-  // TLS-BLOCKS:     store i{{[0-9]+}}* [[G_CPY_VAL]], i{{[0-9]+}}** [[G_CPY_ADDR]]
-  // TLS-BLOCKS:     [[G_CAPTURE_CAST:%.+]] = bitcast [[G_CAPTURE_TY]]* [[G_CAPTURE]] to i8*
-  // TLS-BLOCKS:     call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* [[G_CAPTURE_CAST]]
+  // TLS-BLOCKS:     [[G_CPY_VAL:%.+]] = call i{{[0-9]+}}* [[G_CTOR:@.+]]()
+  // TLS-BLOCKS:     call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i32* [[G_CPY_VAL]])
 
   // TLS-BLOCKS:     define {{.*}}i{{[0-9]+}}* [[G_CTOR]]() {
   // TLS-BLOCKS:     ret i{{[0-9]+}}* [[G]]
   // TLS-BLOCKS:     }
 #pragma omp parallel copyin(g)
   {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
-    // TLS-BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, [[G_CAPTURE_TY]]* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
+    // TLS-BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}})
 
     // threadprivate_g = g;
     // BLOCKS: call {{.*}}i8* @__kmpc_threadprivate_cached({{.+}} [[G]]
@@ -177,8 +168,7 @@ int main() {
     // BLOCKS: store volatile i{{[0-9]+}} %{{.+}}, i{{[0-9]+}}* %{{.+}},
     // BLOCKS: [[DONE]]
 
-    // TLS-BLOCKS-DAG: [[G_CAPTURE_FIELD:%.+]] = getelementptr inbounds [[G_CAPTURE_TY]], [[G_CAPTURE_TY]]* {{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-    // TLS-BLOCKS-DAG: [[G_CAPTURE_SRC:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[G_CAPTURE_FIELD]]
+    // TLS-BLOCKS-DAG: [[G_CAPTURE_SRC:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
     // TLS-BLOCKS-DAG: [[G_CAPTURE_DST:%.+]] = call i{{[0-9]+}}* [[G_CTOR]]()
     // TLS-BLOCKS-DAG: [[G_CAPTURE_SRCC:%.+]] = ptrtoint i{{[0-9]+}}* [[G_CAPTURE_SRC]] to i{{[0-9]+}}
     // TLS-BLOCKS-DAG: [[G_CAPTURE_DSTC:%.+]] = ptrtoint i{{[0-9]+}}* [[G_CAPTURE_DST]] to i{{[0-9]+}}
@@ -241,8 +231,8 @@ int main() {
 // CHECK-LABEL: @main
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_COPY_ASSIGN:@.+]]([[S_FLOAT_TY]]* [[TEST]], [[S_FLOAT_TY]]*
-// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, {{%.+}}*)* [[MAIN_MICROTASK:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
-// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, {{%.+}}*)* [[MAIN_MICROTASK1:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
+// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[MAIN_MICROTASK:@.+]] to void (i32*, i32*, ...)*))
+// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[MAIN_MICROTASK1:@.+]] to void (i32*, i32*, ...)*))
 // CHECK: = call {{.*}}i{{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call {{.*}} [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
@@ -250,28 +240,18 @@ int main() {
 // TLS-CHECK-LABEL: @main
 // TLS-CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // TLS-CHECK: call {{.*}} [[S_FLOAT_TY_COPY_ASSIGN:@.+]]([[S_FLOAT_TY]]* [[TEST]], [[S_FLOAT_TY]]*
-// TLS-CHECK-DAG: store i{{[0-9]+}}* [[T_VAR]], i{{[0-9]+}}** [[T_VAR_FIELD:%.+]],
-// TLS-CHECK-DAG: store [2 x i{{[0-9]+}}]* [[VEC]], [2 x i{{[0-9]+}}]** [[VEC_FIELD:%.+]],
-// TLS-CHECK-DAG: store [2 x [[S_FLOAT_TY]]]* [[S_ARR]], [2 x [[S_FLOAT_TY]]]** [[S_ARR_FIELD:%.+]],
-// TLS-CHECK-DAG: store [[S_FLOAT_TY]]* [[VAR]], [[S_FLOAT_TY]]** [[VAR_FIELD:%.+]],
-// TLS-CHECK-DAG: [[T_VAR_FIELD]] = getelementptr inbounds {{.*}}
-// TLS-CHECK-DAG: [[VEC_FIELD]] = getelementptr inbounds {{.*}}
-// TLS-CHECK-DAG: [[S_ARR_FIELD]] = getelementptr inbounds {{.*}}
-// TLS-CHECK-DAG: [[VAR_FIELD]] = getelementptr inbounds {{.*}}
-// TLS-CHECK:     call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, {{%.+}}*)* [[MAIN_MICROTASK:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
-// TLS-CHECK-DAG: store i{{[0-9]+}}* [[T_VAR]], i{{[0-9]+}}** [[T_VAR_FIELD:%.+]],
-// TLS-CHECK-DAG: [[T_VAR_FIELD]] = getelementptr inbounds {{.*}}
-// TLS-CHECK:     call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, {{%.+}}*)* [[MAIN_MICROTASK1:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
+// TLS-CHECK:     call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [2 x i32]*, [2 x [[S_FLOAT_TY]]]*, [[S_FLOAT_TY]]*)* [[MAIN_MICROTASK:@.+]] to void (i32*, i32*, ...)*),
+// TLS-CHECK:     call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*)* [[MAIN_MICROTASK1:@.+]] to void (i32*, i32*, ...)*),
 // TLS-CHECK: = call {{.*}}i{{.+}} [[TMAIN_INT:@.+]]()
 // TLS-CHECK: call {{.*}} [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // TLS-CHECK: ret
 
-// CHECK: define internal {{.*}}void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, {{%.+}}* %{{.+}})
+// CHECK: define internal {{.*}}void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 // CHECK: [[GTID_ADDR:%.+]] = load i32*, i32** [[GTID_ADDR_ADDR]],
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_ADDR]],
 
-// TLS-CHECK: define internal {{.*}}void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, {{%.+}}* %{{.+}})
+// TLS-CHECK: define internal {{.*}}void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}},
 // TLS-CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
 // threadprivate_t_var = t_var;
@@ -283,8 +263,11 @@ int main() {
 // CHECK: load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR]],
 // CHECK: store i{{[0-9]+}} %{{.+}}, i{{[0-9]+}}* %{{.+}},
 
-// TLS-CHECK: [[MASTER_FIELD:%.+]] = getelementptr inbounds {{.+}}
-// TLS-CHECK: [[MASTER_REF:%.+]] = load i32*, i32** [[MASTER_FIELD]]
+// TLS-CHECK: [[MASTER_REF:%.+]] = load i32*, i32** %
+// TLS-CHECK: [[MASTER_REF2:%.+]] = load [2 x i32]*, [2 x i32]** %
+// TLS-CHECK: [[MASTER_REF3:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** %
+// TLS-CHECK: [[MASTER_REF4:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** %
+
 // TLS-CHECK: [[MASTER_LONG:%.+]] = ptrtoint i32* [[MASTER_REF]] to i{{[0-9]+}}
 // TLS-CHECK: icmp ne i{{[0-9]+}} [[MASTER_LONG]], ptrtoint (i{{[0-9]+}}* [[T_VAR]] to i{{[0-9]+}})
 // TLS-CHECK: br i1 %{{.+}}, label %[[NOT_MASTER:.+]], label %[[DONE:.+]]
@@ -296,9 +279,7 @@ int main() {
 // CHECK: call {{.*}}i8* @__kmpc_threadprivate_cached({{.+}} [[VEC]]
 // CHECK: call void @llvm.memcpy{{.*}}(i8* %{{.+}}, i8* bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*),
 
-// TLS-CHECK: [[MASTER_FIELD:%.+]] = getelementptr inbounds {{.+}}
-// TLS-CHECK: [[MASTER_REF:%.+]] = load [2 x i32]*, [2 x i32]** [[MASTER_FIELD]]
-// TLS-CHECK: [[MASTER_CAST:%.+]] = bitcast [2 x i32]* [[MASTER_REF]] to i8*
+// TLS-CHECK: [[MASTER_CAST:%.+]] = bitcast [2 x i32]* [[MASTER_REF2]] to i8*
 // TLS-CHECK: call void @llvm.memcpy{{.*}}(i8* bitcast ([2 x i{{[0-9]+}}]* [[VEC]] to i8*), i8* [[MASTER_CAST]]
 
 // threadprivate_s_arr = s_arr;
@@ -311,9 +292,7 @@ int main() {
 // CHECK: call {{.*}} [[S_FLOAT_TY_COPY_ASSIGN]]([[S_FLOAT_TY]]* {{.+}}, [[S_FLOAT_TY]]* {{.+}})
 // CHECK: br i1 {{.+}}, label %{{.+}}, label %[[S_ARR_BODY]]
 
-// TLS-CHECK: [[MASTER_FIELD:%.+]] = getelementptr inbounds {{.+}}
-// TLS-CHECK: [[MASTER_REF:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[MASTER_FIELD]]
-// TLS-CHECK: [[MASTER_CAST:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[MASTER_REF]] to [[S_FLOAT_TY]]*
+// TLS-CHECK: [[MASTER_CAST:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[MASTER_REF3]] to [[S_FLOAT_TY]]*
 // TLS-CHECK-DAG: [[S_ARR_SRC_BEGIN:%.+]] = phi [[S_FLOAT_TY]]* {{.*}}[[MASTER_CAST]]
 // TLS-CHECK-DAG: [[S_ARR_DST_BEGIN:%.+]] = phi [[S_FLOAT_TY]]* {{.*}}getelementptr inbounds ([2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR]], i{{[0-9]+}} 0, i{{[0-9]+}} 0)
 // TLS-CHECK: call {{.*}} [[S_FLOAT_TY_COPY_ASSIGN]]([[S_FLOAT_TY]]* {{.+}}, [[S_FLOAT_TY]]* {{.+}})
@@ -328,9 +307,7 @@ int main() {
 // CHECK: call {{.*}} [[S_FLOAT_TY_COPY_ASSIGN]]([[S_FLOAT_TY]]* {{%.+}}, [[S_FLOAT_TY]]* {{.*}}[[VAR]])
 // CHECK: [[DONE]]
 
-// TLS-CHECK: [[MASTER_FIELD:%.+]] = getelementptr inbounds {{.+}}
-// TLS-CHECK: [[MASTER_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[MASTER_FIELD]]
-// TLS-CHECK: call {{.*}} [[S_FLOAT_TY_COPY_ASSIGN]]([[S_FLOAT_TY]]* {{.*}}[[VAR]], [[S_FLOAT_TY]]* {{.*}}[[MASTER_REF]])
+// TLS-CHECK: call {{.*}} [[S_FLOAT_TY_COPY_ASSIGN]]([[S_FLOAT_TY]]* {{.*}}[[VAR]], [[S_FLOAT_TY]]* {{.*}}[[MASTER_REF4]])
 
 // CHECK: call {{.*}}i32 @__kmpc_cancel_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i32 [[GTID]])
 // CHECK: ret void
@@ -340,12 +317,12 @@ int main() {
 // TLS-CHECK: call {{.*}}i32 @__kmpc_cancel_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i32 [[GTID]])
 // TLS-CHECK: ret void
 
-// CHECK: define internal {{.*}}void [[MAIN_MICROTASK1]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, {{%.+}}* %{{.+}})
+// CHECK: define internal {{.*}}void [[MAIN_MICROTASK1]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 // CHECK: [[GTID_ADDR:%.+]] = load i32*, i32** [[GTID_ADDR_ADDR]],
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_ADDR]],
 
-// TLS-CHECK: define internal {{.*}}void [[MAIN_MICROTASK1]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, {{%.+}}* %{{.+}})
+// TLS-CHECK: define internal {{.*}}void [[MAIN_MICROTASK1]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // TLS-CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
 // threadprivate_t_var = t_var;
@@ -358,8 +335,8 @@ int main() {
 // CHECK: store i{{[0-9]+}} %{{.+}}, i{{[0-9]+}}* %{{.+}},
 // CHECK: [[DONE]]
 
-// TLS-CHECK: [[MASTER_FIELD:%.+]] = getelementptr inbounds {{.+}}
-// TLS-CHECK: [[MASTER_REF:%.+]] = load i32*, i32** [[MASTER_FIELD]]
+// TLS-CHECK: [[MASTER_REF:%.+]] = load i32*, i32** %
+
 // TLS-CHECK: [[MASTER_LONG:%.+]] = ptrtoint i32* [[MASTER_REF]] to i{{[0-9]+}}
 // TLS-CHECK: icmp ne i{{[0-9]+}} [[MASTER_LONG]], ptrtoint (i{{[0-9]+}}* [[T_VAR]] to i{{[0-9]+}})
 // TLS-CHECK: br i1 %{{.+}}, label %[[NOT_MASTER:.+]], label %[[DONE:.+]]
@@ -379,33 +356,23 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_COPY_ASSIGN:@.+]]([[S_INT_TY]]* [[TEST]], [[S_INT_TY]]*
-// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, {{%.+}}*)* [[TMAIN_MICROTASK:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
-// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, {{%.+}}*)* [[TMAIN_MICROTASK1:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
+// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[TMAIN_MICROTASK:@.+]] to void (i32*, i32*, ...)*))
+// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[TMAIN_MICROTASK1:@.+]] to void (i32*, i32*, ...)*))
 // CHECK: call {{.*}} [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 
 // TLS-CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // TLS-CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // TLS-CHECK: call {{.*}} [[S_INT_TY_COPY_ASSIGN:@.+]]([[S_INT_TY]]* [[TEST]], [[S_INT_TY]]*
-// TLS-CHECK-DAG: store i{{[0-9]+}}* [[TMAIN_T_VAR]], i{{[0-9]+}}** [[T_VAR_FIELD:%.+]],
-// TLS-CHECK-DAG: store [2 x i{{[0-9]+}}]* [[TMAIN_VEC]], [2 x i{{[0-9]+}}]** [[VEC_FIELD:%.+]],
-// TLS-CHECK-DAG: store [2 x [[S_INT_TY]]]* [[TMAIN_S_ARR]], [2 x [[S_INT_TY]]]** [[S_ARR_FIELD:%.+]],
-// TLS-CHECK-DAG: store [[S_INT_TY]]* [[TMAIN_VAR]], [[S_INT_TY]]** [[VAR_FIELD:%.+]],
-// TLS-CHECK-DAG: [[T_VAR_FIELD]] = getelementptr inbounds {{.*}}
-// TLS-CHECK-DAG: [[VEC_FIELD]] = getelementptr inbounds {{.*}}
-// TLS-CHECK-DAG: [[S_ARR_FIELD]] = getelementptr inbounds {{.*}}
-// TLS-CHECK-DAG: [[VAR_FIELD]] = getelementptr inbounds {{.*}}
-// TLS-CHECK:     call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, {{%.+}}*)* [[TMAIN_MICROTASK:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
-// TLS-CHECK-DAG: store i{{[0-9]+}}* [[TMAIN_T_VAR]], i{{[0-9]+}}** [[T_VAR_FIELD:%.+]],
-// TLS-CHECK-DAG: [[T_VAR_FIELD]] = getelementptr inbounds {{.*}}
-// TLS-CHECK:     call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, {{%.+}}*)* [[TMAIN_MICROTASK1:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
+// TLS-CHECK:     call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [2 x i32]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void (i32*, i32*, ...)*),
+// TLS-CHECK:     call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*)* [[TMAIN_MICROTASK1:@.+]] to void (i32*, i32*, ...)*),
 //
-// CHECK: define internal {{.*}}void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, {{%.+}}* %{{.+}})
+// CHECK: define internal {{.*}}void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 // CHECK: [[GTID_ADDR:%.+]] = load i32*, i32** [[GTID_ADDR_ADDR]],
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_ADDR]],
 //
-// TLS-CHECK: define internal {{.*}}void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, {{%.+}}* %{{.+}})
+// TLS-CHECK: define internal {{.*}}void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // TLS-CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
 // threadprivate_t_var = t_var;
@@ -417,8 +384,11 @@ int main() {
 // CHECK: load i{{[0-9]+}}, i{{[0-9]+}}* [[TMAIN_T_VAR]],
 // CHECK: store i{{[0-9]+}} %{{.+}}, i{{[0-9]+}}* %{{.+}},
 
-// TLS-CHECK: [[MASTER_FIELD:%.+]] = getelementptr inbounds {{.+}}
-// TLS-CHECK: [[MASTER_REF:%.+]] = load i32*, i32** [[MASTER_FIELD]]
+// TLS-CHECK: [[MASTER_REF:%.+]] = load i32*, i32** %
+// TLS-CHECK: [[MASTER_REF1:%.+]] = load [2 x i32]*, [2 x i32]** %
+// TLS-CHECK: [[MASTER_REF2:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** %
+// TLS-CHECK: [[MASTER_REF3:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
+
 // TLS-CHECK: [[MASTER_LONG:%.+]] = ptrtoint i32* [[MASTER_REF]] to i{{[0-9]+}}
 // TLS-CHECK: icmp ne i{{[0-9]+}} [[MASTER_LONG]], ptrtoint (i{{[0-9]+}}* [[TMAIN_T_VAR]] to i{{[0-9]+}})
 // TLS-CHECK: br i1 %{{.+}}, label %[[NOT_MASTER:.+]], label %[[DONE:.+]]
@@ -430,9 +400,7 @@ int main() {
 // CHECK: call {{.*}}i8* @__kmpc_threadprivate_cached({{.+}} [[TMAIN_VEC]]
 // CHECK: call {{.*}}void @llvm.memcpy{{.*}}(i8* %{{.+}}, i8* bitcast ([2 x i{{[0-9]+}}]* [[TMAIN_VEC]] to i8*),
 
-// TLS-CHECK: [[MASTER_FIELD:%.+]] = getelementptr inbounds {{.+}}
-// TLS-CHECK: [[MASTER_REF:%.+]] = load [2 x i32]*, [2 x i32]** [[MASTER_FIELD]]
-// TLS-CHECK: [[MASTER_CAST:%.+]] = bitcast [2 x i32]* [[MASTER_REF]] to i8*
+// TLS-CHECK: [[MASTER_CAST:%.+]] = bitcast [2 x i32]* [[MASTER_REF1]] to i8*
 // TLS-CHECK: call void @llvm.memcpy{{.*}}(i8* bitcast ([2 x i{{[0-9]+}}]* [[TMAIN_VEC]] to i8*), i8* [[MASTER_CAST]]
 
 // threadprivate_s_arr = s_arr;
@@ -445,9 +413,7 @@ int main() {
 // CHECK: call {{.*}} [[S_INT_TY_COPY_ASSIGN]]([[S_INT_TY]]* {{.+}}, [[S_INT_TY]]* {{.+}})
 // CHECK: br i1 {{.+}}, label %{{.+}}, label %[[S_ARR_BODY]]
 
-// TLS-CHECK: [[MASTER_FIELD:%.+]] = getelementptr inbounds {{.+}}
-// TLS-CHECK: [[MASTER_REF:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** [[MASTER_FIELD]]
-// TLS-CHECK: [[MASTER_CAST:%.+]] = bitcast [2 x [[S_INT_TY]]]* [[MASTER_REF]] to [[S_INT_TY]]*
+// TLS-CHECK: [[MASTER_CAST:%.+]] = bitcast [2 x [[S_INT_TY]]]* [[MASTER_REF2]] to [[S_INT_TY]]*
 // TLS-CHECK-DAG: [[S_ARR_SRC_BEGIN:%.+]] = phi [[S_INT_TY]]* {{.*}}[[MASTER_CAST]]
 // TLS-CHECK-DAG: [[S_ARR_DST_BEGIN:%.+]] = phi [[S_INT_TY]]* {{.*}}getelementptr inbounds ([2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[TMAIN_S_ARR]], i{{[0-9]+}} 0, i{{[0-9]+}} 0)
 // TLS-CHECK: call {{.*}} [[S_INT_TY_COPY_ASSIGN]]([[S_INT_TY]]* {{.+}}, [[S_INT_TY]]* {{.+}})
@@ -462,9 +428,7 @@ int main() {
 // CHECK: call {{.*}} [[S_INT_TY_COPY_ASSIGN]]([[S_INT_TY]]* {{%.+}}, [[S_INT_TY]]* {{.*}}[[TMAIN_VAR]])
 // CHECK: [[DONE]]
 
-// TLS-CHECK: [[MASTER_FIELD:%.+]] = getelementptr inbounds {{.+}}
-// TLS-CHECK: [[MASTER_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[MASTER_FIELD]]
-// TLS-CHECK: call {{.*}} [[S_INT_TY_COPY_ASSIGN]]([[S_INT_TY]]* {{.*}}[[TMAIN_VAR]], [[S_INT_TY]]* {{.*}}[[MASTER_REF]])
+// TLS-CHECK: call {{.*}} [[S_INT_TY_COPY_ASSIGN]]([[S_INT_TY]]* {{.*}}[[TMAIN_VAR]], [[S_INT_TY]]* {{.*}}[[MASTER_REF3]])
 
 // CHECK: call {{.*}}i32 @__kmpc_cancel_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i32 [[GTID]])
 // CHECK: ret void
@@ -474,12 +438,12 @@ int main() {
 // TLS-CHECK: call {{.*}}i32 @__kmpc_cancel_barrier(%{{.+}}* [[IMPLICIT_BARRIER_LOC]], i32 [[GTID]])
 // TLS-CHECK: ret void
 
-// CHECK: define internal {{.*}}void [[TMAIN_MICROTASK1]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, {{%.+}}* %{{.+}})
+// CHECK: define internal {{.*}}void [[TMAIN_MICROTASK1]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 // CHECK: [[GTID_ADDR:%.+]] = load i32*, i32** [[GTID_ADDR_ADDR]],
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_ADDR]],
 
-// TLS-CHECK: define internal {{.*}}void [[TMAIN_MICROTASK1]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, {{%.+}}* %{{.+}})
+// TLS-CHECK: define internal {{.*}}void [[TMAIN_MICROTASK1]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}},
 // TLS-CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
 // threadprivate_t_var = t_var;
@@ -492,8 +456,8 @@ int main() {
 // CHECK: store i{{[0-9]+}} %{{.+}}, i{{[0-9]+}}* %{{.+}},
 // CHECK: [[DONE]]
 
-// TLS-CHECK: [[MASTER_FIELD:%.+]] = getelementptr inbounds {{.+}}
-// TLS-CHECK: [[MASTER_REF:%.+]] = load i32*, i32** [[MASTER_FIELD]]
+// TLS-CHECK: [[MASTER_REF:%.+]] = load i32*, i32** %
+
 // TLS-CHECK: [[MASTER_LONG:%.+]] = ptrtoint i32* [[MASTER_REF]] to i{{[0-9]+}}
 // TLS-CHECK: icmp ne i{{[0-9]+}} [[MASTER_LONG]], ptrtoint (i{{[0-9]+}}* [[TMAIN_T_VAR]] to i{{[0-9]+}})
 // TLS-CHECK: br i1 %{{.+}}, label %[[NOT_MASTER:.+]], label %[[DONE:.+]]
@@ -513,7 +477,6 @@ int main() {
 #endif
 #else
 // ARRAY-LABEL: array_func
-// TLS-ARRAY: [[ARR_CAPTURETY:%.+]] = type { [2 x {{.*}}]*, [2 x {{.*}}]* }
 // TLS-ARRAY-LABEL: array_func
 
 struct St {
@@ -531,15 +494,10 @@ void array_func() {
 // ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.+}}, i8* bitcast ([2 x i32]* @{{.+}} to i8*), i64 8, i32 4, i1 false)
 // ARRAY: call dereferenceable(8) %struct.St* @{{.+}}(%struct.St* %{{.+}}, %struct.St* dereferenceable(8) %{{.+}})
 
-// TLS-ARRAY-DAG: [[ARR_CAP1:%.+]] = getelementptr inbounds [[ARR_CAPTURETY]], [[ARR_CAPTURETY]]* {{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// TLS-ARRAY-DAG: [[ARR_CAP2:%.+]] = getelementptr inbounds [[ARR_CAPTURETY]], [[ARR_CAPTURETY]]* {{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// TLS-ARRAY-DAG: store [2 x {{.*}}]* @{{.*}}, [2 x {{.*}}]** [[ARR_CAP1]]
-// TLS-ARRAY-DAG: store [2 x {{.*}}]* @{{.*}}, [2 x {{.*}}]** [[ARR_CAP2]]
 // TLS-ARRAY: @__kmpc_fork_call(
-// TLS-ARRAY-DAG: call void @llvm.memcpy.p0i8.p0i8.i64(i8* bitcast ([2 x i32]* @{{.+}} to i8*), i8* [[REF:%.+]], i64 8, i32 4, i1 false)
-// TLS-ARRAY-DAG: [[REF]] = bitcast [2 x i32]* [[REFT:%.+]] to i8*
-// TLS-ARRAY-DAG: [[REFT]] = load [2 x i32]*, [2 x i32]** [[FIELDADDR:%.+]],
-// TLS-ARRAY-DAG: [[FIELDADDR]] = getelementptr inbounds [[ARR_CAPTURETY]], [[ARR_CAPTURETY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}}
+// TLS-ARRAY: [[REFT:%.+]] = load [2 x i32]*, [2 x i32]** [[ADDR:%.+]],
+// TLS-ARRAY: [[REF:%.+]] = bitcast [2 x i32]* [[REFT]] to i8*
+// TLS-ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* bitcast ([2 x i32]* @{{.+}} to i8*), i8* [[REF]], i64 8, i32 4, i1 false)
 // TLS-ARRAY: call dereferenceable(8) %struct.St* @{{.+}}(%struct.St* %{{.+}}, %struct.St* dereferenceable(8) %{{.+}})
 
 #pragma omp threadprivate(a, s)
diff --git a/test/OpenMP/parallel_firstprivate_codegen.cpp b/test/OpenMP/parallel_firstprivate_codegen.cpp
index 6e0f014f4d..419af7992b 100644
--- a/test/OpenMP/parallel_firstprivate_codegen.cpp
+++ b/test/OpenMP/parallel_firstprivate_codegen.cpp
@@ -31,8 +31,6 @@ struct S {
 // CHECK-DAG: [[S_FLOAT_TY:%.+]] = type { float }
 // CHECK-DAG: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
 // CHECK-DAG: [[ST_TY:%.+]] = type { i{{[0-9]+}}, i{{[0-9]+}} }
-// CHECK-DAG: [[CAP_MAIN_TY:%.+]] = type { [2 x i{{[0-9]+}}]*, i{{[0-9]+}}*, [2 x [[S_FLOAT_TY]]]*, [[S_FLOAT_TY]]* }
-// CHECK-DAG: [[CAP_TMAIN_TY:%.+]] = type { [2 x i{{[0-9]+}}]*, i{{[0-9]+}}*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]* }
 // CHECK-DAG: [[IMPLICIT_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 66, i32 0, i32 0, i8*
 
 template <typename T>
@@ -59,18 +57,12 @@ int main() {
   // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: [[G_LOCAL_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[AGG_CAPTURED:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // LAMBDA: store i{{[0-9]+}}* [[G]], i{{[0-9]+}}** [[G_LOCAL_REF]]
-  // LAMBDA: [[ARG:%.+]] = bitcast %{{.+}}* [[AGG_CAPTURED]] to i8*
-  // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* [[ARG]])
+  // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i32* [[G]])
 #pragma omp parallel firstprivate(g)
   {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}})
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // LAMBDA: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
-    // LAMBDA: [[ARG:%.+]] = load %{{.+}}*, %{{.+}}** [[ARG_REF]]
-    // LAMBDA: [[G_REF_ADDR:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-    // LAMBDA: [[G_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[G_REF_ADDR]]
+    // LAMBDA: [[G_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[G_REF_ADDR:%.+]]
     // LAMBDA: [[G_VAL:%.+]] = load volatile i{{[0-9]+}}, i{{[0-9]+}}* [[G_REF]]
     // LAMBDA: store i{{[0-9]+}} [[G_VAL]], i{{[0-9]+}}* [[G_PRIVATE_ADDR]]
     // LAMBDA: call {{.*}}i32 @__kmpc_cancel_barrier(
@@ -97,18 +89,12 @@ int main() {
   // BLOCKS: call {{.*}}void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: [[G_LOCAL_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[AGG_CAPTURED:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // BLOCKS: store i{{[0-9]+}}* [[G]], i{{[0-9]+}}** [[G_LOCAL_REF]]
-  // BLOCKS: [[ARG:%.+]] = bitcast %{{.+}}* [[AGG_CAPTURED]] to i8*
-  // BLOCKS: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* [[ARG]])
+  // BLOCKS: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i32* [[G]])
 #pragma omp parallel firstprivate(g)
   {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}})
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // BLOCKS: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
-    // BLOCKS: [[ARG:%.+]] = load %{{.+}}*, %{{.+}}** [[ARG_REF]]
-    // BLOCKS: [[G_REF_ADDR:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-    // BLOCKS: [[G_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[G_REF_ADDR]]
+    // BLOCKS: [[G_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[G_REF_ADDR:%.+]]
     // BLOCKS: [[G_VAL:%.+]] = load volatile i{{[0-9]+}}, i{{[0-9]+}}* [[G_REF]]
     // BLOCKS: store i{{[0-9]+}} [[G_VAL]], i{{[0-9]+}}* [[G_PRIVATE_ADDR]]
     // BLOCKS: call {{.*}}i32 @__kmpc_cancel_barrier(
@@ -149,29 +135,28 @@ int main() {
 // CHECK: define {{.*}}i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [2 x i32]*, i32*, [2 x [[S_FLOAT_TY]]]*, [[S_FLOAT_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
 // CHECK: = call {{.*}}i{{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call {{.*}} [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal {{.*}}void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal {{.*}}void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [2 x i32]* dereferenceable(8) %{{.+}}, i32* dereferenceable(4) %{{.+}}, [2 x [[S_FLOAT_TY]]]* dereferenceable(8) %{{.+}}, [[S_FLOAT_TY]]* dereferenceable(4) %{{.+}})
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
 // CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
 // CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]],
 // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_PTR_REF]],
+
+// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** %
+// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+// CHECK: [[S_ARR_REF:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** %
+// CHECK: [[VAR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** %
+
 // CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_REF]],
 // CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_PRIV]],
-// CHECK: [[VEC_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_PTR_REF:%.+]],
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
 // CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
-// CHECK: [[S_ARR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 2
-// CHECK: [[S_ARR_REF:%.+]] = load [2 x [[S_FLOAT_TY]]]*, [2 x [[S_FLOAT_TY]]]** [[S_ARR_REF_PTR]],
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_FLOAT_TY]]], [2 x [[S_FLOAT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
 // CHECK: [[S_ARR_BEGIN:%.+]] = bitcast [2 x [[S_FLOAT_TY]]]* [[S_ARR_REF]] to [[S_FLOAT_TY]]*
 // CHECK: [[S_ARR_PRIV_END:%.+]] = getelementptr [[S_FLOAT_TY]], [[S_FLOAT_TY]]* [[S_ARR_PRIV_BEGIN]], i{{[0-9]+}} 2
@@ -182,8 +167,6 @@ int main() {
 // CHECK: call {{.*}} [[S_FLOAT_TY_COPY_CONSTR:@.+]]([[S_FLOAT_TY]]* {{.+}}, [[S_FLOAT_TY]]* {{.+}}, [[ST_TY]]* [[ST_TY_TEMP]])
 // CHECK: call {{.*}} [[ST_TY_DESTR:@.+]]([[ST_TY]]* [[ST_TY_TEMP]])
 // CHECK: br i1 {{.+}}, label %{{.+}}, label %[[S_ARR_BODY]]
-// CHECK: [[VAR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 3
-// CHECK: [[VAR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[VAR_REF_PTR]],
 // CHECK: call {{.*}} [[ST_TY_DEFAULT_CONSTR]]([[ST_TY]]* [[ST_TY_TEMP:%.+]])
 // CHECK: call {{.*}} [[S_FLOAT_TY_COPY_CONSTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]], [[S_FLOAT_TY]]* {{.*}} [[VAR_REF]], [[ST_TY]]* [[ST_TY_TEMP]])
 // CHECK: call {{.*}} [[ST_TY_DESTR]]([[ST_TY]]* [[ST_TY_TEMP]])
@@ -197,27 +180,27 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call {{.*}}void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [2 x i32]*, i32*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call {{.*}} [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal {{.*}}void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal {{.*}}void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, [2 x i32]* dereferenceable(8) %{{.+}}, i32* dereferenceable(4) %{{.+}}, [2 x [[S_INT_TY]]]* dereferenceable(8) %{{.+}}, [[S_INT_TY]]* dereferenceable(4) %{{.+}})
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
 // CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
 // CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]],
 // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]],
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_PTR_REF]],
+
+// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** %
+// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+// CHECK: [[S_ARR_REF:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** %
+// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
+
 // CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_REF]],
 // CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_PRIV]],
-// CHECK: [[VEC_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_PTR_REF:%.+]],
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
 // CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
-// CHECK: [[S_ARR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 2
-// CHECK: [[S_ARR_REF:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** [[S_ARR_REF_PTR]],
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
 // CHECK: [[S_ARR_BEGIN:%.+]] = bitcast [2 x [[S_INT_TY]]]* [[S_ARR_REF]] to [[S_INT_TY]]*
 // CHECK: [[S_ARR_PRIV_END:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_PRIV_BEGIN]], i{{[0-9]+}} 2
@@ -228,8 +211,6 @@ int main() {
 // CHECK: call {{.*}} [[S_INT_TY_COPY_CONSTR:@.+]]([[S_INT_TY]]* {{.+}}, [[S_INT_TY]]* {{.+}}, [[ST_TY]]* [[ST_TY_TEMP]])
 // CHECK: call {{.*}} [[ST_TY_DESTR]]([[ST_TY]]* [[ST_TY_TEMP]])
 // CHECK: br i1 {{.+}}, label %{{.+}}, label %[[S_ARR_BODY]]
-// CHECK: [[VAR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 3
-// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_REF_PTR]],
 // CHECK: call {{.*}} [[ST_TY_DEFAULT_CONSTR]]([[ST_TY]]* [[ST_TY_TEMP:%.+]])
 // CHECK: call {{.*}} [[S_INT_TY_COPY_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]], [[S_INT_TY]]* {{.*}} [[VAR_REF]], [[ST_TY]]* [[ST_TY_TEMP]])
 // CHECK: call {{.*}} [[ST_TY_DESTR]]([[ST_TY]]* [[ST_TY_TEMP]])
@@ -242,30 +223,50 @@ int main() {
 
 #endif
 #else
-// ARRAY-LABEL: array_func
 struct St {
   int a, b;
   St() : a(0), b(0) {}
   St(const St &) { }
   ~St() {}
+  void St_func(St s[2], int n, long double vla1[n]) {
+    double vla2[n][n];
+    a = b;
+#pragma omp parallel firstprivate(s, vla1, vla2)
+    vla1[b] = vla2[1][n - 1] = a = b;
+  }
 };
 
+// ARRAY-LABEL: array_func
 void array_func(float a[3], St s[2], int n, long double vla1[n]) {
-  double vla2[n];
+  double vla2[n][n];
 // ARRAY: @__kmpc_fork_call(
-// ARRAY: [[PRIV_A:%.+]] = alloca float*
-// ARRAY: [[PRIV_S:%.+]] = alloca %struct.St*
-// ARRAY: [[PRIV_VLA1:%.+]] = alloca x86_fp80*
-// ARRAY: store float* %{{.+}}, float** [[PRIV_A]],
-// ARRAY: store %struct.St* %{{.+}}, %struct.St** [[PRIV_S]],
-// ARRAY: store x86_fp80* %{{.+}}, x86_fp80** [[PRIV_VLA1]],
+// ARRAY-DAG: [[PRIV_A:%.+]] = alloca float**,
+// ARRAY-DAG: [[PRIV_S:%.+]] = alloca %struct.St**,
+// ARRAY-DAG: [[PRIV_VLA1:%.+]] = alloca x86_fp80**,
+// ARRAY-DAG: [[PRIV_VLA2:%.+]] = alloca double*,
+// ARRAY-DAG: store float** %{{.+}}, float*** [[PRIV_A]],
+// ARRAY-DAG: store %struct.St** %{{.+}}, %struct.St*** [[PRIV_S]],
+// ARRAY-DAG: store x86_fp80** %{{.+}}, x86_fp80*** [[PRIV_VLA1]],
+// ARRAY-DAG: store double* %{{.+}}, double** [[PRIV_VLA2]],
 // ARRAY: call i8* @llvm.stacksave()
 // ARRAY: [[SIZE:%.+]] = mul nuw i64 %{{.+}}, 8
 // ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.+}}, i8* %{{.+}}, i64 [[SIZE]], i32 8, i1 false)
-// ARRAY: call void @llvm.stackrestore(i8*
 #pragma omp parallel firstprivate(a, s, vla1, vla2)
+  s[0].St_func(s, n, vla1);
   ;
 }
+
+// ARRAY-LABEL: St_func
+// ARRAY: @__kmpc_fork_call(
+// ARRAY-DAG: [[PRIV_S:%.+]] = alloca %struct.St**,
+// ARRAY-DAG: [[PRIV_VLA1:%.+]] = alloca x86_fp80**,
+// ARRAY-DAG: [[PRIV_VLA2:%.+]] = alloca double*,
+// ARRAY-DAG: store %struct.St** %{{.+}}, %struct.St*** [[PRIV_S]],
+// ARRAY-DAG: store x86_fp80** %{{.+}}, x86_fp80*** [[PRIV_VLA1]],
+// ARRAY-DAG: store double* %{{.+}}, double** [[PRIV_VLA2]],
+// ARRAY: call i8* @llvm.stacksave()
+// ARRAY: [[SIZE:%.+]] = mul nuw i64 %{{.+}}, 8
+// ARRAY: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.+}}, i8* %{{.+}}, i64 [[SIZE]], i32 8, i1 false)
 #endif
 
 
diff --git a/test/OpenMP/parallel_for_codegen.cpp b/test/OpenMP/parallel_for_codegen.cpp
index 314ba43ec6..d024b1914f 100644
--- a/test/OpenMP/parallel_for_codegen.cpp
+++ b/test/OpenMP/parallel_for_codegen.cpp
@@ -9,20 +9,15 @@
 #define HEADER
 
 // CHECK-DAG: [[IDENT_T_TY:%.+]] = type { i32, i32, i32, i32, i8* }
-// CHECK-DAG: [[CAP_TY:%.+]] = type { i8* }
 
 // CHECK-LABEL: with_var_schedule
 void with_var_schedule() {
   double a = 5;
 // CHECK: [[CHUNK_SIZE:%.+]] = fptosi double %{{.+}}to i8
 // CHECK: store i8 %{{.+}}, i8* [[CHUNK:%.+]],
-// CHECK: [[CHUNK_REF:%.+]] = getelementptr inbounds [[CAP_TY]], [[CAP_TY]]* [[CAP_ARG:%.+]], i{{.+}} 0, i{{.+}} 0
-// CHECK: store i8* [[CHUNK]], i8** [[CHUNK_REF]],
-// CHECK: [[BITCAST:%.+]] = bitcast [[CAP_TY]]* [[CAP_ARG]] to i8*
-// CHECK: call void {{.+}} @__kmpc_fork_call({{.+}}, i8* [[BITCAST]])
+// CHECK: call void {{.+}} @__kmpc_fork_call({{.+}}, i8* [[CHUNK]])
 
-// CHECK: [[CHUNK_REF:%.+]] = getelementptr inbounds [[CAP_TY]], [[CAP_TY]]* %{{.+}}, i{{.+}} 0, i{{.+}} 0
-// CHECK: [[CHUNK:%.+]] = load i8*, i8** [[CHUNK_REF]],
+// CHECK: [[CHUNK:%.+]] = load i8*, i8** %
 // CHECK: [[CHUNK_VAL:%.+]] = load i8, i8* [[CHUNK]],
 // CHECK: [[CHUNK_SIZE:%.+]] = sext i8 [[CHUNK_VAL]] to i64
 // CHECK: call void @__kmpc_for_static_init_8u([[IDENT_T_TY]]* [[DEFAULT_LOC:@[^,]+]], i32 [[GTID:%[^,]+]], i32 33, i32* [[IS_LAST:%[^,]+]], i64* [[OMP_LB:%[^,]+]], i64* [[OMP_UB:%[^,]+]], i64* [[OMP_ST:%[^,]+]], i64 1, i64 [[CHUNK_SIZE]])
@@ -36,8 +31,8 @@ void with_var_schedule() {
 // CHECK-LABEL: define {{.*void}} @{{.*}}without_schedule_clause{{.*}}(float* {{.+}}, float* {{.+}}, float* {{.+}}, float* {{.+}})
 void without_schedule_clause(float *a, float *b, float *c, float *d) {
   #pragma omp parallel for
-// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC:[@%].+]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %{{.+}}*)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
-// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* [[GTID_PARAM_ADDR:%.+]], i32* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC:[@%].+]], i32 4, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, float**, float**, float**, float**)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*),
+// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* noalias [[GTID_PARAM_ADDR:%.+]], i32* noalias %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}})
 // CHECK: store i32* [[GTID_PARAM_ADDR]], i32** [[GTID_REF_ADDR:%.+]],
 // CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_REF_ADDR]],
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_REF]],
@@ -80,8 +75,8 @@ void without_schedule_clause(float *a, float *b, float *c, float *d) {
 // CHECK-LABEL: define {{.*void}} @{{.*}}static_not_chunked{{.*}}(float* {{.+}}, float* {{.+}}, float* {{.+}}, float* {{.+}})
 void static_not_chunked(float *a, float *b, float *c, float *d) {
   #pragma omp parallel for schedule(static)
-// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %{{.+}}*)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
-// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* [[GTID_PARAM_ADDR:%.+]], i32* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC:[@%].+]], i32 4, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, float**, float**, float**, float**)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*),
+// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* noalias [[GTID_PARAM_ADDR:%.+]], i32* noalias %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}})
 // CHECK: store i32* [[GTID_PARAM_ADDR]], i32** [[GTID_REF_ADDR:%.+]],
 // CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_REF_ADDR]],
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_REF]],
@@ -124,8 +119,8 @@ void static_not_chunked(float *a, float *b, float *c, float *d) {
 // CHECK-LABEL: define {{.*void}} @{{.*}}static_chunked{{.*}}(float* {{.+}}, float* {{.+}}, float* {{.+}}, float* {{.+}})
 void static_chunked(float *a, float *b, float *c, float *d) {
   #pragma omp parallel for schedule(static, 5)
-// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %{{.+}}*)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
-// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* [[GTID_PARAM_ADDR:%.+]], i32* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC:[@%].+]], i32 4, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, float**, float**, float**, float**)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*),
+// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* noalias [[GTID_PARAM_ADDR:%.+]], i32* noalias %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}})
 // CHECK: store i32* [[GTID_PARAM_ADDR]], i32** [[GTID_REF_ADDR:%.+]],
 // CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_REF_ADDR]],
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_REF]],
@@ -187,8 +182,8 @@ void static_chunked(float *a, float *b, float *c, float *d) {
 // CHECK-LABEL: define {{.*void}} @{{.*}}dynamic1{{.*}}(float* {{.+}}, float* {{.+}}, float* {{.+}}, float* {{.+}})
 void dynamic1(float *a, float *b, float *c, float *d) {
   #pragma omp parallel for schedule(dynamic)
-// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %{{.+}}*)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
-// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* [[GTID_PARAM_ADDR:%.+]], i32* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC:[@%].+]], i32 4, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, float**, float**, float**, float**)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*),
+// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* noalias [[GTID_PARAM_ADDR:%.+]], i32* noalias %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}})
 // CHECK: store i32* [[GTID_PARAM_ADDR]], i32** [[GTID_REF_ADDR:%.+]],
 // CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_REF_ADDR]],
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_REF]],
@@ -232,8 +227,8 @@ void dynamic1(float *a, float *b, float *c, float *d) {
 // CHECK-LABEL: define {{.*void}} @{{.*}}guided7{{.*}}(float* {{.+}}, float* {{.+}}, float* {{.+}}, float* {{.+}})
 void guided7(float *a, float *b, float *c, float *d) {
   #pragma omp parallel for schedule(guided, 7)
-// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %{{.+}}*)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
-// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* [[GTID_PARAM_ADDR:%.+]], i32* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC:[@%].+]], i32 4, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, float**, float**, float**, float**)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*),
+// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* noalias [[GTID_PARAM_ADDR:%.+]], i32* noalias %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}})
 // CHECK: store i32* [[GTID_PARAM_ADDR]], i32** [[GTID_REF_ADDR:%.+]],
 // CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_REF_ADDR]],
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_REF]],
@@ -279,8 +274,8 @@ void test_auto(float *a, float *b, float *c, float *d) {
   unsigned int x = 0;
   unsigned int y = 0;
   #pragma omp parallel for schedule(auto) collapse(2)
-// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %{{.+}}*)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
-// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* [[GTID_PARAM_ADDR:%.+]], i32* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC:[@%].+]], i32 6, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*, i32*, float**, float**, float**, float**)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*),
+// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* noalias [[GTID_PARAM_ADDR:%.+]], i32* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}}, i32* dereferenceable(4) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}})
 // CHECK: store i32* [[GTID_PARAM_ADDR]], i32** [[GTID_REF_ADDR:%.+]],
 // CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_REF_ADDR]],
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_REF]],
@@ -330,8 +325,8 @@ void test_auto(float *a, float *b, float *c, float *d) {
 void runtime(float *a, float *b, float *c, float *d) {
   int x = 0;
   #pragma omp parallel for collapse(2) schedule(runtime)
-// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %{{.+}}*)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
-// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* [[GTID_PARAM_ADDR:%.+]], i32* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: call void ([[IDENT_T_TY]]*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call([[IDENT_T_TY]]* [[DEFAULT_LOC:[@%].+]], i32 5, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i32*, float**, float**, float**, float**)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*),
+// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* noalias [[GTID_PARAM_ADDR:%.+]], i32* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}}, float** dereferenceable(8) %{{.+}})
 // CHECK: store i32* [[GTID_PARAM_ADDR]], i32** [[GTID_REF_ADDR:%.+]],
 // CHECK: [[GTID_REF:%.+]] = load i32*, i32** [[GTID_REF_ADDR]],
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_REF]],
diff --git a/test/OpenMP/parallel_for_linear_codegen.cpp b/test/OpenMP/parallel_for_linear_codegen.cpp
index 7d27efef7b..d980dd34de 100644
--- a/test/OpenMP/parallel_for_linear_codegen.cpp
+++ b/test/OpenMP/parallel_for_linear_codegen.cpp
@@ -23,9 +23,7 @@ float f;
 char cnt;
 
 // CHECK: [[S_FLOAT_TY:%.+]] = type { float }
-// CHECK: [[CAP_MAIN_TY:%.+]] = type { float**, i64* }
 // CHECK: [[S_INT_TY:%.+]] = type { i32 }
-// CHECK: [[CAP_TMAIN_TY:%.+]] = type { i32**, i32* }
 // CHECK-DAG: [[IMPLICIT_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 66, i32 0, i32 0, i8*
 // CHECK-DAG: [[F:@.+]] = global float 0.0
 // CHECK-DAG: [[CNT:@.+]] = global i8 0
@@ -48,10 +46,10 @@ int main() {
   // LAMBDA: call void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i32* [[G]])
 #pragma omp parallel for linear(g:5)
   for (int i = 0; i < 2; ++i) {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}})
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: [[G_START_ADDR:%.+]] = alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
@@ -60,7 +58,6 @@ int main() {
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // LAMBDA: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     // LAMBDA: store i32 0,
     // LAMBDA: [[GTID_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %{{.+}}
     // LAMBDA: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_REF]]
@@ -97,10 +94,10 @@ int main() {
   // BLOCKS: call void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i32* [[G]])
 #pragma omp parallel for linear(g:5)
   for (int i = 0; i < 2; ++i) {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}})
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: [[G_START_ADDR:%.+]] = alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
@@ -109,7 +106,6 @@ int main() {
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // BLOCKS: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     // BLOCKS: store i32 0,
     // BLOCKS: [[GTID_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %{{.+}}
     // BLOCKS: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_REF]]
@@ -156,13 +152,12 @@ int main() {
 // CHECK: define i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 2, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, float**, i64*)* [[MAIN_MICROTASK:@.+]] to void
 // CHECK: = call {{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call void [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 
-// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, float** dereferenceable(8) %{{.+}}, i64* dereferenceable(8) %{{.+}})
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: [[PVAR_START:%.+]] = alloca float*,
 // CHECK: [[LVAR_START:%.+]] = alloca i64,
@@ -175,13 +170,12 @@ int main() {
 // CHECK: [[LVAR_PRIV:%.+]] = alloca i64,
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
 
+// CHECK: [[PVAR_REF:%.+]] = load float**, float*** %
+// CHECK: [[LVAR_REF:%.+]] = load i64*, i64** %
+
 // Check for default initialization.
-// CHECK: [[PVAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[PVAR_REF:%.+]] = load float**, float*** [[PVAR_PTR_REF]],
 // CHECK: [[PVAR_VAL:%.+]] = load float*, float** [[PVAR_REF]],
 // CHECK: store float* [[PVAR_VAL]], float** [[PVAR_START]],
-// CHECK: [[LVAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[LVAR_REF:%.+]] = load i64*, i64** [[LVAR_PTR_REF]],
 // CHECK: [[LVAR_VAL:%.+]] = load i64, i64* [[LVAR_REF]],
 // CHECK: store i64 [[LVAR_VAL]], i64* [[LVAR_START]],
 // CHECK: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 [[GTID:%.+]], i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
@@ -210,11 +204,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 2, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32**, i32*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call void [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i32** dereferenceable(8) %{{.+}}, i32* dereferenceable(4) %{{.+}})
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: [[PVAR_START:%.+]] = alloca i32*,
 // CHECK: [[LVAR_START:%.+]] = alloca i32,
@@ -227,13 +221,12 @@ int main() {
 // CHECK: [[LVAR_PRIV:%.+]] = alloca i32,
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
 
+// CHECK: [[PVAR_REF:%.+]] = load i32**, i32*** %
+// CHECK: [[LVAR_REF:%.+]] = load i32*, i32** %
+
 // Check for default initialization.
-// CHECK: [[PVAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[PVAR_REF:%.+]] = load i32**, i32*** [[PVAR_PTR_REF]],
 // CHECK: [[PVAR_VAL:%.+]] = load i32*, i32** [[PVAR_REF]],
 // CHECK: store i32* [[PVAR_VAL]], i32** [[PVAR_START]],
-// CHECK: [[LVAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[LVAR_REF:%.+]] = load i32*, i32** [[LVAR_PTR_REF]],
 // CHECK: [[LVAR_VAL:%.+]] = load i32, i32* [[LVAR_REF]],
 // CHECK: store i32 [[LVAR_VAL]], i32* [[LVAR_START]],
 // CHECK: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 [[GTID:%.+]], i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
diff --git a/test/OpenMP/parallel_for_simd_codegen.cpp b/test/OpenMP/parallel_for_simd_codegen.cpp
index eb1e493230..4383c0a5b4 100644
--- a/test/OpenMP/parallel_for_simd_codegen.cpp
+++ b/test/OpenMP/parallel_for_simd_codegen.cpp
@@ -119,14 +119,13 @@ void simple(float *a, float *b, float *c, float *d) {
 
 // Init linear private var.
 // CHECK: [[LIN_VAR:%.+]] = load i32*, i32** %
-// CHECK-NEXT: [[LIN_LOAD:%.+]] = load i32, i32* [[LIN_VAR]]
+// CHECK: [[LIN_LOAD:%.+]] = load i32, i32* [[LIN_VAR]]
 // CHECK-NEXT: store i32 [[LIN_LOAD]], i32* [[LIN_START:%[^,]+]]
 // Remember linear step.
 // CHECK: [[CALL_VAL:%.+]] = invoke
 // CHECK: store i64 [[CALL_VAL]], i64* [[LIN_STEP:%[^,]+]]
 
-// CHECK: [[GLIN_VAR:%.+]] = load double**, double*** %
-// CHECK-NEXT: [[GLIN_LOAD:%.+]] = load double*, double** [[GLIN_VAR]]
+// CHECK: [[GLIN_LOAD:%.+]] = load double*, double** [[GLIN_VAR:%.+]],
 // CHECK-NEXT: store double* [[GLIN_LOAD]], double** [[GLIN_START:%[^,]+]]
 
 // CHECK: call void @__kmpc_for_static_init_8u(%ident_t* {{[^,]+}}, i32 %{{[^,]+}}, i32 34, i32* %{{[^,]+}}, i64* [[LB:%[^,]+]], i64* [[UB:%[^,]+]], i64* [[STRIDE:%[^,]+]], i64 1, i64 1)
@@ -179,11 +178,9 @@ void simple(float *a, float *b, float *c, float *d) {
 // CHECK: call void @__kmpc_for_static_fini(%ident_t* {{.+}}, i32 %{{.+}})
 //
 // Linear start and step are used to calculate final value of the linear variables.
-// CHECK: [[LIN_VAR:%.+]] = load i32*, i32** %
 // CHECK: [[LINSTART:.+]] = load i32, i32* [[LIN_START]]
 // CHECK: [[LINSTEP:.+]] = load i64, i64* [[LIN_STEP]]
 // CHECK: store i32 {{.+}}, i32* [[LIN_VAR]],
-// CHECK: [[GLIN_VAR:%.+]] = load double**, double*** %
 // CHECK: [[GLINSTART:.+]] = load double*, double** [[GLIN_START]]
 // CHECK: store double* {{.*}}[[GLIN_VAR]]
 // CHECK: call void @__kmpc_barrier(%ident_t* {{.+}}, i32 %{{.+}})
diff --git a/test/OpenMP/parallel_if_codegen.cpp b/test/OpenMP/parallel_if_codegen.cpp
index e711b5d299..9ecf57211f 100644
--- a/test/OpenMP/parallel_if_codegen.cpp
+++ b/test/OpenMP/parallel_if_codegen.cpp
@@ -16,14 +16,14 @@ int Arg;
 
 // CHECK-LABEL: define {{.*}}void @{{.+}}gtid_test
 void gtid_test() {
-// CHECK:  call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 1, {{.+}}* [[GTID_TEST_REGION1:@.+]] to void
+// CHECK:  call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 0, {{.+}}* [[GTID_TEST_REGION1:@.+]] to void
 #pragma omp parallel
 #pragma omp parallel if (parallel: false)
   gtid_test();
 // CHECK: ret void
 }
 
-// CHECK: define internal {{.*}}void [[GTID_TEST_REGION1]](i{{.+}}* [[GTID_PARAM:%.+]], i
+// CHECK: define internal {{.*}}void [[GTID_TEST_REGION1]](i{{.+}}* noalias [[GTID_PARAM:%.+]], i32* noalias
 // CHECK: store i{{[0-9]+}}* [[GTID_PARAM]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]],
 // CHECK: [[GTID_ADDR:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[GTID_ADDR_REF]]
 // CHECK: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_ADDR]]
@@ -51,7 +51,7 @@ int tmain(T Arg) {
 // CHECK-LABEL: define {{.*}}i{{[0-9]+}} @main()
 int main() {
 // CHECK: [[GTID:%.+]] = call {{.*}}i32 @__kmpc_global_thread_num(
-// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 1, void {{.+}}* [[CAP_FN4:@.+]] to void
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 0, void {{.+}}* [[CAP_FN4:@.+]] to void
 #pragma omp parallel if (true)
   fn4();
 // CHECK: call {{.*}}void @__kmpc_serialized_parallel(%{{.+}}* @{{.+}}, i32 [[GTID]])
@@ -63,7 +63,7 @@ int main() {
 
 // CHECK: br i1 %{{.+}}, label %[[OMP_THEN:.+]], label %[[OMP_ELSE:.+]]
 // CHECK: [[OMP_THEN]]
-// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 1, void {{.+}}* [[CAP_FN6:@.+]] to void
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 0, void {{.+}}* [[CAP_FN6:@.+]] to void
 // CHECK: br label %[[OMP_END:.+]]
 // CHECK: [[OMP_ELSE]]
 // CHECK: call {{.*}}void @__kmpc_serialized_parallel(%{{.+}}* @{{.+}}, i32 [[GTID]])
@@ -92,14 +92,14 @@ int main() {
 
 // CHECK-LABEL: define {{.+}} @{{.+}}tmain
 // CHECK: [[GTID:%.+]] = call {{.*}}i32 @__kmpc_global_thread_num(
-// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 1, void {{.+}}* [[CAP_FN1:@.+]] to void
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 0, void {{.+}}* [[CAP_FN1:@.+]] to void
 // CHECK: call {{.*}}void @__kmpc_serialized_parallel(%{{.+}}* @{{.+}}, i32 [[GTID]])
 // CHECK: store i32 [[GTID]], i32* [[GTID_ADDR:%.+]],
 // CHECK: call void [[CAP_FN2:@.+]](i32* [[GTID_ADDR]],
 // CHECK: call {{.*}}void @__kmpc_end_serialized_parallel(%{{.+}}* @{{.+}}, i32 [[GTID]])
 // CHECK: br i1 %{{.+}}, label %[[OMP_THEN:.+]], label %[[OMP_ELSE:.+]]
 // CHECK: [[OMP_THEN]]
-// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 1, void {{.+}}* [[CAP_FN3:@.+]] to void
+// CHECK: call {{.*}}void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 0, void {{.+}}* [[CAP_FN3:@.+]] to void
 // CHECK: br label %[[OMP_END:.+]]
 // CHECK: [[OMP_ELSE]]
 // CHECK: call {{.*}}void @__kmpc_serialized_parallel(%{{.+}}* @{{.+}}, i32 [[GTID]])
diff --git a/test/OpenMP/parallel_private_codegen.cpp b/test/OpenMP/parallel_private_codegen.cpp
index d2d02a7c3d..616d50fae9 100644
--- a/test/OpenMP/parallel_private_codegen.cpp
+++ b/test/OpenMP/parallel_private_codegen.cpp
@@ -19,9 +19,7 @@ struct S {
 volatile int g = 1212;
 
 // CHECK: [[S_FLOAT_TY:%.+]] = type { float }
-// CHECK: [[CAP_MAIN_TY:%.+]] = type { i8 }
 // CHECK: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
-// CHECK: [[CAP_TMAIN_TY:%.+]] = type { i8 }
 template <typename T>
 T tmain() {
   S<T> test;
@@ -45,13 +43,11 @@ int main() {
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
   // LAMBDA-NOT: = getelementptr inbounds %{{.+}},
-  // LAMBDA: [[ARG:%.+]] = bitcast %{{.+}}* %{{.+}} to i8*
-  // LAMBDA: call{{.*}} void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* [[ARG]])
+  // LAMBDA: call{{.*}} void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel private(g)
   {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // LAMBDA: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     g = 1;
     // LAMBDA: store i{{[0-9]+}} 1, i{{[0-9]+}}* [[G_PRIVATE_ADDR]],
     // LAMBDA: [[G_PRIVATE_ADDR_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
@@ -76,13 +72,11 @@ int main() {
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
   // BLOCKS-NOT: = getelementptr inbounds %{{.+}},
-  // BLOCKS: [[ARG:%.+]] = bitcast %{{.+}}* %{{.+}} to i8*
-  // BLOCKS: call{{.*}} void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* [[ARG]])
+  // BLOCKS: call{{.*}} void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel private(g)
   {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // BLOCKS: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     g = 1;
     // BLOCKS: store i{{[0-9]+}} 1, i{{[0-9]+}}* [[G_PRIVATE_ADDR]],
     // BLOCKS-NOT: [[G]]{{[[^:word:]]}}
@@ -118,13 +112,12 @@ int main() {
 // CHECK: define i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[MAIN_MICROTASK:@.+]] to void
 // CHECK: = call i{{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call void [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
 // CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
 // CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]],
@@ -145,11 +138,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call void [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
 // CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
 // CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]],
diff --git a/test/OpenMP/parallel_reduction_codegen.cpp b/test/OpenMP/parallel_reduction_codegen.cpp
index d04676010c..6ecd42bb94 100644
--- a/test/OpenMP/parallel_reduction_codegen.cpp
+++ b/test/OpenMP/parallel_reduction_codegen.cpp
@@ -22,8 +22,6 @@ struct S {
 
 // CHECK-DAG: [[S_FLOAT_TY:%.+]] = type { float }
 // CHECK-DAG: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
-// CHECK-DAG: [[CAP_MAIN_TY:%.+]] = type { [2 x i{{[0-9]+}}]*, float*, [2 x [[S_FLOAT_TY]]]*, [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]*, float* }
-// CHECK-DAG: [[CAP_TMAIN_TY:%.+]] = type { [2 x i{{[0-9]+}}]*, i{{[0-9]+}}*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*, [[S_INT_TY]]*, i{{[0-9]+}}* }
 // CHECK-DAG: [[IMPLICIT_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 66, i32 0, i32 0, i8*
 // CHECK-DAG: [[REDUCTION_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 18, i32 0, i32 0, i8*
 // CHECK-DAG: [[REDUCTION_LOCK:@.+]] = common global [8 x i32] zeroinitializer
@@ -51,22 +49,16 @@ int main() {
   // LAMBDA: call void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: [[G_LOCAL_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[AGG_CAPTURED:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // LAMBDA: store i{{[0-9]+}}* [[G]], i{{[0-9]+}}** [[G_LOCAL_REF]]
-  // LAMBDA: [[ARG:%.+]] = bitcast %{{.+}}* [[AGG_CAPTURED]] to i8*
-  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* [[ARG]])
+  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i32* [[G]])
 #pragma omp parallel reduction(+:g)
   {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}})
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
 
     // Reduction list for runtime.
     // LAMBDA: [[RED_LIST:%.+]] = alloca [1 x i8*],
 
-    // LAMBDA: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
-    // LAMBDA: [[ARG:%.+]] = load %{{.+}}*, %{{.+}}** [[ARG_REF]]
-    // LAMBDA: [[G_REF_ADDR:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-    // LAMBDA: [[G_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[G_REF_ADDR]]
+    // LAMBDA: [[G_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[G_REF_ADDR:%.+]]
     // LAMBDA: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[G_PRIVATE_ADDR]]
     g = 1;
     // LAMBDA: store i{{[0-9]+}} 1, i{{[0-9]+}}* [[G_PRIVATE_ADDR]],
@@ -112,22 +104,16 @@ int main() {
   // BLOCKS: call void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: [[G_LOCAL_REF:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[AGG_CAPTURED:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-  // BLOCKS: store i{{[0-9]+}}* [[G]], i{{[0-9]+}}** [[G_LOCAL_REF]]
-  // BLOCKS: [[ARG:%.+]] = bitcast %{{.+}}* [[AGG_CAPTURED]] to i8*
-  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* [[ARG]])
+  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i32* [[G]])
 #pragma omp parallel reduction(-:g)
   {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}})
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
 
     // Reduction list for runtime.
     // BLOCKS: [[RED_LIST:%.+]] = alloca [1 x i8*],
 
-    // BLOCKS: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
-    // BLOCKS: [[ARG:%.+]] = load %{{.+}}*, %{{.+}}** [[ARG_REF]]
-    // BLOCKS: [[G_REF_ADDR:%.+]] = getelementptr inbounds %{{.+}}, %{{.+}}* [[ARG]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
-    // BLOCKS: [[G_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[G_REF_ADDR]]
+    // BLOCKS: [[G_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[G_REF_ADDR:%.+]]
     // BLOCKS: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[G_PRIVATE_ADDR]]
     g = 1;
     // BLOCKS: store i{{[0-9]+}} 1, i{{[0-9]+}}* [[G_PRIVATE_ADDR]],
@@ -185,13 +171,12 @@ int main() {
 // CHECK: define {{.*}}i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 6, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [2 x i32]*, float*, [2 x [[S_FLOAT_TY]]]*, [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]*, float*)* [[MAIN_MICROTASK:@.+]] to void
 // CHECK: = call {{.*}}i{{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call {{.*}} [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}},
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca float,
 // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: [[VAR1_PRIV:%.+]] = alloca [[S_FLOAT_TY]],
@@ -202,23 +187,20 @@ int main() {
 
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[T_VAR_REF:%.+]] = load float*, float** [[T_VAR_PTR_REF]],
+// CHECK: [[T_VAR_REF:%.+]] = load float*, float** %
+// CHECK: [[VAR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** %
+// CHECK: [[VAR1_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** %
+// CHECK: [[T_VAR1_REF:%.+]] = load float*, float** %
+
 // For + reduction operation initial value of private variable is 0.
 // CHECK: store float 0.0{{.+}}, float* [[T_VAR_PRIV]],
 
-// CHECK: [[VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[VAR_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[VAR_PTR_REF:%.+]],
 // For & reduction operation initial value of private variable is ones in all bits.
 // CHECK: call {{.*}} [[S_FLOAT_TY_CONSTR:@.+]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
 
-// CHECK: [[VAR1_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[VAR1_REF:%.+]] = load [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]** [[VAR_PTR_REF:%.+]],
 // For && reduction operation initial value of private variable is 1.0.
 // CHECK: call {{.*}} [[S_FLOAT_TY_CONSTR:@.+]]([[S_FLOAT_TY]]* [[VAR1_PRIV]])
 
-// CHECK: [[T_VAR1_PTR_REF:%.+]] = getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[T_VAR1_REF:%.+]] = load float*, float** [[T_VAR1_PTR_REF]],
 // For min reduction operation initial value of private variable is largest repesentable value.
 // CHECK: store float 0x47EFFFFFE0000000, float* [[T_VAR1_PRIV]],
 
@@ -448,12 +430,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_TMAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 6, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [2 x i32]*, i32*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*, [[S_INT_TY]]*, i32*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call {{.*}} [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}},
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
 // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]],
 // CHECK: [[VAR1_PRIV:%.+]] = alloca [[S_INT_TY]],
@@ -464,23 +445,20 @@ int main() {
 
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_PTR_REF]],
+// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
+// CHECK: [[VAR1_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
+// CHECK: [[T_VAR1_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+
 // For + reduction operation initial value of private variable is 0.
 // CHECK: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[T_VAR_PRIV]],
 
-// CHECK: [[VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_PTR_REF:%.+]],
 // For & reduction operation initial value of private variable is ones in all bits.
 // CHECK: call {{.*}} [[S_INT_TY_CONSTR:@.+]]([[S_INT_TY]]* [[VAR_PRIV]])
 
-// CHECK: [[VAR1_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[VAR1_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_PTR_REF:%.+]],
 // For && reduction operation initial value of private variable is 1.0.
 // CHECK: call {{.*}} [[S_INT_TY_CONSTR:@.+]]([[S_INT_TY]]* [[VAR1_PRIV]])
 
-// CHECK: [[T_VAR1_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[T_VAR1_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR1_PTR_REF]],
 // For min reduction operation initial value of private variable is largest repesentable value.
 // CHECK: store i{{[0-9]+}} 2147483647, i{{[0-9]+}}* [[T_VAR1_PRIV]],
 
diff --git a/test/OpenMP/parallel_sections_codegen.cpp b/test/OpenMP/parallel_sections_codegen.cpp
index 2e9e1f95ae..a517062181 100644
--- a/test/OpenMP/parallel_sections_codegen.cpp
+++ b/test/OpenMP/parallel_sections_codegen.cpp
@@ -22,9 +22,9 @@ T tmain() {
 
 // CHECK-LABEL: @main
 int main() {
-// CHECK: call void (%{{.+}}*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, %{{.+}}*)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*), i8* %{{.+}})
+// CHECK: call void (%{{.+}}*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i32 0, void (i32*, i32*, ...)* bitcast (void (i32*, i32*)* [[OMP_PARALLEL_FUNC:@.+]] to void (i32*, i32*, ...)*))
 // CHECK-LABEL: }
-// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* [[GTID_PARAM_ADDR:%.+]], i32* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: define internal void [[OMP_PARALLEL_FUNC]](i32* noalias [[GTID_PARAM_ADDR:%.+]], i32* noalias %{{.+}})
 // CHECK: store i32* [[GTID_PARAM_ADDR]], i32** [[GTID_REF_ADDR:%.+]],
 #pragma omp parallel sections
   {
diff --git a/test/OpenMP/sections_firstprivate_codegen.cpp b/test/OpenMP/sections_firstprivate_codegen.cpp
index af9c1fe5fb..b5db2c5871 100644
--- a/test/OpenMP/sections_firstprivate_codegen.cpp
+++ b/test/OpenMP/sections_firstprivate_codegen.cpp
@@ -30,7 +30,6 @@ struct S {
 // CHECK-DAG: [[S_FLOAT_TY:%.+]] = type { float }
 // CHECK-DAG: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
 // CHECK-DAG: [[ST_TY:%.+]] = type { i{{[0-9]+}}, i{{[0-9]+}} }
-// CHECK-DAG: [[CAP_TMAIN_TY:%.+]] = type { i{{[0-9]+}}*, [2 x i{{[0-9]+}}]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]* }
 
 template <typename T>
 T tmain() {
@@ -71,11 +70,11 @@ int main() {
   // LAMBDA: call void [[OUTER_LAMBDA:@.+]](
   [&]() {
 // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-// LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+// LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp sections firstprivate(g)
   {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // Skip temp vars for loop
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
@@ -113,11 +112,11 @@ int main() {
   // BLOCKS: call void {{%.+}}(i8
   ^{
 // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-// BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+// BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp sections firstprivate(g)
    {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // Skip temp vars for loop
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
@@ -209,11 +208,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [2 x i32]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call {{.*}} [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}},
 // Skip temp vars for loop
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
@@ -226,22 +225,21 @@ int main() {
 // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]],
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
+// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** %
+// CHECK: [[S_ARR:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** %
+// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
+
 // firstprivate t_var(t_var)
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_PTR_REF]],
 // CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_REF]],
 // CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_PRIV]],
 
 // firstprivate vec(vec)
-// CHECK: [[VEC_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_PTR_REF:%.+]],
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
 // CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
 
 // firstprivate s_arr(s_arr)
-// CHECK: [[S_ARR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 2
-// CHECK: [[S_ARR:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** [[S_ARR_REF]],
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
 // CHECK: [[S_ARR_PRIV_END:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_PRIV_BEGIN]], i{{[0-9]+}} 2
 // CHECK: [[IS_EMPTY:%.+]] = icmp eq [[S_INT_TY]]* [[S_ARR_PRIV_BEGIN]], [[S_ARR_PRIV_END]]
@@ -253,8 +251,6 @@ int main() {
 // CHECK: br i1 {{.+}}, label %{{.+}}, label %[[S_ARR_BODY]]
 
 // firstprivate var(var)
-// CHECK: [[VAR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 3
-// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_REF_PTR]],
 // CHECK: call {{.*}} [[ST_TY_DEFAULT_CONSTR]]([[ST_TY]]* [[ST_TY_TEMP:%.+]])
 // CHECK: call {{.*}} [[S_INT_TY_COPY_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]], [[S_INT_TY]]* {{.*}} [[VAR_REF]], [[ST_TY]]* [[ST_TY_TEMP]])
 // CHECK: call {{.*}} [[ST_TY_DESTR]]([[ST_TY]]* [[ST_TY_TEMP]])
diff --git a/test/OpenMP/sections_lastprivate_codegen.cpp b/test/OpenMP/sections_lastprivate_codegen.cpp
index c2f68aa0fe..4aa8c5c17b 100644
--- a/test/OpenMP/sections_lastprivate_codegen.cpp
+++ b/test/OpenMP/sections_lastprivate_codegen.cpp
@@ -21,9 +21,7 @@ struct S {
 volatile int g = 1212;
 
 // CHECK: [[S_FLOAT_TY:%.+]] = type { float }
-// CHECK: [[CAP_MAIN_TY:%.+]] = type { i{{[0-9]+}}*, [2 x i{{[0-9]+}}]*, [2 x [[S_FLOAT_TY]]]*, [[S_FLOAT_TY]]* }
 // CHECK: [[S_INT_TY:%.+]] = type { i32 }
-// CHECK: [[CAP_TMAIN_TY:%.+]] = type { i{{[0-9]+}}*, [2 x i{{[0-9]+}}]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]* }
 // CHECK-DAG: [[IMPLICIT_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 66, i32 0, i32 0, i8*
 // CHECK-DAG: [[SINGLE_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 322, i32 0, i32 0, i8*
 // CHECK-DAG: [[SECTIONS_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 194, i32 0, i32 0, i8*
@@ -59,18 +57,17 @@ int main() {
   // LAMBDA: call void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp sections lastprivate(g)
   {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: alloca i{{[0-9]+}},
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // LAMBDA: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     // LAMBDA: [[GTID_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %{{.+}}
     // LAMBDA: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_REF]]
     // LAMBDA: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 [[GTID]], i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
@@ -112,18 +109,17 @@ int main() {
   // BLOCKS: call void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp sections lastprivate(g)
   {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: alloca i{{[0-9]+}},
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
-    // BLOCKS: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     // BLOCKS: [[GTID_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %{{.+}}
     // BLOCKS: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_REF]]
     // BLOCKS: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 [[GTID]], i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
@@ -187,14 +183,13 @@ int main() {
 // CHECK: define i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, %{{.+}}*)* [[MAIN_MICROTASK1:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [2 x i32]*, [2 x [[S_FLOAT_TY]]]*, [[S_FLOAT_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[MAIN_MICROTASK1:@.+]] to void
 // CHECK: = call {{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call void [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 
-// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}},
 // CHECK-NOT: alloca i{{[0-9]+}},
 // CHECK-NOT: alloca [2 x i{{[0-9]+}}],
 // CHECK-NOT: alloca [2 x [[S_FLOAT_TY]]],
@@ -205,11 +200,6 @@ int main() {
 // CHECK: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_REF]]
 // CHECK: call i32 @__kmpc_single(
 
-// CHECK-DAG: getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK-DAG: getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK-DAG: getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 2
-// CHECK-DAG: getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 3
-
 // <Skip loop body>
 
 // CHECK-NOT: call void [[S_FLOAT_TY_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
@@ -222,7 +212,7 @@ int main() {
 // CHECK: ret void
 
 //
-// CHECK: define internal void [[MAIN_MICROTASK1]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, %{{.+}}* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK1]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: [[X_PRIV:%.+]] = alloca double,
 // CHECK-NOT: alloca double
 
@@ -255,11 +245,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [2 x i32]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call void [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}},
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
@@ -271,19 +261,16 @@ int main() {
 // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]],
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_REF:%.+]]
 
+// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** %
+// CHECK: [[S_ARR_REF:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** %
+// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
+
 // Check for default initialization.
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_PTR_REF]],
 // CHECK-NOT: [[T_VAR_PRIV]]
-// CHECK: [[VEC_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_PTR_REF:%.+]],
 // CHECK-NOT: [[VEC_PRIV]]
-// CHECK: [[S_ARR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 2
-// CHECK: [[S_ARR_REF:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** [[S_ARR_REF_PTR]],
 // CHECK: [[S_ARR_PRIV_ITEM:%.+]] = phi [[S_INT_TY]]*
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[S_ARR_PRIV_ITEM]])
-// CHECK: [[VAR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 3
-// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_REF_PTR]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]])
 // CHECK: call {{.+}} @__kmpc_for_static_init_4(%{{.+}}* @{{.+}}, i32 %{{.+}}, i32 34, i32* [[IS_LAST_ADDR:%.+]], i32* %{{.+}}, i32* %{{.+}}, i32* %{{.+}}, i32 1, i32 1)
 // <Skip loop body>
diff --git a/test/OpenMP/sections_private_codegen.cpp b/test/OpenMP/sections_private_codegen.cpp
index 9b579a2aa6..40712d4922 100644
--- a/test/OpenMP/sections_private_codegen.cpp
+++ b/test/OpenMP/sections_private_codegen.cpp
@@ -20,9 +20,7 @@ struct S {
 volatile double g;
 
 // CHECK: [[S_FLOAT_TY:%.+]] = type { float }
-// CHECK: [[CAP_MAIN_TY:%.+]] = type { i8 }
 // CHECK: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
-// CHECK: [[CAP_TMAIN_TY:%.+]] = type { i8 }
 template <typename T>
 T tmain() {
   S<T> test;
@@ -47,13 +45,12 @@ int main() {
   // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp sections private(g)
   {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double,
-    // LAMBDA: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     g = 1;
     // LAMBDA: call {{.*}}void @__kmpc_for_static_init_4(
     // LAMBDA: store double 1.0{{.+}}, double* [[G_PRIVATE_ADDR]],
@@ -80,13 +77,12 @@ int main() {
   // BLOCKS: call {{.*}}void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* {{.+}})
+  // BLOCKS: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp sections private(g)
     {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca double,
-    // BLOCKS: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     g = 1;
     // BLOCKS: call {{.*}}void @__kmpc_for_static_init_4(
     // BLOCKS: store double 1.0{{.+}}, double* [[G_PRIVATE_ADDR]],
@@ -128,13 +124,12 @@ int main() {
 // CHECK: define i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[MAIN_MICROTASK:@.+]] to void
 // CHECK: = call i{{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call void [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
 // CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
 // CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]],
@@ -159,11 +154,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call void [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: alloca i32,
 // CHECK: alloca i32,
 // CHECK: alloca i32,
diff --git a/test/OpenMP/sections_reduction_codegen.cpp b/test/OpenMP/sections_reduction_codegen.cpp
index 3c2525790b..fee02dca84 100644
--- a/test/OpenMP/sections_reduction_codegen.cpp
+++ b/test/OpenMP/sections_reduction_codegen.cpp
@@ -22,8 +22,6 @@ struct S {
 
 // CHECK-DAG: [[S_FLOAT_TY:%.+]] = type { float }
 // CHECK-DAG: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
-// CHECK-DAG: [[CAP_MAIN_TY:%.+]] = type { float*, [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]*, float*, [2 x i{{[0-9]+}}]*, [2 x [[S_FLOAT_TY]]]* }
-// CHECK-DAG: [[CAP_TMAIN_TY:%.+]] = type { i{{[0-9]+}}*, [[S_INT_TY]]*, [[S_INT_TY]]*, i{{[0-9]+}}*, [2 x i{{[0-9]+}}]*, [2 x [[S_INT_TY]]]* }
 // CHECK-DAG: [[ATOMIC_REDUCE_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 18, i32 0, i32 0, i8*
 // CHECK-DAG: [[IMPLICIT_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 66, i32 0, i32 0, i8*
 // CHECK-DAG: [[SINGLE_BARRIER_LOC:@.+]] = private unnamed_addr constant %{{.+}} { i32 0, i32 322, i32 0, i32 0, i8*
@@ -55,11 +53,11 @@ int main() {
   // LAMBDA: call void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp sections reduction(+:g)
     {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* %{{.+}})
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double,
 
     // Reduction list for runtime.
@@ -115,11 +113,11 @@ int main() {
   // BLOCKS: call void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp sections reduction(-:g)
     {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* %{{.+}})
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca double,
 
     // Reduction list for runtime.
@@ -192,13 +190,12 @@ int main() {
 // CHECK: define {{.*}}i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 6, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, float*, [[S_FLOAT_TY]]*, [[S_FLOAT_TY]]*, float*, [2 x i32]*, [2 x [[S_FLOAT_TY]]]*)* [[MAIN_MICROTASK:@.+]] to void
 // CHECK: = call {{.*}}i{{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call {{.*}} [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}},
 // CHECK-NOT: alloca float,
 // CHECK-NOT: alloca [[S_FLOAT_TY]],
 // CHECK-NOT: alloca [[S_FLOAT_TY]],
@@ -210,11 +207,6 @@ int main() {
 // CHECK: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_REF]]
 // CHECK: call i32 @__kmpc_single(
 
-// CHECK-DAG: getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK-DAG: getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK-DAG: getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 2
-// CHECK-DAG: getelementptr inbounds [[CAP_MAIN_TY]], [[CAP_MAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 3
-
 // CHECK-NOT: call {{.*}} [[S_FLOAT_TY_DESTR]]([[S_FLOAT_TY]]* [[VAR_PRIV]])
 // CHECK-NOT: call {{.*}} [[S_FLOAT_TY_DESTR]]([[S_FLOAT_TY]]*
 
@@ -228,12 +220,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_TMAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 6, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [[S_INT_TY]]*, [[S_INT_TY]]*, i32*, [2 x i32]*, [2 x [[S_INT_TY]]]*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call {{.*}} [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}},
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
 // CHECK: alloca i{{[0-9]+}},
@@ -249,23 +240,20 @@ int main() {
 
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_PTR_REF]],
+// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
+// CHECK: [[VAR1_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
+// CHECK: [[T_VAR1_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+
 // For + reduction operation initial value of private variable is 0.
 // CHECK: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[T_VAR_PRIV]],
 
-// CHECK: [[VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_PTR_REF:%.+]],
 // For & reduction operation initial value of private variable is ones in all bits.
 // CHECK: call {{.*}} [[S_INT_TY_CONSTR:@.+]]([[S_INT_TY]]* [[VAR_PRIV]])
 
-// CHECK: [[VAR1_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[VAR1_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_PTR_REF:%.+]],
 // For && reduction operation initial value of private variable is 1.0.
 // CHECK: call {{.*}} [[S_INT_TY_CONSTR:@.+]]([[S_INT_TY]]* [[VAR1_PRIV]])
 
-// CHECK: [[T_VAR1_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} {{[0-9]+}}
-// CHECK: [[T_VAR1_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR1_PTR_REF]],
 // For min reduction operation initial value of private variable is largest repesentable value.
 // CHECK: store i{{[0-9]+}} 2147483647, i{{[0-9]+}}* [[T_VAR1_PRIV]],
 
diff --git a/test/OpenMP/single_firstprivate_codegen.cpp b/test/OpenMP/single_firstprivate_codegen.cpp
index 01eaea3181..5ea3a967bd 100644
--- a/test/OpenMP/single_firstprivate_codegen.cpp
+++ b/test/OpenMP/single_firstprivate_codegen.cpp
@@ -30,7 +30,6 @@ struct S {
 // CHECK-DAG: [[S_FLOAT_TY:%.+]] = type { float }
 // CHECK-DAG: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
 // CHECK-DAG: [[ST_TY:%.+]] = type { i{{[0-9]+}}, i{{[0-9]+}} }
-// CHECK-DAG: [[CAP_TMAIN_TY:%.+]] = type { i{{[0-9]+}}*, [2 x i{{[0-9]+}}]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]* }
 
 template <typename T>
 T tmain() {
@@ -70,11 +69,11 @@ int main() {
   // LAMBDA: call void [[OUTER_LAMBDA:@.+]](
   [&]() {
 // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-// LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+// LAMBDA: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp single firstprivate(g)
   {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
     // LAMBDA: call i32 @__kmpc_single(
     // LAMBDA: [[G_VAL:%.+]] = load volatile i{{[0-9]+}}, i{{[0-9]+}}* [[G]]
@@ -104,11 +103,11 @@ int main() {
   // BLOCKS: call void {{%.+}}(i8
   ^{
 // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-// BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+// BLOCKS: call void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp single firstprivate(g)
    {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca i{{[0-9]+}},
     // BLOCKS: call i32 @__kmpc_single(
     // BLOCKS: [[G_VAL:%.+]] = load volatile i{{[0-9]+}}, i{{[0-9]+}}* [[G]]
@@ -192,36 +191,36 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 4, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, i32*, [2 x i32]*, [2 x [[S_INT_TY]]]*, [[S_INT_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call {{.*}} [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}}, i32* dereferenceable(4) %{{.+}}, [2 x i32]* dereferenceable(8) %{{.+}}, [2 x [[S_INT_TY]]]* dereferenceable(8) %{{.+}}, [[S_INT_TY]]* dereferenceable(4) %{{.+}})
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
 // CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
 // CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]],
 // CHECK: [[VAR_PRIV:%.+]] = alloca [[S_INT_TY]],
 // CHECK: store i{{[0-9]+}}* [[GTID_ADDR]], i{{[0-9]+}}** [[GTID_ADDR_ADDR:%.+]],
 
+// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** %
+// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** %
+// CHECK: [[S_ARR:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** %
+// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** %
+
 // CHECK: [[GTID_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[GTID_ADDR_ADDR]]
 // CHECK: [[GTID:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[GTID_REF]]
 // CHECK: call i32 @__kmpc_single(
+
 // firstprivate t_var(t_var)
-// CHECK: [[T_VAR_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0
-// CHECK: [[T_VAR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[T_VAR_PTR_REF]],
 // CHECK: [[T_VAR_VAL:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[T_VAR_REF]],
 // CHECK: store i{{[0-9]+}} [[T_VAR_VAL]], i{{[0-9]+}}* [[T_VAR_PRIV]],
 
 // firstprivate vec(vec)
-// CHECK: [[VEC_PTR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 1
-// CHECK: [[VEC_REF:%.+]] = load [2 x i{{[0-9]+}}]*, [2 x i{{[0-9]+}}]** [[VEC_PTR_REF:%.+]],
 // CHECK: [[VEC_DEST:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_PRIV]] to i8*
 // CHECK: [[VEC_SRC:%.+]] = bitcast [2 x i{{[0-9]+}}]* [[VEC_REF]] to i8*
 // CHECK: call void @llvm.memcpy.{{.+}}(i8* [[VEC_DEST]], i8* [[VEC_SRC]],
 
 // firstprivate s_arr(s_arr)
-// CHECK: [[S_ARR_REF:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 2
-// CHECK: [[S_ARR:%.+]] = load [2 x [[S_INT_TY]]]*, [2 x [[S_INT_TY]]]** [[S_ARR_REF]],
 // CHECK: [[S_ARR_PRIV_BEGIN:%.+]] = getelementptr inbounds [2 x [[S_INT_TY]]], [2 x [[S_INT_TY]]]* [[S_ARR_PRIV]], i{{[0-9]+}} 0, i{{[0-9]+}} 0
 // CHECK: [[S_ARR_PRIV_END:%.+]] = getelementptr [[S_INT_TY]], [[S_INT_TY]]* [[S_ARR_PRIV_BEGIN]], i{{[0-9]+}} 2
 // CHECK: [[IS_EMPTY:%.+]] = icmp eq [[S_INT_TY]]* [[S_ARR_PRIV_BEGIN]], [[S_ARR_PRIV_END]]
@@ -233,8 +232,6 @@ int main() {
 // CHECK: br i1 {{.+}}, label %{{.+}}, label %[[S_ARR_BODY]]
 
 // firstprivate var(var)
-// CHECK: [[VAR_REF_PTR:%.+]] = getelementptr inbounds [[CAP_TMAIN_TY]], [[CAP_TMAIN_TY]]* %{{.+}}, i{{[0-9]+}} 0, i{{[0-9]+}} 3
-// CHECK: [[VAR_REF:%.+]] = load [[S_INT_TY]]*, [[S_INT_TY]]** [[VAR_REF_PTR]],
 // CHECK: call {{.*}} [[ST_TY_DEFAULT_CONSTR]]([[ST_TY]]* [[ST_TY_TEMP:%.+]])
 // CHECK: call {{.*}} [[S_INT_TY_COPY_CONSTR]]([[S_INT_TY]]* [[VAR_PRIV]], [[S_INT_TY]]* {{.*}} [[VAR_REF]], [[ST_TY]]* [[ST_TY_TEMP]])
 // CHECK: call {{.*}} [[ST_TY_DESTR]]([[ST_TY]]* [[ST_TY_TEMP]])
diff --git a/test/OpenMP/single_private_codegen.cpp b/test/OpenMP/single_private_codegen.cpp
index e228b1b148..b08c11d2a2 100644
--- a/test/OpenMP/single_private_codegen.cpp
+++ b/test/OpenMP/single_private_codegen.cpp
@@ -20,9 +20,7 @@ struct S {
 volatile double g;
 
 // CHECK: [[S_FLOAT_TY:%.+]] = type { float }
-// CHECK: [[CAP_MAIN_TY:%.+]] = type { i8 }
 // CHECK: [[S_INT_TY:%.+]] = type { i{{[0-9]+}} }
-// CHECK: [[CAP_TMAIN_TY:%.+]] = type { i8 }
 template <typename T>
 T tmain() {
   S<T> test;
@@ -46,13 +44,12 @@ int main() {
   // LAMBDA: call{{.*}} void [[OUTER_LAMBDA:@.+]](
   [&]() {
   // LAMBDA: define{{.*}} internal{{.*}} void [[OUTER_LAMBDA]](
-  // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* %{{.+}})
+  // LAMBDA: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp single private(g)
   {
-    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // LAMBDA: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // LAMBDA: [[G_PRIVATE_ADDR:%.+]] = alloca double,
-    // LAMBDA: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     g = 1;
     // LAMBDA: call {{.*}}i32 @__kmpc_single(
     // LAMBDA: store double 1.0{{.+}}, double* [[G_PRIVATE_ADDR]],
@@ -78,13 +75,12 @@ int main() {
   // BLOCKS: call {{.*}}void {{%.+}}(i8
   ^{
   // BLOCKS: define{{.*}} internal{{.*}} void {{.+}}(i8*
-  // BLOCKS: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 1, {{.+}}* [[OMP_REGION:@.+]] to {{.+}}, i8* {{.+}})
+  // BLOCKS: call {{.*}}void {{.+}} @__kmpc_fork_call({{.+}}, i32 0, {{.+}}* [[OMP_REGION:@.+]] to {{.+}})
 #pragma omp parallel
 #pragma omp single private(g)
   {
-    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* %{{.+}}, i32* %{{.+}}, %{{.+}}* [[ARG:%.+]])
+    // BLOCKS: define{{.*}} internal{{.*}} void [[OMP_REGION]](i32* noalias %{{.+}}, i32* noalias %{{.+}})
     // BLOCKS: [[G_PRIVATE_ADDR:%.+]] = alloca double,
-    // BLOCKS: store %{{.+}}* [[ARG]], %{{.+}}** [[ARG_REF:%.+]],
     g = 1;
     // BLOCKS: call {{.*}}i32 @__kmpc_single(
     // BLOCKS: store double 1.0{{.+}}, double* [[G_PRIVATE_ADDR]],
@@ -123,13 +119,12 @@ int main() {
 // CHECK: define i{{[0-9]+}} @main()
 // CHECK: [[TEST:%.+]] = alloca [[S_FLOAT_TY]],
 // CHECK: call {{.*}} [[S_FLOAT_TY_DEF_CONSTR:@.+]]([[S_FLOAT_TY]]* [[TEST]])
-// CHECK: %{{.+}} = bitcast [[CAP_MAIN_TY]]*
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_MAIN_TY]]*)* [[MAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[MAIN_MICROTASK:@.+]] to void
 // CHECK: = call i{{.+}} [[TMAIN_INT:@.+]]()
 // CHECK: call void [[S_FLOAT_TY_DESTR:@.+]]([[S_FLOAT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_MAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[MAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
 // CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
 // CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_FLOAT_TY]]],
@@ -154,11 +149,11 @@ int main() {
 // CHECK: define {{.*}} i{{[0-9]+}} [[TMAIN_INT]]()
 // CHECK: [[TEST:%.+]] = alloca [[S_INT_TY]],
 // CHECK: call {{.*}} [[S_INT_TY_DEF_CONSTR:@.+]]([[S_INT_TY]]* [[TEST]])
-// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 1, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*, [[CAP_TMAIN_TY]]*)* [[TMAIN_MICROTASK:@.+]] to void
+// CHECK: call void (%{{.+}}*, i{{[0-9]+}}, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)*, ...) @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{[0-9]+}} 0, void (i{{[0-9]+}}*, i{{[0-9]+}}*, ...)* bitcast (void (i{{[0-9]+}}*, i{{[0-9]+}}*)* [[TMAIN_MICROTASK:@.+]] to void
 // CHECK: call void [[S_INT_TY_DESTR:@.+]]([[S_INT_TY]]*
 // CHECK: ret
 //
-// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* [[GTID_ADDR:%.+]], i{{[0-9]+}}* %{{.+}}, [[CAP_TMAIN_TY]]* %{{.+}})
+// CHECK: define internal void [[TMAIN_MICROTASK]](i{{[0-9]+}}* noalias [[GTID_ADDR:%.+]], i{{[0-9]+}}* noalias %{{.+}})
 // CHECK: [[T_VAR_PRIV:%.+]] = alloca i{{[0-9]+}},
 // CHECK: [[VEC_PRIV:%.+]] = alloca [2 x i{{[0-9]+}}],
 // CHECK: [[S_ARR_PRIV:%.+]] = alloca [2 x [[S_INT_TY]]],
diff --git a/test/OpenMP/task_codegen.cpp b/test/OpenMP/task_codegen.cpp
index 433578d681..cbbfc44794 100644
--- a/test/OpenMP/task_codegen.cpp
+++ b/test/OpenMP/task_codegen.cpp
@@ -230,23 +230,23 @@ int main() {
   }
   return a;
 }
-// CHECK: define internal i32 [[TASK_ENTRY1]](i32, [[KMP_TASK_T]]{{.*}}*)
+// CHECK: define internal i32 [[TASK_ENTRY1]](i32, [[KMP_TASK_T]]{{.*}}* noalias)
 // CHECK: store i32 15, i32* [[A_PTR:@.+]]
 // CHECK: [[A_VAL:%.+]] = load i32, i32* [[A_PTR]]
 // CHECK: [[A_VAL_I8:%.+]] = trunc i32 [[A_VAL]] to i8
 // CHECK: store i8 [[A_VAL_I8]], i8* %{{.+}}
 // CHECK: store i32 10, i32* %{{.+}}
 
-// CHECK: define internal i32 [[TASK_ENTRY2]](i32, [[KMP_TASK_T]]{{.*}}*)
+// CHECK: define internal i32 [[TASK_ENTRY2]](i32, [[KMP_TASK_T]]{{.*}}* noalias)
 // CHECK: store i32 1, i32* [[A_PTR:@.+]]
 
-// CHECK: define internal i32 [[TASK_ENTRY3]](i32, [[KMP_TASK_T]]{{.*}}*)
+// CHECK: define internal i32 [[TASK_ENTRY3]](i32, [[KMP_TASK_T]]{{.*}}* noalias)
 // CHECK: store i32 2, i32* [[A_PTR:@.+]]
 
-// CHECK: define internal i32 [[TASK_ENTRY4]](i32, [[KMP_TASK_T]]{{.*}}*)
+// CHECK: define internal i32 [[TASK_ENTRY4]](i32, [[KMP_TASK_T]]{{.*}}* noalias)
 // CHECK: store i32 3, i32* [[A_PTR:@.+]]
 
-// CHECK: define internal i32 [[TASK_ENTRY5]](i32, [[KMP_TASK_T]]{{.*}}*)
+// CHECK: define internal i32 [[TASK_ENTRY5]](i32, [[KMP_TASK_T]]{{.*}}* noalias)
 // CHECK: store i32 4, i32* [[A_PTR:@.+]]
 #endif
 
diff --git a/test/OpenMP/task_firstprivate_codegen.cpp b/test/OpenMP/task_firstprivate_codegen.cpp
index 89aadbeb6a..a527ce63c5 100644
--- a/test/OpenMP/task_firstprivate_codegen.cpp
+++ b/test/OpenMP/task_firstprivate_codegen.cpp
@@ -75,7 +75,7 @@ int main() {
     // LAMBDA: store double 2.0{{.+}}, double* [[G_REF]]
 
     // LAMBDA: store double* %{{.+}}, double** %{{.+}},
-    // LAMBDA: define internal i32 [[TASK_ENTRY]](i32, %{{.+}}*)
+    // LAMBDA: define internal i32 [[TASK_ENTRY]](i32, %{{.+}}* noalias)
     g = 1;
     // LAMBDA: store double 1.0{{.+}}, double* %{{.+}},
     // LAMBDA: call void [[INNER_LAMBDA]](%
@@ -110,7 +110,7 @@ int main() {
     // BLOCKS: ret
 
     // BLOCKS: store double* %{{.+}}, double** %{{.+}},
-    // BLOCKS: define internal i32 [[TASK_ENTRY]](i32, %{{.+}}*)
+    // BLOCKS: define internal i32 [[TASK_ENTRY]](i32, %{{.+}}* noalias)
     g = 1;
     // BLOCKS: store double 1.0{{.+}}, double* %{{.+}},
     // BLOCKS-NOT: [[G]]{{[[^:word:]]}}
@@ -241,7 +241,7 @@ int main() {
 // CHECK: store [2 x i32]* [[PRIV_VEC]], [2 x i32]** [[ARG4]],
 // CHECK: ret void
 
-// CHECK: define internal i32 [[TASK_ENTRY]](i32, [[KMP_TASK_MAIN_TY]]*)
+// CHECK: define internal i32 [[TASK_ENTRY]](i32, [[KMP_TASK_MAIN_TY]]* noalias)
 
 // CHECK: [[PRIV_VAR_ADDR:%.+]] = alloca [[S_DOUBLE_TY]]*,
 // CHECK: [[PRIV_T_VAR_ADDR:%.+]] = alloca i32*,
@@ -263,7 +263,7 @@ int main() {
 
 // CHECK: ret
 
-// CHECK: define internal i32 [[DESTRUCTORS]](i32, [[KMP_TASK_MAIN_TY]]*)
+// CHECK: define internal i32 [[DESTRUCTORS]](i32, [[KMP_TASK_MAIN_TY]]* noalias)
 // CHECK: [[PRIVATES:%.+]] = getelementptr inbounds [[KMP_TASK_MAIN_TY]], [[KMP_TASK_MAIN_TY]]* [[RES_KMP_TASK:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
 // CHECK: [[PRIVATE_VAR_REF:%.+]] = getelementptr inbounds [[PRIVATES_MAIN_TY]], [[PRIVATES_MAIN_TY]]* [[PRIVATES]], i{{.+}} 0, i{{.+}} 0
 // CHECK: [[PRIVATE_S_ARR_REF:%.+]] = getelementptr inbounds [[PRIVATES_MAIN_TY]], [[PRIVATES_MAIN_TY]]* [[PRIVATES]], i{{.+}} 0, i{{.+}} 1
@@ -376,7 +376,7 @@ int main() {
 // CHECK: store [[S_INT_TY]]* [[PRIV_VAR]], [[S_INT_TY]]** [[ARG4]],
 // CHECK: ret void
 
-// CHECK: define internal i32 [[TASK_ENTRY]](i32, [[KMP_TASK_TMAIN_TY]]*)
+// CHECK: define internal i32 [[TASK_ENTRY]](i32, [[KMP_TASK_TMAIN_TY]]* noalias)
 
 // CHECK: [[PRIV_T_VAR_ADDR:%.+]] = alloca i32*,
 // CHECK: [[PRIV_VEC_ADDR:%.+]] = alloca [2 x i32]*,
@@ -398,7 +398,7 @@ int main() {
 
 // CHECK: ret
 
-// CHECK: define internal i32 [[DESTRUCTORS]](i32, [[KMP_TASK_TMAIN_TY]]*)
+// CHECK: define internal i32 [[DESTRUCTORS]](i32, [[KMP_TASK_TMAIN_TY]]* noalias)
 // CHECK: [[PRIVATES:%.+]] = getelementptr inbounds [[KMP_TASK_TMAIN_TY]], [[KMP_TASK_TMAIN_TY]]* [[RES_KMP_TASK:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
 // CHECK: [[PRIVATE_S_ARR_REF:%.+]] = getelementptr inbounds [[PRIVATES_TMAIN_TY]], [[PRIVATES_TMAIN_TY]]* [[PRIVATES]], i{{.+}} 0, i{{.+}} 2
 // CHECK: [[PRIVATE_VAR_REF:%.+]] = getelementptr inbounds [[PRIVATES_TMAIN_TY]], [[PRIVATES_TMAIN_TY]]* [[PRIVATES]], i{{.+}} 0, i{{.+}} 3
diff --git a/test/OpenMP/task_if_codegen.cpp b/test/OpenMP/task_if_codegen.cpp
index cbdf117a05..5992be02a8 100644
--- a/test/OpenMP/task_if_codegen.cpp
+++ b/test/OpenMP/task_if_codegen.cpp
@@ -21,14 +21,14 @@ int Arg;
 
 // CHECK-LABEL: define void @{{.+}}gtid_test
 void gtid_test() {
-// CHECK:  call void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 1, {{.+}}* [[GTID_TEST_REGION1:@.+]] to void
+// CHECK:  call void {{.+}} @__kmpc_fork_call(%{{.+}}* @{{.+}}, i{{.+}} 0, {{.+}}* [[GTID_TEST_REGION1:@.+]] to void
 #pragma omp parallel
 #pragma omp task if (task: false)
   gtid_test();
 // CHECK: ret void
 }
 
-// CHECK: define internal void [[GTID_TEST_REGION1]](i32* [[GTID_PARAM:%.+]], i
+// CHECK: define internal void [[GTID_TEST_REGION1]](i32* noalias [[GTID_PARAM:%.+]], i
 // CHECK: store i32* [[GTID_PARAM]], i32** [[GTID_ADDR_REF:%.+]],
 // CHECK: [[GTID_ADDR:%.+]] = load i32*, i32** [[GTID_ADDR_REF]]
 // CHECK: [[GTID:%.+]] = load i32, i32* [[GTID_ADDR]]
diff --git a/test/OpenMP/task_private_codegen.cpp b/test/OpenMP/task_private_codegen.cpp
index b29d0d39a1..63ca915d7d 100644
--- a/test/OpenMP/task_private_codegen.cpp
+++ b/test/OpenMP/task_private_codegen.cpp
@@ -69,7 +69,7 @@ int main() {
     // LAMBDA: [[G_REF:%.+]] = load double*, double** [[G_PTR_REF]]
     // LAMBDA: store double 2.0{{.+}}, double* [[G_REF]]
 
-    // LAMBDA: define internal i32 [[TASK_ENTRY]](i32, %{{.+}}*)
+    // LAMBDA: define internal i32 [[TASK_ENTRY]](i32, %{{.+}}* noalias)
     g = 1;
     // LAMBDA: store double 1.0{{.+}}, double* %{{.+}},
     // LAMBDA: call void [[INNER_LAMBDA]](%
@@ -99,7 +99,7 @@ int main() {
     // BLOCKS-NOT: [[G]]{{[[^:word:]]}}
     // BLOCKS: ret
 
-    // BLOCKS: define internal i32 [[TASK_ENTRY]](i32, %{{.+}}*)
+    // BLOCKS: define internal i32 [[TASK_ENTRY]](i32, %{{.+}}* noalias)
     g = 1;
     // BLOCKS: store double 1.0{{.+}}, double* %{{.+}},
     // BLOCKS-NOT: [[G]]{{[[^:word:]]}}
@@ -199,7 +199,7 @@ int main() {
 // CHECK: store [2 x i32]* [[PRIV_VEC]], [2 x i32]** [[ARG4]],
 // CHECK: ret void
 
-// CHECK: define internal i32 [[TASK_ENTRY]](i32, [[KMP_TASK_MAIN_TY]]*)
+// CHECK: define internal i32 [[TASK_ENTRY]](i32, [[KMP_TASK_MAIN_TY]]* noalias)
 
 // CHECK: [[PRIV_VAR_ADDR:%.+]] = alloca [[S_DOUBLE_TY]]*,
 // CHECK: [[PRIV_T_VAR_ADDR:%.+]] = alloca i32*,
@@ -221,7 +221,7 @@ int main() {
 
 // CHECK: ret
 
-// CHECK: define internal i32 [[DESTRUCTORS]](i32, [[KMP_TASK_MAIN_TY]]*)
+// CHECK: define internal i32 [[DESTRUCTORS]](i32, [[KMP_TASK_MAIN_TY]]* noalias)
 // CHECK: [[PRIVATES:%.+]] = getelementptr inbounds [[KMP_TASK_MAIN_TY]], [[KMP_TASK_MAIN_TY]]* [[RES_KMP_TASK:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
 // CHECK: [[PRIVATE_VAR_REF:%.+]] = getelementptr inbounds [[PRIVATES_MAIN_TY]], [[PRIVATES_MAIN_TY]]* [[PRIVATES]], i{{.+}} 0, i{{.+}} 0
 // CHECK: [[PRIVATE_S_ARR_REF:%.+]] = getelementptr inbounds [[PRIVATES_MAIN_TY]], [[PRIVATES_MAIN_TY]]* [[PRIVATES]], i{{.+}} 0, i{{.+}} 1
@@ -306,7 +306,7 @@ int main() {
 // CHECK: store [[S_INT_TY]]* [[PRIV_VAR]], [[S_INT_TY]]** [[ARG4]],
 // CHECK: ret void
 
-// CHECK: define internal i32 [[TASK_ENTRY]](i32, [[KMP_TASK_TMAIN_TY]]*)
+// CHECK: define internal i32 [[TASK_ENTRY]](i32, [[KMP_TASK_TMAIN_TY]]* noalias)
 
 // CHECK: [[PRIV_T_VAR_ADDR:%.+]] = alloca i32*,
 // CHECK: [[PRIV_VEC_ADDR:%.+]] = alloca [2 x i32]*,
@@ -328,7 +328,7 @@ int main() {
 
 // CHECK: ret
 
-// CHECK: define internal i32 [[DESTRUCTORS]](i32, [[KMP_TASK_TMAIN_TY]]*)
+// CHECK: define internal i32 [[DESTRUCTORS]](i32, [[KMP_TASK_TMAIN_TY]]* noalias)
 // CHECK: [[PRIVATES:%.+]] = getelementptr inbounds [[KMP_TASK_TMAIN_TY]], [[KMP_TASK_TMAIN_TY]]* [[RES_KMP_TASK:%.+]], i{{[0-9]+}} 0, i{{[0-9]+}} 1
 // CHECK: [[PRIVATE_S_ARR_REF:%.+]] = getelementptr inbounds [[PRIVATES_TMAIN_TY]], [[PRIVATES_TMAIN_TY]]* [[PRIVATES]], i{{.+}} 0, i{{.+}} 2
 // CHECK: [[PRIVATE_VAR_REF:%.+]] = getelementptr inbounds [[PRIVATES_TMAIN_TY]], [[PRIVATES_TMAIN_TY]]* [[PRIVATES]], i{{.+}} 0, i{{.+}} 3