[OpenCL] Emit enqueued block as kernel

In OpenCL the kernel function and non-kernel function has different calling conventions.
For certain targets they have different argument ABIs. Also kernels have special function
attributes and metadata for runtime to launch them.

The blocks passed to enqueue_kernel is supposed to be executed as kernels. As such,
the block invoke function should be emitted as kernel with proper calling convention and
argument ABI.

This patch emits enqueued block as kernel. If a block is both called directly and passed
to enqueue_kernel, separate functions will be generated.

Differential Revision: https://reviews.llvm.org/D38134


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@315804 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 36 insertions, 0 deletions

diff --git a/lib/CodeGen/CGOpenCLRuntime.cpp b/lib/CodeGen/CGOpenCLRuntime.cpp
index 49eab45ba8..4d637615d2 100644
--- a/lib/CodeGen/CGOpenCLRuntime.cpp
+++ b/lib/CodeGen/CGOpenCLRuntime.cpp
@@ -16,6 +16,7 @@
 #include "CGOpenCLRuntime.h"
 #include "CodeGenFunction.h"
 #include "TargetInfo.h"
+#include "clang/CodeGen/ConstantInitBuilder.h"
 #include "llvm/IR/DerivedTypes.h"
 #include "llvm/IR/GlobalValue.h"
 #include <assert.h>
@@ -110,3 +111,38 @@ llvm::PointerType *CGOpenCLRuntime::getGenericVoidPointerType() {
       CGM.getLLVMContext(),
       CGM.getContext().getTargetAddressSpace(LangAS::opencl_generic));
 }
+
+CGOpenCLRuntime::EnqueuedBlockInfo
+CGOpenCLRuntime::emitOpenCLEnqueuedBlock(CodeGenFunction &CGF, const Expr *E) {
+  // The block literal may be assigned to a const variable. Chasing down
+  // to get the block literal.
+  if (auto DR = dyn_cast<DeclRefExpr>(E)) {
+    E = cast<VarDecl>(DR->getDecl())->getInit();
+  }
+  if (auto Cast = dyn_cast<CastExpr>(E)) {
+    E = Cast->getSubExpr();
+  }
+  auto *Block = cast<BlockExpr>(E);
+
+  // The same block literal may be enqueued multiple times. Cache it if
+  // possible.
+  auto Loc = EnqueuedBlockMap.find(Block);
+  if (Loc != EnqueuedBlockMap.end()) {
+    return Loc->second;
+  }
+
+  // Emit block literal as a common block expression and get the block invoke
+  // function.
+  llvm::Function *Invoke;
+  auto *V = CGF.EmitBlockLiteral(cast<BlockExpr>(Block), &Invoke);
+  auto *F = CGF.getTargetHooks().createEnqueuedBlockKernel(
+      CGF, Invoke, V->stripPointerCasts());
+
+  // The common part of the post-processing of the kernel goes here.
+  F->addFnAttr(llvm::Attribute::NoUnwind);
+  F->setCallingConv(
+      CGF.getTypes().ClangCallConvToLLVMCallConv(CallingConv::CC_OpenCLKernel));
+  EnqueuedBlockInfo Info{F, V};
+  EnqueuedBlockMap[Block] = Info;
+  return Info;
+}