diff options
author | Eli Bendersky <eliben@google.com> | 2014-03-18 23:51:07 +0000 |
---|---|---|
committer | Eli Bendersky <eliben@google.com> | 2014-03-18 23:51:07 +0000 |
commit | 21354ec60d0dde4f1995d816701b895c57f806bd (patch) | |
tree | c00e675912eb5ddca01b813594beb3b9d03dd719 /test | |
parent | cae25dcbf7347d1a04f8746aedd6d6600b528a40 (diff) | |
download | llvm-21354ec60d0dde4f1995d816701b895c57f806bd.tar.gz llvm-21354ec60d0dde4f1995d816701b895c57f806bd.tar.bz2 llvm-21354ec60d0dde4f1995d816701b895c57f806bd.tar.xz |
Expose "noduplicate" attribute as a property for intrinsics.
The "noduplicate" function attribute exists to prevent certain optimizations
from duplicating calls to the function. This is important on platforms where
certain function call duplications are unsafe (for example execution barriers
for CUDA and OpenCL).
This patch makes it possible to specify intrinsics as "noduplicate" and
translates that to the appropriate function attribute.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@204200 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'test')
-rw-r--r-- | test/CodeGen/NVPTX/noduplicate-syncthreads.ll | 74 | ||||
-rw-r--r-- | test/Feature/intrinsic-noduplicate.ll | 9 |
2 files changed, 83 insertions, 0 deletions
diff --git a/test/CodeGen/NVPTX/noduplicate-syncthreads.ll b/test/CodeGen/NVPTX/noduplicate-syncthreads.ll new file mode 100644 index 0000000000..64745fcba3 --- /dev/null +++ b/test/CodeGen/NVPTX/noduplicate-syncthreads.ll @@ -0,0 +1,74 @@ +; RUN: opt < %s -O3 -S | FileCheck %s + +; Make sure the call to syncthreads is not duplicate here by the LLVM +; optimizations, because it has the noduplicate attribute set. + +; CHECK: call void @llvm.cuda.syncthreads +; CHECK-NOT: call void @llvm.cuda.syncthreads + +; Function Attrs: nounwind +define void @foo(float* %output) #1 { +entry: + %output.addr = alloca float*, align 8 + store float* %output, float** %output.addr, align 8 + %0 = load float** %output.addr, align 8 + %arrayidx = getelementptr inbounds float* %0, i64 0 + %1 = load float* %arrayidx, align 4 + %conv = fpext float %1 to double + %cmp = fcmp olt double %conv, 1.000000e+01 + br i1 %cmp, label %if.then, label %if.else + +if.then: ; preds = %entry + %2 = load float** %output.addr, align 8 + %3 = load float* %2, align 4 + %conv1 = fpext float %3 to double + %add = fadd double %conv1, 1.000000e+00 + %conv2 = fptrunc double %add to float + store float %conv2, float* %2, align 4 + br label %if.end + +if.else: ; preds = %entry + %4 = load float** %output.addr, align 8 + %5 = load float* %4, align 4 + %conv3 = fpext float %5 to double + %add4 = fadd double %conv3, 2.000000e+00 + %conv5 = fptrunc double %add4 to float + store float %conv5, float* %4, align 4 + br label %if.end + +if.end: ; preds = %if.else, %if.then + call void @llvm.cuda.syncthreads() + %6 = load float** %output.addr, align 8 + %arrayidx6 = getelementptr inbounds float* %6, i64 0 + %7 = load float* %arrayidx6, align 4 + %conv7 = fpext float %7 to double + %cmp8 = fcmp olt double %conv7, 1.000000e+01 + br i1 %cmp8, label %if.then9, label %if.else13 + +if.then9: ; preds = %if.end + %8 = load float** %output.addr, align 8 + %9 = load float* %8, align 4 + %conv10 = fpext float %9 to double + %add11 = fadd double %conv10, 3.000000e+00 + %conv12 = fptrunc double %add11 to float + store float %conv12, float* %8, align 4 + br label %if.end17 + +if.else13: ; preds = %if.end + %10 = load float** %output.addr, align 8 + %11 = load float* %10, align 4 + %conv14 = fpext float %11 to double + %add15 = fadd double %conv14, 4.000000e+00 + %conv16 = fptrunc double %add15 to float + store float %conv16, float* %10, align 4 + br label %if.end17 + +if.end17: ; preds = %if.else13, %if.then9 + ret void +} + +; Function Attrs: noduplicate nounwind +declare void @llvm.cuda.syncthreads() #2 + +!0 = metadata !{void (float*)* @foo, metadata !"kernel", i32 1} +!1 = metadata !{null, metadata !"align", i32 8} diff --git a/test/Feature/intrinsic-noduplicate.ll b/test/Feature/intrinsic-noduplicate.ll new file mode 100644 index 0000000000..9a2b0aba5b --- /dev/null +++ b/test/Feature/intrinsic-noduplicate.ll @@ -0,0 +1,9 @@ +; RUN: llvm-as < %s | llvm-dis | FileCheck %s + +; Make sure LLVM knows about the noduplicate attribute on the +; llvm.cuda.syncthreads intrinsic. + +declare void @llvm.cuda.syncthreads() + +; CHECK: declare void @llvm.cuda.syncthreads() #[[ATTRNUM:[0-9]+]] +; CHECK: attributes #[[ATTRNUM]] = { noduplicate nounwind } |