diff options
| author | Nadav Rotem <nrotem@apple.com> | 2013-09-21 00:27:05 +0000 |
|---|---|---|
| committer | Nadav Rotem <nrotem@apple.com> | 2013-09-21 00:27:05 +0000 |
| commit | 3371172a6761f7e12d0dc80b631dabfd70e0fecf (patch) | |
| tree | 3af688fdd9058d022253ce0ba5ccab0475d77c8a /llvm/test/Transforms/LoopVectorize | |
| parent | f1dfbfdde1245b1f045bf57e3090782fd37db475 (diff) | |
| download | bcm5719-llvm-3371172a6761f7e12d0dc80b631dabfd70e0fecf.tar.gz bcm5719-llvm-3371172a6761f7e12d0dc80b631dabfd70e0fecf.zip | |
LoopVectorizer: Only allow vectorization of intrinsics. We can't know for sure that the functions 'abs' or 'round' are the functions from libm.
rdar://15012650
llvm-svn: 191122
Diffstat (limited to 'llvm/test/Transforms/LoopVectorize')
| -rw-r--r-- | llvm/test/Transforms/LoopVectorize/intrinsic.ll | 31 |
1 files changed, 28 insertions, 3 deletions
diff --git a/llvm/test/Transforms/LoopVectorize/intrinsic.ll b/llvm/test/Transforms/LoopVectorize/intrinsic.ll index 216b937ad6d..99d6646e673 100644 --- a/llvm/test/Transforms/LoopVectorize/intrinsic.ll +++ b/llvm/test/Transforms/LoopVectorize/intrinsic.ll @@ -1018,7 +1018,7 @@ for.body: ; preds = %entry, %for.body %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ] %arrayidx = getelementptr inbounds float* %x, i64 %indvars.iv %0 = load float* %arrayidx, align 4 - %call = tail call float @fabsf(float %0) nounwind readnone + %call = tail call float @llvm.fabs.f32(float %0) nounwind readnone store float %call, float* %arrayidx, align 4 %indvars.iv.next = add i64 %indvars.iv, 1 %lftr.wideiv = trunc i64 %indvars.iv.next to i32 @@ -1029,6 +1029,31 @@ for.end: ; preds = %for.body ret void } -declare float @fabsf(float) nounwind readnone - declare double @llvm.pow.f64(double, double) nounwind readnone + + +;CHECK: @not_intrin +;CHECK: @round +;CHECK-NOT: @round +;CHECK: ret +define void @not_intrin(i32* nocapture %A) nounwind ssp uwtable { + br label %1 + +; <label>:1 ; preds = %1, %0 + %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ] + %2 = getelementptr inbounds i32* %A, i64 %indvars.iv + %3 = load i32* %2, align 4 + %4 = add nsw i32 %3, 3 + store i32 %4, i32* %2, align 4 + %5 = trunc i64 %indvars.iv to i32 + tail call void @round(i32 %5) nounwind + %indvars.iv.next = add i64 %indvars.iv, 1 + %lftr.wideiv = trunc i64 %indvars.iv.next to i32 + %exitcond = icmp eq i32 %lftr.wideiv, 256 + br i1 %exitcond, label %6, label %1 + +; <label>:6 ; preds = %1 + ret void +} + +declare void @round(i32) |
