diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2016-07-26 23:25:44 +0000 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2016-07-26 23:25:44 +0000 |
| commit | e3862cdc933da4864598aa8616a33b076d99077c (patch) | |
| tree | e394808494c9b8404e7afbd6b74d608ee91db503 /llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp | |
| parent | 685e8ff95305b35c7be889d4a0f82d2ca567779c (diff) | |
| download | bcm5719-llvm-e3862cdc933da4864598aa8616a33b076d99077c.tar.gz bcm5719-llvm-e3862cdc933da4864598aa8616a33b076d99077c.zip | |
AMDGPU: Use rcp for fdiv 1, x with fpmath metadata
Using rcp should be OK for safe math usually, so this
should not be replacing the original fdiv.
llvm-svn: 276823
Diffstat (limited to 'llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp')
| -rw-r--r-- | llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp b/llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp index 0627708485c..b955e231699 100644 --- a/llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp +++ b/llvm/lib/Target/AMDGPU/AMDGPUCodeGenPrepare.cpp @@ -76,7 +76,7 @@ static bool shouldKeepFDivF32(Value *Num, bool UnsafeDiv) { return false; // Reciprocal f32 is handled separately without denormals. - return UnsafeDiv && CNum->isExactlyValue(+1.0); + return UnsafeDiv || CNum->isExactlyValue(+1.0); } // Insert an intrinsic for fast fdiv for safe math situations where we can |
