diff options
| author | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | 2017-08-28 18:00:08 +0000 |
|---|---|---|
| committer | Stanislav Mekhanoshin <Stanislav.Mekhanoshin@amd.com> | 2017-08-28 18:00:08 +0000 |
| commit | 312c557b3b1d989faaf4437107cb8487bca6d515 (patch) | |
| tree | c3e18c030d115668c64c61468edd2334456cfcd1 /llvm/lib/Transforms/InstCombine/InstCombineSimplifyDemanded.cpp | |
| parent | ea2683ecb8a6e45da255af1ff84f7e25478ee9b5 (diff) | |
| download | bcm5719-llvm-312c557b3b1d989faaf4437107cb8487bca6d515.tar.gz bcm5719-llvm-312c557b3b1d989faaf4437107cb8487bca6d515.zip | |
[AMDGPU] Fix regression in AMDGPULibCalls allowing native for doubles
Under -cl-fast-relaxed-math we could use native_sqrt, but f64 was
allowed to produce HSAIL's nsqrt instruction. HSAIL is not here
and we stick with non-existing native_sqrt(double) as a result.
Add check for f64 to not return native functions and also remove
handling of f64 case for fold_sqrt.
Differential Revision: https://reviews.llvm.org/D37223
llvm-svn: 311900
Diffstat (limited to 'llvm/lib/Transforms/InstCombine/InstCombineSimplifyDemanded.cpp')
0 files changed, 0 insertions, 0 deletions
