diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2023-05-05 13:16:58 -0400 |
|---|---|---|
| committer | Matt Arsenault <arsenm2@gmail.com> | 2023-05-18 08:34:46 +0100 |
| commit | 8f3e64624c2e49b61ee578aec493260a59a35e80 (patch) | |
| tree | 74129de67a8cc685850164dbbd979ae041f10083 /clang/lib/Sema/SemaModule.cpp | |
| parent | b233eb70cd82ca3c320fac4bb8c2cccd1fe97696 (diff) | |
| download | llvm-8f3e64624c2e49b61ee578aec493260a59a35e80.zip llvm-8f3e64624c2e49b61ee578aec493260a59a35e80.tar.gz llvm-8f3e64624c2e49b61ee578aec493260a59a35e80.tar.bz2 | |
AMDGPU: Fold fmed3 of fpext sources to f16 fmed3
InstCombine already does this for minnum/maxnum. If we
also apply this to fmed3, we don't need to explicitly
use 16-bit fmed3 if we're not sure the target
supports 16-bit instructions yet.
Diffstat (limited to 'clang/lib/Sema/SemaModule.cpp')
0 files changed, 0 insertions, 0 deletions