diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2024-06-27 21:17:16 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-06-27 21:17:16 +0200 |
| commit | a2a73d892a3fd16a1dd2218cf142c7e99cc69ba6 (patch) | |
| tree | f19d0c9aaeaea6eb3e22278837722dddac50fc21 /llvm/lib/CodeGen/TargetLoweringObjectFileImpl.cpp | |
| parent | 918313d17d38d8723d5d81fef008538677abf9cc (diff) | |
| download | llvm-a2a73d892a3fd16a1dd2218cf142c7e99cc69ba6.zip llvm-a2a73d892a3fd16a1dd2218cf142c7e99cc69ba6.tar.gz llvm-a2a73d892a3fd16a1dd2218cf142c7e99cc69ba6.tar.bz2 | |
AMDGPU: Fix no return atomicrmw fadd v2f16 selection for gfx908 (#96948)
We previously would always expand this with a cmpxchg loop, while
it should be the same conditions as the f32 case (except for the
denormal concern).
Diffstat (limited to 'llvm/lib/CodeGen/TargetLoweringObjectFileImpl.cpp')
0 files changed, 0 insertions, 0 deletions