diff options
| author | Krzysztof Drewniak <Krzysztof.Drewniak@amd.com> | 2024-09-11 17:51:07 -0500 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-09-11 17:51:07 -0500 |
| commit | aa60a3e4d0664dedc8ae0ea005459186fdc1aab9 (patch) | |
| tree | 5cafe08afb11db81ba12988356950f3b2ecbcabc /clang/lib/Basic/FileManager.cpp | |
| parent | 7910812414108ed9085548e2704f3ad5c018e970 (diff) | |
| download | llvm-aa60a3e4d0664dedc8ae0ea005459186fdc1aab9.zip llvm-aa60a3e4d0664dedc8ae0ea005459186fdc1aab9.tar.gz llvm-aa60a3e4d0664dedc8ae0ea005459186fdc1aab9.tar.bz2 | |
[mlir][AMDGPU] Support vector<2xf16> inputs to buffer atomic fadd (#108286)
Extend the lowering of atomic.fadd to support the v2f16 variant
avaliable on some AMDGPU chips.
Re-lands #108238 (and addresses review comments from there)
Co-authored-by: Giuseppe Rossini <giuseppe.rossini@amd.com>
Diffstat (limited to 'clang/lib/Basic/FileManager.cpp')
0 files changed, 0 insertions, 0 deletions