diff options
| author | Brox Chen <guochen2@amd.com> | 2025-11-11 01:13:33 -0500 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-11-11 01:13:33 -0500 |
| commit | 356b0ddc6cb02c405632d40433b2706eccaf2410 (patch) | |
| tree | 79a2946bd9045e1357192de34e99341c60371e99 /clang/lib/FrontendTool/ExecuteCompilerInvocation.cpp | |
| parent | 8475a66d4376344d2f3b330f351fe651eaf558dd (diff) | |
| download | llvm-356b0ddc6cb02c405632d40433b2706eccaf2410.zip llvm-356b0ddc6cb02c405632d40433b2706eccaf2410.tar.gz llvm-356b0ddc6cb02c405632d40433b2706eccaf2410.tar.bz2 | |
[AMDGPU][CodeGen] enable D16Writes32BitVgpr for gfx12 (#165587)
The issue in https://github.com/llvm/llvm-project/pull/157795 also
reproducible for gfx12.
This should only be used in true16 mode. Although gfx12 not yet has
true16 mode enable, there could be downstream branch to test true16 mode
manually thus enable this workaround now
Diffstat (limited to 'clang/lib/FrontendTool/ExecuteCompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions