diff options
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | 2020-05-16 13:48:55 -0400 |
|---|---|---|
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | 2020-05-18 15:34:33 -0400 |
| commit | b27a538dda4caf2752ed8c6c731361ddf9458b87 (patch) | |
| tree | 5a080d5e4f3f523230574483e58b5049dc6204d0 /clang/lib/Frontend/CompilerInvocation.cpp | |
| parent | ef649e8fd5d1748764a9afca3ce0b80113a6a239 (diff) | |
| download | llvm-b27a538dda4caf2752ed8c6c731361ddf9458b87.zip llvm-b27a538dda4caf2752ed8c6c731361ddf9458b87.tar.gz llvm-b27a538dda4caf2752ed8c6c731361ddf9458b87.tar.bz2 | |
AMDGPU: Fix illegally constant folding from V_MOV_B32_sdwa
This was assumed to be a simple move, and interpreting the immediate
modifier operand as a materialized immediate. Apparently the SDWA pass
never produces these, but GlobalISel does emit these for some vector
shuffles.
Diffstat (limited to 'clang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions