| Issue |
138766
|
| Summary |
[LLVM][AMDGPU] PR#137930 introduces correctness inconsistency
|
| Labels |
backend:AMDGPU,
llvm:codegen
|
| Assignees |
frederik-h
|
| Reporter |
lialan
|
PR: https://github.com/llvm/llvm-project/pull/137930 shows a correctness issue which is reported here: https://github.com/iree-org/iree/issues/20737
Supplying: 1. source LLVM IR; 2. assembly files with and without #137930:
https://gist.github.com/lialan/0cf658c174d02edef6aebc9c4d248c2e
LLVM IR file is small so I did not try to get a minimal reproducer. The diff of before(left)/after(right) assembly is relatively scoped.
<img width="1342" alt="Image" src="" />
Seems relevant change is to enable `V_CNDMASK_B32_E32` to be converted to SDWA: https://github.com/llvm/llvm-project/pull/137930/files#diff-dc487be6b0b90778163b61e49a83c61afc0dc647a5582f282e610f3344419180L1111-L1114
![]()
_______________________________________________
llvm-bugs mailing list
llvm-bugs@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
