[PATCH] D155050: [AMDGPU] Wave32 CodeGen for amdgcn.ballot.i64

Thu Jul 13 01:28:38 PDT 2023

foad added inline comments.

================
Comment at: llvm/test/CodeGen/AMDGPU/llvm.amdgcn.ballot.i64.wave32.ll:2
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -march=amdgcn -mcpu=gfx1010 -mattr=+wavefrontsize32,-wavefrontsize64 < %s | FileCheck %s
+; RUN: llc -march=amdgcn -mcpu=gfx1100 -amdgpu-enable-delay-alu=0 -mattr=+wavefrontsize32,-wavefrontsize64 < %s | FileCheck %s
----------------
Pierre-vh wrote:
> arsenm wrote:
> > with and without global-isel. Also no reason to spell out both features, just rely on wave32 being the default. Also add a run line with wave32 on a wave64 target if it doesn't fail too horribly
> > Also add a run line with wave32 on a wave64 target if it doesn't fail too horribly
> What target can I use? I think <GFX10 is wave64 default but do they support wave32 mode? There's a comment in `AMDGPURemoveIncompatibleFunction` that says gfx9 and below don't support wave32 mode
Wave32 did not exist until GFX10.

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D155050/new/

https://reviews.llvm.org/D155050