[llvm] [AMDGPU] Max. WG size-induced occupancy limits max. waves/EU (PR #137807)

Lucas Ramirez via llvm-commits llvm-commits at lists.llvm.org
Tue Apr 29 06:30:16 PDT 2025


================
@@ -42,4 +42,4 @@ bb2:
 declare i32 @llvm.amdgcn.workitem.id.x() #0
 
 attributes #0 = { nounwind readnone }
-attributes #1 = { "amdgpu-num-vgpr"="9" "amdgpu-flat-work-group-size"="1024,1024" }
+attributes #1 = { "amdgpu-num-vgpr"="9" }
----------------
lucas-rami wrote:

The load/store interleaving pattern disappears if I leave the attribute (which implies a max. waves/EU of 8, whereas it was 10 before the change or with the default flat workgroup size range).

https://github.com/llvm/llvm-project/pull/137807


More information about the llvm-commits mailing list