We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 01fede0 commit d324fd8Copy full SHA for d324fd8
attentionbench/attention_utils.py
@@ -93,7 +93,7 @@ def get_translation_info(self) -> str:
93
llvm_func_attrs += [f'"denormal-fp-math-f32" = "preserve-sign"']
94
return (
95
f"#iree_codegen.translation_info<"
96
- + f"LLVMGPUVectorDistribute"
+ + f"pipeline = LLVMGPUVectorDistribute"
97
+ f" workgroup_size = [{self.N_warp * self.M_warp * 64}]"
98
+ f" subgroup_size = 64"
99
+ f" ,{{mma_schedule = {self.get_mma_schedule()}"
0 commit comments