[Mlir-commits] [mlir] [MLIR][ROCDL] Add conversion for gpu.subgroup_id to ROCDL (PR #136405)

[Alan Li]

================
@@ -204,6 +204,15 @@ def ROCDL_ReadlaneOp : ROCDL_IntrOp<"readlane", [], [0], [AllTypesMatch<["res",
    }];
 }
 
+// The LLVM intrinsic function name is rather mouthful,
+// so here we opt to use a shorter rocdl name.
+def ROCDL_WaveIdOp :  LLVM_IntrOpBase<ROCDL_Dialect, "wave_id",
----------------
lialan wrote:

I was not sure about this. If generally ROCDL ops should map exactly to LLVM intrinsics, then I am okay with using the long name here. I will make the change.

https://github.com/llvm/llvm-project/pull/136405