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

[Alan Li]

================
@@ -608,6 +608,10 @@ def ConvertGpuOpsToROCDLOps : Pass<"convert-gpu-to-rocdl", "gpu::GPUModuleOp"> {
                clEnumValN(::mlir::gpu::amd::Runtime::HIP, "HIP", "HIP"),
                clEnumValN(::mlir::gpu::amd::Runtime::OpenCL, "OpenCL",
                           "OpenCL"))}]>,
+    Option<"subgroupSize", "subgroup-size", "unsigned",
+           "0",
+           "specify subgroup size for the kernel, if left empty, the default "
+           "value will be decided by the target chipset.">,
----------------
lialan wrote:

@krzysz00 there is the TODO https://github.com/llvm/llvm-project/pull/136405/files#diff-cd4257dddc1cb3043071e5c7641774615ffd685cc779acf70a47a3e83401b515R67. 

Currently wavefrontsize is not exposed to rocdl yet, I prefer to do it after this is merged. Think this is a good idea?


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