[Mlir-commits] [mlir] [mlir][gpu] Add 'cluster_size' attribute to gpu.subgroup_reduce (PR #104851)

[Jakub Kuderski]

================
@@ -1208,11 +1208,21 @@ def GPU_SubgroupReduceOp : GPU_Op<"subgroup_reduce", [SameOperandsAndResultType]
     ```mlir
     %1 = gpu.subgroup_reduce add %a : (f32) -> (f32)
     %2 = gpu.subgroup_reduce add %b : (vector<4xf16>) -> (vector<4xf16>)
+    %3 = gpu.subgroup_reduce add %c cluster_size(4) : (f32) -> (f32)
----------------
kuhar wrote:

nit: we don't need the parens in the result type in any of these examples

https://github.com/llvm/llvm-project/pull/104851