[Mlir-commits] [mlir] [mlir] Implement indexed access op interfaces for memref, vector, gpu, nvgpu (PR #177014)

[Kunwar Grover]

================
@@ -1863,6 +1866,8 @@ def GPU_SubgroupMmaLoadMatrixOp : GPU_Op<"subgroup_mma_load_matrix",
      %0 = gpu.subgroup_mma_load_matrix src[%i,%j] : {leadDimension = 32 : i32}
           : memref<32x32xf16, 3>, !gpu.mma_matrix<16x16xf16, "AOp">
     ```
+
+    Implements IndexedAccessOpInterface.
----------------
Groverkss wrote:

Theres no convention in the codebase for documenting this afaik. It's ok to add it... but it's just weird to add it one off.

https://github.com/llvm/llvm-project/pull/177014