[Mlir-commits] [mlir] [mlir][xegpu] Improve XeGPU op verification logic for SIMT flavor and update tests. (PR #127920)

[Adam Siemieniuk]

================
@@ -73,43 +76,6 @@ static bool isWriteHintOrNone(const CachePolicyAttr &attr) {
          kind == CachePolicy::WRITE_BACK || kind == CachePolicy::WRITE_THROUGH;
 }
 
-// Validations for nd instruction arguments is successful if any of these are
-// true:
-// - tensor descriptor and the output vector shapes exactly match.
-// - tensor descriptor has a sg_map attribute and the distributed vector shape
-//   matches the tensor descriptor shape when scaled using sg_map factors on
-//   each dimension.
-static bool isArgShapesValid(ArrayRef<int64_t> descShape,
----------------
adam-smnk wrote:

Why not keep the validation common?
The new checks are still shared between store and load.

+1 for better error messages but I think they could still be captured by a common helper

https://github.com/llvm/llvm-project/pull/127920