[Mlir-commits] [mlir] [MLIR][XeGPU] Remove create tdesc op from xegpu dialect (PR #182804)

[Jianhui Li]

================
@@ -739,59 +739,6 @@ LogicalResult UpdateNdOffsetOp::verify() {
   return success();
 }
 
-//===----------------------------------------------------------------------===//
-// XeGPU_CreateDescOp
-//===----------------------------------------------------------------------===//
-
-void CreateDescOp::build(OpBuilder &builder, OperationState &state,
-                         TensorDescType TensorDesc, Value source,
-                         llvm::ArrayRef<OpFoldResult> offsets) {
-  auto loc = source.getLoc();
-  int64_t size = static_cast<int64_t>(offsets.size());
-  auto type = VectorType::get(size, builder.getIndexType());
-  auto values = getValueOrCreateConstantIndexOp(builder, loc, offsets);
-  auto offset = vector::FromElementsOp::create(builder, loc, type, values);
-  build(builder, state, TensorDesc, source, offset);
-}
-
-void CreateDescOp::build(OpBuilder &builder, OperationState &state,
-                         TensorDescType TensorDesc, Value source,
-                         llvm::ArrayRef<int64_t> offsets) {
-  auto ofrs = getAsIndexOpFoldResult(builder.getContext(), offsets);
-  build(builder, state, TensorDesc, source, ofrs);
-}
-
-LogicalResult CreateDescOp::verify() {
-  auto tdescTy = getTensorDescType();
----------------
Jianhui-Li wrote:

please remove tensor descriptor related verification in load/store/prefetch. 
remove isValidGatherScatterParams()

https://github.com/llvm/llvm-project/pull/182804