[Mlir-commits] [mlir] [mlir][amdgpu] Add amdgpu.make_dma_descriptor (PR #169407)

[Erick Ochoa Lopez]

================
@@ -705,6 +705,34 @@ LogicalResult TransposeLoadOp::verify() {
   return success();
 }
 
+//===----------------------------------------------------------------------===//
+// MakeDmaDescriptorOp
+//===----------------------------------------------------------------------===//
+
+LogicalResult MakeDmaDescriptorOp::verify() {
+  ArrayRef<int64_t> globalStaticStrides = getGlobalStaticStrides();
+
+  if (globalStaticStrides.empty()) {
+    return emitOpError("strides must not be empty.");
+  }
+  if (globalStaticStrides.back() != 1) {
+    return emitOpError("strides for the innermost dimension must be 1.");
+  }
+
+  ArrayRef<int64_t> globalStaticSizes = getGlobalStaticSizes();
+  size_t rank = globalStaticSizes.size();
+  if (rank != globalStaticStrides.size()) {
+    return emitOpError("strides and sizes must have same rank.");
+  }
+
+  ArrayRef<int64_t> sharedStaticSizes = getSharedStaticSizes();
+  if (rank != sharedStaticSizes.size()) {
+    return emitOpError("tensor must have same rank as tile.");
+  }
+
----------------
amd-eochoalo wrote:

Thanks for the review! https://github.com/llvm/llvm-project/pull/169407/commits/d51853c2f7b2326551835f8660ffdf90af741416 I will merge after the tests finish and continue working on the comments on the next one. Will ping you when ready :)

https://github.com/llvm/llvm-project/pull/169407