[Mlir-commits] [mlir] [mlir][vector] Do not flatten or scalarize region-masked transfers (PR #231799)

[llvmlistbot at llvm.org]

================
@@ -178,3 +178,50 @@ func.func @transfer_read_2d_extract_dynamic(%m: memref<?x?xf32>, %row_idx: index
   %elem = vector.extract %vec[%row_offset, %col_offset] : f32 from vector<10x5xf32>
   return %elem : f32
 }
+
+// -----
+
+// A region-masked transfer cannot be scalarized because the rewrite would
+// create multiple operations inside the vector.mask region.
----------------
Vaisman wrote:

Agreed, reworded to "not supported yet". 
These patterns had no mask-operand tests either, I added `*_with_mask` variants.

https://github.com/llvm/llvm-project/pull/231799