[Mlir-commits] [mlir] [mlir][vector] Update `CastAwayInsertLeadingOneDim` (PR #211777)

[Federico Bruzzone]

================
@@ -594,11 +592,11 @@ func.func @cast_away_insert_leading_one_dims_rank2_scalable(%s: vector<1x[4]xf32
 
 // CHECK-LABEL: func @cast_away_insert_leading_one_dims_rank2_one_dest
 //  CHECK-SAME: (%[[S:.+]]: vector<1x4xf32>, %[[V:.+]]: vector<1x2x1x4xf32>)
-//       CHECK:   %[[EXTRACTS:.+]] = vector.extract %[[S]][0] : vector<4xf32> from vector<1x4xf32>
-//       CHECK:   %[[EXTRACTV:.+]] = vector.extract %[[V]][0] : vector<2x1x4xf32> from vector<1x2x1x4xf32>
-//       CHECK:   %[[INSERT:.+]] = vector.insert %[[EXTRACTS]], %[[EXTRACTV]] [1, 0] : vector<4xf32> into vector<2x1x4xf32>
-//       CHECK:   %[[BCAST:.+]] = vector.broadcast %[[INSERT]] : vector<2x1x4xf32> to vector<1x2x1x4xf32>
-//       CHECK:   return %[[BCAST]]
+//       CHECK:   %[[SCS:.+]] = vector.shape_cast %[[S]] : vector<1x4xf32> to  vector<4xf32>
----------------
FedericoBruzzone wrote:

```suggestion
//       CHECK:   %[[SCS:.+]] = vector.shape_cast %[[S]] : vector<1x4xf32> to vector<4xf32>
```

https://github.com/llvm/llvm-project/pull/211777