[Mlir-commits] [mlir] [mlir][Vector] Fix vector.insert folder for scalar to 0-d inserts (PR #113828)

[llvmlistbot at llvm.org]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-mlir

Author: Kunwar Grover (Groverkss)

<details>
<summary>Changes</summary>

The current vector.insert folder tries to replace a scalar with a 0-rank vector. This patch fixes this crash by not folding unless they types of the result and replacement are same.

---
Full diff: https://github.com/llvm/llvm-project/pull/113828.diff


2 Files Affected:

- (modified) mlir/lib/Dialect/Vector/IR/VectorOps.cpp (+4-4) 
- (modified) mlir/test/Dialect/Vector/canonicalize.mlir (+12) 


``````````diff
diff --git a/mlir/lib/Dialect/Vector/IR/VectorOps.cpp b/mlir/lib/Dialect/Vector/IR/VectorOps.cpp
index d71a236f62f454..03d2409f42c524 100644
--- a/mlir/lib/Dialect/Vector/IR/VectorOps.cpp
+++ b/mlir/lib/Dialect/Vector/IR/VectorOps.cpp
@@ -2951,11 +2951,11 @@ void InsertOp::getCanonicalizationPatterns(RewritePatternSet &results,
               InsertOpConstantFolder>(context);
 }
 
-// Eliminates insert operations that produce values identical to their source
-// value. This happens when the source and destination vectors have identical
-// sizes.
 OpFoldResult vector::InsertOp::fold(FoldAdaptor adaptor) {
-  if (getNumIndices() == 0)
+  // Fold "vector.insert %v, %dest [] : vector<2x2xf32> from vector<2x2xf32>" to
+  // %v. Note: Do not fold "vector.insert %v, %dest [] : f32 into vector<f32>"
+  // (type mismatch).
+  if (getNumIndices() == 0 && getSourceType() == getResult().getType())
     return getSource();
   return {};
 }
diff --git a/mlir/test/Dialect/Vector/canonicalize.mlir b/mlir/test/Dialect/Vector/canonicalize.mlir
index 6d6bc199e601c0..580daa2a13d15e 100644
--- a/mlir/test/Dialect/Vector/canonicalize.mlir
+++ b/mlir/test/Dialect/Vector/canonicalize.mlir
@@ -2745,6 +2745,18 @@ func.func @vector_insert_const_regression(%arg0: i8) -> vector<4xi8> {
 
 // -----
 
+// CHECK-LABEL: func @insert_into_0d_regression(
+//  CHECK-SAME:     %[[v:.*]]: vector<f32>)
+//       CHECK:   %[[extract:.*]] = vector.insert %{{.*}}, %[[v]] [] : f32 into vector<f32>
+//       CHECK:   return %[[extract]]
+func.func @insert_into_0d_regression(%v: vector<f32>) -> vector<f32> {
+  %cst = arith.constant 0.000000e+00 : f32
+  %0 = vector.insert %cst, %v [] : f32 into vector<f32>
+  return %0 : vector<f32>
+}
+
+// -----
+
 // CHECK-LABEL: @contiguous_extract_strided_slices_to_extract
 // CHECK:        %[[EXTRACT:.+]] = vector.extract {{.*}}[0, 0, 0, 0, 0] : vector<4xi32> from vector<8x1x2x1x1x4xi32>
 // CHECK-NEXT:   return %[[EXTRACT]] :  vector<4xi32>

``````````

</details>


https://github.com/llvm/llvm-project/pull/113828