[Mlir-commits] [mlir] 57a0bf4 - [mlir][linalg] Do not set insertion point inside padding function (#165420)

[llvmlistbot at llvm.org]

Author: Kunwar Grover
Date: 2025-10-29T12:26:37Z
New Revision: 57a0bf46ffdab3ae2cbab7cb593ebb90561f7b8e

URL: https://github.com/llvm/llvm-project/commit/57a0bf46ffdab3ae2cbab7cb593ebb90561f7b8e
DIFF: https://github.com/llvm/llvm-project/commit/57a0bf46ffdab3ae2cbab7cb593ebb90561f7b8e.diff

LOG: [mlir][linalg] Do not set insertion point inside padding function (#165420)

Remove insertion point in rewriteAsPaddedOp. There is no gurantee that
the sizes provided by the user are before the operation to pad. It's
better to let the user handle where to insert the newly created
operations, as long as they are after the origin operation to pad.

Added: 
    

Modified: 
    mlir/include/mlir/Dialect/Linalg/Transforms/Transforms.h
    mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
    mlir/lib/Dialect/Linalg/Transforms/PadTilingInterface.cpp

Removed: 
    


################################################################################
diff  --git a/mlir/include/mlir/Dialect/Linalg/Transforms/Transforms.h b/mlir/include/mlir/Dialect/Linalg/Transforms/Transforms.h
index c89fc59c91830..d00183a1e16a1 100644
--- a/mlir/include/mlir/Dialect/Linalg/Transforms/Transforms.h
+++ b/mlir/include/mlir/Dialect/Linalg/Transforms/Transforms.h
@@ -653,6 +653,9 @@ struct PadTilingInterfaceResult {
 //    interpreted as the bounding box (dynamic) value to pad to.
 /// * Use "options.paddingValues" to set the padding value of the created
 //    tensor::PadOp.
+//
+// The transformation assumes that the insertion point is set after the
+// operation to pad.
 FailureOr<PadTilingInterfaceResult>
 rewriteAsPaddedOp(OpBuilder &, TilingInterface toPad,
                   PadTilingInterfaceOptions options,

diff  --git a/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp b/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
index 794dda96d1dfa..8b89244486339 100644
--- a/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
+++ b/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
@@ -2464,6 +2464,8 @@ transform::PadTilingInterfaceOp::apply(transform::TransformRewriter &rewriter,
         .setPaddingSizes(getMixedPaddingSizes())
         .setPadToMultipleOf(getPadToMultipleOf());
 
+    OpBuilder::InsertionGuard g(rewriter);
+    rewriter.setInsertionPointAfter(targetOp);
     auto maybePadOps = rewriteAsPaddedOp(
         rewriter, cast<TilingInterface>(targetOp.getOperation()), options);
     if (failed(maybePadOps)) {

diff  --git a/mlir/lib/Dialect/Linalg/Transforms/PadTilingInterface.cpp b/mlir/lib/Dialect/Linalg/Transforms/PadTilingInterface.cpp
index 3e787a2ad0ef5..52ab92f180575 100644
--- a/mlir/lib/Dialect/Linalg/Transforms/PadTilingInterface.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/PadTilingInterface.cpp
@@ -288,10 +288,6 @@ FailureOr<PadTilingInterfaceResult> linalg::rewriteAsPaddedOp(
     return failure();
   }
 
-  OpBuilder::InsertionGuard g(builder);
-  // Set IP after toPad because we also take the dims of toPad's output.
-  builder.setInsertionPointAfter(toPad);
-
   // 1. Get the loopUpperBounds from the TilingInterface.
   SmallVector<Range> iterationDomain = toPad.getIterationDomain(builder);