[mlir][memref] Add a folder for chained AssumeAlignmentOp ops. #142425
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters. Learn more about bidirectional Unicode characters
The chained ops can be folded away when they have the same alignment. Signed-off-by: hanhanW <hanhan0912@gmail.com>
Member
| @llvm/pr-subscribers-mlir-memref @llvm/pr-subscribers-mlir Author: Han-Chung Wang (hanhanW) ChangesThe chained ops can be folded away when they have the same alignment. Full diff: https://github.com/llvm/llvm-project/pull/142425.diff 3 Files Affected:
diff --git a/mlir/include/mlir/Dialect/MemRef/IR/MemRefOps.td b/mlir/include/mlir/Dialect/MemRef/IR/MemRefOps.td index f33ecb28d27cd..77e3074661abf 100644 --- a/mlir/include/mlir/Dialect/MemRef/IR/MemRefOps.td +++ b/mlir/include/mlir/Dialect/MemRef/IR/MemRefOps.td @@ -174,6 +174,7 @@ def AssumeAlignmentOp : MemRef_Op<"assume_alignment", [ }]; let hasVerifier = 1; + let hasFolder = 1; } //===----------------------------------------------------------------------===// diff --git a/mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp b/mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp index aa9587510670c..d56b32193765e 100644 --- a/mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp +++ b/mlir/lib/Dialect/MemRef/IR/MemRefOps.cpp @@ -533,6 +533,15 @@ void AssumeAlignmentOp::getAsmResultNames( setNameFn(getResult(), "assume_align"); } +OpFoldResult AssumeAlignmentOp::fold(FoldAdaptor adaptor) { + auto source = getMemref().getDefiningOp<AssumeAlignmentOp>(); + if (!source) + return {}; + if (source.getAlignment() != getAlignment()) + return {}; + return getMemref(); +} + //===----------------------------------------------------------------------===// // CastOp //===----------------------------------------------------------------------===// diff --git a/mlir/test/Dialect/MemRef/canonicalize.mlir b/mlir/test/Dialect/MemRef/canonicalize.mlir index da6faf490653d..7a267ae8a2c95 100644 --- a/mlir/test/Dialect/MemRef/canonicalize.mlir +++ b/mlir/test/Dialect/MemRef/canonicalize.mlir @@ -1177,3 +1177,16 @@ func.func @cannot_fold_transpose_cast(%arg0: memref<?x4xf32>) -> memref<?x?xf32, // CHECK: return %[[TRANSPOSE]] return %transpose : memref<?x?xf32, #transpose_map> } + +// ----- + +// CHECK-LABEL: func @fold_assume_alignment_chain +// CHECK-SAME: %[[ARG0:[a-zA-Z0-9]+]] +func.func @fold_assume_alignment_chain(%0: memref<128xf32>) -> memref<128xf32> { + // CHECK: %[[ALIGN:.+]] = memref.assume_alignment %[[ARG0]], 16 + %1 = memref.assume_alignment %0, 16 : memref<128xf32> + // CHECK-NOT: memref.assume_alignment + %2 = memref.assume_alignment %1, 16 : memref<128xf32> + // CHECK: return %[[ALIGN]] + return %2 : memref<128xf32> +} |
qedawkins approved these changes Jun 2, 2025
Collaborator
| LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/116/builds/13650 Here is the relevant piece of the build log for the reference |
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit. This suggestion is invalid because no changes were made to the code. Suggestions cannot be applied while the pull request is closed. Suggestions cannot be applied while viewing a subset of changes. Only one suggestion per line can be applied in a batch. Add this suggestion to a batch that can be applied as a single commit. Applying suggestions on deleted lines is not supported. You must change the existing code in this line in order to create a valid suggestion. Outdated suggestions cannot be applied. This suggestion has been applied or marked resolved. Suggestions cannot be applied from pending reviews. Suggestions cannot be applied on multi-line comments. Suggestions cannot be applied while the pull request is queued to merge. Suggestion cannot be applied right now. Please check back later.
The chained ops can be folded away when they have the same alignment.