Skip to content

Conversation

@fabianmcg
Copy link
Contributor

Use ValueRange instead of SmallVector in tensor::createPadHighOp for the dynOutDims arg.

@llvmbot
Copy link
Member

llvmbot commented Jun 16, 2025

@llvm/pr-subscribers-mlir-tensor

@llvm/pr-subscribers-mlir

Author: Fabian Mora (fabianmcg)

Changes

Use ValueRange instead of SmallVector in tensor::createPadHighOp for the dynOutDims arg.


Full diff: https://github.com/llvm/llvm-project/pull/144397.diff

2 Files Affected:

  • (modified) mlir/include/mlir/Dialect/Tensor/Utils/Utils.h (+1-1)
  • (modified) mlir/lib/Dialect/Tensor/Utils/Utils.cpp (+1-2)
diff --git a/mlir/include/mlir/Dialect/Tensor/Utils/Utils.h b/mlir/include/mlir/Dialect/Tensor/Utils/Utils.h
index 1a4733df3f187..a1ce4e252c2f4 100644
--- a/mlir/include/mlir/Dialect/Tensor/Utils/Utils.h
+++ b/mlir/include/mlir/Dialect/Tensor/Utils/Utils.h
@@ -30,7 +30,7 @@ namespace tensor {
 // for _static_ dimensions.
 PadOp createPadHighOp(RankedTensorType resType, Value source, Value pad,
                       bool nofold, Location loc, OpBuilder &builder,
-                      SmallVector<Value> dynOutDims = {});
+                      ValueRange dynOutDims = std::nullopt);
 
 // Creates dim ops for each dynamic dimension of the ranked tensor argument and
 // returns these as values.
diff --git a/mlir/lib/Dialect/Tensor/Utils/Utils.cpp b/mlir/lib/Dialect/Tensor/Utils/Utils.cpp
index 11ae0108594dd..289296a07d9d3 100644
--- a/mlir/lib/Dialect/Tensor/Utils/Utils.cpp
+++ b/mlir/lib/Dialect/Tensor/Utils/Utils.cpp
@@ -24,8 +24,7 @@ using namespace mlir::tensor;
 
 PadOp mlir::tensor::createPadHighOp(RankedTensorType resType, Value source,
                                     Value pad, bool nofold, Location loc,
-                                    OpBuilder &b,
-                                    SmallVector<Value> dynOutDims) {
+                                    OpBuilder &b, ValueRange dynOutDims) {
 
   // This assumption simplifies the following logic without limiting what's
   // required _today_. If needed, we can relax it in the future.

@fabianmcg fabianmcg merged commit 695c4f2 into llvm:main Jun 16, 2025
10 checks passed
@fabianmcg fabianmcg deleted the users/fabian/pr-nfc-pad-arg branch June 16, 2025 18:04
akuhlens pushed a commit to akuhlens/llvm-project that referenced this pull request Jun 24, 2025
…sor::createPadHighOp` (llvm#144397)

Use `ValueRange` instead of `SmallVector` in `tensor::createPadHighOp`
for the `dynOutDims` arg.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants