[Mlir-commits] [mlir] [MLIR][XEGPU] Add blocking support for scatter ops (PR #144766)
llvmlistbot at llvm.org
llvmlistbot at llvm.org
Wed Jun 18 14:17:04 PDT 2025
github-actions[bot] wrote:
<!--LLVM CODE FORMAT COMMENT: {clang-format}-->
:warning: C/C++ code formatter, clang-format found issues in your code. :warning:
<details>
<summary>
You can test this locally with the following command:
</summary>
``````````bash
git-clang-format --diff HEAD~1 HEAD --extensions cpp -- mlir/lib/Dialect/XeGPU/Transforms/XeGPUBlocking.cpp mlir/test/lib/Dialect/XeGPU/TestXeGPUTransforms.cpp
``````````
</details>
<details>
<summary>
View the diff from clang-format here.
</summary>
``````````diff
diff --git a/mlir/lib/Dialect/XeGPU/Transforms/XeGPUBlocking.cpp b/mlir/lib/Dialect/XeGPU/Transforms/XeGPUBlocking.cpp
index 8e38d8c73..3950e8f70 100644
--- a/mlir/lib/Dialect/XeGPU/Transforms/XeGPUBlocking.cpp
+++ b/mlir/lib/Dialect/XeGPU/Transforms/XeGPUBlocking.cpp
@@ -303,7 +303,8 @@ void XeGPUBlockingPass::runOnOperation() {
// If the encoding is a ScatterTensorDescAttr, we need to
// potentially adjust the chunk size based on the inst_data.
if (tdescTy.isScattered()) {
- auto scatterAttr = llvm::dyn_cast_if_present<xegpu::ScatterTensorDescAttr>(encoding);
+ auto scatterAttr =
+ llvm::dyn_cast_if_present<xegpu::ScatterTensorDescAttr>(encoding);
int64_t chunkSize = scatterAttr.getChunkSize().getInt();
if (chunkSize > 1) {
diff --git a/mlir/test/lib/Dialect/XeGPU/TestXeGPUTransforms.cpp b/mlir/test/lib/Dialect/XeGPU/TestXeGPUTransforms.cpp
index 4aa5c4e8f..6ea4354aa 100644
--- a/mlir/test/lib/Dialect/XeGPU/TestXeGPUTransforms.cpp
+++ b/mlir/test/lib/Dialect/XeGPU/TestXeGPUTransforms.cpp
@@ -118,7 +118,8 @@ struct TestXeGPUUnrollingPatterns
// To create a new attribute with a different chunk_size:
auto newEncoding = xegpu::ScatterTensorDescAttr::get(
- ctx, scatterAttr.getMemorySpace().getValue(), blockedChunkSize);
+ ctx, scatterAttr.getMemorySpace().getValue(),
+ blockedChunkSize);
encoding = newEncoding;
}
``````````
</details>
https://github.com/llvm/llvm-project/pull/144766
More information about the Mlir-commits
mailing list