[Mlir-commits] [mlir] [mlir][xegpu] Resolve layout conflict on scf.for init operands (PR #215911)

[llvmlistbot at llvm.org]

llvmorg-github-actions[bot] wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-mlir

Author: Jianhui Li (Jianhui-Li)

<details>
<summary>Changes</summary>

  Extends XeGPU layout conflict resolution to cover values carried  across region  boundaries. Previously
  `ResolveLayoutConflicts::resolveVectorConsumer` skipped all region-branch operands, so a mismatch between a loop-carried
  value's own layout and the layout its loop position requires was never reconciled and causing lowering issue.
  
  
  assisted-by-claude

---
Full diff: https://github.com/llvm/llvm-project/pull/215911.diff


3 Files Affected:

- (modified) mlir/lib/Dialect/XeGPU/Transforms/XeGPULayoutImpl.cpp (+29) 
- (modified) mlir/lib/Dialect/XeGPU/Transforms/XeGPUPropagateLayout.cpp (+11-19) 
- (modified) mlir/test/Dialect/XeGPU/resolve-layout-conflicts.mlir (+23-3) 


``````````diff

``````````

</details>


https://github.com/llvm/llvm-project/pull/215911