[Mlir-commits] [mlir] [MLIR][XeGPU] Enable one-step subgroup distribution of cross-lane reduction to shuffle op (PR #182698)

[Artem Kroviakov]

https://github.com/akroviakov approved this pull request.

I suppose the source rank limitation will be addressed in subsequent PRs

https://github.com/llvm/llvm-project/pull/182698