[Mlir-commits] [mlir] [MLIR][XeGPU] Legalize elementwise vector width before the XeVM conversions (PR #217131)

[Artem Kroviakov]

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

The change seems modular enough to be easily guarded if needed.
I would like to have the reason for `arith-expand-ops` placement close to the end of the pipeline (instead of pre-blocking) documented.


https://github.com/llvm/llvm-project/pull/217131