[Mlir-commits] [mlir] [mlir][vector-to-gpu] Do not convert ops masked by vector.mask (PR #226548)

[llvmlistbot at llvm.org]

Vaisman wrote:

@fabianmcg Thanks again for the review. I’ve documented the gpu.subgroup_mma/NVGPU masking asymmetry in the PR description as requested. The PR is still green and cleanly mergeable. If this addresses your concern, could you mark it approved, or let me know if anything else is needed? Thanks!

https://github.com/llvm/llvm-project/pull/226548