[clang] [MLIR][NVGPU] Introduce `nvgpu.wargroup.mma.store` Op for Hopper GPUs (PR #65441)

[via cfe-commits]

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

I think the singleton implementation for the warpsize value is broken and anyway overkill.
I believe we still miss a change in a cmake file.

Other than that couple of nits but LGTM.

https://github.com/llvm/llvm-project/pull/65441