[Mlir-commits] [flang] [mlir] [MLIR][NVVM] Update mbarrier.arrive.* Op (PR #168758)

[Valentin Clement バレンタイン クレメン]

clementval wrote:

> Fixed a build failure in `flang` project, in the latest revision.
> 
> @clementval , Please review the fix in `CUDAIntrinsicCall.cpp` and let me know if it looks correct.

Yeah that looks fine. We have one inline ptx using `mbarrier.arrive.expect_tx.*` form so let me know when this feature will become available. 

https://github.com/llvm/llvm-project/pull/168758