[Mlir-commits] [clang] [flang] [mlir] Revert "[flang][flang-driver][mlir][OpenMP] atomic control support" (PR #150504)

[Sergio Afonso]

skatrak wrote:

I imagine we could fix the problem by removing the `-triple amdgcn-amd-amdgpu` option from the RUN lines of the new failing test, or by adding a `!REQUIRES: amdgpu-registered-target` if the option is needed.

https://github.com/llvm/llvm-project/pull/150504