[Openmp-commits] [openmp] [OpenMP] Use new OMPT state and sync kinds for barrier events (PR #95602)

[Jan André Reuter via Openmp-commits]

Thyre wrote:

I've tried the branch with Score-P and a simple `printf` tool and the results looked reasonable. 
Would it make sense to differentiate the barrier types in the tests to ensure that the correct barrier types are returned, if possible? A similar thing was done with the loop types in #97429.

https://github.com/llvm/llvm-project/pull/95602