jtb20 wrote: I've made a change to this patch to adjust the wording used for the deadlock-avoidance error in AMDGPU's `omp_set_lock` implementation, with a view to dropping the follow-on 4/4 patch in this stack. I think the lane serialization for `omp critical` probably covers the cases that the 4/4 patch supports well enough anyway.
https://github.com/llvm/llvm-project/pull/242479 _______________________________________________ llvm-branch-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-branch-commits
