Worse, validity of the cache changes whether it has a barrier (in debug mode) or not. I would prefer to avoid calling PetscCommDuplicate() because it is fewer key lookups and then we can make PetscCommGetThreadComm explicitly NOT collective. Any objections? -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20121020/95eafecb/attachment.html>
- [petsc-dev] PetscCommGetThreadComm() is currently collective Jed Brown
