On Oct 20, 2012, at 3:58 PM, Jed Brown wrote: > Worse, validity of the cache changes whether it has a barrier (in debug mode) > or not.
I don't understand the point you are making here. > I would prefer to avoid calling PetscCommDuplicate() because it is fewer key > lookups and then we can make PetscCommGetThreadComm explicitly NOT > collective. Any objections? Shri
