Jeff Hammond <[email protected]> writes:

At this point, you should subscribe to http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-greq and have further discussion of this issue there. I copied that list to make the transition seemless.

The mpi3-greq list hasn't had any activity in five years, so it's not clear any interested parties are there. I care quite a bit about improving the ability of generalized requests to make progress, mostly because I want to be able to implement domain-specific nonblocking collectives with similar progress semantics to those provided directly by MPI.