You hit upon a problem that Jeff and I were talking about yesterday. Both master and 2.x need to be fixed. The problem is a little more subtle as direct launch under non-PMIx systems doesn’t have access to a non-blocking fence, and thus there is no way to call progress while waiting for the barrier to complete. I’ve proposed a solution, but I’m out for a day or two (eye surgery this morning) and won’t get a chance to fix it right away.
> On Oct 6, 2015, at 11:17 AM, Nysal Jan K A <jny...@gmail.com> wrote: > > In v1.8 there is a RTE barrier in finalize. OMPI_LAZY_WAIT_FOR_COMPLETION > waits for the barrier to complete. Internally opal_progress() is invoked. In > the master branch we call PMIX fence instead. PMIX_WAIT_FOR_COMPLETION seems > to only call usleep. How will ompi progress outstanding operations ? > > Regards > --Nysal > _______________________________________________ > devel mailing list > de...@open-mpi.org > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel > Link to this post: > http://www.open-mpi.org/community/lists/devel/2015/10/18144.php
