Le 7 oct. 2013 à 15:01, Wesley Bland <[email protected]> a écrit :
I'm going through the wiki page and the spec to clean some things up after the last discussion. As I recall, our final decision was to not continue pushing ULFM as 4 separate tickets (main ticket, RMA ticket, I/O ticket, dynamic processes ticket), but to combine them all into one. Unless there is objection, I'll close tickets 325, 326, and 327.
Yes. I'll make another round of review on dynamic process, I think nobody took a serious look at it in ages, then we'll be getting there.
Ticket 336 probably either needs to be closed or undergo a major revision after the discussion on the last con call. My preference would probably be to close it entirely and start a new one as the new idea is very different than the older attribute query mechanism.
I was looking at info objects, so it seems what we need is 1) a new routine MPI_Set_options( MPI_Info ), or MPI_INIT_EX(argc, argv, MPI_Info ), which is necessarily a new ticket (unrelated to FT) and 2) a standardized key-value pair for "WANT_FT". That second part could go in 336, or another new FT ticket.
I'll plan on closing all 4 tickets by the end of the week if I don't hear any objections.
Thanks, Wesley _______________________________________________ mpiwg-ft mailing list [email protected] http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-ft
-- * Dr. Aurélien Bouteiller * Researcher at Innovative Computing Laboratory * University of Tennessee * 1122 Volunteer Boulevard, suite 309b * Knoxville, TN 37996 * 865 974 9375