I'm on the bridge if anyone wants to discuss. Otherwise, I'll drop by quarter past the hour. On Mon, Dec 15, 2014 at 11:07 AM, Jim Dinan <[email protected]> wrote:

I have not yet produced a new version of the proposal with the changes incorporated, but they are all captured in the attached PDF. I'm happy to meet to review these, if folks want to. One item that bears discussion is the change that would cause MPI_Group_compare operations to result in MPI_ALIASED when the group arguments belong to different processes.

~Jim.

On Sun, Dec 14, 2014 at 7:05 AM, Balaji, Pavan <[email protected]> wrote:

...

Jim,

Do you want to discuss the changes suggested at the Forum for the endpoints reading at the call on Monday, or do you need more time?

Since that's the only agenda item for now, we'll cancel the call if you are not ready yet.

There was some reasonable interest in the independent communicators proposal. At least enough to encourage me to pursue it. I'm going to take it offline for a bit to work on it and get some implementation experience before bringing it back up again.

Thanks,

-- Pavan

Sent from my iPhone _______________________________________________ mpiwg-hybridpm mailing list [email protected] http://lists.mpi-forum.org/mailman/listinfo.cgi/mpiwg-hybridpm