[mpiwg-ft] Ticket 324 June 2015 Reading

[Bland, Wesley]

It would get deprecated in favor of a similar error handler which would do similar, but slightly more well-defined things. The replacement would still abort the job. I would assume that we would want to make the replacement default error handler the one that aborts MPI_COMM_WORLD.


In the end, the behavior would be exactly the same. It’s just that the “high quality implementation” specification would be slightly more specific.



On June 1, 2015 at 10:00:12 PM, Bland, Wesley (wesley.bland at intel.com<mailto:wesley.bland at intel.com>) wrote:

Notes from the ticket reading are now posted on the wiki: https://svn.mpi-forum.org/trac/mpi-forum-web/wiki/ftwg2015-06-01

TL;DR - The reading did not “pass”, but we got lots of good feedback to come back with a new version. We should consider splitting this into two or three tickets. One to define new errhandlers that does the new definition (abort communicator) and one that’s a more well defined old definition (abort MPI_COMM_WORLD). Another ticket will deprecate MPI_COMM_ERRORS_ARE_FATAL. Another ticket will consolidate the definitions of all of the error handling text to a single place.

The rest of the details can be found in the wiki.

I’ll be working on some drafts over the next few days to try to get new versions of this ticket out for discussion. My tentative hope is to get this ready for a new plenary in September. There’s going to be enough changes that this should probably get a plenary before another reading.

Comments welcome.

Thanks,
Wesley