On Mar 11 2011, Eugene Loh wrote:
The idea would be to hardwire support for MPI_THREAD_MULTIPLE to be off, just as we have done for progress threads. Threads might still be used for other purposes -- e.g., ORTE, openib async thread, etc.
That's what I was assuming, too. Threads used behind the scene are not a problem - IF you can solve the infernal synchronisation and signalling problems and stop them causing trouble in other ways! Whether this should be done for MPI_THREAD_SERIALIZED, too, is debatable, but it can't be done using a cautious interpretation of any published standards that I know of, and in my experience can't be done reliably. That's not nice, and I don't want to go there :-( Regards, Nick Maclaren.
