Thanks for tracking that down!
Here's where it was removed:
and then it was later applied to the v1.4 branch in r23448.
I'll double check back with Ralph (the ORTE guy), but I don't think that this change matters.
On Nov 18, 2010, at 5:19 AM, tmishima_at_[hidden] wrote:
> I found that ./openmpi-1.4.3/ompi/runtime/ompi_mpi_init.c was changed.
> Calling opal_progress_event_users_decrement was deleted as below.
> $diff openmpi-1.4.2/ompi/runtime/ompi_mpi_init.c openmpi-1.4.3
> < /* Undo ORTE calling opal_progress_event_users_increment() during
> < MPI lifetime, to get better latency when not using TCP. Do
> < this *after* dyn_init, as dyn init uses lots of ORTE
> < communication and we don't want to hinder the performance of
> < that code. */
> < opal_progress_event_users_decrement();
> I confirmed that this change affects SYSTEM CPU & performace decrement.
> Is it safe to replace the routine with older one?
> Anyway, if it's just a mistake, please fix it in the next version.
> Tetsuya Mishima
> users mailing list
For corporate legal information go to: