Does OpenMPI know about the number of CPUS per node for FreeBSD?

DM

On Tue, 7 Apr 2009, Ralph Castain wrote:

I would really suggest looking at George's note first as I think you are chasing your tail here. It sounds like the most likely problem is that OMPI thinks you are oversubscribed and is setting sched_yield accordingly. which would fully account for these diffs.

Note that the methods for setting sched yield are different between the various releases, so something could be catching you here. If you are sure you aren't oversubscribing, then set -mca mpi_yield_when_idle 0 on your cmd line to override any internal settings.

On Apr 7, 2009, at 1:32 PM, Steve Kargl wrote:

On Tue, Apr 07, 2009 at 12:00:55PM -0700, Mostyn Lewis wrote:
Steve,

Did you rebuild 1.2.9? As I see you have static libraries, maybe there's
a lurking phthread or something else that may have changed over time?

DM

Yes.  I downloaded 1.2.9, 1.3, and 1.3.1, all within minutes of each
other.  All are configured with --enable-static --disable-shared.
Each is installed in its own directory under /usr/local and
I use full patch names to mpicc and mpiexec.

I just rebuilt 1.3.1 without --enable-static --disable-shared.
There is no change in performance.  1.3.1 is a factor of 10
slower than 1.2.9 with messages of ~10KB and smaller.

I've used svn to grab trunk.  Once I determine how to build
the trunk, I'll do a binary search to see if I can identify
the revision where the problem first occurs.

--
Steve
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users

_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users

Reply via email to