On Dec 4, 2009, at 7:46 AM, Nicolas Bock wrote: > Hello list, > > when I run the attached example, which spawns a "slave" process with > MPI_Comm_spawn(), I see the following: > > nbock 19911 0.0 0.0 53980 2288 pts/0 S+ 07:42 0:00 > /usr/local/openmpi-1.3.4-gcc-4.4.2/bin/mpirun -np 3 ./master > nbock 19912 92.1 0.0 158964 3868 pts/0 R+ 07:42 0:23 ./master > nbock 19913 0.0 0.0 158960 3812 pts/0 S+ 07:42 0:00 ./master > nbock 19914 0.0 0.0 158960 3800 pts/0 S+ 07:42 0:00 ./master > nbock 19929 91.1 0.0 158964 3896 pts/0 R+ 07:42 0:20 ./slave arg1 > arg2 > nbock 19930 95.8 0.0 158964 3900 pts/0 R+ 07:42 0:22 ./slave arg1 > arg2 > nbock 19931 94.7 0.0 158964 3896 pts/0 R+ 07:42 0:21 ./slave arg1 > arg2 > > The third column is the CPU usage according to top. I notice 3 master > processes, which I attribute to the fact that MPI_Comm_spawn really fork()s > and then spawns, but that's my uneducated guess.
Ummm....if you look at your cmd line /usr/local/openmpi-1.3.4-gcc-4.4.2/bin/mpirun -np 3 ./master you will see that you specified 3 copies of master be run :-) > What I don't understand is why PID 19912 is using any CPU resources at all. > It's supposed to be waiting at the MPI_Barrier() for the slaves to finish. > What is PID 19912 doing? It is polling at the barrier. This is done aggressively by default for performance. You can tell it to be less aggressive if you want via the yield_when_idle mca param. > > Some more information: > > $ uname -a > Linux mujo 2.6.31-gentoo-r6 #2 SMP PREEMPT Fri Dec 4 07:08:07 MST 2009 x86_64 > Intel(R) Core(TM)2 Quad CPU Q8200 @ 2.33GHz GenuineIntel GNU/Linux > > openmpi version 1.3.4 > gcc version 4.4.2 > > nick > > <master.c><slave.c>_______________________________________________ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users
