Hello list,<br><br>when I run the attached example, which spawns a &quot;slave&quot; process with MPI_Comm_spawn(), I see the following:<br><br>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<br>

nbock    19912 92.1  0.0 158964  3868 pts/0    R+   07:42   0:23 ./master<br>nbock    19913  0.0  0.0 158960  3812 pts/0    S+   07:42   0:00 ./master<br>nbock    19914  0.0  0.0 158960  3800 pts/0    S+   07:42   0:00 ./master<br>

nbock    19929 91.1  0.0 158964  3896 pts/0    R+   07:42   0:20 ./slave arg1 arg2<br>nbock    19930 95.8  0.0 158964  3900 pts/0    R+   07:42   0:22 ./slave arg1 arg2<br>nbock    19931 94.7  0.0 158964  3896 pts/0    R+   07:42   0:21 ./slave arg1 arg2<br>

<br>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&#39;s my uneducated guess. What I don&#39;t understand is why PID 19912 is using any CPU resources at all. It&#39;s supposed to be waiting at the MPI_Barrier() for the slaves to finish. What is PID 19912 doing?<br>

<br>Some more information:<br><br>$ uname -a<br>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<br><br>openmpi version 1.3.4<br>

gcc version 4.4.2<br><br>nick<br><br>