In fixing a problem for Mellanox, I noted that we had somehow lost our default binding policy intentions - i.e., we were no longer automatically binding by default. I fixed that, and so you’d get this if the numactl and numactl-devel libs are missing.
I suspect we should eliminate those warnings if we bind by default? On the other matter: that has been the agreed-upon behavior for some time now. If you don’t specify anything, we launch the number of procs equal to the number of slots, with the slots auto-detected using hwloc and equating slots to number of discovered cores. However, if you use -host, then we assume you are telling us the number of slots=1 for every time you mention the host name. This overrides any auto-discovery. HTH Ralph > On Mar 20, 2015, at 8:16 AM, Rolf vandeVaart <rvandeva...@nvidia.com> wrote: > > Greetings: > > I am now seeing the following message for all my calls to mpirun on ompi > master. This started with last night’s MTT run. Is this intentional? > > [rvandevaart@ivy0 ~]$ mpirun -np 1 hostname > -------------------------------------------------------------------------- > WARNING: a request was made to bind a process. While the system > supports binding the process itself, at least one node does NOT > support binding memory to the process location. > > Node: ivy0 > > This usually is due to not having the required NUMA support installed > on the node. In some Linux distributions, the required support is > contained in the libnumactl and libnumactl-devel packages. > This is a warning only; your job will continue, though performance may be > degraded. > -------------------------------------------------------------------------- > ivy0.nvidia.com <http://ivy0.nvidia.com/> > > > On another note, I noticed on both 1.8 and master that we get different > number of nodes if we specify the hostname. This is not too big a deal, but > surprised me. > > [rvandevaart@ivy0 ~]$ /opt/openmpi/v1.8.4/bin/mpirun hostname > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > ivy0.nvidia.com <http://ivy0.nvidia.com/> > [rvandevaart@ivy0 ~]$ /opt/openmpi/v1.8.4/bin/mpirun -host ivy0 hostname > ivy0.nvidia.com <http://ivy0.nvidia.com/> > [rvandevaart@ivy0 ~]$ > > This email message is for the sole use of the intended recipient(s) and may > contain confidential information. Any unauthorized review, use, disclosure > or distribution is prohibited. If you are not the intended recipient, please > contact the sender by reply email and destroy all copies of the original > message. > _______________________________________________ > devel mailing list > de...@open-mpi.org <mailto:de...@open-mpi.org> > Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel > <http://www.open-mpi.org/mailman/listinfo.cgi/devel> > Link to this post: > http://www.open-mpi.org/community/lists/devel/2015/03/17140.php > <http://www.open-mpi.org/community/lists/devel/2015/03/17140.php>
