Hmmm...looks to me like the code should protect against this - unless the system isn't correctly reporting binding support. Could you run this with "-mca ess_base_verbose 10"? This will output the topology we found, including the binding support (which isn't in the usual output).

On Jan 8, 2014, at 8:14 PM, Ralph Castain <rhc@open-mpi.org> wrote:

Hmmm...I see the problem. Looks like binding isn't supported on that system for some reason, so we need to turn "off" our auto-binding when we hit that condition. I'll check to see why that isn't happening (was supposed to do so)


On Jan 8, 2014, at 3:43 PM, Paul Hargrove <phhargrove@lbl.gov> wrote:

While I have yet to get a working build on NetBSD for x86-64 h/w, I *have* successfully built Open MPI's current 1.7.4rc tarball on NetBSD-6 for x86.  However, I can't *run* anything:

Attempting the ring_c example on 2 cores:
-bash-4.2$ mpirun -mca btl sm,self -np 2 examples/ring_c
--------------------------------------------------------------------------
While computing bindings, we found no available cpus on
the following node:

  Node:  pcp-j-17

Please check your allocation.
--------------------------------------------------------------------------

The failure is the same w/o "-mca btl sm,self"
Singleton runs fail just as the np=2 run did.

I've attached compressed output from "ompi_info --all".

Since this is probably an hwloc-related issue, I also build hwloc-1.7.2 from pristine sources.
I have attached compressed output of lstopo which NOTABLY indicates a failure to bind to both of the CPUs.

For now, an explicit "--bind-to none" is working for me.
Please let me know what additional info may be required.

-Paul

--
Paul H. Hargrove                          PHHargrove@lbl.gov
Future Technologies Group
Computer and Data Sciences Department     Tel: +1-510-495-2352
Lawrence Berkeley National Laboratory     Fax: +1-510-486-6900
<ompi_info-netbsd-x86.txt.bz2><lstopo172-netbsd-x86.txt.bz2>_______________________________________________
devel mailing list
devel@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/devel