Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

Subject: Re: [OMPI devel] OMPI 1.6 affinity fixes: PLEASE TEST
From: Jeff Squyres (jsquyres_at_[hidden])
Date: 2012-05-30 07:20:04


On May 30, 2012, at 5:05 AM, Mike Dubman wrote:

> Not good:

@#$%@#%@#!! But I guess this is why we test. :-(

> /labhome/alexm/workspace/openmpi-1.6.1a1hge06c2f2a0859/inst/bin/mpirun --host h-qa-017,h-qa-017,h-qa-017,h-qa-017,h-qa-018,h-qa-018,h-qa-018,h-qa-018 -np 8 --bind-to-core -bynode -display-map /usr/mpi/gcc/mlnx-openmpi-1.6rc4/tests/osu_benchmarks-3.1.1/osu_alltoall
>
> ======================== JOB MAP ========================
>
> Data for node: h-qa-017 Num procs: 4
> Process OMPI jobid: [36855,1] Process rank: 0
> Process OMPI jobid: [36855,1] Process rank: 2
> Process OMPI jobid: [36855,1] Process rank: 4
> Process OMPI jobid: [36855,1] Process rank: 6
>
> Data for node: h-qa-018 Num procs: 4
> Process OMPI jobid: [36855,1] Process rank: 1
> Process OMPI jobid: [36855,1] Process rank: 3
> Process OMPI jobid: [36855,1] Process rank: 5
> Process OMPI jobid: [36855,1] Process rank: 7
>
> =============================================================
> --------------------------------------------------------------------------
> An invalid physical processor ID was returned when attempting to bind
> an MPI process to a unique processor.
> [snip]
> $hwloc-ls --of console
> Machine (32GB)
> NUMANode L#0 (P#0 16GB) + Socket L#0 + L3 L#0 (20MB) + L2 L#0 (256KB) + L1 L#0 (32KB) + Core L#0
> PU L#0 (P#0)
> PU L#1 (P#2)
> NUMANode L#1 (P#1 16GB) + Socket L#1 + L3 L#1 (20MB) + L2 L#1 (256KB) + L1 L#1 (32KB) + Core L#1
> PU L#2 (P#1)
> PU L#3 (P#3)

Is this hwloc output exactly the same on both nodes?

-- 
Jeff Squyres
jsquyres_at_[hidden]
For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/