Short version:
==
OMPI 1.6.soon-to-be-1 will report *logical* hwloc core bitmasks (not PUs!).
The reasons for this are sordid and, frankly, uninteresting. :-\
Perhaps we need to update this to be something a bit more user-friendly before
1.6.1 goes final. Hrm...
More detail:
==
could display hwloc bitmasks to avoid
confusion.
Youri
-Message d'origine-
De : hwloc-users-boun...@open-mpi.org [mailto:hwloc-users-boun...@open-mpi.org]
De la part de Brice Goglin
Envoyé : mercredi 30 mai 2012 16:07
À : Hardware locality user list
Objet : Re: [hwloc-users] Understandin
Jeff,
What is the displayed bitmask in OMPI 1.6? Is it the hwloc bitmask? Or
the OMPI bitmask made of OMPI indexes?
Brice
Le 30/05/2012 16:01, Jeff Squyres a écrit :
> You might want to try the OMPI tarball that is about to become OMPI v1.6.1 --
> we made a bunch of affinity-related fixes, and
You might want to try the OMPI tarball that is about to become OMPI v1.6.1 --
we made a bunch of affinity-related fixes, and it should be much more
predictable / stable in what it does in terms of process binding:
http://www.open-mpi.org/~jsquyres/unofficial/
(these affinity fixes are not y
Hello Youri,
When using openmpi 1.4.4 with --np 2 --bind-to-core --bycore" it reports
the following:
>
> [hostname:03339] [[17125,0],0] odls:default:fork binding child
> [[17125,1],0] to cpus 0001
>
> [hostname:03339] [[17125,0],0] odls:default:fork binding child
> [[17125,1],1] to cpus 0002
>
Bit
Hi,
I'm just trying to make sure that the output of hwloc-ps is consistent
with openmpi's binding report.
For starters here is the output of hwloc-ps:
Machine (48GB)
NUMANode L#0 (P#0 24GB) + Socket L#0 + L3 L#0 (12MB)
L2 L#0 (256KB) + L1 L#0 (32KB) + Core L#0 + PU L#0 (P#0)