Re: [hwloc-users] hwloc 1.5, freebsd and linux output on the same hardware

On Fri, Oct 5, 2012 at 5:01 PM, Samuel Thibault wrote: > Sebastian Kuzminsky, le Sat 06 Oct 2012 00:55:57 +0200, a écrit : > > binding to CPU0 > > could not bind to CPU0: Resource deadlock avoided > > Mmm, from what I read in the freebsd kernel: > > /* > * Create a set

Re: [hwloc-users] hwloc 1.5, freebsd and linux output on the same hardware

Sebastian Kuzminsky, le Sat 06 Oct 2012 00:55:57 +0200, a écrit : > binding to CPU0 > could not bind to CPU0: Resource deadlock avoided Mmm, from what I read in the freebsd kernel: /* * Create a set in the space provided in 'set' with the provided parameters. * The set is returned with a

Re: [hwloc-users] hwloc 1.5, freebsd and linux output on the same hardware

On Thu, Oct 4, 2012 at 6:39 PM, Samuel Thibault wrote: > Sebastian Kuzminsky, le Wed 03 Oct 2012 17:24:55 +0200, a écrit : > > So that's an improvement over the svn trunk > > yesterday, but it's not all the way fixed yet! > > Ok. Apparemently hwloc can't bind itself to

Re: [OMPI users] -output-filename 1234 versus --mca orte_output_filename 1234

No - one is just a shorthand for the other. On Thu, Oct 4, 2012 at 9:54 AM, Sébastien Boisvert < sebastien.boisver...@ulaval.ca> wrote: > Hi, > > Is there any difference in the code path between > > mpiexec -n 1 -output-filename 1234 ./a.out > > and > > mpiexec -n 1 --mca

Re: [OMPI users] question to -cpus-to-proc

I don't recall anyone ever asking for that before - the original people who wanted cpus-per-proc only cared about it at the job level, not the individual app. So it currently is a global value. It can be modified, but it'll be awhile before I get around to it. On Thu, Oct 4, 2012 at 10:53 PM,

Re: [OMPI users] Mellanox MLX4_EVENT_TYPE_SRQ_LIMIT kernel messages

I wrote: > You wrote: > "Reply-to considered harmful" again, sigh.

[OMPI users] too much stack size: _silently_ failback to IPoIB

Dear Open MPI developer, there are often problems with the user limit for the stack size (ulimit -s) on Linux if running Fortran and/or OpenMP(=hybride) programs. In one case we have seen the user has set the stack size in his environment by occasion far too high - to about one TeraByte (on

[OMPI users] question to -cpus-to-proc

Hi, if I start different programs on one command line, I can use different options "-host" and "-np". mpiexec -report-bindings -host linpc0 -np 1 date : \ -host sunpc0,sunpc1 -np 2 hostname Fri Oct 5 07:37:25 CEST 2012 sunpc0 sunpc1 Only the last