Hello List !
The test was a preliminary test:
1] the configuration files are the untouched ones as installed by default by
Debian.
2] the firewall was stopped for the test.
Otherwise, the test job is launched with sbatch:
sbatch phello.sh
This preliminary test was working with the Lenny versi
I can replicate your error; this looks like a Libtool bug.
Open MPI does specifically test each of the C, C++, F77, and F90
compilers for the -pthread flag (and others). When mixing gcc/g++ and
pgf77/pgf90, OMPI's configure script correctly determines that gcc/g++
support -pthread, but pgf
Run with "--mca ess_base_verbose 1000" on the mpirun command line and
send the output, such as:
mpirun --mca ess_base_verbose 1000 rest of your command here...
On Mar 30, 2009, at 5:33 PM, Russell McQueeney wrote:
I only invoked orted manually to see the error message, as it wasn't
sh
(snipping the devel list out of the CC)
Do you have the LSF support libraries installed on all nodes? Are
they all the same version? Do they include the lsb_init symbol? If
memory serves, the name of the library is simply "liblsf" or
"libbat" (.a or .so; I don't know which).
On Mar 30
Jeff Squyres wrote:
On Mar 28, 2009, at 5:31 AM, François PELLEGRINI wrote:
MPI_Comm_size (MPI_COMM_WORLD, &procglbnbr);
MPI_Comm_rank (MPI_COMM_WORLD, &proclocnum);
dataloctab = malloc (2 * (procglbnbr + 1) * sizeof (int));
dataglbtab = dataloctab + 2;
Why +2? Shouldn't it be +pr
Hmm. It *shouldn't* be related to the OS version. I'm using RHEL4
for my tests; RHEL5 performs pretty much the same way with regards to
spawn/connect/accept. But then again, who knows? :-\
Can you try attaching a debugger to the hung processes to see where
exactly they're hung? Perhaps
I only invoked orted manually to see the error message, as it wasn't
showing up on the node's monitor or the xterm window i used to run
mpirun. And no, no prefix command, no aliases, no absolute path,
environment variables set.
Can you supply all the information listed here:
http://www.open-mpi.org/community/help/
We need to know exactly how you are invoking mpirun, what MCA
parameters have been set, etc.
On Mar 28, 2009, at 12:37 PM, Jerome BENOIT wrote:
Hello List,
I have just tried the current SVN Debian p
On Mar 28, 2009, at 5:31 AM, François PELLEGRINI wrote:
MPI_Comm_size (MPI_COMM_WORLD, &procglbnbr);
MPI_Comm_rank (MPI_COMM_WORLD, &proclocnum);
dataloctab = malloc (2 * (procglbnbr + 1) * sizeof (int));
dataglbtab = dataloctab + 2;
Why +2? Shouldn't it be +procglbnbr?
3) Messag
You should not be invoking orted manually -- getting errors when you
invoke orted manually is probably to be expected.
Can you answer the questions from my original mail? Thanks.
On Mar 27, 2009, at 6:29 PM, Russell McQueeney wrote:
Jeff Squyres wrote:
> Hmm -- puzzling -- the error file y
On Mar 27, 2009, at 11:22 AM, Gary Draving wrote:
Thanks for the advice, we tried "-mca btl_openib_ib_min_rnr_timer 25
-mca btl_openib_ib_timeout 20" but we are still getting errors as we
increase the Ns of HPL.dat value into the thousands. Is it ok to just
add these valuse to .openmpi/mca-para
On Mar 30, 2009, at 2:34 PM, Jeff Squyres (jsquyres) wrote:
Yes, it should. However, I'm running into some oddness when I
configure --without-memory-manager.
Thankfully, I can retract this statement. :-)
I found out that the oddness that I was running into was due to my own
testing probl
> Grasping at straws, I suggest putting /bin before /usr/bin in your
> PATH (we may be using different versions of grep, gawk, etc.). Try
> removing Absoft, Intel, and Pathscale compilers from your environment.
I have set .bashrc to give me
~> echo $PATH
/bin:/sbin:/usr/bin:/usr/sbin/:/usr/local/
On Mon, Mar 30, 2009 at 03:15:18PM -0400, Terry Dontje wrote:
> >
> >
> >->ls -l /bin/sh
> >lrwxrwxrwx 1 root root 4 2008-04-02 02:13 /bin/sh -> bash
> >
> I am using tcsh it may take me a while to set up bash for myself.
I too run tcsh as a login shell but the scripts are all shebanged to run
On Mon, Mar 30, 2009 at 02:01:13PM -0500, Ethan Mallove wrote:
> On Mon, Mar/30/2009 07:26:28PM, Kevin McManus wrote:
> > > > you run 'uname -X'?
> > >
> > > uname -X gives me "invalid option" on RHEL {4,5} and SLES {9,10}.
> >
> > which is what I would expect
> > do you also need to supply a pla
Date: Mon, 30 Mar 2009 19:05:25 +0100
From: Kevin McManus
Subject: Re: [OMPI users] Linux opteron infiniband sunstudioconfigure
problem
To: Open MPI Users
Message-ID: <20090330180524.gt13...@gre.ac.uk>
Content-Type: text/plain; charset=us-ascii
> > I will try to reproduce the
On Mon, Mar/30/2009 07:26:28PM, Kevin McManus wrote:
> > > you run 'uname -X'?
> >
> > uname -X gives me "invalid option" on RHEL {4,5} and SLES {9,10}.
>
> which is what I would expect
> do you also need to supply a platform identity/type as an argument?
>
> > Post your config.log file.
>
> at
Just to follow up for the web archives, this is *likely* connected to
the critical bug that we recently announced:
http://www.open-mpi.org/community/lists/announce/2009/03/0029.php
On Mar 24, 2009, at 1:17 PM, Åke Sandgren wrote:
We're having problems with code that uses BLACS and openm
On Mar 27, 2009, at 2:54 PM, Åke Sandgren wrote:
> The Open MPI team is working on a fix; it will be included in the
> v1.3.2 release. Updates will be posted to the ticket.
Will --with-memory-manager=none also work?
I guess it would but I'd like a confirmation of this.
Yes, it should. Howe
> > you run 'uname -X'?
>
> uname -X gives me "invalid option" on RHEL {4,5} and SLES {9,10}.
which is what I would expect
do you also need to supply a platform identity/type as an argument?
> Post your config.log file.
attached to previous relpy
let me know if you need it re-sending
k.mcma.
> cat /etc/*release
->cat /etc/*release
LSB_VERSION="core-2.0-noarch:core-3.0-noarch:core-2.0-x86_64:core-3.0-x86_64"
SUSE LINUX 10.1 (X86-64)
VERSION = 10.1
k.mcma...@gre.ac.uk - http://staffweb.cms.gre.ac.uk/~k.mcmanus
--
Dr Kevin M
On Mon, Mar/30/2009 07:05:25PM, Kevin McManus wrote:
>
> > > I will try to reproduce the problem.
> >
> > I am not able to reproduce this with openmpi-1.3.2a1r20880.tar.gz.
> >
> > $ uname -a
> > Linux ... 2.6.16.46-0.12-smp #1 SMP Thu May 17 14:00:09 UTC 2007 x86_64
> > x86_64 x86_64 GNU/L
FYI - there is a Libtool thread/patch that resolved this issue:
http://lists.gnu.org/archive/html/libtool/2009-03/msg00035.html
-Ethan
On Fri, Mar/20/2009 01:36:58PM, Ethan Mallove wrote:
> On Fri, Mar/20/2009 01:09:56PM, Ethan Mallove wrote:
> > Let me try this again. Below is the error from
> > I will try to reproduce the problem.
>
> I am not able to reproduce this with openmpi-1.3.2a1r20880.tar.gz.
>
> $ uname -a
> Linux ... 2.6.16.46-0.12-smp #1 SMP Thu May 17 14:00:09 UTC 2007 x86_64
> x86_64 x86_64 GNU/Linux
> $ cc -V
> cc: Sun C 5.9 Linux_i386 Patch 124871-01 2007/07
Terry Dontje wrote:
I also was unable to reproduce the configure error with the latest 1.3
tarball. I was on a SLES distribution. What distribution are you on
and can you possibly try and configure using gcc instead of Sun
Studio? I have a feeling this issue is a larger configure issue and
I also was unable to reproduce the configure error with the latest 1.3
tarball. I was on a SLES distribution. What distribution are you on
and can you possibly try and configure using gcc instead of Sun Studio?
I have a feeling this issue is a larger configure issue and not Sun
Studio specif
On Mon, Mar 30, 2009 at 04:24:54PM +0100, John Hearns wrote:
> 2009/3/30 Kevin McManus :
> > >
> > I can find psm libs at...
> >
> > /usr/lib/libpsm_infinipath.so.1.0
> > /usr/lib/libpsm_infinipath.so.1
> > /usr/lib64/libpsm_infinipath.so.1.0
> > /usr/lib64/libpsm_infinipath.so.1
>
>
> On x86_64
On Mon, Mar 30, 2009 at 11:23:33AM -0400, Iain Bason wrote:
>
> On Mar 30, 2009, at 11:02 AM, Kevin McManus wrote:
>
> >configure: error: unrecognized option: --diable-mpi-threads
>
> Typo. That should be --disable-mpi-threads.
oops
silly me :-)
argument accepted but still get MCA bucket prob
Dear Gabriele,
to the best of my knowledge the MPI standard does not provide such a
function.
The reason is that when calling MPI_Gather, the standard requires
matching type signatures (i.e., the sendcount and sendtype argument on
each of the non root processes must match the recvcount and
On Mar 30, 2009, at 11:43 AM, Gabriele Fatigati wrote:
i'm writing an MPI_Gather wrapper to collect void elements. My
queation is: is there a portable way to know the type of received
elements, like MPI_INT or MPI_DOUBLE?
What datatype argument are you giving to MPI_GATHER -- MPI_BYTE?
I've
On Mon, Mar/30/2009 09:04:26AM, Ethan Mallove wrote:
> On Thu, Mar/26/2009 04:52:28PM, Kevin McManus wrote:
> >
> > Hi All,
> >
> > As a complete beginner (to OpenMPI) I am attempting to build on
> > a Linux opteron infiniband platform using SunStudio compilers.
> >
> > My build script looks lik
Dear OpenMPI developers,
i'm writing an MPI_Gather wrapper to collect void elements. My
queation is: is there a portable way to know the type of received
elements, like MPI_INT or MPI_DOUBLE? I've noted that i can retrieve
this information by ompi_datatype_t-> name field, but i think isn't
portable
2009/3/30 Kevin McManus :
> >
> I can find psm libs at...
>
> /usr/lib/libpsm_infinipath.so.1.0
> /usr/lib/libpsm_infinipath.so.1
> /usr/lib64/libpsm_infinipath.so.1.0
> /usr/lib64/libpsm_infinipath.so.1
On x86_64 type systems /usr/lib64 are the 64 bit libraries, /usr/lib
are the 32 bit ones
On Mar 30, 2009, at 11:02 AM, Kevin McManus wrote:
configure: error: unrecognized option: --diable-mpi-threads
Typo. That should be --disable-mpi-threads.
Iain
Hi Jeff, list
Jeff: Thank you for getting back to me.
1) MPI-F90 features
I most likely need the F90 bindings.
The majority of the climate/ocean/atmosphere programs
are written in F90.
I've been using mpif90 to build these codes for a while
(with OpenMPI and MPICH2).
These are mostly "communi
On Mon, Mar 30, 2009 at 09:04:26AM -0500, Ethan Mallove wrote:
> On Thu, Mar/26/2009 04:52:28PM, Kevin McManus wrote:
> >
> > Hi All,
> >
> > As a complete beginner (to OpenMPI) I am attempting to build on
> > a Linux opteron infiniband platform using SunStudio compilers.
> >
> > My build script
Hi guys,
I've a problem with the last stable build and the last nightly snapshot.
When I run a job directly with mpirun no problem.
If I try to submit it with lsf:
bsub -a openmpi -m grid01 mpirun.lsf /mnt/ewd/mpi/fibonacci/fibonacci_mpi
I get the follow error:
mpirun: symbol lookup error: /usr/l
On Thu, Mar/26/2009 04:52:28PM, Kevin McManus wrote:
>
> Hi All,
>
> As a complete beginner (to OpenMPI) I am attempting to build on
> a Linux opteron infiniband platform using SunStudio compilers.
>
> My build script looks like...
>
> #!/bin/sh
>
> ../configure x86_64 \
> CC=cc CXX=CC
Hi Jeff and all members of the list,
You were perfectly right about the wrong string lengths, but even if
corrected, I do still have the same deadlock problems on this simple
child/parent process.
Could it be some bug specifically related to the CentOS 5.2 Linux
distribution ?
Best regards
Lion
Updated link:
http://itss.it/depot/openmpi/
Alex
On Mon, Mar 30, 2009 at 2:52 PM, Alessandro Surace wrote:
> Hi Ralph and all,
> I've find that this question is a known bug
> https://svn.open-mpi.org/trac/ompi/ticket/972 .
>
> This bug is resolved in version 1.3.1.
> I've created the BIN RPM wit
On Mon, Mar 30, 2009 at 08:31:40AM -0400, Terry Dontje wrote:
> Sorry for the delay in response, I was out of the office late last week.
>
> Can you tell me what version of Open MPI you are trying to build (1.2 or
> 1.3 branch)? Are you using the tarball on the Open MPI site or code
> downloade
Hi Ralph and all,
I've find that this question is a known bug
https://svn.open-mpi.org/trac/ompi/ticket/972 .
This bug is resolved in version 1.3.1.
I've created the BIN RPM without the known issue with -D_FORTIFY_SOURCE=2.
You can DL the bin RPM at:
http://www.itss.it/public_ftp/openmpi/openmpi-
Sorry for the delay in response, I was out of the office late last week.
Can you tell me what version of Open MPI you are trying to build (1.2 or 1.3
branch)? Are you using the tarball on the Open MPI site or code downloaded
from the svn repository? Can you tell me which distribution and vers
43 matches
Mail list logo
