Joseph,
Indeed, there was a problem in the MXM rpm.
The fixed MXM has been published at the same location:
http://mellanox.com/downloads/hpc/mxm/v1.1/mxm-latest.tar
-- YK
On 12/4/2012 9:20 AM, Joseph Farran wrote:
> Hi Mike.
>
> Removed the old mxm, downloaded and installed:
>
> /tmp/mxm/v
What does it means ???
So far, I count three people interested in OpenMPI on Windows. That's
not a case for ongoing support.
Damien
On 04/12/2012 11:32 AM, Durga Choudhury wrote:
All
Since I did not see any Microsoft/other 'official' folks pick up the
ball, let me step up. I have been lurking in this list for quit
I've searched the FAQ but not come up with anything about this ... with OpenMPI
1.6.3 and gcc 4.7.2, when I
./configure --prefix=/gpfs/apps/openmpi/1.6.3/gcc_4.7.2-tm-ib \
F77=gfortran FC=gfortran CC=gcc CXX=c++ \
--with-tm=/gpfs/sysapps/torque/4.1.2 \
--with-openib \
2>&1|tee ../logs/co
I have not tested with gfortran 4.7.2.
Can you send the config.log file? (please compress)
On Dec 6, 2012, at 12:42 PM, Paul Hatton wrote:
> I've searched the FAQ but not come up with anything about this ... with
> OpenMPI 1.6.3 and gcc 4.7.2, when I
>
> ./configure --prefix=/gpfs/apps/openm
I just tried with the following:
$ gcc --version
gcc (GCC) 4.7.2
Copyright (C) 2012 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
$ ./configure --prefix=/tmp/do
Thanks. zip-ed config.log attached
--
Paul Hatton
High Performance Computing and Visualisation Specialist
IT Services, The University of Birmingham
Ph: 0121-414-3994 Mob: 07785-977340 Skype: P.S.Hatton
[Service Manager, Birmingham Environment for Academic Research]
[Also Technical Director, IBM
This does not appear to be the right config.log -- the test in question passed:
configure:53489: checking Fortran 90 kind of MPI_INTEGER_KIND
(selected_int_kind(9))
configure:53507: gfortran -o conftest conftestf.f90
configure:53514: $? = 0
On Dec 6, 2012, at 1:11 PM, Paul Hatton wrote:
>
Thanks. Is this with gcc in the default location (/usr/bin/gcc) or elsewhere?
It may not be relevant, but I built v4.7.2 and installed it outside of the
system area:
[appmaint@bb2login04 openmpi-1.6.3]$ which gcc
/gpfs/apps/gcc/v4.7.2/bin/gcc
[appmaint@bb2login04 openmpi-1.6.3]$ gcc -v
Using bui
Oh, sorry - I tried a build with the system gcc and it worked. I'll repeat the
failed one and get it to you. Sorry about that.
--
Paul Hatton
High Performance Computing and Visualisation Specialist
IT Services, The University of Birmingham
Ph: 0121-414-3994 Mob: 07785-977340 Skype: P.S.Hatton
failed one attached.
--
Paul Hatton
High Performance Computing and Visualisation Specialist
IT Services, The University of Birmingham
Ph: 0121-414-3994 Mob: 07785-977340 Skype: P.S.Hatton
[Service Manager, Birmingham Environment for Academic Research]
[Also Technical Director, IBM Visual and Sp
It looks like your gfortran compiler is broken...?
configure:53489: checking Fortran 90 kind of MPI_INTEGER_KIND
(selected_int_kind(9))
configure:53507: gfortran -o conftest conftestf.f90
configure:53514: $? = 0
configure:53554: ./conftest
./conftest: /usr/lib64/libgfortran.so.3: version `GFO
Mine was with a non-default-location gcc-4.7.2 as well.
$which gfortran
/usr/projects/hpcsoft/moonlight/gcc/4.7.2/bin/gfortran
$ gfortran --version
GNU Fortran (GCC) 4.7.2
Copyright (C) 2012 Free Software Foundation, Inc.
GNU Fortran comes with NO WARRANTY, to the extent permitted by law.
You may
Ahh, it's looking in /usr/lib64 and finding libgfortran.so.3 which is older
than the one expected by gcc 4.7 (system gcc is 4.4.6). Looking at my gcc
installation I don't have any shared libraries in there, just static ones.
David - if you don't mind a slight diversion, where is your libgfortran
Eric,
You are correct, our builds do not show lustre support:
MCA io: information "io_romio_user_configure_params" (value: , data
source: default value)
So to enable this, when I build OpenMPI I should pass:
--with-io-romio-flags='--with-file-system=testfs+ufs+nfs+lustre'
We have Lustre, loc
My gcc 4.7.2 build appears to be ignoring the one in /usr/lib64 and using the
one in its own install tree:
/usr/projects/hpcsoft/moonlight/gcc/4.7.2/lib64/libgfortran.so.3
-david
--
David Gunter
HPC-3: Infrastructure Team
Los Alamos National Laboratory
On Dec 6, 2012, at 2:50 PM, Paul Hatton
Thanks. This is obviously (now) a problem with my gcc build which isn't
appropriate for this list. I'll re-visit this and post a solution once I've
(hopefully) got this working. I don't have any shared libraries (*.so.*) in my
gcc tree so something went badly wrong ...
Thanks for your help.
--
17 matches
Mail list logo
