It appears you have found a bug in Libtool, the package we use for
hiding all the portability issues in building libraries. Thankfully,
Open MPI has been befriended by a Libtool developer and he provided a
workaround for your problem (Thanks Ralf!). If you configure with
the option LD="ld -64", you should be able to properly build. Can
you give this a whirl and let us know if it works?
Thanks,
Brian
On Dec 16, 2005, at 8:36 AM, Pierre Valiron wrote:
Dear Brian and George,
We could not check your fix because the nightly tarball was not
updated accordingly ;-).
I have discovered today the 1.0.1 version on the open-mpi web page
and Francoise Roch tested it. The make goes a little further but
still fails. Plesase find the logs in attachment.
Good luck !
Pierre.
Brian Barrett wrote:
Thanks for the update. I've fixed the next bug in subversion on
the trunk and it should be in tonight's nightly tarball. I
should also make it into v1.0.1, once that is made available.
Brian
On Dec 7, 2005, at 11:41 AM, Pierre Valiron wrote:
Thanks Brian and George,
Francoise Roch tried the following version as you suggested
http://www.open-mpi.org/~brbarret/download/
openmpi-1.1a1r8384.tar.gz
Things go a little further but the make still fails.
Please find the logs attached.
Pierre.
Brian Barrett wrote:
On Dec 5, 2005, at 4:05 PM, Pierre Valiron wrote:
I tried to experiment open-mpi on our Solaris10 v40z cluster
hoping it might surpass lam/mpi 7.1.2b28...
I used the following script to compile in 64 bit mode:
<snip>
The configure runs fine and the make aborts very rapidly.
I attach the log for configure+make and the config.log file.
George spotted the problem - it looks like config.guess/
config.sub always report i386 as the architecture when running
on an Opteron. This isn't totally unexpected -- generally
config.guess / config.sub reports the architecture target that
is the default for the given platform and, as you said,
Solaris on Opteron defaults to emitting 32 bit code. I think
I have a fix for this issue, but I don't have any Opteron
machines running Solaris to verify. Can you grab the Open MPI
tarball at the URL below and see if it builds properly?
http://www.open-mpi.org/~brbarret/download/
openmpi-1.1a1r8384.tar.gz
If that works for you, we'll push the change into Open MPI
1.0.1 (it's a very small change).
Thanks,
Brian
--
Soutenez le mouvement SAUVONS LA RECHERCHE :
http://recherche-en-danger.apinc.org/
_/_/_/_/ _/ _/ Dr. Pierre VALIRON
_/ _/ _/ _/ Laboratoire d'Astrophysique
_/ _/ _/ _/ Observatoire de Grenoble / UJF
_/_/_/_/ _/ _/ BP 53 F-38041 Grenoble Cedex 9 (France)
_/ _/ _/ http://www-laog.obs.ujf-grenoble.fr/
~valiron/
_/ _/ _/ Mail: pierre.vali...@obs.ujf-grenoble.fr
_/ _/ _/ Phone: +33 4 7651 4787 Fax: +33 4 7644 8821
_/ _/_/
<make.output.bz2>
<Makefile.openmpi.bz2>
<config.log.bz2>
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
--
Soutenez le mouvement SAUVONS LA RECHERCHE :
http://recherche-en-danger.apinc.org/
_/_/_/_/ _/ _/ Dr. Pierre VALIRON
_/ _/ _/ _/ Laboratoire d'Astrophysique
_/ _/ _/ _/ Observatoire de Grenoble / UJF
_/_/_/_/ _/ _/ BP 53 F-38041 Grenoble Cedex 9 (France)
_/ _/ _/ http://www-laog.obs.ujf-grenoble.fr/~valiron/
_/ _/ _/ Mail: pierre.vali...@obs.ujf-grenoble.fr
_/ _/ _/ Phone: +33 4 7651 4787 Fax: +33 4 7644 8821
_/ _/_/
<make.out.tar.gz>
<config.log.tar.gz>
_______________________________________________
users mailing list
us...@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
