I will be out of the office starting 17/10/2007 and will not return until
29/10/2007.
I will respond to your message when I return. For urgent queries you can
contact my manager Ignesius Thambyraj. For publication/patent related
queries you can contact my colleague Rajan Ravindran.
Oops! Looks like we missed handling the --program-suffix stuff
properly when making our sym links. I've filed a ticket on the issue:
https://svn.open-mpi.org/trac/ompi/ticket/1168
On Oct 16, 2007, at 1:51 PM, Victor Prosolin wrote:
Hi all.
This is a bug report, though minor.
Briefly:
Hi all.
This is a bug report, though minor.
Briefly: when I configure openmpi with the option
--program-suffix=-gcc-4.3 I get a bunch of symbolic links that point to
non-existing files.
Here is then configure command:
FC=gfortran-4.3 F77=gfortran-4.3 CC=gcc-4.3 ./configure
--enable-mpi-f77 --ena
On 10/16/07, Jeff Squyres wrote:
> On Oct 16, 2007, at 11:20 AM, Brian Granger wrote:
>
> > Wow, that is quite a study of the different options. I will spend
> > some time looking over things to better understand the (complex)
> > situation. I will also talk with Lisandro Dalcin about what he th
On Oct 16, 2007, at 11:56 AM, Jeff Squyres wrote:
On Oct 16, 2007, at 11:20 AM, Brian Granger wrote:
Wow, that is quite a study of the different options. I will spend
some time looking over things to better understand the (complex)
situation. I will also talk with Lisandro Dalcin about what
On Oct 16, 2007, at 11:20 AM, Brian Granger wrote:
Wow, that is quite a study of the different options. I will spend
some time looking over things to better understand the (complex)
situation. I will also talk with Lisandro Dalcin about what he thinks
the best approach is for mpi4py. One ques
Jeff,
Wow, that is quite a study of the different options. I will spend
some time looking over things to better understand the (complex)
situation. I will also talk with Lisandro Dalcin about what he thinks
the best approach is for mpi4py. One question though. You said that
nothing had changed
On Oct 12, 2007, at 3:5 PM, Brian Granger wrote:
> My guess is that Rmpi is dynamically loading libmpi.so, but not
> specifying the RTLD_GLOBAL flag. This means that libmpi.so is not
> available to the components the way it should be, and all goes
> downhill from there. It only mostly works becau
Jeff,
Thank you for your thoughts. Some more comments inlined.
Regards,
--Oleg
On 10/9/07, Jeff Squyres wrote:
>
> Interesting idea.
>
> One obvious solution would be to mpirun your controller tasks and, as
> you mentioned, use MPI to communicate between them. Then you can use
> MPI_COMM_SPAWN
