[MITgcm-support] Problem with parallel build: No. of processes not equal to nPx*nPy

[Jean-Michel Campin]

Hi,

Just a comment here: It's not necessary to add "usingMPI=.TRUE."
in "eedata" as long as you generate the executable using
genmake2 with the "-mpi" option: 
> ../tools/genmake2 -mpi -of [] ...
The "-mpi" option of genmake2 turns on the 2 CPP options 
ALLOW_USE_MPI and ALWAYS_USE_MPI
and in this case the setting of the logical "usingMPI" is ignored.

Jean-Michel

On Sat, Nov 12, 2011 at 05:46:52AM -0800, Yuan Lian wrote:
> Did you run "openmpi -np 2 ./mitgcmuv"? If so, could you provide the
> messages right after you ran
> "genmake2 -mpi -mods your_code_directory -of
> linux_amd64_gfortran+mpi_generic"? Also make sure
> you are running with "usingMPI=.TRUE." in "eedata".
> 
> Yuan
> 
> 
> 
> On 11/12/11 5:32 AM, Chun-Yan Zhou wrote:
> >Dear Lian,
> > I tried as you suggested. But I got the same error:
> >No. of processes not equal to nPx*nPy 1 2
> >  Any other advice?
> > Thank you very much!
> > chunyan
> >---------------------------------------------------
> >On Nov 12, 2011 4:17 AM, lian at ashimaresearch.com
> ><lian at ashimaresearch.com> wrote
> >Try nPx=1 and sNx=80. The implementation of parallel in longitude
> >had problems, not sure if it has some last changes though
> >
> >>>> Chun-Yan Zhou 2011/11/12 11:14 >>>
> >
> >Hi Gus,
> >
> > Yes, it is the 'SIZE.h' file, type mistake. Any other idea?
> >
> > chunyan
> >
> >
> >
> >The University of Dundee is a registered Scottish charity, No: SC015096
> >
> >
> >
> >_______________________________________________
> >MITgcm-support mailing list
> >MITgcm-support at mitgcm.org
> >http://mitgcm.org/mailman/listinfo/mitgcm-support
> 

> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://mitgcm.org/mailman/listinfo/mitgcm-support