compiling PETSC with mvapich

[Matthew Knepley]

On 6/28/07, Randall Mackie <randy at geosystem.us> wrote:
> I was able to run it on one processor, using the mpirun_rsh command as follows:
>
> /usr/local/ofed/mpi/intel/mvapich-0.9.7-mlx2.2.0/bin/mpirun_rsh -np 1 hpcluster.local ./conftest
>
> I am now running ./reconfigure.py as instructed.
>
> What I don't understand is why the configure tool is unable to execute binaries created
> with C.

Send the original configure.log to petsc-maint and we can try and figure it out.

  MAtt

> Has anyone else used mvapich? (this is for use on infiniband).
>
> Randy
>
> Matthew Knepley wrote:
> > On 6/28/07, Randall Mackie <randy at geosystem.us> wrote:
> >> I am trying to build PETSC with mvapich, and in the configure step, I
> >> am getting the message
> >>
> >> *********************************************************************************
> >>
> >>                      UNABLE to EXECUTE BINARIES for config/configure.py
> >> ---------------------------------------------------------------------------------------
> >>
> >> Cannot run executables created with C. It is likely that you will need
> >> to configure using --with-batch which allows
> >> configuration without interactive sessions.
> >>
> >>
> >>
> >> So if I add the --with-batch, and then --with-mpi-shared=1, then I get
> >> the following messages:
> >>
> >> =================================================================================
> >>
> >>      Since your compute nodes require use of a batch system or mpiexec
> >> you must:
> >>   1) Submit ./conftest to 1 processor of your batch system or system
> >> you are
> >>      cross-compiling for; this will generate the file reconfigure.py
> >>   2) Run ./reconfigure.py (to complete the configure process).
> >> =================================================================================
> >>
> >>
> >>
> >> However, using mvapich requires at least 2 processors, so I appear to
> >> be stuck.
> >
> > Wow, that sounds like a really broken way to operate. However, running
> > on two
> > processors might be alright, depending on whether the file system is
> > shared. Each
> > process will open 'reconfigure.py'. If they do not overwrite each
> > other, you are fine.
> > Are you sure things cnanot be run on 1 proc?
> >
> >  Thanks,
> >
> >   Matt
> >
> >> Any advice?
> >>
> >> Thanks, Randy
> >>
> >>
> >> --
> >> Randall Mackie
> >> GSY-USA, Inc.
> >> PMB# 643
> >> 2261 Market St.,
> >> San Francisco, CA 94114-1600
> >> Tel (415) 469-8649
> >> Fax (415) 469-5044
> >>
> >> California Registered Geophysicist
> >> License No. GP 1034
> >>
> >>
> >
> >
>
> --
> Randall Mackie
> GSY-USA, Inc.
> PMB# 643
> 2261 Market St.,
> San Francisco, CA 94114-1600
> Tel (415) 469-8649
> Fax (415) 469-5044
>
> California Registered Geophysicist
> License No. GP 1034
>
>


-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which
their experiments lead.
-- Norbert Wiener