Hi,
On Thu, 2 Mar 2006, David A. Case wrote:
> On Thu, Mar 02, 2006, B. Lachele Foley wrote:
>
> > The main complaint seems to be that certain variables are not
> > defined in the parallel libraries.
> > I set MPICH_HOME to /usr/local/programs/mpich-1.2.6.
>
> (Just a note, to everyone: the latest configure scripts follow the pmemd
> model, and just use a single environment variable (MPI_HOME) for all
forms
> of MPI.)
This may be a Minor nusiance (sorry spelling) for some because some
platforms already have environment variables defined, eg
module load mpich2
defines MPICH2_HOME, etc
> >
/usr/local/programs/mpich-1.2.6/lib/libmpich.a(comm_split.o)(.text+0xf7):
> > In function `MPI_Comm_split':
> > : undefined reference to `PMPI_Allreduce'
>
> It looks like you need to add -lpmpich to the LOADLIB string in
config.h.
> I'm guessing that some versions of mpich need this (earlier ones?
different
> architectures?, different communicators?) and others don't.
>
> I see that Bob Duke just includes "-lmpich -lpmpich" for mpich, whereas
> sander is asking for "-lfmpich -lmpich". Does anyone know if there is a
> universal correct answer to this, or a way (as with lam and openmpi) to
> tell from the installation itself what is required?
Yes, use -show, which still works for mpich2,
and this used to be in the configure:
from distributed amber8:
echo "MPICH_HOME is set to $MPICH_HOME"
mpichlibs=`$MPICH_HOME/bin/mpif90 -show | perl -p -e
's/(-[lL]\S+\s)|\S+\s/$1/g'`
loadlib="$loadlib $mpichlibs"
But it has been removed from both the mpich and mpich2 sections
in amber9's configure.
> [Aside: Bob's libraries for MPICH2 are also different than sander's.
Yet
> both seem to work for me on the same machine.... A significant number
of
> posts on the reflector have to do with parallel installation, so it
would
> be great to see if we can clean this up somehow.]
Several years ago there was talk, maybe mostly by John Mongan,
to support a parallel configure option that used whatever mpif90, etc
wrappers were in the user's path.
Naturally, this is double edged because it will
1] eliminate complaints from those installers who (perhaps magically to
them) have the right mpi stuff ready to go but otherwise wouldnt be able
to handle amber's simple configure requirements;
2] however, when it doesnt work it may encourage even more even dumber
reflector questions.
Aside, my philosophy in the COLUMBUS world has been to cater to
ready-to-go environments because I see those environments as
becoming more common. However, for COLUMBUS we need Global
Arrays installed which adds another level of installation
prerequsites.
Scott
Received on Wed Apr 05 2006 - 23:49:42 PDT
