On Thu, Mar 19, 2009 at 3:51 PM, John Peterson <[email protected]> wrote:
> I'm still waiting to see if this will link and actually run a program,
> but if you have a simple-to-set-up benchmark case you'd like me to try
> out on this machine I can give it a shot.  We have about 100k SUs
> there for testing/fooling around purposes.

Bah, it looks like they have a weird PETSc build.  It doesn't define
PACKAGES_LIBS in petscconf, and so we try to "guess" package
dependencies in Make.common.   And our guess is way wrong because it
appears that they have used a non-standard naming system for petsc
libraries.  Instead of the usual libpetscmat.a, libpetscvec.a, etc. in
$PETSC_DIR/lib I found instead:

libcmumps.a
libcraypetsc.a
libdmumps.a
libHYPRE.a
etc...

no vec or mat in sight, but the libcraypetsc.a looks promising...

-- 
John

------------------------------------------------------------------------------
Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
easily build your RIAs with Flex Builder, the Eclipse(TM)based development
software that enables intelligent coding and step-through debugging.
Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com
_______________________________________________
Libmesh-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-devel

Reply via email to