I had tried the icc both static and dynamic (with same result). But,
I hadn't tried the gcc dynamic -- that was an omission due to a
dropped neuron or something and not intentional. I thought I'd done
it until I looked in my, er, "notes." :-)
gcc w/ no static compiled (I'm about to make test.parallel).
Yay, so far.
Thanks!
:-) Lachele
On 9/11/06, David A. Case <case.scripps.edu> wrote:
> On Mon, Sep 11, 2006, Lachele Foley (Lists) wrote:
>
> > Amber9 on an itanium machine (RHEL 3 build 4):
> > 2.4.21-27.EL #1 SMP Wed Dec 1 21:54:21 EST 2004 ia64 ia64 ia64 GNU/Linux
> >
> > The parallel compile fails with the following:
> >
> > ...
> > qm_div.o force.o \
> > ../lmod/lmod.a ../lapack/lapack.a ../blas/blas.a \
> > ../lib/nxtsec.o ../lib/sys.a -L/opt/scali/lib -lfmpi -lmpi
> > /opt/intel/mkl721//lib/64/libmkl_lapack.a
> > /opt/intel/mkl721//lib/64/libmkl_ipf.a -L/opt/intel/mkl721//lib/64
> > -lguide -lpthread -ldl
> > ld: cannot find -lfmpi
> >
> > The file libfmpi.so exists in /opt/scali/lib, and is
> > world-readable/executable.
>
> Just a guess: is it possible that you had the -static flag on the load line
> when the above error occurred? If so, remove it (to allow the linker to look
> for shared libraries). I gather you tried this, but maybe you get some
> different error(?) when -static is not present.
>
> ...dac
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber.scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
>
--
:-) Lachele
Lachele Foley
CCRC/UGA
2-0263
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Wed Sep 13 2006 - 06:07:11 PDT