Hi,
On Tue, Aug 24, 2010 at 05:45:25PM -0400, Scott Brozell wrote:
> On Tue, Aug 24, 2010 at 03:48:32PM -0400, case wrote:
> > On Tue, Aug 24, 2010, Ross Walker wrote:
> >
> > > I am guessing you are using 11.0 or
> > > 11.1 which probably doesn't do this.
> >
> > Yes: I am using Intel 11.1. So it's good to know that this only affects a
> > subset of Intel users.
> >
> > I'll put the fix into git, and prepare bugfixes; but will hold off posting
> > them until others weigh in on the following point:
> >
> > > So yes we should probably change this to FOPTFLAGS and issue the bugfix. I
> > > am at ACS right now so it is a little hard to fully check this and make sure
> > > it doesn't break anything else - PGF90 or Gfortran for example but I can
> > > check later in the week if someone else doesn't have an opportunity to
> > > quickly try it.
>
> I'll test with these compilers tonight;
> and I'll communicate this bug to Tomko, Panda, et al.
I built and tested pmemd with
pgcc 9.0-4 64-bit target on x86-64 Linux -tp k8-64e
serial and parallel
gcc version 4.1.2 20071124 (Red Hat 4.1.2-42)
serial only.
I can confirm that this patch corrected the link line oversight
and didnt cause problems.
The parallel tests had some hangs:
mpiexec -n 2 ../../bin/pmemd.MPI -O -i gbin -c md12.x -o mdout.dhfr.noshake
mpiexec -n 2 ../../bin/pmemd.MPI -O -i mdin -p NMA_scee+scnb_box.prmtop -c NMA_box.inpcrd -o variable_14_ntb1.mdout
and sander.MPI hung in this test
cd evb/poh_dbonds_umb_dg_UFF_9DG_nmpimd_full_TST-freqf && ./Run.evb
I havent investigated these.
scott
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Fri Aug 27 2010 - 20:00:03 PDT
