[gmx-users] GROMACS 4.5.1 mdrun re-compile for MPI

[Carsten Kutzner]

Hi,

you could try a make clean, and then configure again with --enable-threads.
It seems for some reason you only build the serial mdrun version.

Carsten


On Mar 24, 2011, at 1:51 PM, Adam Herbst wrote:

> Dear GROMACS users,
> I successfully installed GROMACS 4.5.1 several months ago on a Mac Pro with 12 CPUs, and the "mdrun" command (not "mpirun mdrun_mpi") allows parallel simulations--it automatically uses multiple processors, while the number of processors can be manually specified as N with the flag "mdrun -nt N".  I understand that this is a feature of GROMACS 4 and later.  Now I am making minor changes to the mdrun source code, and I want to recompile such that the parallel version of mdrun is updated with my changes.  But when I run:
> 
>   make mdrun (or just make)
>   make install-mdrun (or just make install)
> 
> from the top-level source directory, the only executables that are updated are the ones with the _mpi suffix, such as mdrun_mpi.  The version of mdrun in src/kernel/ is updated, but this one has no -nt flag and cannot seem to run on multiple processors.  And when I run
> 
>   mpirun -np N mdrun_mpi [options],
> 
> the same simulation is started separately on each processor, leading to a crash.  If I use
> 
>   mpirun -np 1 -cpus-per-proc N mdrun_mpi [options],
> 
> I get an error message that this is not supported on my computer ("An attempt to set processor affinity has failed").
> 
> I can't configure the input .tpr file for parallel because grompp doesn't have the -np flag in GROMACS 4.
> 
> How can I update the parallel-capable "mdrun" executable with my changes?
> Thanks in advance,
> 
> Adam
> -- 
> gmx-users mailing list    gmx-users at gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists


--
Dr. Carsten Kutzner
Max Planck Institute for Biophysical Chemistry
Theoretical and Computational Biophysics
Am Fassberg 11, 37077 Goettingen, Germany
Tel. +49-551-2012313, Fax: +49-551-2012302
http://www.mpibpc.mpg.de/home/grubmueller/ihp/ckutzne