[gmx-users] XT3 porting
David van der Spoel
spoel at xray.bmc.uu.se
Tue Apr 18 07:26:27 CEST 2006
Tom Joseph wrote:
> Providing a backtrace or two would really help. Given the exact same
> input files and run conditions, does it sometimes segfault and sometimes
> not? Does this happen on a login node (full glibc and such) or just the
> compute nodes (POSIX subset)?
> Separately, is it a better idea to use fftw2 since it supports MPI?
No, gromacs now does the parallel FFT itself, and fftw3 is slightly faster.
> (Hopefully soon my Teragrid access will extend to bigben, so I'm
> especially interested in this...)
> On Apr 17, 2006, at 10:06 PM, Shawn T. Brown wrote:
>> Hello, I am trying to port GROMACS 3.3.1 to the CRAY XT3, which is
>> essentially an Opteron machine with the PGI compilers. I can get
>> everything to compile, but when I run mdrun, I am “randomly” getting
>> segmentation faults. This seems to happen no matter what input file I
>> use. It happens in init_forcerec (line 1001), in add_gbond, and in
>> solvent_check. Some times it runs fine. I am configuring with the
>> following command:
>> ./configure --with-fft=fftw3 --disable-nice --enable-mpi
>> --program-suffix="_xt3" --enable-float --with-external-blas
>> --with-external-lapack --disable-threads
>> I am wondering if anyone else has seen these types of problems and
>> knows how to fix them. I have tried compiling the code without
>> optimization, to no avail.
>> Thanks in advance,
> gmx-users mailing list gmx-users at gromacs.org
> 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/mailing_lists/users.php
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596, 75124 Uppsala, Sweden
phone: 46 18 471 4205 fax: 46 18 511 755
spoel at xray.bmc.uu.se spoel at gromacs.org http://folding.bmc.uu.se
More information about the gromacs.org_gmx-users