Hi everyone.
just my build/testing results.
amd64 linux (gentoo); gcc 4.3.4 (with some patches probably),
ifort 9.1.036, mpich-1.2.7p1-ch_shmem
1. to speed up the process and adapt the configure to my system,
used the attached configure.diff followed by "configure -mpi intel"
2. cd src/sander && make sander.MPI && cp -p sander.MPI ../../exe
cd src/pmemd/src && make pmemd.MPI && cp -p pmemd.MPI ../../../exe
[or something similar]
3. ran the tests with 4 threads:
214 file comparisons passed
5 file comparisons failed
8 tests experienced errors
some errors are simply due to the incomplete build (no LES, etc),
but there were some "failures" that look more or less real:
==============================================================
cd tgtmd/change_target.ntr && ./Run.tgtmd
SANDER: Targeted MD with changing target and restraints
Must have more residues than processors!
Must have more residues than processors!
Must have more residues than processors!
[2] MPI Abort by user Aborting program !
[1] MPI Abort by user Aborting program !
[3] MPI Abort by user Aborting program !
==============================================================
cd softcore && ./Run_sc
Running the Softcore potential tests
==============================================================
Minimization test
Running multisander version of sander Amber11
Total processors = 4
Number of groups = 2
[0] MPI Abort by user Aborting program !
[0] Aborting program!
forrtl: error (69): process interrupted (SIGINT)
./Run.min: Program error
forrtl: error (69): process interrupted (SIGINT)
Protein-Ligand complex test
forrtl: error (69): process interrupted (SIGINT)
Running multisander version of sander Amber11
Total processors = 4
it looks like not a failure, but still:
SANDER BOMB in subroutine imin > 0 and numtasks > 1
TI minimizations cannot be performed with > 2 CPUs
==============================================================
cd ti_decomp; ./Run.ti_decomp
Running multisander version of sander Amber11
Total processors = 4
Number of groups = 2
diffing ti_decomp_1.out.save with ti_decomp_1.out
possible FAILURE: check ti_decomp_1.out.dif
(please find it attached; it's ~1M ... sorry)
Have a nice day,
Volodymyr
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Mon Mar 29 2010 - 10:30:02 PDT
