<table cellspacing="0" cellpadding="0" border="0" ><tr><td valign="top" style="font: inherit;">Hi all,<br><br><br>Now I have been trying to make the parallel version work<br>I used the arch file (I used ifort 10.0 and mkl 10)<br><br>CC = icc<br>CPP = cpp<br>FC = /opt/ompi_ifort10/bin/mpif90 -FR<br>FC_fixed = /opt/ompi_ifort10/bin/mpif90 -FI<br>LD = /opt/ompi_ifort10/bin/mpif90 -i_dynamic -openmp<br>AR = ar -r<br>DFLAGS = -D__INTEL -D__FFTSG -D__parallel -D__BLACS -D__SCALAPACK -D__FFTW3<br>CPPFLAGS = -traditional -C $(DFLAGS) -P<br>FCFLAGS = $(DFLAGS) -O3 -Vaxlib<br>LDFLAGS = $(FCFLAGS) -L/opt/intel/mkl/10.0.1.014/lib/em64t/<br>LIBS=/opt/intel/mkl/10.0.1.014/lib/em64t/libmkl_scalapack.a /opt/intel/mkl/10.0.1.014/lib/em64t/libmkl_blacs_openmpi_ilp64.a
/opt/intel/mkl/10.0.1.014/lib/em64t/libmkl_em64t.a /opt/FFTW/fftw-3.1.3/.libs/libfftw3.a -lguide -lpthread<br><br><br>OBJECTS_ARCHITECTURE = machine_intel.o<br>
<br><br><br><br>The compilation was fine<br><br><br>I produced the binary cp2k.popt<br><br><br><br>Then I went to the tests directory<br><br>and I use the script<br><br>#!/bin/sh<br>###########################################################<br>#PBS -N TEST<br>#PBS -l walltime=100:00:00<br>#PBS -l nodes=1:ppn=1:node0<br>############################################################<br>cd ${PBS_O_WORKDIR}<br>/opt/ompi_ifort10/bin/mpirun -np 1 -npernode 1 -x OMP_NUM_THREADS=1 ~/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt 3H2O-ep.inp > out<br>exit 0<br>############################################################<br>~<br>~<br><br><br>But I got the following error <br><br>[shizuwo:27227] *** Process received signal ***<br>[shizuwo:27227] Signal: Segmentation fault (11)<br>[shizuwo:27227] Signal code: Address not mapped (1)<br>[shizuwo:27227] Failing at address: 0x100<br>[shizuwo:27227] [ 0] /lib64/tls/libpthread.so.0
[0x2a957bd650]<br>[shizuwo:27227] [ 1] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(cp_blacs_env_mp_cp_blacs_env_create_+0x37f) [0xe0325d]<br>[shizuwo:27227] [ 2] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(qs_environment_types_mp_qs_env_create_+0x3bc) [0x849c2c]<br>[shizuwo:27227] [ 3] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(qs_main_mp_quickstep_create_force_env_+0x7d) [0x8e1ec9]<br>[shizuwo:27227] [ 4] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(f77_interface_mp_create_force_env_+0x1431) [0x4c2ce3]<br>[shizuwo:27227] [ 5] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(ep_methods_mp_ep_env_init_low_+0x66a) [0x13fa502]<br>[shizuwo:27227] [ 6] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(cp_ep_init_+0x8) [0xf35cbc]<br>[shizuwo:27227] [ 7]
/home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(ep_types_mp_ep_env_create_+0x21b) [0xf0b783]<br>[shizuwo:27227] [ 8] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(force_env_methods_mp_ep_create_force_env_+0x46) [0x511210]<br>[shizuwo:27227] [ 9] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(f77_interface_mp_create_force_env_+0x1532) [0x4c2de4]<br>[shizuwo:27227] [10] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(cp2k_runs_mp_cp2k_run_+0x1498) [0x48dc5c]<br>[shizuwo:27227] [11] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(cp2k_runs_mp_run_input_+0x10a) [0x48c762]<br>[shizuwo:27227] [12] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(MAIN__+0xda5) [0x48b7c5]<br>[shizuwo:27227] [13] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(main+0x2a) [0x48aa12]<br>[shizuwo:27227] [14]
/lib64/tls/libc.so.6(__libc_start_main+0xea) [0x2a96d91aaa]<br>[shizuwo:27227] [15] /home/robechan/cp2k_new2/cp2k/exe/TEST_64_2_ifort10_mkl10/cp2k.popt(for_write_seq_fmt_xmit+0x42) [0x48a94a]<br>[shizuwo:27227] *** End of error message ***<br>mpirun noticed that job rank 0 with PID 27227 on node shizuwo exited on signal 11 (Segmentation fault).<br>~<br><br>and the following output in the file out<br><br><br>----------------------------------------------------------------------------------------------------------------<br><br><br>@shizuwo has created process number 0 27227<br><br> **** **** ****** ** PROGRAM STARTED AT Wed
Oct 22 15:24:26 2008<br> ***** ** *** *** ** PROGRAM STARTED ON shizuwo<br> ** **** ****** PROGRAM STARTED BY<br> ***** ** ** ** ** PROGRAM PROCESS ID 27227<br> **** ** ******* ** PROGRAM STARTED IN
/home/robechan/cp2k_new2/cp2k/tests/E<br> P<br><br> CP2K| version string: CP2K version 2.0.0 (Development Version)<br> CP2K| is freely available from http://cp2k.berlios.de/<br> CP2K| Program compiled at Wed Oct 22 15:11:53 JST 2008<br> CP2K| Program compiled
on shizuwo<br> CP2K| Program compiled for TEST_64_2_ifort10_mkl10<br> CP2K| Last CVS entry xc_xwpbe.F/1.12/Tue Aug 26 23:46:48 2008<br> CP2K| Input file
name 3H2O-ep.inp<br><br> GLOBAL| Force Environment number 1<br> GLOBAL| Basis set file name BASIS_SET<br> GLOBAL| Potential file
name POTENTIAL<br> GLOBAL| MM Potential file name MM_POTENTIAL<br> GLOBAL| Coordinate file name __STD_INPUT__<br> GLOBAL| Method
name CP2K<br> GLOBAL| Project name H2O<br> GLOBAL| Preferred FFT
library FFTW3<br> GLOBAL| Run type MD<br> GLOBAL| All-to-all communication in single precision F<br> GLOBAL| FFTs using library dependent
lengths F<br> GLOBAL| Global print level 1<br> GLOBAL| Total number of message passing processes 1<br> GLOBAL| Number of threads for this
process 1<br> GLOBAL| This output is from process 0<br><br><br> 15:24:26 NOTE in ep_methods:ep_env_init_low ep_methods:ep_env_init_low<br> creating main_qs_env<br><br>-------------------------------------------------------------------------------<br><br><br>The segmentation occurs because the programs is probably looking for something and does not find it any idea ?<br><br><br>Thanks<br><br><br><br>Roberto<br><br><br><br><br></td></tr></table><br>
<hr size=1><font face="Arial" size="2">Scopri il <a href="http://us.rd.yahoo.com/mail/it/taglines/yahoo/ymail/SIG=11djrg460/**http%3A%2F%2Fwww.ymailblogit.com%2Fblog%2F"> Blog di Yahoo! Mail</a>: trucchi, novità , consigli... e scrivi la tua opinione!</font>