[gmx-users] Segfault using Verlet-list with Gromacs 4.6.5 - MPI only

Sébastien Côté sebastien.cote.4 at umontreal.ca
Wed Apr 16 16:40:50 CEST 2014


Dear Mark,
I have opened an issue at http://redmine.gromacs.org and provided the requested files.
Thanks,
Sebastien 

> Date: Tue, 15 Apr 2014 12:17:05 +0200
> From: mark.j.abraham at gmail.com
> To: gmx-users at gromacs.org
> Subject: Re: [gmx-users] Segfault using Verlet-list with Gromacs 4.6.5 - MPI	only
> 
> On Apr 15, 2014 1:01 AM, "Sébastien Côté" <sebastien.cote.4 at umontreal.ca>
> wrote:
> >
> > Dear gmx developers,
> > I am trying to use Verlet list for MD simulations using MPI
> parallelization only with Gromacs 4.6.5. However, I am getting a segfault
> during the initialization of my system before any time step is executed.
> The simulation runs perfectly when I am either using Group list or Gromacs
> 4.6.1 instead.
> > Is their a problem known for verlet list in Gromacs 4.6.5 for some
> architecture?
> 
> No known problem on the information given. Can you please open an issue at
> http://redmine.gromacs.org, including a tarball of the log, tpr, and the
> files to build the tpr?
> 
> Mark
> 
> > Thanks for your help,
> > Sebastien
> > Below, you can find more specific information:
> > The error message thrown by the system:
> > starting mdrun 'bla'100 steps,      0.2 ps.[node-c3-42:09559] *** Process
> received signal ***[node-c3-42:09559] Signal: Segmentation fault
> (11)[node-c3-42:09559] Signal code:  (128)[node-c3-42:09559] Failing at
> address: (nil)[node-c3-42:09559] [ 0] /lib64/libpthread.so.0()
> [0x3a05c0f500][node-c3-42:09559] [ 1]
> /home/apps/Logiciels/gromacs/gromacs-4.6.5-no-plumed/bin/../lib/libmd_mpi.so.8(nbnxn_kernel_simd_4xn_tab_comb_lb_energrp+0x2223)
> [0x7f676aedc983][node-c3-42:09559] [ 2]
> /home/apps/Logiciels/gromacs/gromacs-4.6.5-no-plumed/bin/../lib/libmd_mpi.so.8(nbnxn_kernel_simd_4xn+0x4b1)
> [0x7f676aea9901][node-c3-42:09559] [ 3]
> /home/apps/intel/composerxe-2011.4.191/mkl/../compiler/lib/intel64/libiomp5.so(__kmp_invoke_microtask+0x93)
> [0x7f6768380323][node-c3-42:09559] *** End of error message
> ***[node-c3-42:09554] *** Process received signal ***
> > The log file finishes as follow:
> > Center of mass motion removal mode is LinearWe have the following groups
> for center of mass motion removal:  0:  Protein  1:  non-Protein
> > ++++ PLEASE READ AND CITE THE FOLLOWING REFERENCE ++++G. Bussi, D.
> Donadio and M. ParrinelloCanonical sampling through velocity rescalingJ.
> Chem. Phys. 126 (2007) pp. 014101-------- -------- --- Thank You ---
> -------- --------
> > System info in the log file:
> > Log file opened on Mon Apr 14 18:51:35 2014Host: node-c3-42  pid: 9551
>  nodeid: 0  nnodes:  12Gromacs version:    VERSION 4.6.5Precision:
>  singleMemory model:       64 bitMPI library:        MPIOpenMP support:
> enabledGPU support:        disabledinvsqrt routine:
>  gmx_software_invsqrt(x)CPU acceleration:   SSE4.1FFT library:
>  MKLLarge file support: enabledRDTSCP usage:       enabledBuilt on:
>   Mon Apr 14 09:57:30 EDT 2014Built by:
> rqchpbib at briaree1[CMAKE]Build OS/arch:      Linux 2.6.32-71.el6.x86_64
> x86_64Build CPU
> vendor:   GenuineIntelBuild CPU brand:    Intel(R) Xeon(R) CPU
> X5650  @ 2.67GHzBuild CPU family:   6   Model: 44   Stepping: 2Build CPU
> features: aes apic clfsh cmov cx8 cx16 htt lahf_lm mmx msr nonstop_tsc pcid
> pclmuldq pdcm pdpe1gb popcnt pse rdtscp sse2 sse3 sse4.1 sse4.2 ssse3C
> compiler:
> /RQusagers/apps/intel/composerxe-2011.4.191/bin/intel64/icc Intel icc (ICC)
> 12.0.4 20110427C compiler flags:   -msse4.1
> >     -mkl=sequential -std=gnu99 -Wall   -ip -funroll-all-loops  -O3
> -DNDEBUGLinked with Intel MKL version 10.3.4.
> >
> >
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
> -- 
> Gromacs Users mailing list
> 
> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
> 
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> 
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-request at gromacs.org.
 		 	   		  


More information about the gromacs.org_gmx-users mailing list