[gmx-users] reg GPU Mdrun Error

[vidhya sankar]

respected  mark sir  ,
                 Thank you fro your previous reply 

When i run the production Mdrun  I have got the following error


job is terminating with segmentation fault error

Back Off! I just backed up CNTPEPRSOLIONSfullplumedGPUtest2.log to ./#CNTPEPRSOLIONSfullplumedGPUtest2.log.5#

Reading file CNTPEPRSOLIONSfullGPUtest2.tpr, VERSION 4.6.1 (single precision)
Using 3 MPI threads
Using 5 OpenMP threads per tMPI thread

3 GPUs detected:
  #0: NVIDIA Tesla K20m, compute cap.: 3.5, ECC:  no, stat: compatible
  #1: NVIDIA Tesla K20m, compute cap.: 3.5, ECC:  no, stat: compatible
  #2: NVIDIA Tesla K20m, compute cap.: 3.5, ECC:  no, stat: compatible

3 GPUs auto-selected for this run: #0, #1, #2


NOTE: The number of threads is not equal to the number of (logical) cores
      and the -pin option is set to auto: will not pin thread to cores.
      This can lead to significant performance degradation.
      Consider using -pin on (and -pinoffset in case you run multiple jobs).


Back Off! I just backed up CNTPEPRSOLIONSfullplumedGPUtest2.trr to ./#CNTPEPRSOLIONSfullplumedGPUtest2.trr.5#

Back Off! I just backed up CNTPEPRSOLIONSfullplumedGPUtest2.edr to ./#CNTPEPRSOLIONSfullplumedGPUtest2.edr.5#

WARNING: This run will generate roughly 27348 Mb of data

starting mdrun 'C225N99O45 in water'
10000000 steps,  20000.0 ps.
Segmentation fault (core dumped)

What is the meaning of the note ?

How to Give the command  to run? i gave as follows

mdrun -s CNTPEPRSOLIONSfullplumed.tpr   -plumed plumed1.dat     -nb  gpu    -v  -deffnm  CNTPEPRSOLIONSfull    -cpt 2



Thanks in Advance