Neighbor list builds = 6 Dangerous builds = 0 , error

LAMMPS LAMMPS General Discussion
1

Hi Everyone,

I am runing lammps code for polymer chain. I am giving the input file command in the regular way adding some specific condition like

neighbor 3.5 multi
neigh_modify every 1 delay 0 check yes

Relax sim by performing minimisation

min_style fire
timestep 0.0000001
minimize 0.0 1.0e-8 1000 100000
timestep 0.00001
minimize 0.0 1.0e-8 1000 100000
timestep 0.1
minimize 0.0 1.0e-8 1000 100000
timestep 10
minimize 0.0 1.0e-8 1000 100000

fix fxnve all nve
fix fxlange all langevin 200 200 5000.0 548669595

I am getting error like this…

Total # of neighbors = 425088
Ave neighs/atom = 6.5197546
Ave special neighs/atom = 1.9877301
Neighbor list builds = 6
Dangerous builds = 0
malloc(): memory corruption
[loki:10938] *** Process received signal ***
[loki:10938] Signal: Aborted (6)
[loki:10938] Signal code: (-6)
[loki:10938] [ 0] /lib/x86_64-linux-gnu/libpthread.so.0(+0x12980)[0x7fa0320ff980]
[loki:10938] [ 1] /lib/x86_64-linux-gnu/libc.so.6(gsignal+0xc7)[0x7fa031d3ae87]
[loki:10938] [ 2] /lib/x86_64-linux-gnu/libc.so.6(abort+0x141)[0x7fa031d3c7f1]
[loki:10938] [ 3] /lib/x86_64-linux-gnu/libc.so.6(+0x89837)[0x7fa031d85837]
[loki:10938] [ 4] /lib/x86_64-linux-gnu/libc.so.6(+0x908ba)[0x7fa031d8c8ba]
[loki:10938] [ 5] /lib/x86_64-linux-gnu/libc.so.6(+0x94a04)[0x7fa031d90a04]
[loki:10938] [ 6] /lib/x86_64-linux-gnu/libc.so.6(+0x95965)[0x7fa031d91965]
[loki:10938] [ 7] /lib/x86_64-linux-gnu/libc.so.6(realloc+0x36b)[0x7fa031d94fbb]
[loki:10938] [ 8] lmp_mpi(+0x1b1c19)[0x557f34fffc19]
[loki:10938] [ 9] lmp_mpi(+0xb6f25)[0x557f34f04f25]
[loki:10938] [10] lmp_mpi(+0xbf940)[0x557f34f0d940]
[loki:10938] [11] lmp_mpi(+0x9711a)[0x557f34ee511a]
[loki:10938] [12] /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xe7)[0x7fa031d1dc87]
[loki:10938] [13] lmp_mpi(+0x9736a)[0x557f34ee536a]
[loki:10938] *** End of error message ***
Aborted (core dumped)

what this error mean?? is it need to install some package ?
Can you please help me to resolve this issue?

Santosh

2

You have triggered a bug in your LAMMPS version.

There is not enough information here to make any specific statements about what the bug is and how to fix it.

3

hey akohlmey…

thank you… I just recompile the lammps and now it is working.

Santosh

4

I don’t think that this will fix the bug, it may just lead to it not being detected or not causing memory corruption anymore due to stack randomization or similar.

5

it may be possible that bug is not resolve. now simulation is running . i am waitting for the simulation to complete or after analyze trajectrory , I will say something or inform you.

6

That is not how you can identify memory corruption. Typically, one would set up a test calculation and run it with valgrind’s memory checker tool or compile an instrumented binary with some sanitizer library and will check all memory accesses. If you have an off-by-one access issue for example, there will rarely be a crash or an incorrect result since many times the out of range access will be to unallocated, but otherwise unused memory. Also an uninitialized pointer variable that is not set to a null pointer in the constructor can lead to the kind of memory corruption report you have seen, but will not automatically happen since it may be (randomly) point to a viable location.