floating point exception (core dump)

Pikachu

unread,

Mar 31, 2016, 10:04:36 AM3/31/16

to hoomd-users

Hi all,

I tried to run a system with ~150000 particles in a simulation box (box size=80*80*80), DPD combined with rigid body method.

I have successfully run my script on a GeForce GTX680 (hoomd 1.0.5 cuda7.0 centos7.1 gcc-4.8.3).

However, when I run this script on a GeForce GTX780 (hoomd 1.0.5[the same version], cuda7.0 centos7.1 gcc-4.8.3),

I get a 'Floating point exception (core dumped)' error.

The output is as follows. I would be grateful for any suggestions.

0.75.py:003 | init.read_xml('particlesna.xml')

HOOMD-blue is running on the following GPU(s):

[0] GeForce GTX 780 12 SM_3.5 @ 0.901 GHz, 3071 MiB DRAM

notice(2): Reading particlesna.xml...

notice(2): --- hoomd_xml file read summary

notice(2): 1546100 positions at timestep 0

notice(2): 6 particle types

notice(2): 1546100 particle body values

notice(2): 24000 bonds

notice(2): Group "all" created containing 1546100 particles

0.75.py:005 | dpd = pair.dpd(r_cut=1.0, T=1.5)

0.75.py:006 | dpd.pair_coeff.set('A', 'A', A=25.0, gamma = 4.5)

0.75.py:007 | dpd.pair_coeff.set('A', 'B', A=60.0, gamma = 4.5)

0.75.py:008 | dpd.pair_coeff.set('A', 'N', A=20.0, gamma = 4.5)

0.75.py:009 | dpd.pair_coeff.set('A', 'W', A=20.0, gamma = 4.5)

0.75.py:010 | dpd.pair_coeff.set('B', 'B', A=25.0, gamma = 4.5)

0.75.py:011 | dpd.pair_coeff.set('B', 'N', A=60.0, gamma = 4.5)

0.75.py:012 | dpd.pair_coeff.set('B', 'W', A=60.0, gamma = 4.5)

0.75.py:013 | dpd.pair_coeff.set('N', 'W', A=26.0, gamma = 4.5)

0.75.py:014 | dpd.pair_coeff.set('N', 'N', A=25.0, gamma = 4.5)

0.75.py:015 | dpd.pair_coeff.set('W', 'W', A=25.0, gamma = 4.5)

0.75.py:016 | dpd.pair_coeff.set('I', 'A', A=200.0, gamma = 4.5)

0.75.py:017 | dpd.pair_coeff.set('I', 'B', A=200.0, gamma = 4.5)

0.75.py:018 | dpd.pair_coeff.set('I', 'N', A=200.0, gamma = 4.5)

0.75.py:019 | dpd.pair_coeff.set('I', 'W', A=200.0, gamma = 4.5)

0.75.py:020 | dpd.pair_coeff.set('I', 'I', A=25.0, gamma = 4.5)

0.75.py:021 | dpd.pair_coeff.set('Z', 'A', A=200.0, gamma = 4.5)

0.75.py:022 | dpd.pair_coeff.set('Z', 'B', A=200.0, gamma = 4.5)

0.75.py:023 | dpd.pair_coeff.set('Z', 'N', A=200.0, gamma = 4.5)

0.75.py:024 | dpd.pair_coeff.set('Z', 'W', A=200.0, gamma = 4.5)

0.75.py:025 | dpd.pair_coeff.set('Z', 'I', A=200.0, gamma = 4.5)

0.75.py:026 | dpd.pair_coeff.set('Z', 'Z', A=25.0, gamma = 4.5)

0.75.py:027 | dpd.set_params(T = 1.0)

0.75.py:028 | nlist.reset_exclusions(exclusions = ['body'])

notice(2): -- Neighborlist exclusion statistics -- :

notice(2): Particles with 0 exclusions : 1546100

notice(2): Neighbors excluded by diameter (slj) : no

notice(2): Neighbors excluded when in the same body: yes

0.75.py:033 | rigid = group.rigid()

notice(2): Group "rigid" created containing 20100 particles

0.75.py:035 | harmonic = bond.harmonic()

0.75.py:038 | harmonic.bond_coeff.set('polymer', k=6.0, r0=0)

0.75.py:041 | integrate.mode_standard(dt=0.02)

0.75.py:042 | dpdp = integrate.nve(group=group.nonrigid())

notice(2): Group "nonrigid" created containing 1526000 particles

0.75.py:042 | dpdp = integrate.nve(group=group.nonrigid())

0.75.py:043 | dpdr = integrate.nve_rigid(group=rigid)

0.75.py:047 | sorter.set_period(100)

0.75.py:048 | zeroer= update.zero_momentum(period=1E2)

0.75.py:051 | mol2 = dump.mol2()

0.75.py:052 | mol2.write(filename='xxf.mol2')

0.75.py:053 | dcd = dump.dcd(filename="trajectory.dcd", period=1E4, overwrite=True)

0.75.py:055 | loggerE = analyze.log(filename='energy.log', period=1000, quantities=['pair_dpd_energy','bond_harmonic_energy', 'kinetic_energy'], overwrite=True)

0.75.py:056 | loggerT = analyze.log(filename = 'temperature.log', period=1E3, quantities=['temperature'], overwrite=True)

0.75.py:058 | loggerM = analyze.log(filename = 'momentum.log', period = 100, quantities=['momentum'],header_prefix='#', overwrite=True)

0.75.py:060 | loggerETC = analyze.log(filename = 'etc.log', period = 100, quantities=['volume', 'pressure','pressure_xx', 'pressure_yy', 'pressure_zz','pressure_xy','pressure_xz','pressure_yz'],header_prefix='#', overwrite=True)

0.75.py:064 | dump.bin(file1="restart.1.bin.gz", file2="restart.2.bin.gz", period=1e5)

*Warning*: dump.bin is deprecated and will be removed in the next release*Warning*: dump.bin does not support triclinic boxes.

*Warning*: dump.bin is deprecated and will be replaced in v1.1.0

0.75.py:065 | run(300000)

** starting run **

Floating point exception (core dumped)

Joshua Anderson

unread,

Mar 31, 2016, 10:08:46 AM3/31/16

to hoomd...@googlegroups.com

Please upgrade to the latest version of hoomd and let us know if that helps.

------
Joshua A. Anderson, Ph.D.
Research Area Specialist, Chemical Engineering, University of Michigan
Phone: 734-647-8244
http://www-personal.umich.edu/~joaander/

--
You received this message because you are subscribed to the Google Groups "hoomd-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hoomd-users...@googlegroups.com.
To post to this group, send email to hoomd...@googlegroups.com.
Visit this group at https://groups.google.com/group/hoomd-users.
For more options, visit https://groups.google.com/d/optout.

Pikachu

unread,

Mar 31, 2016, 11:11:05 PM3/31/16

to hoomd-users

I have intstalled the lastest verision of hoomd (glotzer-hoomd-blue-c2220dd51450.tar.bz2) today, but still get a "floating point exception" error on GeForce GTX 780 just like before.

However, the script can be successfully executed on a GeForce GTX 680 card without any error or warning.

I run hoomd like this: hoomd 0.75.py --gpu=1 > a.log&.

Then the job crashes and give me a message in the terminal:

[1]+ Floating point exception (core dumped) hoomd 0.75.py --gpu=1 > a.log,

in the a.log file, there are just some normal information, no this line.

I was thinking that this error is caused by different gcc version, but the 680 and 780 machine installed the same CentOS7 linux with the same version of gcc.