[gmx-users] Re: gmx-users Digest, Vol 49, Issue 11

[David van der Spoel]

Ramadas Sunil Pophale wrote:
> Ramadas Sunil Pophale wrote:
>> Dear gromacs users,
>>
>> After configuring and installing gromacs, I tested it using 'make
> tests'.
>> Got the following errors.
>>
>> *** glibc detected *** realloc(): invalid next size: 0x08255310 ***
>> sh: line 1:  7535 Aborted       (core dumped) mdrun >mdrun.out 2>&1
>> FAILED. Check files in rb1
>> 1 out of 16 simple tests FAILED
>>
>> I checked the simple/rb1 directory for the .out files. I could not quite
>> see the problem described in those. Seems like it's related to 'allowing
> a
>> memory location that has already been freed to be used again'.
>>
>> Among the files I checked (checkptr.out, grompp.out, mdrun.out), the
>> closest thing to a warning seemed to be
>>
>> gcq#288: "I Feel a Great Disturbance in the Force" (The Emperor Strikes
>> Back) in grompp.out :-)
>>
>> is there anything in the script file (gmxtest.pl), I need to fix?
>>
>>
> -----------------------------------------------------------------------------
>> FAILED. Check files in dec+water
>> FAILED. Check files in nacl
>> FAILED. Check files in tip4p
>> 3 out of 14 complex tests FAILED
>>
>> As for the errors with complex tests, I had 3 of those. I checked the
>> checkpot.out files. Only in one case (dec+water) were the LJ numbers
> close
>> to zero (thus justifying the 'error due to numerical accuracy in adding
> up
>> many interactions' on wiki). In other 2 cases, the deviations in LJ and
> coloumb
>> energies were not small.
>>
>> for example
>>
>> LJ (SR)          step  50:       14440.2,  step  50:      13501.5
>> Coulomb (SR)     step  50:      -90740.2,  step  50:     -84154.3
>>
>> Not sure if I can simply ignore these. Any suggestions?
> 
> -----------------------------------------------------------------------
> 
> How did you compile? What platform are you running on?
> 
> -----------------------------------------------------------------------
> 
> Hi David,
> 
> I used RedHat-linux 4.0 with intel mkl library for fftw.
> The compilation went smoothly. Gromacs was installed in a non-standard
> location as I am not the root.

It seems that nacl and tip4p are the two tests using PME, and hence the 
mkl library could be the problem. Could you try it with FFTW?

> 
> Regards,
> 
> Ramdas.
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please search the archive at http://www.gromacs.org/search before posting!
> Please don't post (un)subscribe requests to the list. Use the 
> www interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php


-- 
David van der Spoel, Ph.D.
Molec. Biophys. group, Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205. Fax: +4618511755.
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se