[gmx-users] BUG: Free energy calculation

[Sabine Reisser]

Hi Mark,

with FE, without PR : same error
without FE, with PR: stable
without FE, without PR: stable

I've never had this error before.

Logfile says:
[...]
Initializing Domain Decomposition on 2 nodes
Dynamic load balancing: no
Will sort the charge groups at every domain (re)decomposition
Initial maximum inter charge-group distances:
     two-body bonded interactions: 3.341 nm, LJC Pairs NB, atoms 22476 22761
   multi-body bonded interactions: 0.649 nm, Angle, atoms 1668 1686
Minimum cell size due to bonded interactions: 3.675 nm
Using 0 separate PME nodes
Optimizing the DD grid for 2 cells with a minimum initial size of 3.675 nm
The maximum allowed number of cells is: X 1 Y 1 Z 2
Domain decomposition grid 1 x 1 x 2, separate PME nodes 0
PME domain decomposition: 2 x 1 x 1
Domain decomposition nodeid 0, coordinates 0 0 0
[...]
Linking all bonded interactions to atoms
There are 55376 inter charge-group exclusions,
will use an extra communication step for exclusion forces for PME

The initial number of communication pulses is: Z 1
The initial domain decomposition cell size is: Z 5.09 nm

The maximum allowed distance for charge groups involved in interactions is:
                  non-bonded interactions           1.200 nm
             two-body bonded interactions  (-rdd)   4.213 nm
           multi-body bonded interactions  (-rdd)   4.213 nm





There it stops. In the 1-thread case, the second part is replaced by 
"Initiating Steepest Descents" and then writing out the energies for 
every step.

Gromacs version is 4.5.5.

I also attached the whole logfile.

Cheers
Sabine




On 03/23/2012 11:52 AM, Mark Abraham wrote:
> On 23/03/2012 9:17 PM, Sabine Reisser wrote:
>    
>> Dear gromacs users/developers,
>>
>> when trying to couple in a peptide into a membrane with:
>>
>> ; Define position restraints for peptide
>> define          = -DPOSRES
>>
>> ; couple in peptide
>> free_energy     = yes
>> init_lambda     = 0.05
>> sc_alpha        = 0.7
>> sc_power        = 1
>> couple-moltype  = Protein
>> couple-lambda0  = none
>> couple-lambda1  = vdw-q
>>
>>
>> grompp works fine, but mdrun (2 threads) gives me
>>
>> Making 1D domain decomposition 1 x 1 x 2
>> *** glibc detected *** mdrun: realloc(): invalid next size:
>> 0x00007f0f30305810 ***
>>
>> and breaks up.
>>
>>
>> When running "mdrun -nt 1 " on only one thread, it works fine.
>>
>> Is this a known bug?
>>      
> First, is it likely not to be a problem with your setup... is your
> system stable in parallel without FE code? Without position restraints?
> What does your .log file say? What GROMACS version is it?
>
> Mark
>    

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: em_005_test.log
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120323/59efdbee/attachment.ksh>