[CP2K:1779] Grimme correction, segmentation fault

[Fawzi Mohamed]

seems to work here (I did 3 Optimization steps).
I suspect a badly compiled binary/library problem.
Try to run the regtests...

In general to see more output one can try to add
&GLOBAL
    FLUSH_SHOULD_FLUSH
&END

or run interactively without redirecting the output

Fawzi
On 17-feb-09, at 15:31, Andrzej wrote:

>
> Dear users,
> when I try to optimize a crystal structure using recently implemented
> Grimme correction and CG optimizer I got an error:
> /opt/torque/mom_priv/jobs/14329.arwen.SC: line 2: 24165 Segmentation
> fault      ./cp2k_d 60.in >60.log.
> Similar case, where Segmentation fault appeared during CG optimization
> has been discussed here, but for a large system treated with classical
> forces. It was suggested that initial structure was unphysical or
> gradients very large.
> In my case system has been pre-optimized and CG worked well for
> energies calculated without Grimme correction. The problem appears
> immediately when I introduce vdWPOTENTIAL. In fact, I can't find any
> clue in output, as a buffer is not released when the error appears.
>
> Can anybody be so kind giving me any advice?
>
> Andrzej
>