Re: Re: namd 2.7b2 or later - amberff/prmtop causes segfault during startup phase

From: Marcello Sega (sega_at_science.unitn.it)
Date: Sat Apr 03 2010 - 19:35:01 CDT

Dear all,

I also had the same problem trying to use amber files with 2.7b2. I
had a cursory look at the code and it seems that the problem is the
variable numenerentries (Parameteres.C), which is never initialized
when reading amber parameters.

To me the quick fix

Parameters::Parameters(Ambertoppar *amber_data, BigReal vdw14)
{
   initialize();
   numenerentries=0;
   // Read in parm parameters
   read_parm(amber_data,vdw14);
}

in the constructor for reading amber parameter data did the job (i.e.
a simple test run gives the same results using b2 and b1)

*Beware* that I'm not sure if this is a proper solution or if on the
contrary it breaks something else (I'm using namd only since a couple
of days, and I don't know the internals: mine is just a wild guess) so
use it at your own risk...

Maybe somebody else could check this?

Cheers,

Marcello

On Wed, Mar 31, 2010 at 2:02 PM, Vlad Cojocaru <Vlad.Cojocaru_at_h-its.org> wrote:
> Dear Peter,
>
> I did not have the time to further test new CVS codes but if you still have
> the problem with the newest CVS codes .. I am afraid that the problem is not
> fixed yet.
> I post my reply to the list as well  maybe there are some updates from the
> NAMD developers on this .. Of course, this problem is a very important one
> for everybody who is using NAMD with AMBER ff ..
>
> Cheers
> Vlad
>
> Peter Schmidtke wrote:
>>
>> Dear Vlad,
>>
>> I saw your posts on the namd mailing list and wondered if you finally
>> solved the problem, I have the same right now and I did not find any
>> follow
>> up of the problem on the namd mailing list.
>> The amber ff part of my conf file looks like this :
>>
>> amber          on
>> parmfile       ../s_comp1.top
>> ambercoor      ../s_comp1.crd
>> outputname     s_comp1_min
>> exclude        scaled1-4
>> 1-4scaling     0.833333
>>
>> where s_comp1.top is a amber topology, and the crd file is an amber
>> coordinate file. Like in your case it yields a segmentation fault during
>> the starting phase.
>>
>> Thanks in advance for your help
>>
>> Best regards.
>>
>>
>>
>
> --
> ----------------------------------------------------------------------------
> Dr. Vlad Cojocaru
>
> Heidelberg Institute for Theoretical Studies
> HITS gGmbH
> Molecular and Cellular Modelling Group
> Schloss-Wolfsbrunnenweg 35
> 69118 Heidelberg
>
> Tel: ++49-6221-533202
> Fax: ++49-6221-533298
>
> e-mail:Vlad.Cojocaru[at]h-its.org
>
> http://projects.villa-bosch.de/mcm/people/cojocaru/
>
> ----------------------------------------------------------------------------
> Amtgericht Mannheim / HRB 337446
> Managing Directors: Dr. h.c. Klaus Tschira, Prof. Dr.-Ing. Andreas Reuter
> http://www.h-its.org (http://eml-research.de)
> ----------------------------------------------------------------------------
>
>

-- 
Dept. Physics, Trento University
via Sommarive 14 loc. Povo, I-38100 Trento
phone: +39 0461 88 1539

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:53:58 CST