[CP2K:567] Re: Cygwin port to windows command-line

Jack Shultz jackyg... at gmail.com
Sun Jan 13 20:36:03 UTC 2008


By the way the input was this example which worked fine when I ran it
through cygwin
C:\cygwin\usr\src\cp2k\exe\Cygwin-i686-g95>cp2k.sopt..\..\tests\NEB\regtest-1\2gly_B-
NEB.inp

I was just wondering if anyone had done something similar in terms of
porting it.

Jack


On 1/13/08, Jack Shultz <jackyg... at gmail.com> wrote:
>
> Looks like they are going to give me a fee waiver for Amber, plus I'm
> optimistic since they are telling me they are moving towards a license that
> will allow grid computing project such as mine.
>
> I also got to chating with Gromacs folk and they tell me there is a a GPL
> mopack version which would allow me to try some of the analysis I'm
> proposing.
>
> I may be superfically looking at a bunch of these chemistry packages for
> now. When I feel confident about one, I'll get more meticulous about all the
> examples as you point out. For now I am try to at least stumble around.
>
> Jack
>
>
>  On 1/13/08, Jack Shultz <jackyg... at gmail.com> wrote:
> >
> > Thank you Axel,
> >
> > Actually it is a necessary complication to implement on windows. I have
> > to answer two questions immediately:
> > 1) Whether this is an appropriate application for modeling reactions
> > 2) Whether it is feasible to integrate into my DC project. Like it or
> > not, windows computers are the dominant platform
> >
> > If I don't have answers to these two questions, I may regret spending
> > lots of time understanding how it works and designing my own simulation.
> >
> > Jack
> >
> > On 1/13/08, Axel <akoh... at gmail.com> wrote:
> > >
> > >
> > >
> > >
> > > On Jan 13, 5:07 am, "Jack Shultz" < jackyg... at gmail.com > wrote:
> > > > Hello,
> > > >
> > > > I am attempting to port the cygwin platform to windows command line.
> > > I've
> > > > done this with several other scientific apps by including the
> > > necessary
> > > > cygwin DLLs in the directory of the binary. But I get this error
> > > with CP2k.
> > > > I've also tried changing the stack size
> > >
> > > jack,
> > >
> > > this looks like an error in the input (or a problem with the input
> > > parser).
> > > what is the input file?
> > >
> > > i also suspect that there is nobody around here with significant
> > > experience (or interest) to debug cp2k on windows machines
> > > (it is _so_ much easier to run on linux). cp2k is rather demanding
> > > in terms of standards compliance and correctness of the compilers,
> > > libraries and input, that i'd strongly suggest to first try to get
> > > started
> > > with a linux machine (or virtual machine) and have one of those always
> > >
> > > ready to cross check. i generally compile/test on two different
> > > platforms
> > > and with two different compilers, to make sure that when i suspect a
> > > bug in cp2k, that is is not a bug in the compiler or library that is
> > > being
> > > exposed by cp2k. this has happened to me and others quite often.
> > >
> > > trying this on windows is simply adding unnecessary complications,
> > > IMO.
> > >
> > > cheers,
> > >    axel.
> > > >
> > > > C:\Program Files\Microsoft Visual Studio 8\VC\bin>editbin
> > > /STACK:33554432
> > > > cp2k.s
> > > > opt
> > > > Microsoft (R) COFF/PE Editor Version 8.00.50727.42
> > > > Copyright (C) Microsoft Corporation.  All rights reserved.
> > > >
> > > > Run time error:
> > > >
> > > >  ===== Routine Calling Stack =====
> > > >
> > > >             5 coordinate_control
> > > >             4 topology_control
> > > >             3 fist_init
> > > >             2 fist_create_force_env
> > > >             1 CP2K
> > > >  CP2K| Abnormal program termination, stopped by process number 0
> > > >     370 [sig] cp2k.sopt 288 open_stackdumpfile: Dumping stack trace
> > > to
> > > > cp2k.sopt
> > > > .stackdump
> > > >
> > > > C:\cygwin\usr\src\cp2k\exe\Cygwin-i686-g95>
> > > > > >
> > >
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20080113/f327e006/attachment.htm>


More information about the CP2K-user mailing list