[Pw_forum] Problem with running CP
Amos Leffler
amosleffler at gmail.com
Mon Jun 15 18:44:53 CEST 2009
Hi Axel,
I tried your suggestions and the program ran just fine! I
traced the problem back to your comment about the tab characters.
When I replaced this input with spaces between input values it cured
the problem. Looking at the Users Group I noted a similar suggestion
by Paoli G. to somebody about a graphite problem.
This suggests that the input instructions eg. "INPUT_CP" should
be more specific about the input specifications. I am still new to
Fortran 95 which is different from the old Fortran 77 and so I didn't
realize this. This could be done without a lot of work and would be
helpful.
In any event my real thanks in solving this problem.
Amos Leffler
unaffiliated
On Fri, Jun 12, 2009 at 2:52 PM, Axel
Kohlmeyer<akohlmey at cmm.chem.upenn.edu> wrote:
> On Fri, 2009-06-12 at 14:12 -0700, Amos Leffler wrote:
>> Dear Axel,
>
> dear amos,
>
>
>> You are correct about the placement of CELL_PARAMETERS. I put
>> it after the ATOMIC_POSITIONS and just tried to run the first part of
>> the program but it still wont run. Do you know if anyone has tried to
>> use cp with a cubic cell?
>
> many have. me too. works just fine.
>
> it seems strange to me, however, that you use an explicit cell
> dimension statement for a primitive cubic cell, where specifying
> ibrav=1 and celldm(1)=16.0 would do the trick just fine.
>
>> Attached is the first input file and the output file so you can
>> see. It looks proper to me but it is easy to miss small things.There
>
> the first issue is that you put the EOF statement at the wrong
> place, it would have come _directly_ after the input.
>
> another problem, is that your file has tab characters instead
> of spaces in the coordinate section and that seems to throw off
> the inline parser for the atom positions.
>
> the next issue is, that you cannot run cp dynamics
> without an &ions namelist, even if you keep the ion
> positions fixed.
>
>
>> is no out file.
>
> there is no out file, because the line executing the
> code and redirecting the output is included in the
> .in file due to wrong placement of the EOF statement.
>
> please read up on shell scripting, if you mess with
> shell scripts, or modify and run .in files directly
> without using the wrapper scripts of the QE distribution.
>
> cheers,
> axel.
>
>> Amos Leffler
>> unaffiliated
>>
>> On Thu, Jun 11, 2009 at 3:07 PM, Axel
>> Kohlmeyer<akohlmey at cmm.chem.upenn.edu> wrote:
>> > On Thu, 2009-06-11 at 15:01 -0700, Amos Leffler wrote:
>> >> Dear Axel,
>> >> If what you suggest that the problem with my file is in the
>> >> the "bin/sh" statement which does the program run properly with the
>> >> file from example18?
>> >
>> > amos,
>> >
>> > good point. nevertheless, you must not remove the EOF entries.
>> > they are essential and that is what probably threw me off-track.
>> >
>> >> Another person sent me the output when he ran my file and it
>> >> did give a large number of errors all saying that the lines below EOF
>> >
>> > that e-mail is a complete fluke, since that person was obviously
>> > running the script as input to cp.x and not as a shell script
>> > in the proper location.
>> >
>> >> were ignored. I did get an error message saying that the program
>> >> expected celldm(1) but I thought this was not used with ibrav =0. ?
>> >
>> > please have a closer look at INPUT_CP.
>> >
>> > you are placing the CELL_PARAMETERS section _before_ ATOMIC_SPECIES,
>> > whereas the input description requires it to come _after_
>> > ATOMIC_POSITIONS. if you do not follow the input syntax,
>> > calculations won't start.
>> >
>> > cheers,
>> > axel.
>> >
>> >
>> >
>> >> Amos Leffler
>> >> unaffiliated
>> >>
>> >> On Tue, Jun 9, 2009 at 2:26 PM, Axel
>> >> Kohlmeyer<akohlmey at cmm.chem.upenn.edu> wrote:
>> >> > On Tue, 2009-06-09 at 13:34 -0700, Amos Leffler wrote:
>> >> >> Dear forum,
>> >> >
>> >> > dear amos,
>> >> >
>> >> >> I have been trying to run the attached file which is based on
>> >> >> that in Example18. The flle in Example18 runs properly but my
>> >> >
>> >> > what platform are you running on: OS type/version, CPU, fortran
>> >> > compiler?
>> >> >
>> >> >> modified version does not. I have noted the information in the "Users
>> >> >> Guide" , p28ff. If I leave out the EOF entries after each of the
>> >> >> "Atomic Positions" the entire file reads in but no output is
>> >> >
>> >> > this _cannot_ work. this is due to shell scripting and has little
>> >> > to do with pw.x/cp.x itself. there are some linux versions that use
>> >> > a /bin/sh implementation that is very minimalistic (but standard
>> >> > compliant), while the run_example scripts tend to depend some
>> >> > some common extensions. you have to leave the EOFs in, but you
>> >> > can try replacing /bin/sh with /bin/bash.
>> >> >
>> >> >> generated. Adding the EOF after the first Atomic_Positions, the output
>> >> >> stops with the message "running the calculation with fixed ions...\c"
>> >> >> . Is there more detailed information on using CP either in espresso
>> >> >> itself or elsewhere?
>> >> >
>> >> > there should be a large number of error messages.
>> >> >
>> >> > this has nothing to do with cp.x. the run_example script
>> >> > just automates the process of generating input files and
>> >> > running them. in general, it is better for manual runs to
>> >> > modify the individual input files, if you are not an expert
>> >> > in bourne shell scripting.
>> >> >
>> >> > cheers,
>> >> > axel.
>> >> >
>> >> >> Thanks for your help.
>> >> >>
>> >> >> Amos Leffler
>> >> >> _______________________________________________
>> >> >> Pw_forum mailing list
>> >> >> Pw_forum at pwscf.org
>> >> >> http://www.democritos.it/mailman/listinfo/pw_forum
>> >> >
>> >> > --
>> >> > =======================================================================
>> >> > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu
>> >> > Center for Molecular Modeling -- University of Pennsylvania
>> >> > Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
>> >> > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425
>> >> > =======================================================================
>> >> > If you make something idiot-proof, the universe creates a better idiot.
>> >> >
>> >> > _______________________________________________
>> >> > Pw_forum mailing list
>> >> > Pw_forum at pwscf.org
>> >> > http://www.democritos.it/mailman/listinfo/pw_forum
>> >> >
>> >> _______________________________________________
>> >> Pw_forum mailing list
>> >> Pw_forum at pwscf.org
>> >> http://www.democritos.it/mailman/listinfo/pw_forum
>> >
>> > --
>> > =======================================================================
>> > Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu
>> > Center for Molecular Modeling -- University of Pennsylvania
>> > Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
>> > tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425
>> > =======================================================================
>> > If you make something idiot-proof, the universe creates a better idiot.
>> >
>> > _______________________________________________
>> > Pw_forum mailing list
>> > Pw_forum at pwscf.org
>> > http://www.democritos.it/mailman/listinfo/pw_forum
>> >
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
>
> --
> =======================================================================
> Axel Kohlmeyer akohlmey at cmm.chem.upenn.edu http://www.cmm.upenn.edu
> Center for Molecular Modeling -- University of Pennsylvania
> Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
> tel: 1-215-898-1582, fax: 1-215-573-6233, office-tel: 1-215-898-5425
> =======================================================================
> If you make something idiot-proof, the universe creates a better idiot.
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
>
More information about the Pw_forum
mailing list