The ABINIT Users Mailing List ( CLOSED )

[Alain Jacques <ajs.jacques@gmail.com>]

Hello Christian,

I have little experience with xlf so take the rest with a pinch of salt. Disabling wannier90 should prevent the processing of the corresponding module. But you have to be sure to start from a clean config. I recommend to work in a "build" directory - i.e. create a build dir, cd and ../configure from there - and use a make distclean each time the Makefiles have to be modified.
So do you still have that module error message if you untar abinit, cd abinit-xxx, mkdir build, cd build, ../configure --disable-wannier90 ... (add your own xlf flags) and make? If the build process stops, try to go on with another make. Another thing to try is to build sequentially.

Kind regards,

Alain

Ludwig, Christian wrote:

CB0DEF182A5E5644ACC163C5655ED2F0016F18BFAA56@EXCHANGE-01.zdv.uni-mainz.de" type="cite">

I tried ./configure --disable-wannier90 but I still get the same error message. And I did not find these build examples in the forum, although I searched for xlf and for Alain Jacques. Could you send me a link or copy/paste the relevant post?

Christian
________________________________________
Von: matthieu verstraete [matthieu.jean.verstraete@gmail.com]
Gesendet: Donnerstag, 23. April 2009 09:23
An: forum@abinit.org
Betreff: Re: [abinit-forum] problems with compilation on ibm power

--disable-wannier90 is what you want. Do a forum search on xlf and you
will get the build examples that Alain Jacques has posted.

type ../configure --help for all options.

Matthieu

On Wed, Apr 22, 2009 at 11:07 AM,  <ludwigc@uni-mainz.de> wrote:
  

Hello,

I am trying to compile abinit 5.4.4p on an IBM POWER5 machine under Linux and I
use the file powerpc64-ibm_fock.ac. Configure gives me the information, that my
compiler (XLF) has been wrapped.
Then, make runs until it gets to defs_wannier90_cpp.f90. It says "Unable to
create module symbol file defs_wannier90.mod. Please check path, directory and
permissions". This I did, path, dir and permissions are fine.

I suspect that it has something to do with preprocessing, since configure
complained about the highly non-standard preprocessing interface of XLF. But I
have no idea how to solve the problem. Do you have any ideas what I should do?

Cheers,
Christian

    

--
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Matthieu Verstraete

European Theoretical Spectroscopy Facility (ETSF)
Dpto. Fisica de Materiales,
U. del Pais Vasco,
Centro Joxe Mari Korta, Av. de Tolosa, 72,   Phone: +34-943018393
E-20018 Donostia-San Sebastian, Spain        Fax  : +34-943018390

Mail : matthieu.jean.verstraete@gmail.com
http://www-users.york.ac.uk/~mjv500