[Pw_forum] AIX + XML files

[Konstantin Kudin]

 Sorry, this is not a solution post, just another data point. On an
ancient IBM (aix 4.2 and xlf 6.1.1 or something, 32-bit compile) I
tried QE-3.1 and the CP code would always hang on the 2nd write to the
xml restart files (1st time the restart directory was saved fine). The
program kept consuming the resources though.

 With the version QE-3.1.1 things started working fine, and I had not
tried anything since.

 So perhaps someone could patch all these clues for IBM problems
together ...

 Kostya


--- Gianni Profeta <Gianni.Profeta at aquila.infn.it> wrote:

> Dear users,
> 
> I find the following problem on IBM AIX Version 5.1  running 
> parallel version of pw.x from versions > 3.1 (including the CVS
> version).
> 
> 1) Configure and compilation proceed without problems (as usual for
> this 
> type of machine and for all the versions of the code)
> 
> 2) running a test input (one k-point) the code stops with the
> following 
> error message just before the first scf iteration:
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>       from write_rho_xml : error #         1
>       cannot opentmp/Al.save/charge-density.xml file for writing
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
> 
> 3) increasing the k-point mesh, the code does not stop but remains 
> running (even if it does not produce outputs)
> 
> I think the problem is related to the iotk library, indeed 
> I found a post of Paolo regarding the incompatibility of iotk library
> in 
> AIX enviroment.
>
http://www.democritos.it/pipermail/pw_forum/2006-September/004912.html
> But the proposed solution does not work, in my case.
> 
> Any suggestions?
> 
> Thank you.
> 
> Gianni
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum
> 



 
____________________________________________________________________________________
Do you Yahoo!?
Everyone is raving about the all-new Yahoo! Mail beta.
http://new.mail.yahoo.com