The ABINIT Users Mailing List ( CLOSED )

[James Raynolds <JRaynolds@uamail.albany.edu>]

Marc,

Thank you very much.  I get the same error with version_4.3.3.  After the
problem with 4.2.4 I started fresh with 4.3.3 thinking I must have corrupted
something in the process of porting to the Mac.

One thing to note...The compiler complains that in Src_8seqpar there is an
inconsistent call to vresfo from scfcv.f (by the way, with the NAG compiler,
the fortran 90 routines need to have the suffix .f90 or .ff90 if invoking
the pre-processor).  The error is that argument 47 of vresfo, "pawfgr" has
the wrong type.  By comparing other calls to the same routine I assumed that
argument 47 should have been "pawfgr%ph1d" rather than "pawfgr" so I changed
it.

This is the only place where I have made a significant change to the source
code.  I'm betting that somehow this is related to the error I'm getting.

Also, the NAG compiler has its own pre-processor and in turn doesn't like
the output from the cc pre-processor.  I use the NAG pre-processor simply by
using the suffix .ff90

Again, thank you very much for your help.

Sincerely,
Jim Raynolds

-----Original Message-----
From: marc.torrent@cea.fr [mailto:marc.torrent@cea.fr] 
Sent: Monday, August 09, 2004 7:52 AM
To: forum@abinit.org
Subject: [abinit-forum] Re: abinit on Mac G5

Dear Jim,

The script you attached shows that you are using the 4.2.4 version of
Abinit...
PAWRAD is an array used in the development of PAW features... it changes
continuously.
So I think you may try the last version of Abinit.
If thew problem persists I will try to find its origin.


Marc Torrent
CEA- Bruyeres-le-Chatel
France

>Dear Mac G5 users,
>
>
>
>I have successfully compiled abinit on the Mac G5 using the NAG f95
compiler >(with a bit of effort).  However, when I run it on the test case
>./Test_in/test1.files I get the following error message complaining about
the >array pawrad:
>
>===========================================================================
=====
>== DATASET  1
==================================================================
>
>  getdim_nloc : COMMENT -
>  Despite there is only a local part to pseudopotential(s),
>  lmnmax and lnmax are set to 1.
> iofn2 : deduce lmnmax  =   1, lnmax  =   1,
>                lmnmaxso=   1, lnmaxso=   1.
>ALLOCATABLE array PAWRAD is not currently allocated
>Program terminated by fatal error
>Abort trap
>
>
>Anyone have an idea as to what is going wrong?
>I've attached the script file to this e-mail.
> 
>Sincerely,
>
>Jim Raynolds