[Wien] IBM AIX error

Wed Mar 20 23:30:06 CET 2013

Peter may have seen this before, I wonder if you have a standard csh.
Try  csh --version. For instance I get

tcsh 6.14.00 (Astron) 2005-03-25 (x86_64-unknown-linux) options
wide,nls,dl,al,kan,sm,rh,color,filec

Also, check the first line of case.in1 -- is it sensible or does it
have something odd (e.g. NaN) after EF=

The above may or may not be an issue.

I've forgotten where we are with this in terms of the mixer, can you
please tell me again what happens. Any difference with MSEC1 (or
MSEC3)? You will probably have to delete the *bro* files then run at
least 3 iterations before anything happens.

Perhaps also send the relevant case.scfm & case.outputm to my private email.

On Wed, Mar 20, 2013 at 5:18 PM, Oliver Albertini <ora at georgetown.edu> wrote:
> Here is the relevant info from adding the verbose flag to the x_lapw file:
>
>
> set lapw2test=`echo $exe | cut -c 1-5`
> if ( "$lapw2test" == 'lapw2' ) then
> if ( ! $?in1orig & ! $?band & ! $?emin1 & ! $?emax1 ) then
> if ( -e $file.scf2$sc$updn && ! -z $file.scf2$sc$updn ) then
> if ( $cmplx == 'c' ) then
> set eferm=`grep ':FER  :' $file.scf2$sc$updn|cut -f 2 -d=`
> set efold=`head -1 $file.in1$cmplx${sc}|cut -c11-17 | sed -e
> "s/[a-zA-Z(,=]//g"`
> if ( $#efold == 0 ) set efold=0.5
> set eferm=`echo " if($eferm > $efold) {if ($eferm > $efold + 0.2) {$efold +
> 0.2} else {$eferm} } else {if($eferm < $efold - 0.2) {$efold - 0.2} else
> ($eferm*0
> .75+$efold*0.25) }" | bc -l`
> syntax error on line 1 stdin
> set modus=`head -1 $file.in1$cmplx${sc} | cut -c1-5`
> echo "$modus  EF=$eferm   (WFFIL, WFPRI, ENFIL, SUPWF) " >
> $file.in1$cmplx${sc}_tmp
> tail -n +2 $file.in1$cmplx${sc} >> $file.in1$cmplx${sc}_tmp
> mv $file.in1$cmplx${sc}_tmp $file.in1$cmplx${sc}
> endif
> endif
> endif
>
>
>
> On Wed, Mar 20, 2013 at 3:13 PM, Oliver Albertini <ora at georgetown.edu>
> wrote:
>>
>> So after making a fresh initialization, the error remains the same. The
>> file uplapw2.error is indeed empty. The default on AIX is ksh, so I tried
>> both #!/bin/ksh -vp
>> and #!/bin/bsh -vf. The -p for Kourne Shell means to ignore .profile.
>>
>>
>> On Wed, Mar 20, 2013 at 12:58 PM, Laurence Marks
>> <L-marks at northwestern.edu> wrote:
>>>
>>> Did you run from a fresh initialization or use the old files? It could
>>> be that something went wrong before so perhaps everything is OK.
>>>
>>> Is the file uplapw2.error empty? If it is then it is something after
>>> that in lapw2para. You can edit the first line to be
>>> #!/bin/csh -vf
>>>
>>> (assuming that your csh is standard) and then do x lapw2 -up. You will
>>> get a mess of output but the bottom will give you an idea where it is
>>> failing. (You might want to do something like "x lapw2 -up >> junk"
>>> and then look at the file junk, and/or a different piping of the
>>> output which will depend upon what shell you are using.)
>>>
>>> On Wed, Mar 20, 2013 at 2:49 PM, Oliver Albertini <ora at georgetown.edu>
>>> wrote:
>>> > Thanks to you both,
>>> >
>>> > I am currently using a standard lapack library. You are right, the
>>> > routines
>>> > dggglm and dgelsy were not present in the SRC_lib version. Removing
>>> > essl
>>> > resulted in even more missing routines, so I tried replacing it with
>>> > standard 64 bit BLAS. It compiled, and I tried with the new mixer, but
>>> > the
>>> > same messages appeared as before.
>>> >
>>> > I am not sure what the meaning of 'syntax error on line 1 stdin' is.
>>> > Yes it
>>> > still happens when running x lapw2 -up.
>>> >
>>> >
>>> >
>>> >
>>> > On Wed, Mar 20, 2013 at 1:07 AM, Peter Blaha
>>> > <pblaha at theochem.tuwien.ac.at>
>>> > wrote:
>>> >>
>>> >> Unfortunately I do not have access to an AIX system, so cannot do
>>> >> anything
>>> >> myself.
>>> >>
>>> >>
>>> >>> I recently compiled 12.1 on AIX (v 6.1) pwr6. Like Luis, I also had
>>> >>> to
>>> >>> make some changes to SRC's in order to finish the compilation. These
>>> >>> were mostly issues with xlf like syntax. 9.2 was the most recent
>>> >>> version
>>> >>> before this.
>>> >>
>>> >>
>>> >> If these changes are different from those already mentioned on the
>>> >> mailing
>>> >> list (for gfortran; like comparing two logical variables with .eq.;
>>> >> ...),
>>> >> I'm always interested to know about them and if these changes are
>>> >> because of
>>> >> violations of the f90 standard (accepted  by ifort), I'll put them
>>> >> into the
>>> >> next release of WIEN2k.
>>> >>
>>> >>
>>> >>> STOP  LAPW2 END
>>> >>> syntax error on line 1 stdin
>>> >>
>>> >>
>>> >> You should try to find out what causes this message.
>>> >>
>>> >> x lapw2 -up
>>> >>
>>> >> does it also occur ? Then the problem is most likely in  x_lapw,
>>> >> otherwise
>>> >> in runsp_lapw
>>> >>
>>> >> -------------
>>> >> mixer: since mixer is not time-critical, I'd remove the essl, but link
>>> >> only with a "standard-lapack" from the web (or with -llapack_lapw from
>>> >> SRC_lib (not 100% sure, all required routines are in there, since this
>>> >> library contains only a subset of Lapack).
>>> >>
>>> >>
>>> >>
>>> >> --
>>> >>
>>> >>                                       P.Blaha
>>> >>
>>> >> --------------------------------------------------------------------------
>>> >> Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
>>> >> Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
>>> >> Email: blaha at theochem.tuwien.ac.at    WWW:
>>> >> http://info.tuwien.ac.at/theochem/
>>> >>
>>> >> --------------------------------------------------------------------------
>>> >> _______________________________________________
>>> >> Wien mailing list
>>> >> Wien at zeus.theochem.tuwien.ac.at
>>> >> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>> >
>>> >
>>>
>>>
>>>
>>> --
>>> Professor Laurence Marks
>>> Department of Materials Science and Engineering
>>> Northwestern University
>>> www.numis.northwestern.edu 1-847-491-3996
>>> "Research is to see what everybody else has seen, and to think what
>>> nobody else has thought"
>>> Albert Szent-Gyorgi
>>> _______________________________________________
>>> Wien mailing list
>>> Wien at zeus.theochem.tuwien.ac.at
>>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>
>>
>

-- 
Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu 1-847-491-3996
"Research is to see what everybody else has seen, and to think what
nobody else has thought"
Albert Szent-Gyorgi