[Wien] Problem in using different k-list!

[Nandan Tandon]

--
Reader
Department of Physics
University of Pune
Pune 411 007


---------------------------- Original Message ----------------------------
Subject: Problem in using different k-list
From:    "Dr. Anjali Kshirsagar" <anjali at physics.unipune.ernet.in> Date:
 Sat, November 3, 2007 4:58 pm
To:      wien at zeus.theochem.tuwien.ac.at
--------------------------------------------------------------------------

Dear Wien Users

I am facing a problem in using a different k-list than generated by the
kgen program in the 07 version of wien. I will elaborate the problem and I
urgently seek your help.

1. We wanted to use Monkhost mesh /special k-pony list for performing a
supercell calculation. We searched the mailing list but could not get an
easy solution to this problem, typically is there a list available with
the package or we have to generate it separately?

2. In earlier versions we have used the option of unit (itape) =4 to read
the k-points from the case.klist and   unit (itape) = 5 to read the
k-points from case.in1(c) file. But somehow this is not working for the
07.3 version. Does any one encounter this problem?

We get following error and the run stops.

==========================================================
forrtl: severe (64): input conversion error, unit 5, file
/horses/wien2k/working/CdSe/unit_5/unit_5.in1c
Image              PC        Routine            Line        Source lapw1c
          089DE4AC  Unknown               Unknown  Unknown lapw1c
   089DC9AD  Unknown               Unknown  Unknown lapw1c
089B066A  Unknown               Unknown  Unknown lapw1c
0898693E  Unknown               Unknown  Unknown lapw1c
08999FE9  Unknown               Unknown  Unknown lapw1c
080813F3  Unknown               Unknown  Unknown lapw1c
08081891  Unknown               Unknown  Unknown lapw1c
08057AF6  Unknown               Unknown  Unknown libc.so.6
00B81F2C  Unknown               Unknown  Unknown lapw1c
08057A31  Unknown               Unknown  Unknown 0.075u 0.006s 0:00.34
20.5%     0+0k 0+0io 1pf+0w
error: command   /horses/wien2k/wien2k07/lapw1c lapw1.def   failed
================================================================

Most probably, the ios flag is not zero and the program asks unit itape to
rewind. For case.klist file rewind starts reading from the start but when
it rewinds case.in1(c), the relevant data is not in place. This is my
analysis. Please comment.

3. I tried to modify the read statements and made them consistent with the
earlier version of the package. Then the lapw1 runs correctly but the
lapw2 gives the following error. It seems to be in the fourir.F file. But
I am not able to get why deallocate is not executed before allocating the
array again.

The only change we have is in the case. in2(c) file and it is for efmod
from TETRA to TEMP.

The program runs well with TETRA for the klist generated by kgen.

===============================================
 LAPW0 END
 LAPW1 END
forrtl: severe (151): allocatable array is already allocated
Image              PC        Routine            Line        Source lapw2c
          089CAF54  Unknown               Unknown  Unknown lapw2c
   089C96ED  Unknown               Unknown  Unknown lapw2c
0899D86A  Unknown               Unknown  Unknown lapw2c
08974DB8  Unknown               Unknown  Unknown lapw2c
0898A51C  Unknown               Unknown  Unknown lapw2c
0807D291  Unknown               Unknown  Unknown lapw2c
08096043  Unknown               Unknown  Unknown lapw2c
080571B6  Unknown               Unknown  Unknown libc.so.6
00B81F2C  Unknown               Unknown  Unknown lapw2c
080570F1  Unknown               Unknown  Unknown

>   stop error
====================================

Thanking you for any help

Regards

Anjali Kshirsagar
--
Reader
Department of Physics
University of Pune
Pune 411 007
India