Hello Laurence,
thanks for your comments. I am the wrong address to do that. I am really
a beginner in reading and writing FORTRAN source codes. I tried to
insert your suggestions in lapw0.F. The compilation failed. I think
other users are much more dedicated and while writing it I just saw that
Gavin has added a patch for atom_write.f and lapw0.f.
Best regards,
Michael
Am 11.07.2025 um 18:07 schrieb Laurence Marks:
Ifx is built to offload to different types of cpu's, including Nvidia
etc. Hence I think it has a different omp structure. There are
indications of stacksize issues with omp which can be patched by using
the environmental variable OMP_STACKSIZE or the iIntel Fortran
extensions kmp_get_stacksize() and kmp_set_stacksize(). If you look in
lapw1.F you will see that the threads stacksize is increased for that
code.
I suggest adding early
#if defined(_OPENMP)
INCLUDE 'omp_lib.h'
#endif
and a bit later
istat = kmp_get_stacksize()
write(*,*)istat
then try using different OMP_STACKSIZE environmental variable values.
If my hunch is right then this is the root cause, and something along
the lines of a kmp_set_stacksize will be a permanent solution.
N.B., if you try this please let me know as I think the listserver may
be failing to distribute my messages.
On Fri, Jul 11, 2025 at 9:55 AM Michael Fechtelkord
<[email protected]> wrote:
Thanks Peter,
so you mean the altering of atom_write.f in SRC_dstart with
write(51,2022) ncom1 and
2022 FORMAT(1x,'LM-max:'1x,I3)
So I will use that.
Thanks again!
Am 11.07.2025 um 15:29 schrieb Peter Blaha:
> No,
> this is not the best solution. There are low symmetries (and a
long LM
> list in case.in2) where NCOM=261 is necessary.
>
> I suggest the fix in dstart as I wrote before.
>
> I will also change the lapw0 code and put lmmaxx=121 (if it was
zero),
> but no longer use NCOM.
>
>> I changed param.inc in SRC_lapw0. I changed the line
>>
>> PARAMETER (NCOM= 261) !181
>>
>> to
>>
>> PARAMETER (NCOM= 121) !181--
>
-----------------------------------------------------------------------
> Peter Blaha, Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna
> Phone: +43-158801165300
> Email: [email protected]
> WWW: http://www.imc.tuwien.ac.at WIEN2k: http://www.wien2k.at
>
-------------------------------------------------------------------------
>
> _______________________________________________
> Wien mailing list
> [email protected]
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:
>
http://www.mail-archive.com/[email protected]/index.html
--
Dr. Michael Fechtelkord
Institut für Geologie, Mineralogie und Geophysik
Ruhr-Universität Bochum
Universitätsstr. 150
D-44780 Bochum
Phone: +49 (234) 32-24380
Fax: +49 (234) 32-04380
Email: [email protected]
Web Page:
https://www.ruhr-uni-bochum.de/kristallographie/kc/mitarbeiter/fechtelkord/
_______________________________________________
Wien mailing list
[email protected]
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:
http://www.mail-archive.com/[email protected]/index.html
--
Emeritus Professor Laurence Marks (Laurie)
Northwestern University
Webpage <http://www.numis.northwestern.edu> and Google Scholar link
<http://scholar.google.com/citations?user=zmHhI9gAAAAJ&hl=en>
"Research is to see what everybody else has seen, and to think what
nobody else has thought", Albert Szent-Györgyi
_______________________________________________
Wien mailing list
[email protected]
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST
at:http://www.mail-archive.com/[email protected]/index.html
--
Dr. Michael Fechtelkord
Institut für Geologie, Mineralogie und Geophysik
Ruhr-Universität Bochum
Universitätsstr. 150
D-44780 Bochum
Phone: +49 (234) 32-24380
Fax: +49 (234) 32-04380
Email:[email protected]
Web
Page:https://www.ruhr-uni-bochum.de/kristallographie/kc/mitarbeiter/fechtelkord/
_______________________________________________
Wien mailing list
[email protected]
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:
http://www.mail-archive.com/[email protected]/index.html
