[Wien] Problems with mpi for Wien12.1

Laurence Marks L-marks at northwestern.edu
Wed Aug 29 05:39:04 CEST 2012


N.b., I meant lapw0 everywhere as I believe you said that is where the
problem is. If it is in lapw1, then change everything to lapw1 in my email.

---------------------------
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
 On Aug 28, 2012 10:34 PM, "Laurence Marks" <L-marks at northwestern.edu>
wrote:

> Hmmm. I was hoping for something human readable like a traceback showing
> where it died. Please check both the lapw0.error files and case.dayfile to
> see if they gave anything useful. Also, what are the last few lines of
> case.output0000?
>
> You may get somewhere by running the mpirun command by hand, I have seen
> this help. If you understand csh then you want to add an echo $tt at the
> relevant location in lapw0para.
>
> If not you can change the first line of lapw1para to "-xf" rather than
> just "-f". Then do x lapw0 -p again. You will get a hundred or so lines of
> output one of which towards the end will be something like
>
> "mpirun -np 12 ..."
>
> Then paste this line by itself in a terminal. Maybe then something human
> readable will emerge.
>
> Unfortunately debugging mpi is not trivial, and a SIGSEV can also be non
> trivial as the error may not appear at the right place, making life more
> fun.
>
> Do you gave totalview or a similar mpi debugger available? You can get a
> demo version of totalview free for I believe 30 days.
>
> ---------------------------
> 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
>  On Aug 28, 2012 10:09 PM, "Paul Fons" <paul-fons at aist.go.jp> wrote:
>
>>  I compiled fftw3 using the Intel suite as well.  The appropriate line
>> from config.log reads
>>
>> ./configure CC=icc F77=ifort MPICC=mpiicc --prefix=/opt/local
>> --enable-mpi --enable-threads --prefix=/opt/local/fftw3
>>
>>  I note that the configuration file only calls for a mpicc compiler (and
>> I used the Intel compiler) and not a fortran compiler.   The compiled code
>> (mpi-bench does work fine with the Intel mpirun).
>>
>>
>>  After commenting out the call W2kinit subroutine and recompiling lapw0
>> (via the siteconfig script), I attempted to run run_lapw in both serial and
>> parallel forms as you can see below.  The serial form worked fine
>>
>>  Paul
>>
>>  matstud at ursa:~/WienDisk/Fons/GaAs> run_lapw
>>  LAPW0 END
>>  LAPW1 END
>>  LAPW2 END
>>  CORE  END
>>  MIXER END
>> ec cc and fc_conv 1 1 1
>>
>>  >   stop
>>
>>
>>  matstud at ursa:~/WienDisk/Fons/GaAs> run_lapw -p
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>>
>>  >   stop error
>>
>>
>>
>>  On Aug 28, 2012, at 9:16 AM, Laurence Marks wrote:
>>
>>  One suggestion: comment out the line towards the top of lapw0.F
>>
>>      call W2kinit
>>
>> You should get a more human readable error message.
>>
>> As an addendum, was fftw3 compiled with mpiifort? I assume from your
>> email that it was, just checking.
>>
>> N.B., there is a small chance that this will hang your computer.
>> _______________________________________________
>> Wien mailing list
>> Wien at zeus.theochem.tuwien.ac.at
>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>>
>>
>>   Dr. Paul Fons
>>  Senior Research Scientist
>>  Functional Nano-phase-change Research Team
>>  Nanoelectronics Research Institute
>>  National Institute for Advanced Industrial Science & Technology
>>  METI
>>
>>  AIST Central 4, Higashi 1-1-1
>>  Tsukuba, Ibaraki JAPAN 305-8568
>>
>>  tel. +81-298-61-5636
>>  fax. +81-298-61-2939
>>
>>  email: *paul-fons at aist.go.jp*
>>
>>  The following lines are in a Japanese font
>>
>>  〒305-8562 茨城県つくば市つくば中央東 1-1-1
>>  産業技術総合研究所
>>  ナノエレクトロニクス研究部門
>>  相変化新規機能デバイス研究チーム
>>  主任研究員
>>  ポール・フォンス
>>
>>
>>
>>
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20120828/5c5ca9e3/attachment.htm>


More information about the Wien mailing list