[Pw_forum] pp.x crashing with fixed occupations

Conor Hogan conor.hogan at roma2.infn.it
Wed Mar 21 19:44:34 CET 2007


Dear forum,

I'm using pp.x from QE3.2 to plot the wavefunctions for a spin polarized
calculation, with smearing and with "from_input" occupations. System is
the Cray XD1 at Cineca, mpich 1.2.6 and pgf90 6.0.1. I'm forced to use
more than one processor as the system is large. Code is compiled with
D__XD1 D__PGI D__MPI.

I get the good old Cray MPI message:
Fatal Error: message truncated. ask 4 got 1168 at line 863 in file
/tmp/igorodet/rpm/BUILD/mpich-1.2.6/mpid/rai/raifma.c

usually around the point where the OCCUPATIONS section is read. Everything
works ok if smearing is used, but not if FIXED_OCCUPATIONS is true (and
hence, f_inp is present). I can work around the problem by cutting the
whole OCCUPATIONS section out and replacing it with a section from a
smearing run, but I don't know if this will cause damage later on (can
it?). Trying to reduce the MAXSZ XD1 MPI parameter in Modules/mp.f90 didnt
help.

I had problems like this before with ifort and earlier versions of pp.x,
but not yet with pgf90. The only thing I can come up with is that
f_inp(nbnd,nspin) is the only module-declared quantity allocated in
pw_restart.f90 : I tried moving the allocation to an earlier section of
the code, but then I get other MPI errors when passing e.g. lgauss at this
point (here I got lost and gave up!).

Any suggestions/workarounds?

Best regards,
Conor

----

Dr. Conor Hogan                                --- -----  ---   ---
Dipartimento di Fisica e CNR-INFM             |      :   |     :
Universita' di Roma "Tor Vergata"             |--    :    ---  |---
Tel: +39 06 72594908                          |      :       | :
Fax: +39 06 2023507                            ---   :    ---  :
http://www.fisica.uniroma2.it/~cmtheo-group/  European Theoretical
                                              Spectroscopy Facility
                                              http://www.etsf.eu

The early bird gets the worm, but the second mouse gets the cheese. - S.
Wright

If you go through a lot of hammers each month, I don't think it necessarily
means you're a hard worker. It may just mean that you have a lot to learn
about proper hammer maintenance - J. Handey



More information about the users mailing list