problem with lam-mpi and ifort

Questions regarding the compilation of VASP on various platforms: hardware, compilers and libraries, etc.


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
admin
Administrator
Administrator
Posts: 2921
Joined: Tue Aug 03, 2004 8:18 am
License Nr.: 458

problem with lam-mpi and ifort

#1 Post by admin » Mon Oct 25, 2004 1:06 pm

alternatively to switching back to an old ifc version, you could choose ZPOTRF
from a different LAPACK package:
usually, the package delivered with vasp (lapack_double.o in vasp.4.lib) works fine,
but depending on your installation, you might also try a version
like mkl-lapack,...
Last edited by admin on Mon Oct 25, 2004 1:06 pm, edited 1 time in total.

ata
Newbie
Newbie
Posts: 2
Joined: Thu Oct 21, 2004 10:27 pm
License Nr.: 156

problem with lam-mpi and ifort

#2 Post by ata » Mon Oct 25, 2004 8:32 pm

I also have this problem. I compiled the program using ifort (ifort-8.1) and when I run it in parallel mode I get the following message:

running on 4 nodes
distr: one band on 1 nodes, 4 groups
vasp.4.6.21 23Feb03 complex
POSCAR found : 2 types and 17 ions

-----------------------------------------------------------------------------
| |
| ADVICE TO THIS USER RUNNING 'VASP/VAMP' (HEAR YOUR MASTER'S VOICE ...): |
| |
| You have a (more or less) 'large supercell' and for larger cells |
| it might be more efficient to use real space projection operators |
| So try LREAL=.TRUE. in the INCAR file. |
| Mind: If you want to do an extremely accurate calculations keep the |
| reciprocal projection scheme (i.e. LREAL=.FALSE.) |
| |
-----------------------------------------------------------------------------

LDA part: xc-table for Ceperly-Alder, standard interpolation
POSCAR, INCAR and KPOINTS ok, starting setup
FFT: planning ... 1
reading WAVECAR
WARNING: random wavefunctions but no delay for mixing, default for NELMDL
LAPACK: Routine ZPOTRF failed! 7
LAPACK: Routine ZPOTRF failed! 7
LAPACK: Routine ZPOTRF failed! 7
LAPACK: Routine ZPOTRF failed! 7


The point is that I have used both atlas optimized part of lapack and lapack_double.o in vasp.4.lib and it didn't disappear and it is very hard for me to get access to ifc7.1 do you have anty suggestion please?
Last edited by ata on Mon Oct 25, 2004 8:32 pm, edited 1 time in total.

Anonymous

problem with lam-mpi and ifort

#3 Post by Anonymous » Fri Nov 05, 2004 7:47 pm

add the following to your ./vasp.4.6/makefile.linux_ifc_P4 file:
mpi.o : mpi.F
$(CPP)
$(FC) -FR -lowercase -O0 -c $*$(SUFFIX)
it seems to work if mpi gets compiled with less optimization
Last edited by Anonymous on Fri Nov 05, 2004 7:47 pm, edited 1 time in total.

masato
Newbie
Newbie
Posts: 3
Joined: Tue Mar 08, 2005 2:02 am
License Nr.: 85

problem with lam-mpi and ifort

#4 Post by masato » Tue Mar 08, 2005 2:16 am

Above reply by Anonymous helped a lot. Thanks.
Last edited by masato on Tue Mar 08, 2005 2:16 am, edited 1 time in total.

Tony

problem with lam-mpi and ifort

#5 Post by Tony » Thu Mar 31, 2005 3:21 pm

[blockquote][/blockquote] [quote="admin"]alternatively to switching back to an old ifc version, you could choose ZPOTRF
from a different LAPACK package:
usually, the package delivered with vasp (lapack_double.o in vasp.4.lib) works fine,
but depending on your installation, you might also try a version
like mkl-lapack,...[/quote]
Last edited by Tony on Thu Mar 31, 2005 3:21 pm, edited 1 time in total.

Post Reply