[CPMD-list] Corrupted memory message on IBM-SP3
Juerg Hutter
hutter at pci.unizh.ch
Wed Nov 13 08:20:56 CET 2002
Hi
DIPOLE DYNAMICS (i.e. the calculation of the dipole moment
using the Berry phase algorithm) is not implemented
with Vanderbilt ultra-soft PP.
Juerg
----------------------------------------------------------
Juerg Hutter Phone : ++41 1 635 4491
Physical Chemistry Institute FAX : ++41 1 635 6838
University of Zurich E-mail: hutter at pci.unizh.ch
Winterthurerstrasse 190
CH-8057 Zurich, Switzerland
----------------------------------------------------------
On Tue, 12 Nov 2002, Jung Goo Lee wrote:
>
> Hi!
>
> I was running a DIPOLE DYNAMICS for 64 water molecules in order to compute
> the dipole moment on IBM-SP3 using 8 CPU's.
> The part of my input looks like;
>
> ********
> &CPMD
> MOLECULAR DYNAMICS
> RESTART WAVEFUNCTION COORDINATES LATEST
> TEMPERATURE
> 0
> QUENCH IONS
> QUENCH BO
> DIPOLE DYNAMICS WANNIER
> WANNIER OPTIMIZATION
> TRAJECTORY SAMPLE
> 1
> MAXSTEP
> 10
> STRUCTURE BONDS ANGLES
> TIMESTEP
> 4
> EMASS
> 900
> &END
>
>
> &ATOMS
> *008-O-gpbe--bm.uspp NEWF BINARY
> LMAX=P
> 64
> -0.17126 -5.41610 0.28145
>
> *******
>
> But I got a CORRUPTED MEMORY message
> (PROGRAM STOPS IN SUBROUTINE MEMORY_CHECK| CORRUPTED MEMORY [PROC= 3]) .
>
> What shoud be changed?
>
> P.S. Part of Makefile of CPMDv3.5.3 looks like this;
> ***
> #--------------- Default Configuration for IBM-SP3 ---------------
> SRC = .
> DEST = .
> BIN = .
> #QMMM_FLAGS = -D__QMECHCOUPL
> #QMMM_LIBS = -L. -lmm
> FFLAGS = -O3 -qstrict -qxlf77=leadzero -qmaxmem=32768 -qarch=pwr3
> LFLAGS = -L/usr/local/lapack -llapack_pwr3 -lessl -llapack_pwr3 \
> -bbinder:/usr/lib/bind -bmaxdata:2048000000 -qarch=pwr3
> $(QMMM_LIBS)
> CFLAGS = -O3 -qstrict -qarch=pwr3
> CPP = /usr/ccs/lib/cpp -P
> CPPFLAGS = -D__IBM -DLAPACK -DPARALLEL -DMP_LIBRARY=__MPI -DFFT_ESSL
> CC = cc
> FC = mpxlf -c -O3 -qstrict
> LD = mpxlf -O3 -qstrict
> AR = /usr/bin/ar
> #----------------------------------------------------------------------------
> ***
>
> Jung G. Lee
>
> _______________________________________________
> CPMD-list mailing list
> CPMD-list at cpmd.org
> http://www.cpmd.org/mailman/listinfo/cpmd-list
>
More information about the CPMD-list
mailing list