[CPMD-list] error when using k-points

[Axel Kohlmeyer]

On Sun, 29 Apr 2007, qfzhang wrote:

QZ> Hi,Dear all!
QZ>   I want to use 16 CPUs for geometry optimizition using 3*4*4
QZ> MONKHORST-PACK kpoints.But it gives the error message as below.But
QZ> when I use 8 CPUs for calculations,no that problem.And if I don't
QZ> use kpoints, no that problem as well.What's wrong?

i don't know. there are several (open source) codes out there,
that can do this kind of calculation (plane wave pseudopotential with
k-points) so much more efficient, that i rarely bother to try this
with cpmd (only to verify that gamma point is sufficient, which 
requires only single point calculations).

however, what you are seeing is clearly a bug and would make me worried
about all your geometry optimization with kpoints, as there may be 
silent memory corruption, too. however, the minimal requirement in order
to locate and correct the problem is that you make your input available.

cheers,
   axel.


   LSCR=   114225 2*NUMSHEL                     =   130226
 PROGRAM STOPS IN SUBROUTINE RHOOFR_C| SCRATCH ARRAY TOO SMALL [PROC=  10]
LSCR=   114211 2*NUMSHEL                     =   130226
 PROGRAM STOPS IN SUBROUTINE RHOOFR_C| SCRATCH ARRAY TOO SMALL [PROC=  14]
LSCR=   114229 2*NUMSHEL                     =   130226

 PROGRAM STOPS IN SUBROUTINE RHOOFR_C| SCRATCH ARRAY TOO SMALL [PROC=   3]

 ****************************************************************
 *                      ATOMIC COORDINATES                      *
 ****************************************************************
       1      Ba           0.000000       0.000000       0.000000
       2      Ba           7.842363       0.000000       0.000000
       3      Zr           3.921181       3.921181       3.921181
       4      Zr          11.763544       3.921181       3.921181
       5       O           3.921181       3.921181       0.000000
       6       O           3.921181       0.000000       3.921181
       7       O           0.000000       3.921181       3.921181
       8       O          11.763544       3.921181       0.000000
       9       O          11.763544       0.000000       3.921181
      10       O           7.842363       3.921181       3.921181
      11       H           6.446422       1.395941       3.921181
 ****************************************************************

LSCR=   114215 2*NUMSHEL                     =   130226

 PROGRAM STOPS IN SUBROUTINE RHOOFR_C| SCRATCH ARRAY TOO SMALL [PROC=  11]
LSCR=   114225 2*NUMSHEL                     =   130226

 PROGRAM STOPS IN SUBROUTINE RHOOFR_C| SCRATCH ARRAY TOO SMALL [PROC=   1]
LSCR=   114187 2*NUMSHEL                     =   130226

 PROGRAM STOPS IN SUBROUTINE RHOOFR_C| SCRATCH ARRAY TOO SMALL [PROC=  15]
LSCR=   114225 2*NUMSHEL                     =   130226

 PROGRAM STOPS IN SUBROUTINE RHOOFR_C| SCRATCH ARRAY TOO SMALL [PROC=   2]
LSCR=   114225 2*NUMSHEL                     =   130226
LSCR=   114225 2*NUMSHEL                     =   130226
LSCR=   114225 2*NUMSHEL                     =   130226

 PROGRAM STOPS IN SUBROUTINE RHOOFR_C| SCRATCH ARRAY TOO SMALL [PROC=   0]

 PROGRAM STOPS IN SUBROUTINE RHOOFR_C| SCRATCH ARRAY TOO SMALL [PROC=   5]

 PROGRAM STOPS IN SUBROUTINE RHOOFR_C| SCRATCH ARRAY TOO SMALL [PROC=   4]

  Best wishes
                                                   Qianfan Zhang
                                                    2007-04-29




-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.