[CPMD-list] Segmentation Fault

[Axel Kohlmeyer]

On Tue, 1 Apr 2008, Chris Abraham wrote:

CA> Dear All,

dear chris,

CA> I am testing different functionals/pseudo-potentials to see which one
CA> exactly fits my kind of systems. In this process I am facing a "SEGMENTATION
CA> FAULT" error when I use PBE functional with UPF pseudo potential functions
CA> like "Na.pbe-sp-van_ak.UPF" etc and the job stops. The same input runs fine

CPMD does not support ultra-soft pseudopotentials in UPF format.
you have to use either native binary (BINARY) or text format
(FORMATTED) pseudopotential files. as you can see the code is able
to read the info section, but crashes on the data part of the 
pseudopotential file. the next version of CPMD will detect this
cleanly and abort properly.


CA> &SYSTEM
CA>  SYMMETRY
CA>  14
CA>  ANGSTROM
CA>  CELL ABSOLUTE DEGREE
CA>  4.364 4.364 5.862  90 90 90

why symmetry 14?

you have an orthorhombic box, 
why not take advantage of that?

CA>  CUTOFF
CA>  50
CA>  POINT GROUP
CA>  AUTO
CA>  KPOINTS MONKHORST-PACK
CA>  3 3 3

also, there is no support for k-points with ultra-soft
pseudopotentials. since you already have the UPF files,
you may just move on to use quantum-espresso directly.
their support for ultra-soft potentials is very good.

CA>  TESR
CA>  8

why this?? given your box size this would accommodate
a real space cutoff of over 35 angstrom... 

cheers,
   axel.

CA> &END
CA> 
CA> &ATOMS
CA> *H.pbe-van_bm.UPF  UPF
CA>  LMAX=S
CA>  1
CA>  0.011   0.7694    0.3803
CA> 
CA> *Na.pbe-sp-van_ak.UPF UPF
CA>  LMAX=P
CA>  1
CA>  0.000   0.000   0.000
CA> 
CA> *B.pbe-n-van.UPF UPF
CA>  LMAX=P
CA>  1
CA>  0.000   0.000   0.5000
CA> &END
CA> -----------------------------------------------------------------------------------------------------------------------------------------------------------
CA> 


CA> Chris.
CA> 

-- 
=======================================================================
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.