The ABINIT Users Mailing List ( CLOSED )

["Anglade Pierre-Matthieu" <anglade@gmail.com>]

Hi,
I've tryed to triggered this bug on my own version of abinit. Sadly it
fails somewhere else beforehand. I'll try again later with a more
stable version. In the meanwhile, would you mind trying an other
abinit compiled with a different compiler ?  It may works better...
PMA

On 7/17/06, Deyu Lu <dylu@ucdavis.edu> wrote:
> Dear Fabien:
>      Thank you for your help. After setting "ulimit -s unlimited", the
> problem still persists. Abinit 5.1.2 quit while computing vxc matrix. I
> listed some relevant info below.
>
>      Deyu
>
> [root@zinfandel si5h12]# ulimit -a
> core file size          (blocks, -c) unlimited
> data seg size           (kbytes, -d) unlimited
> file size               (blocks, -f) unlimited
> pending signals                 (-i) 1024
> max locked memory       (kbytes, -l) 32
> max memory size         (kbytes, -m) unlimited
> open files                      (-n) 1024
> pipe size            (512 bytes, -p) 8
> POSIX message queues     (bytes, -q) 819200
> stack size              (kbytes, -s) unlimited
> cpu time               (seconds, -t) unlimited
> max user processes              (-u) unlimited
> virtual memory          (kbytes, -v) unlimited
> file locks                      (-x) unlimited
>
> Error message:
> [root@zinfandel si5h12]# abinis < si5h12x.files > epsilon2.log
> forrtl: severe (174): SIGSEGV, segmentation fault occurred
> Image              PC                Routine            Line
> Source
> libpthread.so.0    0000003A4EA0C430  Unknown               Unknown
> Unknown
> libpthread.so.0    0000003A4EA0C2CE  Unknown               Unknown
> Unknown
> libguide.so        0000002A9557CA1A  Unknown               Unknown
> Unknown
>
>
> The end of the log file:
>  cvxclda: calculating Vxc using ixc =   7
>  cvxclda: calling rhohxc to calculate Vxc[n_val] (excluding non-linear
> core corrections)
>  cvxclda: rhohxc returned  Exc[n_val]  =  -9.5730 [Ha]
>                       and <Vxc[n_val]> =  -0.0519 [Ha]
>
>
>  vxc(1:3,1:3)=
>  (-0.3894028,-9.3494627E-31) (0.0000000E+00,0.0000000E+00)
>
>
> On Mon, 2006-07-17 at 09:37 +0200, Fabien Bruneval wrote:
> > In Paris, we encountered the same kind of problems with the intel
> > fortran compiler. It seems that ifort does nasty things with the stack
> > memory.
> >
> > Can you try to set the stack size of your machine to maximum with the
> > statement:
> > ulimit -s unlimited
> >
> > and then run again your calculations in the same sesssion.
> >
> > Tell me if it solves the problem.
> >
> >
> > Fabien
> >
> >
> >
> > dylu@ucdavis.edu wrote:
> > > Indeed it is the memory issue. I ran a test job with smaller npwwfn, nband, and 
> npwsigx. Abinit 5.1.2 finished the job, while Abinit 4.6.5 quit when reading the KSS file 
> giving an error message "Segmentation fault (core dumped)". I
> > > guess that the KSS file reader handles memeory differently in these two 
> versions.
> > >
> > > Deyu
> > >
> > > # Si5H12 in a SC box
> > > # convergence test against ecuteps
> > >
> > > ndtset 2
> > > acell 25 25 25
> > > #ecut     15.0
> > > ecut     3.5
> > > istwfk   1
> > >
> > > # DATASET 1: Calculation of the screening (epsilon^-1 matrix)
> > > optdriver1  3
> > > getkss     1
> > > #npwwfn11    1503 #1503               1.579
> > > #npwwfn21    2109 #2109               2.021
> > > #npwwfn31    2969 #2969               2.527
> > > #npwwfn41    3431 #3431               2.779
> > > #npwwfn51    3911 #3911               3.032
> > > npwwfn1     895
> > > nband1      32
> > > nsheps      25 #25                    305               0.537
> > > #nqptdm     1
> > > #qptdm      0.000010    0.000020    0.000030
> > >
> > > #DATASET 2: Calculation of the Self-Energy matrix elements (GW 
> corrections)
> > >
> > > optdriver2  4
> > > getscr2     -1
> > > npwwfn2     895
> > > npwsigx2    895
> > > nband2      32
> > > nkptgw        1
> > > kptgw        0.000    0.000    0.000
> > > bdgw         14 18
> > >
> > > # GW calculation general parameters
> > > ppmfrq      4.3 eV
> > >
> > > #Definition of the atom types
> > > ntypat 2          # There are 2 types of atom
> > > znucl 14  1       # The keyword "znucl" refers to the atomic number of the
> > >                   # possible type(s) of atom. The pseudopotential(s)
> > >                   # mentioned in the "files" file must correspond
> > >                   # to the type(s) of atom. Here, the only type is 
> Hydrogen.
> > > natom 17          # There are two atoms
> > > typat 5*1 12*2    # Silicon: 1-5; Hydrogen: 6-17
> > >
> > > #Definition of the k-point grid
> > > nkpt 1            # Only one k point is needed for isolated system,
> > >                   # taken by default to be 0.0 0.0 0.0
> > > diemac 1.0        # Although this is not mandatory, it is worth to
> > >                   # precondition the SCF cycle. The model dielectric
> > > diemix 0.5        # function used as the standard preconditioner
> > >                   # is described in the "dielng" input variable section.
> > >                   # Here, we follow the prescriptions for molecules
> > >                   # in a big box
> > >
> > > #Definition of the atoms
> > > xcart  0.0  0.0  0.0   # EXP Si-H bond length 2.797 Bohr (1.48 A)
> > >    2.5168172178E+00   -2.5168172178E+00   -2.5168172178E+00
> > >    -2.5168172178E+00   2.5168172178E+00   -2.5168172178E+00
> > >    -2.5168172178E+00   -2.5168172178E+00   2.5168172178E+00
> > >    2.5168172178E+00   2.5168172178E+00   2.5168172178E+00
> > >    4.1600819556E+00   -4.1600819556E+00   -9.1988967829E-01
> > >    4.1600819556E+00   -9.1988967829E-01   -4.1600819556E+00
> > >    9.1988967829E-01   -4.1600819556E+00   -4.1600819556E+00
> > >    -9.1988967829E-01   4.1600819556E+00   -4.1600819556E+00
> > >    -4.1600819556E+00   9.1988967829E-01   -4.1600819556E+00
> > >    -4.1600819556E+00   4.1600819556E+00   -9.1988967829E-01
> > >    -4.1600819556E+00   -4.1600819556E+00   9.1988967829E-01
> > >    -4.1600819556E+00   -9.1988967829E-01   4.1600819556E+00
> > >    -9.1988967829E-01   -4.1600819556E+00   4.1600819556E+00
> > >    9.1988967829E-01   4.1600819556E+00   4.1600819556E+00
> > >    4.1600819556E+00   9.1988967829E-01   4.1600819556E+00
> > >    4.1600819556E+00   4.1600819556E+00   9.1988967829E-01
> > >
> > > # Use only symmorphic operations
> > > symmorphi 0
> > >


--
Pierre-Matthieu Anglade