[Pw_forum] Segmentation Fault in Subroutine divide_class.f90

[Andrea Dal Corso]

Sorry it is the 0.866025 in cell parameters that has too few digits. It
should be \sqrt{3}/2. 0.8660254 is sufficient in my PC to find the
correct classes. 

Andrea

On Tue, 2015-12-15 at 17:19 +0530, Surender wrote:
> Dear Andrea,
> Thank you for your reply, as per your suggestion, I tried the following
> atomic positions
> 
> ATOMIC_POSITIONS crystal
>  Ni   0.0000000   0.0000000   0.000000
>  Ni   0.0000000   0.0000000   0.500000
>  As   0.3333333   0.6666667   0.250000
>  As   0.6666667   0.3333333   0.750000
> 
> ATOMIC_POSITIONS crystal
>  Ni   0.00000000   0.00000000   0.000000
>  Ni   0.00000000   0.00000000   0.500000
>  As   0.33333333   0.66666667   0.250000
>  As   0.66666667   0.33333333   0.750000
> 
> ATOMIC_POSITIONS crystal
>  Ni   0.000000000   0.000000000   0.000000
>  Ni   0.000000000   0.000000000   0.500000
>  As   0.333333333   0.666666667   0.250000
>  As   0.666666667   0.333333333   0.750000
> 
> ATOMIC_POSITIONS crystal
>  Ni   0.0000000000   0.0000000000   0.000000
>  Ni   0.0000000000   0.0000000000   0.500000
>  As   0.3333333333   0.6666666667   0.250000
>  As   0.6666666667   0.3333333333   0.750000
> 
> But unfortunately I got the same error (for all the atomic positions), did
> I miss something?
> 
> 
> > Please add more digits to 1/3 and 2/3.
> >
> > HTH,
> >
> > Andrea
> >
> > On Tue, 2015-12-15 at 16:34 +0530, Surender wrote:
> >> Dear All,
> >>
> >> I am trying to run the following input file using QE-5.1.2 (compiled
> >> with
> >> Intel Compilers + Intel MKL + FFTW + OpenMPI) and running on Fedora 17
> >> (64-bit)
> >>
> >> //================================//
> >>
> >> &control
> >>     prefix='nias'
> >>     calculation = 'scf'
> >>     restart_mode='from_scratch'
> >>     verbosity='high'
> >>     pseudo_dir = './'
> >>     outdir='./tmp'
> >>     tprnfor=.true.
> >>     tstress=.true.
> >> /
> >> &system
> >>     ibrav= 0
> >>     celldm(1)=6.8067935
> >>     nat= 4
> >>     ntyp= 2
> >> !   nosym=.true.
> >>     ecutwfc=50.0
> >>     ecutrho=500.0
> >>     occupations='smearing'
> >>     smearing='mv'
> >>     degauss=0.005
> >> /
> >> &electrons
> >>     mixing_beta = 0.5
> >>     conv_thr = 1.0D-08
> >> /
> >> ATOMIC_SPECIES
> >>  Ni  10.0  Ni.pbe-n-rrkjus_psl.0.1.UPF
> >>  As  10.0  As.pbe-n-rrkjus_psl.0.2.UPF
> >> ATOMIC_POSITIONS crystal
> >>  Ni   0.000000   0.000000   0.000000
> >>  Ni   0.000000   0.000000   0.500000
> >>  As   0.333333   0.666667   0.250000
> >>  As   0.666667   0.333333   0.750000
> >> K_POINTS automatic
> >>  10 10 10 1 1 1
> >> CELL_PARAMETERS alat
> >>   1.000000   0.000000   0.000000
> >>  -0.500000   0.866025   0.000000
> >>   0.000000   0.000000   1.390616
> >>
> >> //================================//
> >>
> >> Unfortunately, QE crashes with segmentation fault in subroutine
> >> divide_class.f90, for exact message see the attached scf.out file. I
> >> tried
> >> the same input file with QE-5.2.0 and QE-5.2.1 as well but received the
> >> same error. To me, it appears to be related to symmetry because if I use
> >> the flag nosym=true , pw.x works fine but takes too long to complete.
> >> Please help.
> >>
> >> Regards,
> >> Surender Kumar
> >> IIT Bombay, India
> >> _______________________________________________
> >> Pw_forum mailing list
> >> Pw_forum at pwscf.org
> >> http://pwscf.org/mailman/listinfo/pw_forum
> >
> > --
> > Andrea Dal Corso                    Tel. 0039-040-3787428
> > SISSA, Via Bonomea 265              Fax. 0039-040-3787249
> > I-34136 Trieste (Italy)             e-mail: dalcorso at sissa.it
> >
> >
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://pwscf.org/mailman/listinfo/pw_forum
> >
> 
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum

-- 
Andrea Dal Corso                    Tel. 0039-040-3787428
SISSA, Via Bonomea 265              Fax. 0039-040-3787249
I-34136 Trieste (Italy)             e-mail: dalcorso at sissa.it