Page 1 of 1

Segmentation fault (core dumped) problem in NLO calculation

Posted: Wed Nov 02, 2011 7:26 am
by Zhang Xiaole
Hello Dear abinit usrs,
I am trying to repeat the NLO calculation of BaTiO3 recently. After much effort, i still can not obtain the results. Hope someone could help me out.
I am sorry i cannot using the 'Upload attachment' for my input file, so i have to paste them instead.
This is my structure optimization input file
############################################
spgroup 99
ngkpt 10 10 10
natom 5
ntypat 3
znucl 56 22 8
brvltt -1
natrd 4
typat 1 2 3 3
acell 3.994 3.994 4.033 Angstrom
angdeg 90.000 90.000 90.000
xred 0.000 0.000 0.000
0.500 0.500 0.512
0.500 0.500 0.006
0.000 0.500 0.500
ixc 1

#Definition of the scf procedure
#******************************
iscf 5
nstep 1500

#Definition of the plane wave basis
#************************************
ecut 50
ecutsm 0.5
dilatmx 1.05

nshiftk 1
shiftk 0.5 0.5 0.5

#Structural relaxation
#*********************
ionmov 2
optcell 0
tolvrs 1.0d-15
tolmxf 1.0d-5
ntime 1500

# add to conserve old < 6.7.2 behavior for calculating forces at each SCF step
optforces 1
########################################################


and the output seems everything is OK except several WARNINGS about ‘Potential-based CG line minimization has trouble to converge.’,and then i search the forum one post said it is not a problem,because the wavefuction is actually converged
then i do the NLO calculation using the optimized parameter and the following is my input files



###################################################
spgroup 99
natom 5
ntypat 3
znucl 56 22 8
brvltt -1
natrd 4
typat 1 2 3 3 3
acell 3.944 3.944 4.033 Angstrom
angdeg 90.000 90.000 90.000
xred -7.9601674852E-19 -7.9601674853E-19 1.2874227760E-02
5.0000000000E-01 5.0000000000E-01 5.2638115439E-01
5.0000000000E-01 5.0000000000E-01 -1.3986167188E-02
2.2472869684E-19 5.0000000000E-01 4.9636539252E-01
5.0000000000E-01 2.2472869683E-19 4.9636539252E-01
ixc 3

#Parameters of the SCF cycles
#****************************
iscf 5
nstep 1000


#Plane wave basis and k-point grid
#*********************************
ecut 50
ecutsm 0.5
dilatmx 1.05
ngkpt 10 10 10
nshiftk 1
shiftk 0.5 0.5 0.5

ndtset 5 jdtset 1 2 3 4 5

#DATASET1 : scf calculation: GS WF in the BZ
#********************************************
prtden1 1
kptopt1 1
toldfe1 1.0d-10

#DATASET2 : non scf calculation: GS WF in the whole BZ
#*****************************************************
getden2 1
kptopt2 2
iscf2 -2
getwfk2 1
tolwfr2 1.0d-20
nband2 20


#DATASET3 : ddk on the MP grid of k-points
#*****************************************
getwfk3 2
rfdir3 1 1 1
rfelfd3 2
tolwfr3 1.0d-20
nband3 20
kptopt3 2

#DATASET4 : ddE on the MP grid of k-points
#*****************************************
prtden4 1
getwfk4 2
getddk4 3
rfdir4 1 1 1
rfphon4 1
rfatpol4 1 2
tolvrs4 1.0d-10
rfelfd4 3
nband4 20
kptopt4 2
prepanl4 1

#DATASET5 : 3DTE calculation
#***************************
getden5 1
get1den5 4
getwfk5 2
get1wf5 4
nband5 20
kptopt5 2
optdriver5 5
rf1elfd5 1
rf1phon5 1
rf1atpol5 1 5
rf1dir5 1 1 1
rf2elfd5 1
rf2dir5 1 1 1
rf3elfd5 1
rf3dir5 1 1 1

nbdbuf 0
##############################################


when it finished , the log file come up with thounds of WARNING but still finished, and the WARNING most concentrate in the 'cgwf3.F90:702:WARNING' and the last of my log file it said 'The dynamical matrix was incomplete : phonon frequencies may be wrong ...' and 'matrix of third-order energies incomplete changes in the dielectric susceptibility may be wrong.' According to the Frecuently asked questions' ,mverstra said the dynamical matrix was incomplete owing to not all the purturbation was included, so i just neglected this warning and go ahead. after merging the add files i run anaddb it come up with the following.
(except for gamma, not yet able to use time-reversal symmetry)
Now the whole DDB is in central memory
gtblk9 : enter gtblk9
gtblk9 :
Unable to find block corresponding to the following specifications :
Type (rfmeth) = 1
ider qphon(3) qphnrm rfphon rfelfd rfstrs
1 0.00 0.00 0.00 0.00 1 0 0
2 -0.00 -0.00 -0.00 0.00 1 0 0
gtblk9 : enter gtblk9
gtblk9 :
Unable to find block corresponding to the following specifications :
Type (rfmeth) = 1
ider qphon(3) qphnrm rfphon rfelfd rfstrs
1 0.00 0.00 0.00 0.00 1 2 0
2 -0.00 -0.00 -0.00 0.00 1 2 0
gtblk9 : enter gtblk9
gtblk9 :
Unable to find block corresponding to the following specifications :
Type (rfmeth) = 3
ider qphon(3) qphnrm rfphon rfelfd rfstrs
1 0.00 0.00 0.00 1.00 1 2 0
2 0.00 0.00 0.00 1.00 0 2 0
3 0.00 0.00 0.00 1.00 0 2 0
Segmentation fault (core dumped)
###############################################

and no output at al, i have no idea what should do next to fix these problem,please help me. Thanks in advanced


Xiaole

Re: Segmentation fault (core dumped) problem in NLO calculat

Posted: Tue Jun 21, 2016 12:14 pm
by mazibi
Hi Xiaole
i have the same problem for calculate NLO of TiO2, have you solved this problem?

Re: Segmentation fault (core dumped) problem in NLO calculat

Posted: Sat Jul 16, 2016 4:14 pm
by otahiri
i cant't have the nonlinear result and i have this
!ERROR
message: |
Aborting now
src_file: m_optic_tools.F90
src_line: 856
...

#0 0xB6A20133
#1 0x815BDE8 in show_backtrace at m_errors.F90:951
#2 0x811B19A in __m_optic_tools_MOD_nlinopt at m_optic_tools.F90:856
#3 0x804DCC3 in optic at optic.F90:423

Re: Segmentation fault (core dumped) problem in NLO calculat

Posted: Mon Jul 23, 2018 9:19 am
by new_986
Hi
I know this post is not new, but have anybody solved the problem? I have the same problem with abinit 8.6.3

Regards

Nawzad

Re: Segmentation fault (core dumped) problem in NLO calculat

Posted: Fri Oct 05, 2018 7:18 am
by wangxy
In non-linear calculation, the nband should be equal to the number of valence band. I want to know how to confirm the value?

Re: Segmentation fault (core dumped) problem in NLO calculat  [SOLVED]

Posted: Fri Oct 05, 2018 10:01 am
by mverstra
Hello

A first problem is rfatpol in dtset 4. It needs to be "1 natom" to be sure all of the needed perturbations are done. Those are the blocks missing from the dynamical matrix (at gamma only here) and so the Raman coefficients and born effective charges are incomplete. Please do all of the simpler tutorials (rf 1 & 2) before embarking on the others.

Please forget iscf 5 ever existed - the new default (7 or 17) is much better and the tutorials will be updated if that's not already done.