My Community

I use the following INCAR file to relax a 20 atoms system:
Electronic minimisation
ISTART= 1
ICHARG= 0
ENCUT = 500
ISIF = 3
NELM = 40
NELMIN = 4
NELMDL = -12
LREAL = A
EDIFF = 1E-5
MAXMIX= 80
DOS related values
ISMEAR = -5

Ionic relaxation
ALGO = Fast
NSW = 30
IBRION = 1
NFREE = 10
EDIFFG = -0.01

LPLANE = .TRUE.
NPAR = 15
LSCALU =.FALSE.
NSIM = 4

then the stout file present a error massege like this:

N E dE d eps ncg rms rms(c)
DAV: 1 -0.167332233367E+03 -0.16733E+03 -0.23302E+00 15360 0.594E+00 0.100E+00
RMM: 2 -0.167321577232E+03 0.10656E-01 -0.36609E-01 16237 0.162E+00 0.764E-01
RMM: 3 -0.167295498666E+03 0.26079E-01 -0.13828E-01 14507 0.941E-01 0.325E-01
RMM: 4 -0.167292283054E+03 0.32156E-02 -0.80783E-03 15879 0.244E-01 0.120E-01
RMM: 5 -0.167292225916E+03 0.57138E-04 -0.26397E-03 15925 0.111E-01 0.251E-02
RMM: 6 -0.167292189929E+03 0.35988E-04 -0.36745E-04 18915 0.395E-02 0.264E-02
Error EDDRMM: Call to ZHEGV failed. Returncode = 6 3 4
RMM: 7 -0.167292193873E+03 -0.39443E-05 -0.96448E-05 14558 0.226E-02
2 F= -.16729219E+03 E0= -.16729219E+03 d E =0.149910E-01

Can anybody explain the meaning of this error and tell me how to deal with it?
Thanks!

the error is due to a LAPCK call (ZHEGV):
ZHEGV computes all the eigenvalues, and optionally, the eigenvectors
of a complex generalized Hermitian-definite eigenproblem .
there may be several reasons for that error:

1) the RMM-DIIS diagonalisation algorithm is not stable for your specific
setup of the calculation.
--> use ALGO = Normal (blocked Davidson) or
ALGO = Fast (5 steps blocked Davidson, then RMM-DIIS)

2) maybe your input geometry was not reasonable (error occurs at the
very first ionic step) of the last ionic relaxation step lead to an
unreasonable geometry (compare the input and output geometries of the
last ionic relaxation steps). In that case it can be helpful to
--> switch to a different relaxation algorithm (IBRION-tag)
--> reduce the step size of the first step by setting POTIM smaller than
the default value

3) The installation of the LAPACK on your machine was not done properly:
use the LAPACK delivered with the code (vasp.4.lib/lapack_double.o)

4) on some architectures (especially SGI) some Lapack
routines are not working properly. However, it is
possible to avoid the usage of the ZHEGV subroutine
by commenting the line #define USE_ZHEEVX in
rmm-diis.F, davidson.F, subrot.F, and wavpre_noio.F and recompiling
VASP.