[molpro-user] memory error in a do loop variable scan
Jason Byrd
byrd at phys.uconn.edu
Tue Nov 4 14:53:58 GMT 2008
Molpro users,
I am getting a memory error after every iteration of a do loop which
goes over a cas and fci calculation. The error is:
USED MEMORY IN cislow: 32383 36581 36581 36581
FREE MEMORY IN cislow: 104825117 104820919 104820919 104820919
? Error
? Inconsistent memory
? The problem occurs in check_address
We are using molpro 2008.1 with the latest patch. The pertinant part of
the input file is as follows:
!Li3
***,SBKJC Polarized (p,2d) - cpp-sbk lfk scaled C2V symmetry---- A1 symm
file,1,li3-2.1
file,2,li3-2.2
file,3,li3-2.3
!define constants and such here:
pesgrid=[1.8,1.6,1.4,11.0]
show,#pesgrid
thresh,energy=1.d-8,optgrad=1.d-5,gradient=1.d-5
S1 = 0.617700000
S2 = 0.143400000
S3 = 0.050480000
S4 = 0.019230000
P1 = 0.006572900
P2 = 0.617700000
P3 = 0.143400000
P4 = 0.050480000
P5 = 0.019230000
D2 = 0.054686000
D3 = 0.018035500
D4 = 0.007688200
basis={
ECP, li, 2, 1 ;
1;
! p potential
1,1.34306,-.73063;
2;
! s-p potential
0,0.61284,1.80131;
2,1.64881,3.54971;
s,Li,S1,S2,S3,S4
p,Li,P1,P2,P3,P4,P5
d,Li,D1,D2,D3,D4
}
geometry={ang;
X,Y
q1
Li1 1 Radius
q1 1 1.0 2 90.0
Li2 1 Radius 3 90.0 2 180.0
Li3 1 RC 4 90.0 3 A
}
! asymptotic case
geomtype=zmat
Radius=1.29712657
RC=100.d0
A=90.d0
!c2v!
ass=-0.6314624143;
DO I=1,#pesgrid
togo=#pesgrid-I
show,togo
RC=pesgrid(I)
scattering(I)=RC
angle(I)=A
! calculate energy at R
bond(I)=2*Radius
cpp,init,1;
li,1,0.1915,0.0,0.0,0.831;
! 3,1,1: three electron, A1 symm, doublet
{rhf
wf,3,1,1}
{casscf,maxit=600,failsafe
wf,3,1,1}
{optg,space=zmat
active,Radius
numerical,fourpoint}
{fci;
wf,3,1,1}
{optg,space=zmat
active,Radius
numerical
step,0.015,0.015,0.5,0.03,1
}
put,molden,li3.molden
li3energy(I)=energy;
pes(I)=energy(1)-ass
cmpes(I)=pes(I)*tocm
show,Radius,RC,A,pes(I)
text,A1 pes
{table,bond,scattering,angle,cmpes
digits,10,10
type,text
save,A1_fci-cm.dat}
{table,bond,scattering,angle,li3energy
digits,10,10
type,text
save,A1_fci-energy.dat}
{table,bond,scattering,angle,pes
digits,10,10
type,text
save,A1_fci-hartree.dat}
ENDDO
!I
--
Jason Byrd
Graduate Student
Department of Physics
University of Connecticut U-3046
2152 Hillside Road
Storrs, CT 06269-3046
Office: P209 Tel: 860-486-3484
Email: byrd at phys.uconn.edu
More information about the Molpro-user
mailing list