Error comes for Spin-Orbit calculation

Problems running VASP: crashes, internal errors, "wrong" results.


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
Arti
Newbie
Newbie
Posts: 1
Joined: Wed Mar 31, 2010 4:59 am

Error comes for Spin-Orbit calculation

#1 Post by Arti » Wed Jun 30, 2010 7:01 pm

Hello,
I have compiled VASP 5.2 with pgf90 compiler on 64 bit machine. I compiled it both for SO calculation. It gets compiled properly. I is working properly for no SO calculation. But when I run S-O calculation for any system It shows the following error:

vasp.5.2.2 15Apr09 complex
POSCAR found : 1 types and 2 ions
LDA part: xc-table for Pade appr. of Perdew
PGFIO-F-235/formatted write/unit=8/edit descriptor does not match item type.
File name = OUTCAR formatted, sequential access record = 1669
In source file relativistic.f, at line number 487

Does anyone has any idea how to tackle with this problem.
Thanks in advance.
Last edited by Arti on Wed Jun 30, 2010 7:01 pm, edited 1 time in total.

admin
Administrator
Administrator
Posts: 2921
Joined: Tue Aug 03, 2004 8:18 am
License Nr.: 458

Error comes for Spin-Orbit calculation

#2 Post by admin » Thu Jul 01, 2010 7:26 am

sorry, I cannot reproduce this error on our machines, so it must be compiler-dependent.
Please check if it helps to replace the lines following 566 in relativistic.F:
WRITE(IU6,10) ALPHA, BETA,COS(BETA)*COS(ALPHA),-SIN(ALPHA),SIN(BETA)*COS(ALPHA), & COS(BETA)*SIN(ALPHA),COS(ALPHA),SIN(BETA)*SIN(ALPHA), & -SIN(BETA),0,COS(BETA), & COS(BETA)*COS(ALPHA),COS(BETA)*SIN(ALPHA),-SIN(BETA), & -SIN(ALPHA),COS(ALPHA),0, & SIN(BETA)*COS(ALPHA),SIN(BETA)*SIN(ALPHA),COS(BETA)
by
WRITE(IU6,10) ALPHA, BETA,COS(BETA)*COS(ALPHA),-SIN(ALPHA),SIN(BET
A)*COS(ALPHA), &
COS(BETA)*SIN(ALPHA),COS(ALPHA),SIN(BETA)*SIN(ALPHA), &
-SIN(BETA),0.0D0,COS(BETA), &
COS(BETA)*COS(ALPHA),COS(BETA)*SIN(ALPHA),-SIN(BETA), &
-SIN(ALPHA),COS(ALPHA),0.0D0, &
SIN(BETA)*COS(ALPHA),SIN(BETA)*SIN(ALPHA),COS(BETA)

(i.e, replace the 0 by 0.0D0), maybe the Portland compiler is more strict than the Intel compiler does not automatically convert the integer numbers o by real numbers 0.0 In any case, these 2 numbers should be real numbers to match the FORMAT statement
Last edited by admin on Thu Jul 01, 2010 7:26 am, edited 1 time in total.

Post Reply