[QE-users] Error: Segmentation fault - invalid memory reference.

Paolo Giannozzi p.giannozzi at gmail.com
Sat Mar 3 19:01:34 CET 2018


On Sat, Mar 3, 2018 at 1:37 PM, BENYAHIA NEZHA <BENYAHIA-NEZHA at hotmail.fr>
wrote:

       I think that the LDA+1/2 method is not implemented for PAW, because
> when i made another lda-1/2 test for  USPP pseudopotential for Sb atom, it
> worked normally without any segmentation fault. However, iam interested in
> the PAW pseudopotential!!!! Is there any solution to correct the band gap
> calculated with this type of pseudopotential?
>

just implement it. I don't know how difficult it is, maybe it is simple,
maybe it is not.


> In addition, in the last message you posted, you've said that there were
> an error which occurs in the line 227 of file export_upf.f90:
> if(lpaw) qvanl(1:grid%mesh,:,:,:) = pawsetup%augfun(1:grid%mesh,:,:,:)
>
How to correct this error?
>

figure out why it crashes. Most likely the code is accessing an unallocated
variable. It might be sufficient to properly allocated PAW variables to
avoid yhe crash.
-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20180303/d4d17ee2/attachment.html>


More information about the users mailing list