[Q-e-developers] Variable Cell of EXX w/ Custom-FFT

[Paolo Giannozzi]

I think you are right. exx_fft%ggt is never used but exx_fft%gt is. Thank
you for reporting this

Paolo

On Thu, Aug 31, 2017 at 4:01 PM, Satomichi Nishihara <nisihara225 at gmail.com>
wrote:

> Dear developers of QE,
>
>
>
> when variable cell calculation is performed with exact exchange,
>
> it would be required to scale G-vectors of Custom-FFT in “scale_h.f90”.
>
> However, that is not implemented in the current code.
>
> I think it is useful to modify the subroutine “exx_grid_reinit” of exx.f90
> as:
>
>
>
>   SUBROUTINE exx_grid_reinit(at_old)
>
>     IMPLICIT NONE
>
>     REAL(DP), INTENT(IN) :: at_old(:,:)
>
>     !
>
>     INTEGER  :: ig
>
>     REAL(DP) :: gx, gy, gz
>
>     !
>
>     DEALLOCATE(xkq_collect,index_xk,index_sym)
>
>     exx_grid_initialized = .false.
>
>     nkqs = 0
>
>     CALL exx_grid_init()
>
>     !
>
>     DEALLOCATE(working_pool)
>
>     CALL exx_mp_init()
>
>     !
>
>     ! … scale g-vectors
>
>     CALL cryst_to_cart(exx_fft%ngmt, exx_fft%gt, at_old, -1)
>
>     CALL cryst_to_cart(exx_fft%ngmt, exx_fft%gt, bg,     +1)
>
>     !
>
>     DO ig = 1, exx_fft%ngmt
>
>       gx = exx_fft%gt(1, ig)
>
>       gy = exx_fft%gt(2, ig)
>
>       gz = exx_fft%gt(3, ig)
>
>       exx_fft%ggt(ig) = gx * gx + gy * gy + gz * gz
>
>     END DO
>
>     !
>
>   END SUBROUTINE exx_grid_reinit
>
>
>
> Best,
>
> Satomichi Nishihara
>
>
>
> _______________________________________________
> Q-e-developers mailing list
> Q-e-developers at qe-forge.org
> http://qe-forge.org/mailman/listinfo/q-e-developers
>
>


-- 
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/developers/attachments/20170901/c1a10922/attachment.html>