[QE-developers] Commutator between r and nonlocal potential in compute_ppsi.f90

[Paolo Giannozzi]

Are you referring to code like the following?
        CALL zgemm( 'N', 'N', npw, nbnd_occ, nkb, &
             (0.d0,0.5d0), vkb(1,1), npwx, ps2(1,1,1), nkb, (1.d0,0.0d0), &
             ppsi, npwx )
It's old-style fortran: ps2 is not passed as an array but as a pointer.
ps2(1,1,1) points to the first element of the array. In this particular
case, it is the same as ps2 with no indices

Paolo

On Thu, Jul 23, 2020 at 10:49 PM Lun Yue <xifntx at gmail.com> wrote:

> Hi QE developers,
>
> I have a question that I hope you have the time for.
>
> The momentum matrix elements "write_p_avg.f90" includes the commutator
> [r, V_nonlocal], which is implemented in "compute_ppsi.f90".
> In this module, the call to "zgemm" performs the sum between p*psi and
> the nonlocal part included in the array "ps2". I was wondering why only
> a single element of "ps2" is added, even though it is huge array
> calculated in the previous step (it is used in zgemm and deallocated
> right afterwards). Is this a bug, or am I missing something?
>
> Also, is this part of the code well tested?
>
>
> Best regards,
>
> Lun Yue
>
> Louisiana State University
>
> _______________________________________________
> developers mailing list
> developers at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/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/20200804/df5a0d21/attachment.html>