[QE-users] Tracing MPI buidl of QE 7.0 malfuction (gfortranm OpenMPI).

[Ye Luo]

build QE without MPI without OpenMP, run test-suite. To check if your
compiler and library works.
build QE with MPI without OpenMP, run test-suite. To check if your MPI
library works properly.
build QE without MPI but with OpenMP, run test-suite. To check if OpenMP is
the issue.
build QE with MPI and with OpenMP, run test-suite. To check if OpenMP+MPI
works.
If any of the above case shows massive failure, you'd better check the
software environment. I think this is likely the case.
If all the above cases pass, but your own tests fail. Provide a full
reproducer (how you build the code, all the input/output/log of run)
Ye
===================
Ye Luo, Ph.D.
Computational Science Division & Leadership Computing Facility
Argonne National Laboratory


On Thu, Feb 3, 2022 at 3:27 PM Husak Michal <Michal.Husak at vscht.cz> wrote:

> Hi
>
> I am trying to find why the MPI build (gfortan + OpenMP) does not work for
> QE 6.6-7.0
> The code stops in the routine
> good_fft_order (in fft_support.f90)
> on following statement
>  IF (np <= 0 .OR. np > nr) & ....
>
> This is called from
> realspace_grid_init (in fft_types.f90) ...
>
> Following command generate nonsense input generating code stop (np=2,nr=1)
>  dfft%nr1 = good_fft_order( dfft%nr1, fft_fact(1)
>
> I am not an Fortran expert not familiar with the QE architecture so
> tracing this issue more deeply is
> behind my knowledge ....
>
> Any advice I can try ?
> Run some test, debug mode ?
>
> Michal Husak
> UCT Prague ...
>
>
>
>
>
>
>
>
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20220203/8714d06c/attachment.html>