[QE-users] Compilation on a server ends with error 1

Giovanni Pizzi giovanni.pizzi at epfl.ch
Tue Apr 9 21:15:26 CEST 2019


Dear Fabio,
Quantum ESPRESSO automatically downloads Wannier90 and prepares a make.inc file inside the folder it creates.
In there, you can change the variables defining the compilers to use.
Before recompiling, I suggest to do a ‘make clean’ inside the Wannier90 folder to remove old compiled objects, if any.

Best,
Giovanni

--
Giovanni Pizzi
Theory and Simulation of Materials and MARVEL, EPFL
http://people.epfl.ch/giovanni.pizzi
http://nccr-marvel.ch/en/people/profile/giovanni-pizzi



On 9 Apr 2019, at 18:39, Paolo Giannozzi <p.giannozzi at gmail.com<mailto:p.giannozzi at gmail.com>> wrote:

Wannier90 uses its own build system, so you will have to figure out what has to be changed there

Paoo

On Tue, Apr 9, 2019 at 5:59 PM Fabio Costa <fabiocosta4 at hotmail.com<mailto:fabiocosta4 at hotmail.com>> wrote:
Dear Pietro

Thank you very much for your help so far, things are starting to work out. Your guess was actually right, and the mpif90 was using the old version. After a few attempts, The installation was successful, and now I have both mpif90 and gfortran v-5.4 .

I still need to indicate the path of these updated executables when compiling QE, so Im starting with the command "./configure MPIF90=/home/fabio/local/bin/mpif90 F90=/home/fabio/local/bin/gfortran ". This step runs successfully, but when making pw, but eventually it stops with a fatal error warning, about some fox module that could not be opened because it was created with a different GNU fortran version. At this point I found out that inside make.inc, the path I provided to mpif90 was accurate, but F90 contained only gfortran, which I think would link to the native (not-updated) executable. To correct this, I added to F90 the path to the updated executable. After this change, the make pw, ph and w90 finish successfully, and after some quick tests, it seems that the executables are running as expected (no more segmentation faults).

Unfortunately, when I try to make epw, the fatal error happens again (https://imgur.com/ukHDZYc). This should be happening because something was compiled with the old gfortran, even when I manually changed the MPIF90 and F90 to link to the updated versions. Do you have any insights on what could have went wrong?

Thanks for the assistance

Fábio Costa
 MSc student in physics
 Federal University of Bahia, Brazil


_______________________________________________
users mailing list
users at lists.quantum-espresso.org<mailto:users at lists.quantum-espresso.org>
https://lists.quantum-espresso.org/mailman/listinfo/users


--
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

_______________________________________________
users mailing list
users at lists.quantum-espresso.org<mailto: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/20190409/4ef079e0/attachment.html>


More information about the users mailing list