Bugfix: lower bound phase array

Description

Fix the lower bound of the phase argument. This bug went silent as the routine is not called in k-point parallelization, as this is not implemented at the moment.

News snippet

Bug fix not affecting the code results.

Checklist

• I have checked that my code follows the Octopus coding standards
• I have added tests for all the new features added in this request.

changed milestone to %10.0

added Bug label

Can we add or modify a test to cover this?

assigned to @martin.lueders

At the moment this won't be possible to add the k-point parallelization. However the routine is already used in one test and is covered by the testfarm, if this is your question.

So why did you then say that the bug was not detected?

This is a bug that would appear only in k-point parallelization...

OK. Just out of curiousity: What is missing for the k-point parallelization?

The routine computes the Coulomb integrals. There are four wavefunctions involved and all the combinations of k1, k2, k3, k4 can be involved. The distribution of this calculation in memory would be quite involved, as we would need to communicate all the wavefunctions in a smart way, and this would still be very expensive I guess...

We might need to do it at some point for the GKBA implementation that Angel is pushing right now.

OK. Thanks.

approved this merge request

mentioned in commit bf406b75

merged

added Bugfix in Changelog labels