On Wed, Jan 07, 2026 at 04:33:36PM +0100, Michael Banck wrote: > Were those failures always about "grid" and "CONTIGUOUS"?
Yes, I think so. > And did the gfortran version change between those runs? You mean between failed and successful ones? No, I have failed logs with fortran 15.2.0-4 and then I also have successful logs with 15.2.0-4. Looks like this is not a bug in cp2k. I tried using debbisect to see if I could guess the root cause, but debbisect does not work as expected. I've put some build logs here (failed first and successful ones later) just in case somebody wants to look at it. https://people.debian.org/~sanvila/build-logs/cp2k/ But at this point I think it's ok to close the bug. Thanks.
