Hi Marius, Thanks for the bug report. If you can reliably reproduce this issue and can provide a patch which fixes it, I can add that patch to the next set of 5.15 backports. I am a bit limited for time at the moment though and trying to work on the backlog of backports into 6.1.y as well, so I am not sure what the exact timeline would look for a 5.15.y set yet. - Leah On Tue, Apr 23, 2024 at 8:42 PM Darrick J. Wong <djwong@xxxxxxxxxx> wrote: > > On Tue, Apr 23, 2024 at 08:05:22AM -0700, Marius Fleischer wrote: > > Hi Dave, > > > > Thank you so much for your detailed explanations! > > > > > > Do you think it would make sense to backport that patch to 5.15.156 (the > > > > LTS kernel)? > > > > > > That's up to the 5.15 LTS maintainers to decide. They also need to > > > weigh up the fact that xlog_cil_kvmalloc() doesn't exist anymore in > > > the upstream code base. i.e. we found other places that had the same > > > kvmalloc() performance problems, and so lifted that code up further > > > and used it in other places in XFS.... > > > > > > > Understood. Please allow me one last follow-up question. Did I add the correct > > maintainers/mailing lists to reach the 5.15 LTS maintainers? If not, > > how do I find > > the correct addresses? I used scripts/get_maintainer.pl as suggested here > > (https://www.kernel.org/doc/html/latest/admin-guide/reporting-issues.html) > > to locate maintainers/mailing lists. > > You got it correct -- Leah is the 5.15 maintainer of record. > > --D > > > Wishing you a lovely week! > > > > Best, > > Marius
