On Sat, Feb 08, 2025 at 12:00:57PM +0100, Andreas Gruenbacher wrote: > Kent, > > here is another update. Changes: > > * Added a self test for eytzinger0_find(). > > * The patches are now in fully bisectable order, with slightly improved > patch descriptions. > > * The patch that runs the self tests on modprobe has been moved to the > end of the patch queue. This patch is useful for testing, but it > SHOULD NOT BE MERGED. > > > A git tree of the patches can again be found here: > > > https://git.kernel.org/pub/scm/linux/kernel/git/agruen/linux.git/log/?h=bcachefs > > Base: v6.13 > Head: d79258df8791 ("bcachefs: Run the eytzinger tests on modprobe") > > > The previous CI test results at: > > https://evilpiepirate.org/~testdashboard/ci?user=kmo&branch=eytzinger > > against the previous version of this patch queue and with head: > > 94865378e1a3 ("bcachefs: eytzinger1_{next,prev} cleanup") > > show a large number of passed as well as failed tests even for upstream > v6.13, so I don't know what to conclude from that. > > In any case, there is one failure reporting that eytzinger0_find_ge() > didn't find the first matching element in an array with duplicate > elements: > > > https://evilpiepirate.org/~testdashboard/ci?user=kmo&branch=eytzinger&commit=c31c7c659278bc2581ef0c7c4119fe14c7476ea3 > > This is imtentional and expected because at this point in the patch > queue, eytzinger0_find_ge() hasn't been fixed to always return the first > matching element. The current version of the patch queue adds the fix > before adding the test, so this failure will be gone. > > That being said, as per request, the updated patch queue only enables > the self tests in the last patch, so we wouldn't see failures of that > kind anymore, anyway.
Thanks! Applied to my master & for-next branches.
