Two new stable kernels

[Posted May 29, 2025 by jake]

The 6.14.9 and 6.12.31 stable kernels have been released. Each contains an unusually large number of important fixes all over the kernel tree.

Changelogs

Posted May 30, 2025 19:26 UTC (Fri) by npws (subscriber, #168248) [Link] (5 responses)

I know that personal attribution is very important for open source projects, however I find the grouping and ordering by author the pretty much worst imaginable way to represent a shortlog. For deciding whether an update is necessary for me, I care about what changed, not who changed it.

Changelogs

Posted May 31, 2025 8:18 UTC (Sat) by gregkh (subscriber, #8) [Link] (3 responses)

This is the default 'git shortlog' output that the kernel community has been using since the beginning of git. There are many other ways of displaying the information if you want it, 'git log' can take loads of different format options to create custom reports in whatever way you want if the shortlog output isn't useful for you.

Changelogs

Posted May 31, 2025 15:47 UTC (Sat) by npws (subscriber, #168248) [Link] (2 responses)

I'm aware of that of course, however there is obviously the intent to communicate the changes by posting a changelog, and I doubt that this is really the best presentation for that. Maybe its just me who finds this suboptimal, then I'll happily stick to dealing with it myself.

Changelogs

Posted May 31, 2025 16:54 UTC (Sat) by corbet (editor, #1) [Link] (1 responses)

Do you have any thoughts on what would be the best presentation for this information? The stable maintainers might just be interested and, if not, other online sources of information about Linux kernel development might just pay attention...

Changelogs

Posted Jun 1, 2025 17:44 UTC (Sun) by cmm (guest, #81305) [Link]

Presently the actionable part of a stable kernel update announcement is basically "did gregkh write should or must". Which is obviously suboptimal as a way to describe the changes therein (and the shortlog doesn't help), but on the other hand totally sufficient as a tl;dr. :)

Changelogs

Posted Jun 1, 2025 18:14 UTC (Sun) by dezgeg (subscriber, #92243) [Link]

You just haven't figured the way authorship information can be utilized - the paper "The Sound of Silence: Mining Security Vulnerabilities from Secret Integration Channels in Open-Source Projects" suggests some markers for identifying security fixes, such an "patch is authored by Jann Horn" ;)