The 6.4 kernel has been released

[Posted June 26, 2023 by corbet]

Most of the stuff in my mailbox the last week has been about upcoming things for 6.5, and I already have 15 pull requests pending. I appreciate all you proactive people.
But that's for tomorrow. Today we're all busy build-testing the newest kernel release, and checking that it's all good. Right?

Headline features in this release include: generic iterators for BPF, the removal of the SELinux runtime disable knob, the removal of the SLOB memory allocator, linear address masking support on Intel CPUs, process-level samepage merging control, support for user trace events, more infrastructure for writing kernel modules in Rust, per-VMA locks, and much more. See the LWN merge-window summaries (part 1, part 2), and the (in-progress) KernelNewbies 6.4 page for the details.

The 6.4 kernel has been released

Posted Jun 26, 2023 12:42 UTC (Mon) by gmgod (guest, #143864) [Link] (2 responses)

> the removal of the SELinux runtime disable knob

With bugs like this:

https://github.com/fedora-silverblue/issue-tracker/issues...

(note this one is on Silverblue but there are similar reports for coreos and workstation), I am not sure this will go well...

The 6.4 kernel has been released

Posted Jun 26, 2023 14:01 UTC (Mon) by rahulsundaram (subscriber, #21946) [Link] (1 responses)

> (note this one is on Silverblue but there are similar reports for coreos and workstation), I am not sure this will go well...

I think you misunderstand this change. It doesn't prevent runtime changes to make SELinux permissive which is all you need to debug issues. It only removes support for disabling of SELinux completely (without using a boot option) and Fedora did this a long time back.

https://fedoraproject.org/wiki/Changes/Remove_Support_For...

The 6.4 kernel has been released

Posted Jun 26, 2023 17:53 UTC (Mon) by tialaramex (subscriber, #21167) [Link]

It's an XKCD 1172 moment. All behaviours are subject to Hyrum's law, regardless of whether anybody consciously added them, whether they're a typo, or even a really wild coincidence, if you've got enough users somebody is relying on the very stupid bug you wrote, and sees it as a valuable feature. This isn't a bug, but it's not a very good feature, and so it's sensible to remove it.

In a few cases Hyrum's Law means you're stuck with it in practice e.g. HTTP "referer" headers. But in most cases it's just not that many people, and they'll get over it, and if this was the most annoying change for you in 2023 so far you've had a very good year and maybe shouldn't draw attention to that.