Any info on kernel 6.12?

JBinFla · December 7, 2024, 1:59am

Curious how to see what the status of updating the kernel is? Normally I really don’t care but I heard some anecdotal comments by those running it that the desktop environment felt snappier, and I’d love some of that goodness too. I’m not interested in running out of tree so I’m waiting. I found the kernel:stable on opensuse.org (openSUSE Software) and I see that 6.12.3 is there but listed as experimental. Since I’ve never “followed” the kernel much I don’t know how long we typically lag. No, I’m not wanting to rush it, but I did see 6.11 reached EOL so I’m certain it’s “soon”.

I guess I’d like to learn how to see what’s going on. Is there some kind of mailing list, or discord or mailing list that goes with this that I could read thru and just satisfy my own curiosity? Kinda see whats going on? I’m newish to all of this and really would like to see how this stuff goes so I can be more informed. Thanks!

malcolmlewis · December 7, 2024, 2:14am

@JBinFla 6.12.3 was submitted to Factory a number of hours ago https://build.opensuse.org/requests/1228718

ENTPRESTIGIOUS · December 7, 2024, 4:22am

I want to know too because it feels like they’re holding back LOL

mburge · December 7, 2024, 4:23am

It’s taking longer due to an issue with systemd-boot

https://bugzilla.opensuse.org/show_bug.cgi?id=1233752

https://lore.kernel.org/all/c119df40-5cd9-40f8-b7b7-4085b87e90e5@kernel.org/

github.com/systemd/systemd

sdb's zero padding of initrd causes measurements to fail

opened 07:29AM - 02 Dec 24 UTC

jirislaby

bug 🐛 sd-boot/sd-stub/bootctl regression ⚠️ rc-blocker 🚧

https://github.com/systemd/systemd/blob/3f3b4959e2cb9bca1e1ed527a0692c9a8b6a18ea…/src/boot/boot.c#L2498-L2504 appears to cause this downstream issue: https://bugzilla.suse.com/show_bug.cgi?id=1233752 It looked like related to the 6.11..6.12 kernel update. But now, that appears like a pure coincidence. Instead, the initrd size likely (and its modulo-ness by 4) matters… From https://lore.kernel.org/all/7773891b-b699-4f1d-b9ba-220c649aee9d@kernel.org/: > Note that ovmf logged: > Called TcgDxeHashLogExtendEvent 0 58683000 1B1E78C 5FE63C00 5E3492AA > Data 28 B5 2F FD ... E1 29 FE 0 > > But initrd on disk is 1B1E78B long, not 1B1E78C. So the excessive 0 at the end above brews the mismatch. See: > https://bugzilla.suse.com/show_bug.cgi?id=1233752#c14 > "By adding the 0 byte I can replicate the measured digest." > > So there is something aligning the initrd. kernel's libstub just uses and passes load_file2's size down to TcgDxeHashLogExtendEvent, AIUI. So it'd be sdb, ovmf or something. BTW how are sizes stored in/fetched from vfat? But well, fs/fat/ received no significant changes either. In particular, from https://bugzilla.suse.com/show_bug.cgi?id=1233752#c17: > You're missing the fact that systemd-boot installs its own Load File2 protocol which the kernel then uses: > > https://github.com/systemd/systemd/blob/3f3b4959e2cb9bca1e1ed527a0692c9a8b6a18ea/src/boot/initrd.c#L100-L105 > > This protocol currently zero pads the binary to a multiple of 4: > > https://github.com/systemd/systemd/blob/3f3b4959e2cb9bca1e1ed527a0692c9a8b6a18ea/src/boot/boot.c#L2498-L2504 > > This was actually a measurement breaking change introduced into v257-rc1 by f8fa4222c9 ("boot: Make initrd_prepare() semantically equivalent to combine_initrds()") -- @DaanDeMeyer.

github.com/systemd/systemd

boot: exclude the trailing padding from initrd size in initrd_prepare()

systemd:main ← keszybz:boot-padding-exclusion

opened 12:39PM - 06 Dec 24 UTC

keszybz

+11 -4

This is a follow-up for f8fa4222c9ac3e74e91c64e25e9532c99559cf99 (boot: Make ini…trd_prepare() semantically equivalent to combine_initrds()). That commit changed the way that the initrd is prepared, but it also effectively rounded up the size to a multiple of 4. In [1], the users report that the extra 0 results in the calculated hash being different. Let's try to avoid changing the hash calculation results in that case by not including the last pad in the reported size. Fixes https://github.com/systemd/systemd/issues/35439. The other changes to the way the initrd is constructed are kept, so hopefully the original issue with booting in qemu is not recreated. [1] https://bugzilla.suse.com/show_bug.cgi?id=1233752

ENTPRESTIGIOUS · December 7, 2024, 5:37am

No wonder.

JBinFla · December 8, 2024, 3:56am

Thanks for the info. Although it’s over my headi do enjoy reading the details. Such a simple thing as padding bits to get to an even multiple of 4-bytes throwing off the checksum. My hats off to these guys.

tpatsiouras · December 8, 2024, 7:00am

Hello everyone

@JBinFla I have installed Linux 6.12.3-2-xanmod-v3 in Tumbleweed.
So far so good… Didn’t notice any problems during use…

Theodore

system · December 15, 2024, 7:01am

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.