Leap 15.5 - Kernel 5.14.21-150500.55.44.1 breaks i915 drivers

Running a Dell XPS 9315 using i915 drivers, after boot (If it even makes it to plasma running), start getting the kernel message: BUG: workqueue lockup-pool cpus=1 node=0 flags 0x0 nice=0 stuck. Issue appears to be with i915 driver (To which there were a ton of updates). Runs fine on previous kernel (What I’m using). Before filling out an official bug report, I figured I’d see if anyone else with an XPS 9315 was having this problem.


I have provided more information to the bug report. The person assigned would still like your full var/log/messages file. I emailed them mine. Hopefully it is useful. Looks like a problem in usbcore, on my machine due to a reference to null pointer that later devolves into an issue with the i915 driver (probably due to second external monitor on usbc), but possibly a second bug. I don’t know enough to make that diagnosis with confidence, but it seems related. In any event, both issues in my messages showed up on the same kernel update.

I have similar problem on a important (for me) server. 4 tries and thank you snapper. Updates went fine on other Pc’s. Added some info to Bugzilla. Regards from Sweden.

You all probably already have thought of this, but just in case you haven’t, until the bug is clear, I’ve set multiversion.kernels in zypp.conf to retain 5.14.21-150500.55.39, so that I’ll have that to fall back on just in case when a bug fix is pushed (So I can test the fix on this machine before the kernel is deleted).

For anyone following this, just thought I’d bring some info over from the bug report. The problem ended up being in the ax88179 driver. Much thanks to Takashi for effort in fixing the issue. He has pushed an update which should fix the issue (Probably will see it in update sometime in February). If you need a working kernel now, you can boot off of 39 and it will not lockup (It still has some errors in the ax88179 driver though). The update he pushed fixes these errors with the driver. For more detailed information please follow Svyatko’s link above to the bug report.

Using " name -a
Linux ghost 5.14.21-150500.233.g2097397-default #1 SMP PREEMPT_DYNAMIC Thu Feb 1 06:5
2:42 UTC 2024 (2097397) x86_64 x86_64 x86_64 GNU/Linux "
from “https://download.opensuse.org/repositories/Kernel:/SLE15-SP5/pool/x86_64/” Did see some changes yesterday/today. Still working fine. Otherwise “5.14.21-150500.55.39” is working fine… :grinning: