[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#1107521: linux-image-6.12.27+bpo-amd64: ath12k_pci errors and eventual kernel panic starting with linux-image-6.12.27+bpo-amd64



After several suspends and plenty of uptime on battery, I'm fairly confident that linux-image-6.12.12+bpo-amd64_6.12.12-1~bpo12+1_amd64 paired with firmware-atheros_20250410-2~bpo12+1_all is stable. There are only 2 commits to the ath12k driver between 6.12.12 and 6.12.22:

44de00e8bc8f wifi: ath12k: fix handling of 6 GHz rules
425f6a38173b wifi: ath12k: fix tx power, max reg power update to firmware

So, I'll next test 6.12.22 with reverts of one or both of these.

On Wed, Jun 11, 2025 at 9:50 PM Matt Mower <mowerm@gmail.com> wrote:
For what it's worth, if you check the replies after the original report, you can see that I tested 6.12.32 as well as 6.12.27 with ath12k patches reverted. Unfortunately, I can reproduce it in 6.12.22 now as well; the bug only seems to appear when my laptop is not connected to a power source. I reproduce it most often when the device shuts down or goes to sleep, at which point it freezes completely. I snapped a picture when I saw more debugging information that normal on the screen (attached). I've tried downgrading to firmware-atheros_20250410-2~bpo12+1_all as well (dmesg reports firmware fw_version 0x100301e1 fw_build_timestamp 2023-12-06 04:05), and that didn't resolve the issue.

I'm continuing to test older kernel and firmware versions until I find something stable.

On Wed, Jun 11, 2025 at 1:49 PM Uwe Kleine-König <u.kleine-koenig@baylibre.com> wrote:
Control: tag -1 +moreinfo

Hello Matt,

On Sun, Jun 08, 2025 at 08:00:43AM -0700, Matt Mower wrote:
> After updating from linux-image-6.12.22+bpo-amd64 to linux-image-6.12.27+bpo-
> amd64, I am seeing lots of IO_PAGE_FAULT events for device ath12k_pci, and
> twice my system has completely frozen.

There are a few ath12k patches between 6.12.22 and 6.12.27 and a few
more in later stable releases in the 6.12.x series. Can you please test
on the latest 6.12 kernel? There is currently no backport kernel for
that version, but in my experience the kernel from testing should
install fine on a stable box.

If the issue also happens on 6.12.32, someone has to bring that forward
to upstream's attention.

Best regards
Uwe

Reply to: