Control: found -1 6.12.12-1 Control: notfound -1 6.12.27-1~bpo12+1 Hello Matt, On Sun, Jun 15, 2025 at 11:25:54PM -0700, Matt Mower wrote: > I have not found a specific commit or kernel config item or debian patch > that causes this issue. I've confirmed it happens in v6.12.12 very rarely > and it continues to happen in v6.12.33, where it always occurs within the > first few minutes of boot. Most often, my network restarts after the > IO_PAGE_FAULT errors appear. Occasionally, when rebooting, the computer > hangs (the caps lock light flashes and I have to hold power for 10 sec to > turn off the machine). > > This has been difficult to pin down a culprit because of how rare and > intermittently the issue occurs in v6.12.12 and v6.12.22. Sometimes, I can > leave the laptop on for an hour without issue. Another time, it completely > freezes after a few minutes. I think I need assistance identifying some > tests I could run or extra debugging I could provide. > > I've attached the most relevant dmesg logs from recent tests. These tests > were done with firmware-atheros 20250410-2~bpo12+1. Recall, my first bug > report was using firmware from > https://git.codelinaro.org/clo/ath-firmware/ath12k-firmware as of ref > 5f5f6d65. > > linux-image-6.12.27+bpo-amd64 (bookworm-backports): dmesg-6.12.22bpo.log > inux-image-6.12.32-amd64 (sid): dmesg-6.12.32sid.log > Built v6.12.12 (config from 6.12.12 bpo image): dmesg-6.12.12built.log > Built v6.12.33 (config from 6.12.32 sid image): dmesg-6.12.33built.log > Built v6.12.33 but did not apply any debian patches (config from 6.12.32 > sid image): dmesg-6.12.33built-no-debian-patches.log thanks for your extensive testing, that's very helpful. I think this should be forwarded to upstream. I bet they will ask to test on a newer kernel. You can preempt that request by already testing a kernel from Debian experimental which has 6.15.2. Best regards Uwe
Attachment:
signature.asc
Description: PGP signature