#1118508 linux-image-6.12.38+deb13-amd64: Frequent kernel panic during/just after boot on Lenovo P15v Gen1 with external USB hub in Samsung display

Package:
src:linux
Source:
src:linux
Submitter:
Rimas Kudelis
Date:
2026-02-25 18:55:02 UTC
Severity:
normal
Tags:
#1118508#5
Date:
2025-10-21 12:37:25 UTC
From:
To:
Dear Maintainer,

quite often, when booting my PC, I get a kernel panic (the computer
hangs and the caps lock led blinks) at boot time, or just after it ends.
I suspect this might be related with me typing my password, or moving a
mouse at an unfortunate time (too early, something still initializing?),
but I'm not certain that this is it.

Once the system boots successfully and I log in, the kernel panic
doesn't happen.

For example, today, the kernel panic happened when I submitted my
password to the GDM login screen. I think I also had it happen while
entering my LUKS password at least once. Also, the PC actually crashed
the first time I booted it after installing Trixie, causing me to delay
trying Trixie out for a couple weeks.

My setup is a Lenovo P15v laptop, which is connected via USB-C to a
Samsung LS27A600UUU (27" ViewFinity S60UA QHD, USB-C Monitor), to which
my keyboard and mouse combo are connected via USB-A port, as well as a
second 27" Samsung display via the DP-OUT port. I suspect that maybe
this complexity could have something to do with the freezes and be the
reason why they haven't been noticed and caught until now.

I don't think it's a faulty hardware issue however, because I have used
Pop!_OS 22.04 for several years now (it still resides on my second SSD
disk), which *never* exhibited such behavior.

Cheers!

#1118508#14
Date:
2025-10-21 18:45:25 UTC
From:
To:
Hi,

6.12.38-1 is not the most current kernel available in trixie, pliease
do update to 6.12.48-1 and let us please know if the problem persist.

If yes, then we would need (ideally a full) boot log, but if nothing
get recorded in logs, then the netconsole[1] approach might help

[1] https://docs.kernel.org/networking/netconsole.html

On UEFI system pstore alterantively might help us as well.

But first upgrade your kernel to the most current one in trixie, to
determine if the problem persists.

Regards,
Salvatore

#1118508#21
Date:
2025-10-21 18:54:39 UTC
From:
To:
Hi,

disregard the original subject line, I actually upgraded to 6.12.48 this
morning and reported the bug with that kernel already in use, as
indicated in the System Information block.

I'll see about the logs. I don't tend to reboot often, and when I
checked with `journalctl -b-1` (and other numbers) today, I didn't see
anything weird at the end of these logs: they all indicated successful
logouts. I suspect they aren't being saved for some reason. I'll look at
them again the next time Trixie crashes on boot.

Regards,
Rimas


2025-10-21 21:45, Salvatore Bonaccorso rašė:

#1118508#28
Date:
2025-10-21 19:50:20 UTC
From:
To:
Hi Rimas,

Thanks, I have updated the metadata.

Okay let's see if we can get further information on the crashes.

Regards,
Salvatore

#1118508#33
Date:
2026-02-17 10:49:03 UTC
From:
To:
Sorry for keeping this unchecked for a while. I just don't reboot very
often.

I think this is still happening to me with
linux-image-6.12.63+deb13-amd64.

Attaching three journalctl logs from sessions which at least seemed to
crash.

- jctl-3 felt frozen after telling me that it was installing updates,
- jctl-1 became unresponsive right after it informed me that my
encrypted disk had been successfully set up, and
- jctl-0 became unresponsive after a mouse click in GDM login screen.

In both former cases, pressing Esc did nothing to take me away from the
graphical boot screen, so I killed them by long-pressing the Power
button, but I forgot to first check if CapsLock was blinking. In case of
jctl-0, I can confirm that CapsLock was indeed blinking, and I killed
the session by long-pressing Power too.

I hope these logs can help figure out what is happening.

Rimas

#1118508#38
Date:
2026-02-25 18:52:56 UTC
From:
To:
Hi,

Unfortunately this won't help really, because we do not see
information from the crash in this way. We really would need to gather
logs via netconsole attached, as described in https://bugs.debian.org/1118508#14

Do you have the possibility to do so, and hopefully we get a
netconsole log with some crash information from kernel.

Regards,
Salvatore