Your message dated Fri, 07 Mar 2025 22:40:50 +0100
with message-id <14031876.O9o76ZdvQC@debian>
and subject line
has caused the Debian Bug report #1088682,
regarding linux-image-6.11.9-amd64: Abnormal delay at boot time (related to
unstable tsc), also on installer image
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact [email protected]
immediately.)
--
1088682: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1088682
Debian Bug Tracking System
Contact [email protected] with problems
--- Begin Message ---
Package: linux-image-6.11.9-amd64
Version: 6.11.9-1
Severity: normal
Dear Maintainer,
Since bookworm which runs kernel 6.1 I have an extended boot delay which can
last for a few minutes. Dmesg showed that the issue is with the clock source.
Setting the kernel parameter 'tsc=unstable' at boot time fixes the issue. I
have the same problem with Trixie which now has kernel 6.11 and of course the
issue is also with the Debian installer media. (Bookworm and Trixie)
The issue is that it is difficult to find out what is really making the boot
be
stuck for a long time and when booting the installation medium one would
believe that the system is completely frozen since that can last for a few
minutes. The system doesn't react to any user input when in that state.
Moreover if I remove the 'quiet' parameter from the kernel it seems stuck on
detecting USB. This is misleading. It is only by looking at the kernel logs
that I found it was suggested to use tsc=unstable.
How could we prevent users to be stuck at boot time for such a long time
without having a clue on what is happening? Here are a few ideas:
- run the installer by default with tsc=unstable, but maybe this would pass
the
value to the installed system, even when not appropriate? And if it doesn't
pass it to the installed system, that would just move the problem to 1st boot
time and reveal the delay at that time which isn't ideal either.
- build the kernel for the installer medium with
CONFIG_HAVE_UNSTABLE_SCHED_CLOCK ?
- write a notice in the installation guide?
In the meantime I sent a report to the kernel mailing list.
<https://lore.kernel.org/all/[email protected]/>
Regards
-- System Information:
Debian Release: 12.8
APT prefers stable-updates
APT policy: (991, 'stable-updates'), (991, 'stable-security'), (991,
'stable'), (990, 'proposed-updates'), (390, 'oldstable-security'), (390,
'oldstable'), (389, 'oldstable-updates'), (380, 'oldoldstable'), (379,
'oldoldstable-updates'), (370, 'oldoldstable'), (95, 'testing'), (94,
'unstable'), (93, 'experimental')
Architecture: amd64 (x86_64)
Foreign Architectures: i386
Kernel: Linux 6.1.0-27-amd64 (SMP w/4 CPU threads; PREEMPT)
Kernel taint flags: TAINT_PROPRIETARY_MODULE, TAINT_OOT_MODULE,
TAINT_UNSIGNED_MODULE
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8),
LANGUAGE=fr:en_US
Shell: /bin/sh linked to /usr/bin/dash
Init: systemd (via /run/systemd/system)
LSM: AppArmor: enabled
Versions of packages linux-image-6.11.9-amd64 depends on:
ii initramfs-tools [linux-initramfs-tool] 0.142+deb12u1
ii kmod 30+20221128-1
ii linux-base 4.9
Versions of packages linux-image-6.11.9-amd64 recommends:
ii apparmor 3.0.8-3
Versions of packages linux-image-6.11.9-amd64 suggests:
pn debian-kernel-handbook <none>
ii firmware-linux-free 20200122-1
ii grub-efi-amd64 2.06-13+deb12u1
pn linux-doc-6.11 <none>
--- End Message ---
--- Begin Message ---
This is fixed upstream since kernel:
>= 6.1.130
>= 6.6.81
>= 6.12.18
>= 6.13.6
>= 6.14-rc5
https://web.git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/
commit/?id=c157d351460bcf202970e97e611cb6b54a3dd4a4
So this will hopefully soon reach bookworm & trixie, and later on, probably
also bullseye.
Patch is also queued to enter next stable release of longterm releases 5.15 &
5.10, which should be 5.15.179 & 5.10.235
Regards
--- End Message ---
