linux-kernel - [REGRESSION] ? system is stuck in clocksource, >60s delay at boot time without tsc=unstable

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <10cf96aa-1276-4bd4-8966-c890377030c3@yahoo.fr>
Date: Wed, 27 Nov 2024 08:18:41 +0100
From: Fab Stz <fabstz-it@...oo.fr>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: linux-kernel@...r.kernel.org
Subject: [REGRESSION] ? system is stuck in clocksource, >60s delay at boot
 time without tsc=unstable

Hi,

While upgrading from Debian bullseye (kernel 5.10) to bookworm (6.1) I 
noticed that the newer kernel is at the beginning of the boot stuck for 
more than 60 seconds.

This is apparently related to the clocksource module. If I boot with 
tsc=unstable there is no more delay.

In the kernel logs, I have:

clocksource: Long readout interval, skipping watchdog check: cs_nsec: 
512010551 wd_nsec: 39243763320
clocksource: timekeeping watchdog on CPU0: Marking clocksource 'tsc' as 
unstable because the skew is too large:
clocksource:                       'hpet' wd_nsec: 537773520 wd_now: 
3f0f7632 wd_last: 3e425140 mask: ffffffff
clocksource:                       'tsc' cs_nsec: 511996079 cs_now: 
18b0866e6a cs_last: 185f8d68ba mask: ffffffffffffffff
clocksource:                       'tsc' is current clocksource.
tsc: Marking TSC unstable due to clocksource watchdog
TSC found unstable after boot, most likely due to broken BIOS. Use 
'tsc=unstable'.
sched_clock: Marking unstable (3765559657, 1276001)<-(3775071370, -8235646)
clocksource: Checking clocksource tsc synchronization from CPU 1 to CPUs 0.
clocksource: Switched to clocksource hpet

I already had such a warning with 5.10, but there was no >60sec freeze 
with it like with 6.1

The delay is still present in kernel 6.11.

I attach the output of dmesg. On screen, the last line that is displayed 
before being frozen for >60sec is

[    2.898668] hub 2-0:1.0: 5 ports detected

Because of that, I initially thought that there is a problem with USB, 
but actually not since tsc=unstable "fixes" the delay. The kernel 
mentions clocksource only later in the logs, but the timestamps don't 
reflect the delay. My last measure is a freeze of about 70seconds.

I also noticed that if during that period I do a short press on the 
power button of the system, it stops the freeze and the boot process 
continues immediately.

Regards

==== System information ====

It's an Apple iMac 9.1

# cat /proc/version
Linux version 6.1.0-28-amd64 (debian-kernel@...ts.debian.org) (gcc-12 
(Debian 12.2.0-14) 12.2.0, GNU ld (GNU Binutils for Debian) 2.40) #1 SMP 
PREEMPT_DYNAMIC Debian 6.1.119-1 (2024-11-22)

# hostnamectl | grep "Operating System"
Operating System: Debian GNU/Linux 12 (bookworm)

# uname -mi
x86_64 unknown

# lspci -nn
00:00.0 Host bridge [0600]: NVIDIA Corporation MCP79 Host Bridge 
[10de:0a82] (rev b1)
00:00.1 RAM memory [0500]: NVIDIA Corporation MCP79 Memory Controller 
[10de:0a88] (rev b1)
00:03.0 ISA bridge [0601]: NVIDIA Corporation MCP79 LPC Bridge 
[10de:0aae] (rev b2)
00:03.1 RAM memory [0500]: NVIDIA Corporation MCP79 Memory Controller 
[10de:0aa4] (rev b1)
00:03.2 SMBus [0c05]: NVIDIA Corporation MCP79 SMBus [10de:0aa2] (rev b1)
00:03.3 RAM memory [0500]: NVIDIA Corporation MCP79 Memory Controller 
[10de:0a89] (rev b1)
00:03.4 RAM memory [0500]: NVIDIA Corporation MCP79 Memory Controller 
[10de:0a98] (rev b1)
00:03.5 Co-processor [0b40]: NVIDIA Corporation MCP79 Co-processor 
[10de:0aa3] (rev b1)
00:04.0 USB controller [0c03]: NVIDIA Corporation MCP79 OHCI USB 1.1 
Controller [10de:0aa5] (rev b1)
00:04.1 USB controller [0c03]: NVIDIA Corporation MCP79 EHCI USB 2.0 
Controller [10de:0aa6] (rev b1)
00:06.0 USB controller [0c03]: NVIDIA Corporation MCP79 OHCI USB 1.1 
Controller [10de:0aa7] (rev b1)
00:06.1 USB controller [0c03]: NVIDIA Corporation MCP79 EHCI USB 2.0 
Controller [10de:0aa9] (rev b1)
00:08.0 Audio device [0403]: NVIDIA Corporation MCP79 High Definition 
Audio [10de:0ac0] (rev b1)
00:09.0 PCI bridge [0604]: NVIDIA Corporation MCP79 PCI Bridge 
[10de:0aab] (rev b1)
00:0a.0 Ethernet controller [0200]: NVIDIA Corporation MCP79 Ethernet 
[10de:0ab0] (rev b1)
00:0b.0 SATA controller [0106]: NVIDIA Corporation MCP79 AHCI Controller 
[10de:0ab9] (rev b1)
00:0c.0 PCI bridge [0604]: NVIDIA Corporation MCP79 PCI Express Bridge 
[10de:0ac4] (rev b1)
00:10.0 PCI bridge [0604]: NVIDIA Corporation MCP79 PCI Express Bridge 
[10de:0aa0] (rev b1)
00:15.0 PCI bridge [0604]: NVIDIA Corporation MCP79 PCI Express Bridge 
[10de:0ac6] (rev b1)
00:16.0 PCI bridge [0604]: NVIDIA Corporation MCP79 PCI Express Bridge 
[10de:0ac7] (rev b1)
03:00.0 VGA compatible controller [0300]: NVIDIA Corporation C79 
[GeForce 9400] [10de:0867] (rev b1)
04:00.0 Network controller [0280]: Broadcom Inc. and subsidiaries 
BCM4322 802.11a/b/g/n Wireless LAN Controller [14e4:432b] (rev 01)
05:00.0 FireWire (IEEE 1394) [0c00]: LSI Corporation FW643 [TrueFire] 
PCIe 1394b Controller [11c1:5901] (rev 07)

Download attachment "dmesg.noparam.txt.gz" of type "application/gzip" (23540 bytes)