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

Bug#414322: linux-source-2.6.18: kernel freeze on disk i/o (regression)



This is due to the clocsourse set up. (Bug#412194: linux-image-2.6.18-3-k7: total system freeze after max. 10 minutes uptime).
Add boot option: clocksource=tsc


See below for GRUB /boot/grub/menu.lst

--------------snip------------------------------------------------
Hello all,
I've done as Alexander has suggested:
added a boot option in /boot/grub/menu.lst
----snip---->
title        Debian GNU/Linux, kernel 2.6.18-4-k7
root        (hd0,0)
kernel /vmlinuz-2.6.18-4-k7 root=/dev/sda2 ro clocksource=tsc initrd /initrd.img-2.6.18-4-k7
savedefault
rebooted and:
># echo "tsc" > /sys/devices/system/clocksource/clocksource0/current_clocksource <
># cat  /sys/devices/system/clocksource/clocksource0/* <
jiffies tsc pit
tsc
We are up and running now (45min.). Alex. has been up for some hours (9+) using this method. So maybe others can use this and help us understand later the "why" of it.
Anyway, fingers crossed let us see........
matthew

Alexander Schories wrote:
Hello Steve,


uptime with "nosmp" so far: 57 min. Everything runs fine and smoothly. Of course overall system load is about 5 percent (average) to 20 percent (peak) higher than usual - man, i do love and miss smp.. :D

Just kidding, i can easily live happily either with "nosmp" or particularly with 2.6.17 for now. Maybe other affected users might agree to lower the initial severity to "important"?


But guess who appears on stage now again (dmesg log):

#Time: pit clocksource has been installed.
#...
#Real Time Clock Driver v1.12ac
#Generic RTC Driver v1.07

RTC is back for good!

This makes me so curios about the mysticals of linux time again:

http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=734efb467b31e56c2f9430590a9aa867ecf3eea1

After reading, re-reading and hopefully understanding i will try to find a solution, why 2.6.18-k7 decides to use "Programmable Interval Timer" of 1981(!) instead of our even younger and trusted friend "rtc" or even the fancy "acpi_pm" ..

Once i find helpful information or even a solution, i will post it here.


Thank you very much again!


Alexander Schories
Tuebingen, Germany


Sure, it could be, but I don't know anything about the architecture of the
kernel's clocksource stuff.

Good luck,
--Steve Langasek Give me a lever long enough and a Free OS Debian Developer to set it on, and I can move the world. vorlon@debian.org http://www.debian.org/
---------------------------------------------------------------------------------------------------------------------------------------------------------------------
 Iv'e been running for 2 weeks since.
Thank you,
matthew


THR4K wrote:
Package: linux-source-2.6.18
Version: 2.6.18.dfsg.1-11
Severity: important


Hello,

I have recently dist-upgraded my workstation from Sarge to Etch. As usual,
this operation worked fine, but sadly I get some trouble with the 2.6.18
kernel included into Etch.


To begin, I have tried the linux-image-2.6.18-4-k7 but at the first boot
the kernel freeze completely (the MagicSysReq keys doesn't work). So I
build my own image from the current linux-source-2.6.18 to test if the
problem could be eventualy solved with more specific options related to
my hardware configuration.

After rebooting without problem on my own linux-image kernel, for a moment
all seemed to work fine... until the whole system freeze again during the
configuration process of some new packages I have planned to install.

Despite the log files contains nothing useful, investigating a bit (eg.
more reboots and freezes later :-D show that the problem always occurs
during disk i/o operations, specially when loading the system at boot or
installing/deinstalling packages (heavy disk usage = faster freeze).


Of course, I did not have any problem when I use my previous kernel as
fallback (an old 2.6.8 version build with the Debian Sarge sources).

Perhaps a disk driver (I use the AIC7XXX kernel module) bug ?


my lspci :
00:00.0 Host bridge: Advanced Micro Devices [AMD] AMD-760 MP [IGD4-2P]
System Controller (rev 11)
00:01.0 PCI bridge: Advanced Micro Devices [AMD] AMD-760 MP [IGD4-2P]
AGP Bridge
00:07.0 ISA bridge: Advanced Micro Devices [AMD] AMD-766 [ViperPlus] ISA
(rev 02)
00:07.1 IDE interface: Advanced Micro Devices [AMD] AMD-766 [ViperPlus]
IDE (rev 01)
00:07.3 Bridge: Advanced Micro Devices [AMD] AMD-766 [ViperPlus] ACPI
(rev 01)
00:07.4 USB Controller: Advanced Micro Devices [AMD] AMD-766 [ViperPlus]
USB (rev 07)
00:0a.0 SCSI storage controller: Adaptec AHA-3960D / AIC-7899A U160/m
(rev 01)
00:0a.1 SCSI storage controller: Adaptec AHA-3960D / AIC-7899A U160/m
(rev 01)
00:0c.0 Multimedia audio controller: Ensoniq 5880 AudioPCI (rev 02)
00:0d.0 Ethernet controller: 3Com Corporation 3c905C-TX/TX-M [Tornado]
(rev 78)
01:05.0 VGA compatible controller: nVidia Corporation NV25 [GeForce4 Ti
4400] (rev a2)


Thanx for your help,
THR4K



-- System Information:
Debian Release: 4.0
  APT prefers testing
  APT policy: (500, 'testing')
Architecture: i386 (i686)
Shell:  /bin/sh linked to /bin/bash
Kernel: Linux 2.6.8-16.6-tigermp
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)

Versions of packages linux-source-2.6.18 depends on:
ii  binutils                      2.17-3     The GNU assembler, linker and bina
ii  bzip2                         1.0.3-6    high-quality block-sorting file co

Versions of packages linux-source-2.6.18 recommends:
ii  gcc                         4:4.1.1-15   The GNU C compiler
ii  libc6-dev [libc-dev]        2.3.6.ds1-13 GNU C Library: Development Librari
ii  make                        3.81-2       The GNU version of the "make" util

-- no debconf information






Reply to: