ksoftirqd eats all cpu

Bug #49141 reported by Norbert Tretkowski
8
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Incomplete
Low
Unassigned

Bug Description

Binary package hint: linux-source-2.6.15

When my notebook (an IBM T42) is up for a few hours, and was already suspended, ksoftirqd starts eating 100% CPU:

  PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
    3 root 39 19 0 0 0 R 91.5 0.0 40:07.05 ksoftirqd/0

This is not reproducible with Debian.

Tags: linux
Changed in linux-source-2.6.15:
importance: Untriaged → Low
Revision history for this message
Steve White (stevan-white) wrote :

I see this effect when I turn my wireless radio off.

Revision history for this message
Steve White (stevan-white) wrote :
Revision history for this message
sandoz (sandoz) wrote :

It happens on my laptop (IBM Thinkpad T43) too. Today I figured out, that if I turn my wireless device on again (via Fn+F5) the cpu usage drops down. After that I can turn if off without having the problems again. Strange.

Revision history for this message
Gareth Fitzworthington (mapping-gp-deactivatedaccount) wrote :

This bug has had no activity for a considerable period. This is a check to see if there is still interest in investigating this bug report.

Changed in linux-source-2.6.15:
status: New → Incomplete
Revision history for this message
stephen84s (stephen84s) wrote :

I am currently on an HP Pavilion dv6000 (AMD Turion 64x2,1 GB RAM, Broadcom Wireless) and have the same problem,
ksoftirqd is eating almost 50% of the CPU, sometimes it even hits 90-99% making it almost unusable...

Revision history for this message
stephen84s (stephen84s) wrote :

Sorry forgot to mention I am using Gutsy Gibbon.

Revision history for this message
Gareth Fitzworthington (mapping-gp-deactivatedaccount) wrote :

Can you post the result of:
uname -a
& also attach the file resulting from the following: lspci > lspci.txt
Thanks.

Revision history for this message
stephen84s (stephen84s) wrote :

Sorry for the late reply ,

Well this is a result of the uname -a command :-

"Linux stephen 2.6.22-14-generic #1 SMP Tue Feb 12 02:46:46 UTC 2008 x86_64 GNU/Linux"

And attached is the file generated by the lspci command.

Revision history for this message
Gareth Fitzworthington (mapping-gp-deactivatedaccount) wrote :

Can you try booting with the following kernel boot option:
nohz=off

This disables the 'tickless' aspect of the tickless kernel (enabled by default in Gutsy & perhaps other releases). ksoftirqd appears to be getting bombarded by by requests (probably from not quite fully polished hardware drivers).

Revision history for this message
stephen84s (stephen84s) wrote :

I tried booting with the nohz=off setting and this is what I observed,

Initially after logging in, the CPU usage was perfectly normal,

However after a couple of hours, the laptop went into power saving mode due to inactivity and then when I resumed it,
ksoftirqd was back in action taking 80-90% of my CPU,
following is the output of the "top" command in my terminal: -

  PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
    7 root 39 19 0 0 0 S 100 0.0 5:21.58 ksoftirqd/1

Revision history for this message
stephen84s (stephen84s) wrote :

Hi Gareth,
I m sorry for my last post I had not set the nohz=off parameter while booting correctly,

This time however I set it permanently it by editing the /boot/grub/menu.lst file and decided to observe for a couple of days before giving my feed back, and based on my observation over the past two days, "ksoftirqd" has not returned to eat my CPU ......... at least not yet.

Thanks for the fix.

Revision history for this message
Gareth Fitzworthington (mapping-gp-deactivatedaccount) wrote :

Marking as duplicate of Bug #153195
If anyone thinks this referral is errant then please re-mark as NEW.
Thanks.

Revision history for this message
Launchpad Janitor (janitor) wrote : This bug is now reported against the 'linux' package

Beginning with the Hardy Heron 8.04 development cycle, all open Ubuntu kernel bugs need to be reported against the "linux" kernel package. We are automatically migrating this linux-source-2.6.15 kernel bug to the new "linux" package. We appreciate your patience and understanding as we make this transition. Also, if you would be interested in testing the upcoming Intrepid Ibex 8.10 release, it is available at http://www.ubuntu.com/testing . Please let us know your results. Thanks!

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.