LinuxQuestions.org
LinuxAnswers - the LQ Linux tutorial section.
Go Back   LinuxQuestions.org > Forums > Linux Forums > Linux - Software > Linux - Kernel
User Name
Password
Linux - Kernel This forum is for all discussion relating to the Linux kernel.

Notices

Reply
 
Search this Thread
Old 08-10-2011, 06:50 PM   #1
steevven1
LQ Newbie
 
Registered: Aug 2011
Posts: 2

Rep: Reputation: Disabled
Linux Kernel 3.0 Battery Life/Power Consumption


Hello all,

I have just upgraded from Linux kernel 2.6.38 to Linux 3.0 by installing the .deb packages from Oneiric (new Ubuntu version), and I am experiencing a major drop in battery life (about 25% less time) on my laptop. It is also running noticeably hotter. Reverting to the older kernel brings back the good battery life and cool temperatures.

I need the 3.0 kernel because of some of the drivers it contains. Does anyone have any idea what to do to fix this power issue? Is the 3.0 kernel just supposed to be that much worse?

I'm running Xubuntu 11.04 64-bit.

Thank you!


Additional info:

sudo powertop yields the following interesting information:

Top causes for wakeups:
52.9% (317. [Rescheduling interrupts] <kernel IPI>

What is that?
 
Old 08-10-2011, 07:50 PM   #2
pljvaldez
Guru
 
Registered: Dec 2005
Location: Somewhere on the String
Distribution: Debian Squeeze (x86)
Posts: 6,092

Rep: Reputation: 269Reputation: 269Reputation: 269
You might search around the phoronix website for "kernel regressions". I recall some possible power related regression issues in July which were correctable by passing different kernel parameters at boot.
 
Old 08-11-2011, 12:28 AM   #3
steevven1
LQ Newbie
 
Registered: Aug 2011
Posts: 2

Original Poster
Rep: Reputation: Disabled
Thanks for the suggestion. I did try what they said, as well as some other things. Here were my results:

I tried all of the following kernel parameters (result in italics):

pcie_aspm=force
No improvement

acpi=noirq
No improvement

noapic
Command not recognized. Won't boot.

nolapic
Command not recognized. Won't boot.

acpi=off
No improvement

More ideas?
 
Old 10-06-2011, 04:05 AM   #4
salasi
Senior Member
 
Registered: Jul 2007
Location: Directly above centre of the earth, UK
Distribution: SuSE, plus some hopping
Posts: 3,884

Rep: Reputation: 774Reputation: 774Reputation: 774Reputation: 774Reputation: 774Reputation: 774Reputation: 774
I have doubts that this directly helps the OP, but I want to draw attention to this blog post.

Essentially, this argues that there is no regression, but a bug fix that has some unfortunate side effects in some circumstances, and that there may be work-arounds, again, in some circumstances.

To summarise, it is a semantic argument; that is, it argues that to call it a regression is to misuse the term, in this case. (And, as far as can be told, the Phorons are pretty much unrepentant about their use of the word regression.)

While this doesn't affect the amount of power drawn by the OP's computer by even a milliampere, I tend to agree with the blog post (fewt) argument, but I will freely admit that if you are sitting in front of a computer that suddenly has half the battery life that it had previously, it will seem very like a regression.
 
Old 10-06-2011, 12:10 PM   #5
pljvaldez
Guru
 
Registered: Dec 2005
Location: Somewhere on the String
Distribution: Debian Squeeze (x86)
Posts: 6,092

Rep: Reputation: 269Reputation: 269Reputation: 269
Quote:
Originally Posted by salasi View Post
I have doubts that this directly helps the OP, but I want to draw attention to this blog post.

Essentially, this argues that there is no regression, but a bug fix that has some unfortunate side effects in some circumstances, and that there may be work-arounds, again, in some circumstances.
...

I tend to agree with the blog post (fewt) argument, but I will freely admit that if you are sitting in front of a computer that suddenly has half the battery life that it had previously, it will seem very like a regression.
I saw that recently as well and fully agree with your statement. I posted the Phoronix stuff back in August when the power issues were just coming around.

Here's hoping the brilliant kernel maintainers can figure out another way to handle this so there isn't a problem for some people. Or I can dream hardware vendors will quit doing stupid things - but that might be a pipe dream.
 
  


Reply


Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Laptop Battery life.. Does Distro affect battery life?? jeevanism Linux - Hardware 6 07-09-2011 02:16 AM
Linux machine power consumption kbarani Linux - Software 2 11-05-2009 08:37 AM
LXer: Linux: Reducing Power Consumption LXer Syndicated Linux News 0 07-06-2007 04:31 PM
Windows: 4-5 hour battery life, Gentoo : ~2:30 battery life numerous Linux - Laptop and Netbook 3 01-17-2005 08:39 PM
windows, good battery life -> fedora, bad battery life daedalusq1115 Linux - Laptop and Netbook 1 01-17-2005 06:40 AM


All times are GMT -5. The time now is 05:53 AM.

Main Menu
My LQ
Write for LQ
LinuxQuestions.org is looking for people interested in writing Editorials, Articles, Reviews, and more. If you'd like to contribute content, let us know.
Main Menu
Syndicate
RSS1  Latest Threads
RSS1  LQ News
Twitter: @linuxquestions
identi.ca: @linuxquestions
Facebook: linuxquestions Google+: linuxquestions
Open Source Consulting | Domain Registration