I just installed powertop on my laptop and i have to recognize that my laptop wont go into any c-state. Here is a screenshot
http://mitglied.lycos.de/frog1/etc/powertop.jpg
Can i force my cpu to go into c-states?
install powernowd or cpudynd etc?
]]>Does anyone know if the firefox patch is implemented on repository version?
While there are many things that can keep the Firefox browser active, such as animations in websites you are looking at, Firefox also had a few things that would keep Firefox active even when looking at a blank page. We've fixed two of the most visible bugs in this space, and are collecting more fixes over time.
Mike Hommey has filed a cleaner fix for these in the Firefox bugzilla as bug 380558.
You can download Mike's Firefox 2.0 patchkit here.
If not It would be nice to have the patched firefox version in the repos.
At one stage they were applied, but they caused more problems than they fixed. I think it was decided to wait for them to be merged upstream or stabilise.
Issue: No. You can't "force" it into a sleep state. A sleep state means it's effectively doing nothing, and thus, your computer will do nothing. Fix the programs that are preventing it from sleeping between tasks instead.
James
]]>While there are many things that can keep the Firefox browser active, such as animations in websites you are looking at, Firefox also had a few things that would keep Firefox active even when looking at a blank page. We've fixed two of the most visible bugs in this space, and are collecting more fixes over time.
Mike Hommey has filed a cleaner fix for these in the Firefox bugzilla as bug 380558.
You can download Mike's Firefox 2.0 patchkit here.
If not It would be nice to have the patched firefox version in the repos.
]]>well, whatever, this thing is great, except that I have to cap max_cstate (via sysfs) to 2, else my core 2 duo whines and hisses. so, I can not gain much by squeezing the 1000+ wake up events I have!
anyway, this thing is great to know which processes sporadically, but very often, wake up (which can not be seen in top). e.g I had a bad python process, waking more often than nvidia...
]]>rats, what USB problems would enabling usb suspend as a module cause? I could set it up to only be active when on battery power. 1W of power savings is significant when my computer only uses ~19W with wireless and bluetooth on (in Windows.. I haven't done much setting up power management in Arch yet, even though I haven't run windows on any of my personal computers since Vista Business was released). Every little bit helps
rmmod uhci-hcd
rmmod ohci-hcd
rmmod ehci-hcd
power saved.
(one or two of the above will barf out, depending on what USB chipsets your computer has, that's ok)
]]>PowerTOP is in community: pacman -S powertop
For me, the kernel wakeups are mainly from nvidia (shame on you blob!), kicker (from kde-mod), usb and ndiswrapper.
Without this tool, I think I would not have blamed kicker for eating my battery, but now I can check this!
]]>Use the viper kernel
i don't see the relevance of that to this thread.
g0t rice?
]]>attila wrote:From my view you can find the reason in the doku about hrtimer in the kernel sources. The kernel devs don't suggest it and speaks about a significant overhead if it is on.
Ok I missed that part. Not I'm satisfied .
For what its worth, there are apparently also worse problems than just the overhead:
http://archlinux.org/pipermail/arch/200 … 14482.html
From my view you can find the reason in the doku about hrtimer in the kernel sources. The kernel devs don't suggest it and speaks about a significant overhead if it is on.
Ok I missed that part. Not I'm satisfied .
A little comment about "building your own kernel": The building system of arch is the best understandable package management what i know and having an own kernel package is never stupid. Try it one time and you will understand why me personally don't see it as an problem to do it. But it is really okay if you don't want it.
I'm not building packages, my server does that automatically and pacman/abs is the only reason I stick with Arch, so I know it's easy. I just didn't knew the device adds so much overheat to the code.
]]>And if it's a big issue, give some reasons for rejecting the bug instead.
Didn't they quote the docs that it introduces fair amount of overhead, arr? Isn't that a reason decent enough?
]]>