Linux Kernel Power Regression - A Serious Issue for Portable Computing
21 replies, posted
[img]http://dl.dropbox.com/u/13181909/linux%20kernel%20power.jpg[/img]
Linux has been suffering from power regression in the kernel since v2.6.38, and is only getting worse with the upcoming 3.1 kernel. This is leading to shorter battery runtimes for portable Linux devices, such as laptops, netbooks, and even some Android phones. For the Linux kernel power regressions that were found a few months ago, and hit in Ubuntu 11.04, Phoronix has found the regression that's still present in the Linux 3.0 kernel. The power regression is caused by a change in ASPM, the Active-State Power Management, for PCI Express support.
[img]http://www.phoronix.com/data/img/results/linux_mobile_uffda/1.png[/img]
[img]http://www.phoronix.com/data/img/results/linux_mobile_uffda/2.png[/img]
[quote=Phoronix]If you were hoping that the Linux 3.1 kernel would fix the big power regression problem that's caused by PCI Express Active State Power Management (ASPM) being disabled on more systems since the release of the Linux 2.6.38 kernel, you're not in luck. There has not been any active work in this area. Making things worse though for mobile Linux users interested in a long lasting battery is another new regression in the Linux 3.1 kernel. Affected systems can easily see a 30% increase in power consumption simply when comparing the Linux 3.0 kernel to the current code being assembled for Linux 3.1. For an Intel Sandy Bridge notebook, the power consumption is up by 76% just over the course of this year from Linux kernel regressions.
Besides the new Linux 3.1 kernel power regression, there's also a power regression introduced in the Linux 3.0 kernel that has previously not been talked about on Phoronix. The Linux 3.0 power draw is up by 24% over the Linux 2.6.39 kernel. With all of these regressions, going from the Linux 2.6.38 to 3.1 kernel on an Intel Sandy Bridge notebook [b]increases the power consumption by 76% under Ubuntu Linux and a dramatically shorter battery life.[/b]
The notebook where this regression was spotted has an Intel Core i5 2520M quad-core Sandy Bridge CPU, Intel HD 3000 integrated graphics, 4GB of RAM, and 160GB Intel X-25 Extreme SSD. On the software side was the x86_64 Ubuntu 11.10 development snapshot as of 21 August with Unity 4.8.2, X.Org Server 1.10.3 RC2, xf86-video-intel 2.15.901, GCC 4.6.1, and an EXT4 file-system. The Linux 2.6.38, 2.6.39, 3.0, and 3.1 kernels were obtained from the Ubuntu mainline kernel repository. The Linux 3.1 kernel was a daily snapshot as of 21 August with a post-rc2 kernel.
The Phoronix Test Suite was used for benchmarking the system on each kernel release in an automated manner, complete with battery polling.[/quote]
[b][url=http://www.phoronix.com/scan.php?page=article&item=linux_mobile_uffda&num=1]Source[/url][/b]
and
[b][url=http://www.phoronix.com/scan.php?page=article&item=linux_31_power_regress&num=1]Source[/url][/b]
No wonder I was getting 1 hour battery with a 7800mAh battery on my laptop. Rated informative, and is there anything we can do to fix it?
"The New World Order" is behind this...
when they are going to fix it?
I was about to post a new thread to ask if anyone knew of any power saving techniques I can use for my new laptop. I ran a dual boot setup with win 7 and arch a few days ago and the battery life on arch was worse than what my old P4 laptop used to get with a ubuntu.
I hope this gets fixed soon, I've been dying to run linux on this thing.
Also found some tweaks if anyone needs them
[url]http://www.phoronix.com/scan.php?page=article&item=intel_i915_power&num=1[/url]
I've yet to try them out, but they claim they can dampen the effects of the newer kernel by forcing link state power management and some other things.
They are for Intel machines btw
Strange, I thought Mint was doing better than 7 in this area. To be fair my battery is busted, heh.
[QUOTE=wauterboi;32479811]Strange, I thought Mint was doing better than 7 in this area. To be fair my battery is busted, heh.[/QUOTE]
Mine is too. Everytime I boot into Gnome I'm greeted with "Battery capacity is 18.8%. Your battery may be old or broken."
My netbook running a 3.0.1 kernel (custom compile) is getting ~3 hours with a 2 year old battery... Anyway, hopefully this increased power drain get fixed.
my old laptop got a 15 minute battery life on 2.6.38
also fuck I could have sworn this was the drugs discussion forum for a second
[QUOTE=P320;32460351]3.1 kernel.[/QUOTE]
Whoa. 3.1?
"2.6.<odd>: still a stable kernel, but accept bigger changes leading up to it (timeframe: a month or two).
2.<odd>.x: aim for big changes that may destabilize the kernel for several releases (timeframe: a year or two)
<odd>.x.x: Linus went crazy, broke absolutely everything, and rewrote the kernel to be a microkernel using a special message-passing version of Visual Basic. (timeframe: "we expect that he will be released from the mental institution in a decade or two")." - [url=https://lkml.org/lkml/2005/3/2/247]Torvalds[/url]
[url=http://www.fewt.com/2011/09/about-kernel-30-power-regression-myth.html]About the Kernel 3.0 "Power Regression" Myth[/url]
I'll be sure to use Jupiter when I get a laptop.
[QUOTE=Jookia;32580985][url=http://www.fewt.com/2011/09/about-kernel-30-power-regression-myth.html]About the Kernel 3.0 "Power Regression" Myth[/url][/QUOTE]
I never really took the whole "Power Regression" thing serious though, as I haven't had any trouble at all on my netbook. If anything, my netbook lasts an hour or so longer in Fedora than when using Windows.
[QUOTE=Jookia;32580985][url=http://www.fewt.com/2011/09/about-kernel-30-power-regression-myth.html]About the Kernel 3.0 "Power Regression" Myth[/url][/QUOTE]
I can't take anything they say serious on that site. They just seem like elitist self-proclaimed douchebags stirring up shit.
[QUOTE=P320;32620604]I can't take anything they say serious on that site. They just seem like elitist self-proclaimed douchebags stirring up shit.[/QUOTE]
On Phoronix? The same site that says that Steam for Linux is official?
On the site I linked? The guys that actually explained the situation rather than just showing benchmarks?
Yeah Phoronix is really sensationalist. Every time there is a minuscule increase in power consumption, the Phoronix editor makes about 50 articles about it, and each one is like "It's been a week, and it's still not fixed", followed by "it's been 8 days and it's still not fixed!!!!".
[editline]4th October 2011[/editline]
And then he posts 8 pages of impossible to read graphs.
[QUOTE=PvtCupcakes;32628463]Yeah Phoronix is really sensationalist. Every time there is a minuscule increase in power consumption, the Phoronix editor makes about 50 articles about it, and each one is like "It's been a week, and it's still not fixed", followed by "it's been 8 days and it's still not fixed!!!!".
[editline]4th October 2011[/editline]
And then he posts 8 pages of impossible to read graphs.[/QUOTE]
Ubuntu 11.10 boot time is up, funny that that should happen in a newer OS!
Well there's no reason the power consumption should go up... if ANYTHING, it needs to go DOWN. Phoronix isn't the only site to state that it's a serious problem, so quit being so god-damn ignorant. My laptop runs hotter than it should with the 3.0.4 kernel than it did with the 2.6.37 kernel. My battery is broken, so I can't measure that, but the huge temp increase is an obvious sign of proof that it's a real-world problem.
[QUOTE=P320;32695313]Well there's no reason the power consumption should go up... if ANYTHING, it needs to go DOWN. Phoronix isn't the only site to state that it's a serious problem, so quit being so god-damn ignorant. My laptop runs hotter than it should with the 3.0.4 kernel than it did with the 2.6.37 kernel. My battery is broken, so I can't measure that, but the huge temp increase is an obvious sign of proof that it's a real-world problem.[/QUOTE]
Actually, there was a reason and it was linked a few posts above ([url=http://www.fewt.com/2011/09/about-kernel-30-power-regression-myth.html]here[/url]). The devs fixed a feature that was on by default because it was causing problems in some machines that couldn't support it. I don't know about you, but I'd rather a power increase than a defunct kernel.
Also, like the guy in the blog, I also spent loads of time optimising my eeepc netbook and therefore, I was also looking at temps, battery life, etc constantly and the change (if there was one) was negligible.
[QUOTE=Jookia;32628534]Ubuntu 11.10 boot time is up, funny that that should happen in a newer OS![/QUOTE]
You totally called it.
[B]The Other Issue With Ubuntu 11.10: Boot Speed[/B]
[url]http://www.phoronix.com/scan.php?page=article&item=ubuntu_oneiric_oktoberfest&num=2[/url]
Published today (Oct 11, 2011)
Not really surprising because he has done the same exact article on every single Ubuntu release since the dawn of time.
[editline]11th October 2011[/editline]
Oh my bad. Phoronix already did:
[B]Ubuntu 11.10 Boot Performance: It's Slowing Down[/B] on September 28.
[url]http://www.phoronix.com/scan.php?page=article&item=ubuntu_1110_bootchart&num=1[/url]
I almost forgot that he repeats the same crap once every two weeks.
[editline]11th October 2011[/editline]
I'm waiting for the next "Still waiting for Unreal Tournament 3 on Linux" or "Steam confirmed for Linux" article.
[QUOTE]Ubuntu 11.04: 5.80 seconds
Ubuntu 11.10: 6.76 seconds[/QUOTE]
Not another second of boot time in exchange for up to date software! Dear god, soon we'll be waiting tens of seconds for our computers to load Ubuntu!
[QUOTE=Jookia;32739037]Not another second of boot time in exchange for up to date software! Dear god, soon we'll be waiting tens of seconds for our computers to load Ubuntu![/QUOTE]
And it'll still be faster than Windows.
Why did I ever trust Phoronix?
[QUOTE=T3hGamerDK;32742461]And it'll still be faster than Windows.
Why did I ever trust Phoronix?[/QUOTE]
Dude, I use to trust TechRights.org, which turned out to be conspiracy heaven.
Sorry, you need to Log In to post a reply to this thread.