Talk:How to make use of Dynamic Frequency Scaling

From ThinkWiki
Revision as of 22:33, 25 September 2007 by Matsch (Talk | contribs) (Turn off one Core (Core Duo on T60, Debian SID))
Jump to: navigation, search

CPUfreq "stuck"

Using the "acpi-cpufreq" and "processor" modules, I can use the performance and ondemand governor with great success on a T43, and it switches between 2.1 GHz and ~700 MHz without incident. However, sometimes the processor becomes "stuck" at ~700 MHz, and when I switch to the performance governor "cat /proc/cpuinfo" notes it is still at ~700 MHz.

I have not been able to precisely reproduce these conditions, but they have happened several times. It is cured by a reboot. I'm not running any userspace frequency governers. Anybody else experienced this peculiar behavior? gsmenden 11:20, 10 JAN 2006 (CEST)


I had something similar on my T43. It seems that BIOS interfers with cpufreqd's operation. In the end I set BIOS to "maximum performance" when the laptop is on AC, and let cpufreqd keep track of the speed. It seems to work for me (T43, 2669, 2.6.15-kernel)


Verfied for my T43. Article amended.


Note that this problem seems to disappear completely when using the speedstep-centrino module instead of acpi-cpufreq in a T43p. gsmenden 21:24 2 MAR 2006 (EST)


On T60p with Core 2 Duo 2Gz CPU gets "stuck" at the lowest possible setting (1Gz) after mache wakes up from sleep (standby). I have tried all possible options but nothing seems to be working right now.

CPU Speedstep management activation

I could not find the "processor" and "acpi-cpufreq" modules, thus leading to an empty /sys/devices/system/cpu/cpu0/ and preventing to set cpu throttling. I found the speedstep-centrino module which enables the feature. Environment : X41 (Pentium M), Debian Sid with custom 2.6.12 kernel. Is the Debian part of the article outdated ? Hope this helps, Vincent

speedstep-smi for T22

I had to use the speedstep-smi driver for my T22, not the speedstep-ich driver as stated in the how-to. Thomas


Yes, it was a mistake. Thanks for the note. Wyrfel 21:49, 27 Oct 2005 (CEST)


Extremely low freq on a T22

About an hour ago I made Speedstep work on a T22 running Ubuntu Breezy (5.10). Before that I had the machine randomly boot at 700MHz or 900MHz. That is nothing special. But, earlier today, when I booted it, it was running at 187MHz, according to both /proc/cpuinfo and Gnome's CPU frequency applet. It also took about 4 times as long to do some CPU-intensive processing than usually (grepping and sorting a known amount of text), so I'm still thinking that my Thinkpad really was running at 187MHz until I rebooted it.

Has anyone else noticed anything like this? Is there a way to replicate this behavior? Is there a way to "enable" this "step"?

-- _sd


Yes, i brought my X20 down to similarly low frequencies also with Ubuntu. I think it's possible through ACPI throttling, but I'm not sure if that was actually how i did it.

Wyrfel 23:51, 9 Jan 2006 (CET)


This is a BIOS problem. Disable any Powermanagment in BIOS, else it will boot with lower frequencies if your on battery or any other reason if didnt figure out. ACPI-Throttling does not change the frequency (read mhz) but something else.

echo 4:0 >/proc/acpi/processor/CPU/throttling

This will not change the mhz but will make it slower. I didnt understand how to make any use of it as it does not give more battery.

nusse Tue Mar 21 06:26:31 CET 2006

Obsolete daemons

Removed the note about daemons being obsolete. Using ondemand/conservate is *not* a replacement for daemons, they are generally smarter than a fixed governor and can adapt to different situations better.


I reinserted the note, but changed it a little. The point about that note is that most users get good results and less confusion with those two governors. Feel free to extend the section by some remarks about why one might want to use a deamon instead.

Wyrfel 10:40, 16 January 2006 (CET)


Elaborated a bit on it.

--Earthwings 23:15, 17 January 2006 (CET)


Thx, reads good.

Wyrfel 23:25, 17 January 2006 (CET)


Throttling useless?

I just removed the following comment:

==A note about CPU throttling==
On a modern CPU, throttling is useless, even it can increase power consumption instead of decreasing it. By forcing the CPU to sleep using throttling the CPU will reach a state higher as C2 less often. On a T43 it is a difference of more then 100mW.

I don't know what counts as a "modern CPU" but my desktop Athlon 64 3000+ 768 and my laptop 1.1GHz Pentium M both run *much* cooler at low frequencies than at high frequencies; it is just indisputable that they are using considerably less energy when throttled than before. So I think the above statement needs to be at the very least clarified before it goes into the main article. Ciphergoth 11:06, 20 June 2006 (CEST)


You seem to be very confused about this issue. CPU throttling does not change the clock, as it has nothing to do with clock speed. I will readd the comment, with some extra explanations to avoid the confusion. Downclocking a CPU will definately cause it to consume less power, that's the whole point of the governors.

--hmh 03:48, 27 October 2006 (CEST)

Turn off one Core (Core Duo on T60, Debian SID)

Is it possible to disable one core of the dual core processor? --Matsch 23:36, 18 January 2007 (CET)

Yes. If you have CONFIG_HOTPLUG_CPU=y, CONFIG_HOTPLUG=y, CONFIG_ACPI_HOTPLUG_CPU=y in kernel config. You can disable the second CPU with this command: echo 0 > /sys/devices/system/cpu/cpu1/online and enable it with echo 1 > /sys/devices/system/cpu/cpu1/online.

What the above commentator said is correct. However when I played with this on my T60p w/ T7600 core2 duo, both core thermal sensors reported an increase in temperature even with the machine idle. Re-enabling the second core eventually reduced the overall temperature of the processor. I didn't measure to see if there was any decrease in watts consumed.

Disabling one core with the suggested method makes the remaining core jump to 99-100% usage constantly without any other processes actually consuming more cpu. this is really strange and could explain the above observation. (Debian Linux 2.6.22-2-686 #1 SMP). --Matsch 21:33, 25 September 2007 (UTC)

Related question: I have an R61i and when I resume from sleep the second CPU is disabled. I tried enabling it manually:

root@lappy:/# echo 1 > /sys/devices/system/cpu/cpu1/online bash: echo: write error: Invalid argument 1

I haven't checked the temps but it definitely drains my battery faster with one cpu disabled, which would support the above observation.--Trogdor282 16:56, 9 September 2007 (UTC)

BIOS powersave disable/enable

This is for those who, like me, had earlier read that disabling powersave in the BIOS was a good thing. That worked on a stock Debian kernel 2.6.18-4-686. Doing this with the latest and greatest 2.6.22-1-686 made the entire cpufreq interface in /sys/devices/system/cpu/cpu0(1) disappear. (Re-) enabling the powersave in the BIOS solved this. Note: Thinkpad Z61m

--Gijs 09:30, 17 August 2007 (CEST)