Mihal K wrote: > Jose Rodriguez wrote: >>>> I installed 2.6.24 and everything seems OK except that my CPU >>>> spends about 75% of its time in C2 [snip ipw2200 problem]So to recap, >>>> 2.6.24 >> compiled from source as well as from Debian repos (hpet=force) do >> show a behavior not seen before with 2.6.22 with the mentioned patches.
The patches are i)HPET force and ii)http://www.lesswatts.org/projects/powertop/patches/kernel-2.6.22-rc1.patch >>From http://www.lesswatts.org/projects/powertop/faq.php > > Why did my power usage increase with dynticks? Why do I never reach > sleep states deeper than C2? > > The current ACPI sleep function requires wakeups to reach deeper sleep > states. Certain drivers, such as ipw2200, limit the initial sleep state > to C2 due to DMA activity. With dynticks, which tries to sleep as long > as possible, this means that your processor will, most of the time, only > reach C2 and thus waste power compared to a non-dynticks kernel if it is > mostly idle. This will be solved by an upcoming cpuidle patch. Until > this hits mainline, you should compare your power usage with and without > dynticks (and maybe with different HZ values) to figure out which is > currently the best for your workload. > ------------------------------------------------ > Though it looks like exactly the opposite has happened - 2.6.22 (without > cpuidle) had no problems and 2.6.24 (with cpuidle) has trouble going to > the sleep states deeper than C2. OK, I'm a little bit confused with all these dynticks, tickless kernel, HPET, cpuidle and friends. What is it new now that is causing the problem? Is there any configuration option in 2.6.24 I can play with to recompile it to obtain the nice behaviour I had with 2.6.22 regarding ipw2200? This is, not being stuck at C2 when the wifi radio is on? Thanks _______________________________________________ Power mailing list [email protected] http://www.bughost.org/mailman/listinfo/power
