Am Donnerstag, den 04.02.2010, 16:37 +0100 schrieb Adam Sloboda:
> At Thu, 04 Feb 2010 07:25:03 -0800,
> Arjan van de Ven wrote:
> > 
> > >
> > > I am sorry, are you trying to convince me that my problem is not real?
> > > You can compare to my completely idle system (I even closed the
> > > browser but with these residency values it doesn't really matter
> > > anymore in terms of power saving):
> > >
> > > Cn                  Avg residency
> > > C0 (cpu running)        ( 4.7%)
> > > C0                  0.1ms ( 0.0%)
> > > C1 mwait    0.0ms ( 0.0%)
> > > C2 mwait    0.1ms ( 0.5%)
> > > C6 mwait    2.9ms (94.8%)
> > > P-states (frequencies)
> > >    2.41 Ghz     2.3%
> > >    2.40 Ghz     0.4%
> > >    1.60 Ghz     0.9%
> > >     800 Mhz    96.5%
> > > Wakeups-from-idle per second : 369.8      interval: 15.0s
> > > Power usage (ACPI estimate): 12.2W (3.0 hours)
> > > Top causes for wakeups:
> > >    73.0% (363.4)<kernel IPI>  : Rescheduling interrupts
> > >     9.2% ( 45.7)<interrupt>  : extra timer interrupt
> > >     6.1% ( 30.3)<interrupt>  : iwlagn
> > >     4.0% ( 19.8)<interrupt>  : acpi
> > >     2.1% ( 10.5)           gkrellm : schedule_hrtimeout_range 
> > > (hrtimer_wakeup)
> > >     2.0% ( 10.1)             xfwm4 : schedule_hrtimeout_range 
> > > (hrtimer_wakeup)
> > >
> > 
> > 
> > so one of the things of this rescheduling interrupt is that it is sort of a 
> > side effect of other wakeups/code running.
> > I bet that if you, say, stop gkrellm, you'll see it go down proportionally 
> > ....
> 
> Not really.  As I wrote in the first mail, it's all ok until I run any
> X processes (running just GDM is fine).
> 
> Cn              Avg residency
> C0 (cpu running)        ( 2.3%)
> C0              0.1ms ( 0.0%)
> C1 mwait        0.0ms ( 0.0%)
> C2 mwait        0.1ms ( 0.2%)
> C6 mwait        5.6ms (97.5%)
> P-states (frequencies)
>   2.41 Ghz     3.4%
>   2.40 Ghz     0.0%
>   1.60 Ghz     4.0%
>    800 Mhz    92.7%
> Wakeups-from-idle per second : 195.6  interval: 15.0s
> Power usage (ACPI estimate): 12.2W (2.9 hours) 
> Top causes for wakeups:
>   76.6% (242.6)      <kernel IPI> : Rescheduling interrupts 
>    9.3% ( 29.5)       <interrupt> : iwlagn 
>    5.9% ( 18.7)       <interrupt> : acpi 
>    2.0% (  6.3)       <interrupt> : extra timer interrupt 
>    1.3% (  4.2)       <interrupt> : i...@pci:0000:00:02.0 
> 
> 
> And here with networking turned off:
> 
> Cn              Avg residency
> C0 (cpu running)        ( 2.3%)
> C0              0.0ms ( 0.0%)
> C1 mwait        0.0ms ( 0.0%)
> C2 mwait        0.1ms ( 0.2%)
> C6 mwait        5.9ms (97.5%)
> P-states (frequencies)
>   2.41 Ghz     4.3%
>   2.40 Ghz     3.2%
>   1.60 Ghz     4.3%
>    800 Mhz    88.1%
> Wakeups-from-idle per second : 189.1  interval: 15.0s
> Power usage (ACPI estimate): 11.1W (3.2 hours) 
> Top causes for wakeups:
>   82.9% (254.8)      <kernel IPI> : Rescheduling interrupts 
>    7.2% ( 22.0)       <interrupt> : acpi 
^^^^^^^^^^^^^^ this is not normal. Try different acpi options, apic and
so on. Here yyou can find some of them:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=Documentation/kernel-parameters.txt;h=736d45602886ea7b08024f246f06e9c7c6dc1a05;hb=HEAD

And first of all: try newest git kernel. 

This is definitely not IPI bug. You have some timer/acpi issue, this is
why powertop cant see it.

>    2.7% (  8.3)       <interrupt> : extra timer interrupt 
>    1.1% (  3.3)       <interrupt> : i...@pci:0000:00:02.0 
>    0.7% (  2.3)       <interrupt> : ahci 
>    0.7% (  2.0)       xfce4-hdaps : schedule_hrtimeout_range (hrtimer_wakeup) 
> 
> _______________________________________________
> Power mailing list
> [email protected]
> http://www.bughost.org/mailman/listinfo/power


_______________________________________________
Power mailing list
[email protected]
http://www.bughost.org/mailman/listinfo/power

Reply via email to