Dave Russell wrote:
> 
> Cn          Avg residency (10s)         P-states (frequencies)
> C0 (cpu running)        (12.4%)
> C1                0.0ms ( 0.0%)         2.00 Ghz     9.8%
> C2                1.5ms (81.9%)         1.60 Ghz     0.2%
> C3                1.5ms ( 3.4%)         1333 Mhz     0.4%
> C4                1.3ms ( 2.3%)          800 Mhz    89.4%

these timings are... really bad. you can save quite a few watts on 
this box.

> 
> Wakeups-from-idle per second : 592.1
> Power usage (5 minute ACPI estimate) :   0.5 W (221.6 hours left)
> 
> Top causes for wakeups:
>    27.4% (214.6)              java : futex_wait (hrtimer_wakeup)

I don't know what you're doing with java, but it's severely 
misbehaving....
> ############# Scenario 3 #############
>       PowerTOP version 1.7       (C) 2007 Intel Corporation
> 
> Cn          Avg residency (3s)          P-states (frequencies)
> C0 (cpu running)        (48.4%)
> C1                0.0ms ( 0.0%)         2.00 Ghz    26.5%
> C2                3.4ms ( 2.3%)         1333 Mhz     1.9%
> C3                2.2ms (10.1%)         1067 Mhz     3.8%
> C4                2.6ms (39.2%)          800 Mhz    67.4%

still really bad...

>    31.9% (133.3)       <interrupt> : uhci_hcd:usb1, ehci_hcd:usb5, Intel
> ICH6, eth0, [EMAIL PROTECTED]:0000:01:00.0

looks like you don't have USB suspend enabled; current powertop can 
help you enable that..


>    11.5% ( 48.0)             totem : do_nanosleep (hrtimer_wakeup)
 >     2.9% ( 12.0)             totem : schedule_timeout (process_timeout)

... and totem isn't being so nice

>     2.3% (  9.7)      strigidaemon : do_nanosleep (hrtimer_wakeup)

and neither is this thing (I've not seen this one before)

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

Reply via email to