Hi Auke, How I can run on both cores is there any way? I am not sure if what i have modified is working fine as explained in my previous mail both core's c states shows 100%. As you said I can estimate but estimation has to be based on some model is there any standard way or one has to come up its own estimation model if any one knows? I have also seen in the code the USB and SATA related power estimation but not sure how that can be tested?
Thanks and Regards, Nisha On Mon, Dec 7, 2009 at 12:13 PM, Kok, Auke <[email protected]> wrote: > nisha jain wrote: > > Hi All, > > > > Thanks I am able to successfully get the cpu1 p-states though I am having > trouble cpu cores c states. > > I understand that in the "read_data" function it is accomplished though > after modifying for c1 core i am not > > able to get proper c states update. It always shows c0 and c1 core c0 > state 100%. How to validate both cores c states? > > run something on both cores? > > > Also how these core states can be mapped to power consumption, if at > least over all cpu power consumption can be calculated? > > if you can measure power consumption of each core, you could get those > numbers > > unfortunately we only know the power draw from the battery, so there is no > reliable way to measure core state consumption (you can only estimate). > > > > > > -------------------------------------------------------------------------------------------------------------------------- > > Output > > > --------------------------------------------------------------------------------------------------------------------------- > > Cn Avg residency P0-states > (frequencies) > > C0 (cpu running) (100.0%) 2.21 Ghz > 3.1% > > polling 0.0ms ( 0.0%) 1.60 Ghz > 0.0% > > C1 mwait 0.0ms ( 0.0%) 1200 Mhz > 96.9% > > C2 mwait 0.0ms ( 0.0%) > > C4 mwait 0.0ms ( 0.0%) > > > > C1 (cpu running) (100.0%) P1-states > (frequencies) > > polling 0.0ms ( 0.0%) 2.21 Ghz > 0.0% > > C1 mwait 0.0ms ( 0.0%) 1.60 Ghz > 0.0% > > C2 mwait 0.0ms ( 0.0%) 1200 Mhz > 100.0% > > C4 mwait 0.0ms ( 0.0%) > > Wakeups-from-idle per second : 451.7 interval: 3.0s > > no Battery ACPI power usage estimate available > > > > Top causes for wakeups: > > 38.6% (333.3) <interrupt> : extra timer interrupt > > 22.7% (195.7) npviewer.bin : hrtimer_start_range_ns > (hrtimer_wakeup) > > 7.4% ( 64.0) <interrupt> : PS/2 keyboard/mouse/touchpad > > 7.1% ( 61.0) <interrupt> : nvidia > > 5.7% ( 49.3) pulseaudio : hrtimer_start_range_ns > (hrtimer_wakeup) > > 3.1% ( 26.7) <kernel core> : hrtimer_start_range_ns > (tick_sched_timer) > > 2.4% ( 20.3) gnome-panel : hrtimer_start_range_ns > (hrtimer_wakeup) > > 2.4% ( 20.3) firefox : hrtimer_start_range_ns > (hrtimer_wakeup) > > 2.2% ( 19.3) <interrupt> : ath > > > > > > > -------------------------------------------------------------------------------------------------------------------------- > > > > Thanks and Regards, > > Nisha > > > > > > > >
_______________________________________________ Power mailing list [email protected] http://www.bughost.org/mailman/listinfo/power
