You have to check if 'CONFIG_NO_HZ' is set to Y.
Important: If you are using a kernel older than 2.6.21 this feature
won't be available. Until now it is not available on 64-bit system like
Intels Core 2 or AMDs k8, though there are already patches available ( I
think they can be found at http://www.tglx.de/projects/hrtimers/ , but
I'm not 100% sure if those are the right ones ). This feature is
expected to be available for 64-bit as of kernel 2.6.23.

Peter


On Sun, 2007-05-27 at 11:29 +0200, Antoine Caillau wrote:
> Hi
> 
> how can I check if my kernel is tickless. What is the kernel option ? 
> 
> Thanks for your answer,
> 
> Antoine
> 
> On 27/05/07, Peter Ganzhorn <[EMAIL PROTECTED]> wrote:
>         Are you running a tickless kernel?
>         I'd guess that the "extra timer interrupt" is the former timer
>         interrupt, which will cause 1000 wakeups if you're running a
>         1000hz
>         kernel. Tickless kernels do not wake up because of that
>         interrupt. 
>         
>         Your CPU has only 1 ms in C3 state - this could be caused by a
>         1000hz
>         kernel as well, because it will wake up every ms.
>         
>         Peter
>         
>         
>         On Sun, 2007-05-27 at 10:36 +0200, Antoine Caillau wrote:
>         > Hi, 
>         >
>         > Since 1.4 (or may be 1.3, I do not tested) I've a process
>         who wake up
>         > my CPU a lot. It is called "extra timer interrupt". Can
>         anyone tell me
>         > what it is ?
>         >
>         > Thanks a lot for your help. 
>         >
>         > Antoine
>         >
>         >      PowerTOP version 1.4       (C) 2007 Intel Corporation
>         >
>         > Cn          Avg residency (5s)  Long term residency avg
>         > C0 (cpu running)        (19.6%)
>         > C1                 0.0ms ( 0.0%)                   0.0ms
>         > C2                3.4ms ( 3.7%)                   2.2ms
>         > C3                1.0ms (76.7%)                   1.0ms
>         >
>         >
>         > Wakeups-from-idle per second : 815.9
>         > Power usage (ACPI estimate) :  11.3 W (5.4 hours left)
>         >
>         > Top causes for wakeups:
>         >   95.6% (1166.6)       <interrupt> : extra timer interrupt
>         >    1.1% ( 13.8)       <interrupt> : ohci1394, ipw3945 
>         >    0.7% (  8.8)       <interrupt> : ehci_hcd:usb1,
>         uhci_hcd:usb2
>         >    0.3% (  3.4)   <kernel module> : usb_hcd_poll_rh_status
>         > (rh_timer_func)
>         >    0.3% (  3.4)             pcscd : do_nanosleep
>         (hrtimer_wakeup) 
>         >    0.2% (  2.8)        powersaved : queue_delayed_work_on
>         > (delayed_work_timer_fn
>         > )
>         >
>         > _______________________________________________
>         > 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