Hello there,

I'm facing strange (at least for me, that is) problem with unknown
wakeups. And huge amount of them, because I have average of 4000-5000
when using X. Here's example of my powertop -d output:

Cn                Avg residency
C0 (CPU running)        ( 5,5%)
C1                0,0ms ( 0,0%)
C2                0,1ms (43,4%)
C3               13,3ms (22,9%)
C4               13,7ms (28,1%)
P-states (frequencies)
  1200 MHz     0,0%
  1000 MHz     0,0%
   800 MHz     0,0%
   600 MHz   100,0%
Wakeups-from-idle per second : 4238,9   interval: 15,0s

Top causes for wakeups:
  58,5% ( 51,2)       <interrupt> : ohci_hcd:usb1, ohci_hcd:usb2,
ehci_hcd:usb3, eth0, [EMAIL PROTECTED]:0000:01:05.0
   7,7% (  6,7)            compiz : schedule_timeout (process_timeout)
   6,1% (  5,3)                 X : do_setitimer (it_real_fn)
   5,2% (  4,5)   xfce4-systemloa : schedule_timeout (process_timeout)
   2,9% (  2,5)       <interrupt> : acpi
   2,9% (  2,5)   xfce4-clipman-p : schedule_timeout (process_timeout)
   2,2% (  1,9)     <kernel core> : ehci_work (ehci_watchdog)
   1,3% (  1,1)          Terminal : schedule_timeout (process_timeout)
(...)

I have my linux (2.6.21.4) installed on hp nc4000 laptop, with pentium
m 1.4. NO_HZ enabled, kernel patched with patch from Arjan van de Ven
for aligned timers, radeon module patched for no vblank irq, even
tried HPET force patch, but it's not effective here.
I can achieve ~50 wps, but only in this way:
- kill X (this alone lowers wps only to ~1300)
- kill hald-addon-storage for my dvd (usb)
But then powertop shows 50 wps only in normal mode, when run with -d
it collects again above 1000. I'm running it on a vesafb console, and
only visible difference between these two is that the latter leaves
blinking cursor on.

Anyway, I would greatly appreciate if someone could give me some hint.

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

Reply via email to