Comment 41 for bug 177895

Revision history for this message
Steffen Heyne (steffen-heyne) wrote :

same here: Dell D630 Core2 Duo 2 GHz with hardy alpha 6 and kernel 2.6.24-12.20-i386

Wakeups-from-idle per second : 5426,0 interval: 3,0s
Power usage (5 minute ACPI estimate) : 0,0 W (1602,1 hours left)

Top causes for wakeups:
  33,9% (155,3) <kernel IPI> : Rescheduling interrupts
  14,1% ( 64,7) <interrupt> : extra timer interrupt
  13,6% ( 62,3) <interrupt> : i915@pci:0000:00:02.0
   8,7% ( 39,7) USB device 3-1 : USB-PS/2 Optical Mouse (Logitech)
   8,6% ( 39,3) <interrupt> : uhci_hcd:usb1, uhci_hcd:usb3, ehci_hcd:usb7
   4,3% ( 19,7) <interrupt> : iwl3945
   2,6% ( 12,0) <kernel module> : usb_hcd_poll_rh_status (rh_timer_func)
   2,5% ( 11,3) geyes_applet2 : schedule_timeout (process_timeout)
   2,2% ( 10,0) cpufreq-selecto : cpufreq_governor_dbs (delayed_work_timer_fn)

turning off one core with

   echo 0 > /sys/devices/system/cpu/cpu1/online

completely changes the things:

Wakeups-from-idle per second : 192,3 interval: 3,0s
Power usage (5 minute ACPI estimate) : 0,0 W (1969,8 hours left)

Top causes for wakeups:
  25,6% ( 93,0) <interrupt> : i915@pci:0000:00:02.0
  22,8% ( 83,0) USB device 3-1 : USB-PS/2 Optical Mouse (Logitech)
  22,7% ( 82,3) <interrupt> : uhci_hcd:usb1, uhci_hcd:usb3, ehci_hcd:usb7
   5,4% ( 19,7) <interrupt> : iwl3945
   5,3% ( 19,3) compiz.real : schedule_timeout (process_timeout)
   3,3% ( 12,0) <kernel module> : usb_hcd_poll_rh_status (rh_timer_func)
   2,9% ( 10,7) geyes_applet2 : schedule_timeout (process_timeout)
   2,8% ( 10,0) cpufreq-selecto : cpufreq_governor_dbs (delayed_work_timer_fn)

that's an extreme difference!