Arjan van de Ven wrote:
> Hi,
>
> I just released PowerTOP version 1.9
>
> new in this release:
> * support for the power save mode of the iwl3945/iwl4965 Intel
> wireless drivers
> * support for 2 new 2.6.24(-rc) features
> - PowerTOP now reports USB device activity by mentioning which USB
> device is causing the activity (and not just the interrupt number
> that is shared between all USB and other devices)
> - PowerTOP now reports new wakeup causes that previously were
> hidden; these are for kernel caused wakeups. In older versions
> of PowerTOP these were "ghost" wakeups that weren't attributed
> to a specific cause
>
Great! Unfortunately my ghost wakeups haven't gone away.
PowerTOP 1.9 (C) 2007 Intel Corporation
Collecting data for 60 seconds
< Detailed C-state information is only available on Mobile CPUs (laptops) >
P-states (frequencies)
2.14 Ghz 0.0%
1.60 Ghz 100.0%
Wakeups-from-idle per second : 5.1 interval: 60.0s
Top causes for wakeups:
20.4% ( 0.3) thunderbird-bin : __neigh_event_send (neigh_timer_handler)
12.2% ( 0.2) kded : schedule_timeout (process_timeout)
10.2% ( 0.2) <interrupt> : ahci
7.1% ( 0.1) Xorg : do_setitimer (it_real_fn)
7.1% ( 0.1) <kernel core> : neigh_table_init_no_netlink
(neigh_periodic_timer)
6.1% ( 0.1) konsole : schedule_timeout (process_timeout)
6.1% ( 0.1) klauncher : schedule_timeout (process_timeout)
5.1% ( 0.1) dhclient3 : schedule_timeout (process_timeout)
4.1% ( 0.1) <kernel core> : page_writeback_init (wb_timer_fn)
4.1% ( 0.1) touch : start_this_handle (commit_timeout)
2.0% ( 0.0) Xorg : schedule_timeout (process_timeout)
2.0% ( 0.0) syslogd : do_setitimer (it_real_fn)
2.0% ( 0.0) hald : schedule_timeout (process_timeout)
2.0% ( 0.0) kicker : schedule_timeout (process_timeout)
2.0% ( 0.0) akregator : schedule_timeout (process_timeout)
1.0% ( 0.0) <interrupt> : PS/2 keyboard/mouse/touchpad
1.0% ( 0.0) <kernel core> : __dst_free (delayed_work_timer_fn)
1.0% ( 0.0) pdflush : __make_request (blk_unplug_timeout)
1.0% ( 0.0) <kernel core> : ip_rt_init (delayed_work_timer_fn)
1.0% ( 0.0) kdesktop : schedule_timeout (process_timeout)
1.0% ( 0.0) <kernel core> : inet_initpeers (peer_check_expire)
1.0% ( 0.0) cron : do_nanosleep (hrtimer_wakeup)
Suggestion: Enable the CONFIG_ACPI_BATTERY kernel configuration option.
This option is required to get power estimages from PowerTOP
$ uname -a
Linux singularity 2.6.24-rc1-gcf409d1d #4 SMP Mon Oct 29 23:00:21 GMT
2007 x86_64 GNU/Linux
[-gcf409d1d, I hear you ask?
I have stable TSC's. I don't see why anyone's computer would need to
wake up
twice a second, for the sole purpose of checking the TSC, but there you go.
]
$ git-diff v2.6.24-rc1
diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c
index c8a9d13..a03a06f 100644
--- a/kernel/time/clocksource.c
+++ b/kernel/time/clocksource.c
@@ -175,7 +175,7 @@ static void clocksource_check_watchdog(struct
clocksource *cs)
if (watchdog)
del_timer(&watchdog_timer);
watchdog = cs;
- init_timer(&watchdog_timer);
+ init_timer_deferrable(&watchdog_timer);
watchdog_timer.function = clocksource_watchdog;
/* Reset watchdog cycles */
_______________________________________________
Power mailing list
[email protected]
http://www.bughost.org/mailman/listinfo/power
