Hi, Well, I removed both s3cmci.ko and ar6000.ko and here's what powertop has to say
Wakeups-from-idle per second : 217.8 interval: 5.0s Top causes for wakeups: 92.4% (201.0) <interrupt> : S3C2410 Timer Tick 4.4% ( 9.6) python : schedule_hrtimeout_range (hrtimer_wakeup) 1.1% ( 2.4) <interrupt> : s3c2410_udc 0.5% ( 1.0) gsm0710muxd : schedule_hrtimeout_range (hrtimer_wakeup) 0.5% ( 1.0) Xglamo : hrtimer_start (it_real_fn) 0.4% ( 0.8) enlightenment : schedule_hrtimeout_range (hrtimer_wakeup) 0.2% ( 0.4) batget : schedule_hrtimeout_range (hrtimer_wakeup) 0.2% ( 0.4) <kernel core> : schedule_delayed_work_on (delayed_work_timer_fn) 0.1% ( 0.2) dropbear : sk_reset_timer (tcp_write_timer) 0.1% ( 0.2) <kernel core> : page_writeback_init (wb_timer_fn) 0.1% ( 0.2) <kernel module> : neigh_table_init_no_netlink (neigh_periodic_timer) 0.1% ( 0.2) Xglamo : schedule_hrtimeout_range (hrtimer_wakeup) 0.1% ( 0.2) <kernel core> : neigh_table_init_no_netlink (neigh_periodic_timer) seems like there is still a clock ticking away at 40Hz. Or is this not the WLAN clock? I got about 8 hrs through the night (no wakes) and lost 12%. This is pretty much the same as earlier. Also, as I has asked earlier, how do I __really__ turn off bluetooth and GPS? Is there something lurking ;-) underneath too? Also - just a thought - can't powertop be modified to handle the freerunner better? Or rather a diff utility derived from powertop for the freerunner/openmoko phones. That will really help see where the power is going. Of course you'd need someone with a pretty thorough understanding of the hardware to do this. Hope I'm not diverting too much :-) -- View this message in context: http://n2.nabble.com/Isn%27t-Wifi-supposed-to-be-off-by-default--tp2154678p2171969.html Sent from the Openmoko Kernel mailing list archive at Nabble.com.
