Reply
Thread Tools
Posts: 306 | Thanked: 38 times | Joined on Dec 2009 @ Austria
#31
hi, you can find attached the output file

following widgets i have runing:

foreca weather: update every hour
calendar home widget
simple fmtx widget
media player widget
personal data plan monitor widget
Attached Files
File Type: txt top_out.txt (8.7 KB, 121 views)
 
Posts: 2,829 | Thanked: 1,459 times | Joined on Dec 2009 @ Finland
#32
/rant
Sorry guys I do not have any more juice to write again all fck stuff that has been said across this board already and that can be found from maemo wiki. It just pisses me off.

Most of stuff is not obvious but after reading wiki and other user use cases and reports you tend to get picture what to do. Thatīs called researching, learning and respecting others.

To orginal poster, Iīm facepalming while Iīm saying this:
- Remove widgets
- Revert back to default speed
- Check for background daemons or apps that install stuff that stays on in background or alter maemo OS or jump starts from d-bus hooks (apmefo, recaller etc. you get the point), use top and htop apps for this.
- Take as much as possible modules/features off before running powertop (I have listed these in other thread please read it)
- After you have disabled most of stuff and powertop still gives you device that doesnīt enter sleep then itīs probably time to flash it. If not then start one-by-one enabling stuff and see when battery consumption jumps. E.g. put half of widgets on and test for couple of days -> if consumption high then again disable half of widgets....reapeat until you find your misbehaving app/widget/daemon etc.

- I do not even understand why Iīm repeating this stuff that should be obvious for people who start overclocking and most probably installing stuff from testing/devel.

I hope that you find resolution and post some helpful info here. Good luck.
__________________
TMO links: [iSpy] - [Power search] - [Most thanked] - [Cordia - Maemo5 UI on top MeeGo Core] - [CommunitySSU]

Last edited by slender; 2011-03-01 at 09:23.
 

The Following User Says Thank You to slender For This Useful Post:
Posts: 306 | Thanked: 38 times | Joined on Dec 2009 @ Austria
#33
thanks slender,
but i had these problems long before overclocking, cssu and other things...
if we find the reason, we will post it here, thanks
 
Posts: 2,829 | Thanked: 1,459 times | Joined on Dec 2009 @ Finland
#34
Originally Posted by StocChr View Post
thanks slender,
but i had these problems long before overclocking, cssu and other things...
if we find the reason, we will post it here, thanks
Sorry but it doesnīt matter what you assume to know. For "standardized" testing and easiness for comparing to other people N900s it is always good to start from "zero"-level. It might be troublesome at first but if we repeat it* n-times it tends to be best way to debug stuff. Just start from "blank".


*Meaning different kind of people with different kind of cases where they have problem with cpu not entering to sleep state or unusual high battery consumption.
__________________
TMO links: [iSpy] - [Power search] - [Most thanked] - [Cordia - Maemo5 UI on top MeeGo Core] - [CommunitySSU]
 
Posts: 1,680 | Thanked: 3,685 times | Joined on Jan 2011
#35
Originally Posted by vi_ View Post
-_-' dude, run as root.

1. Set phone to offline mode
2. Shutdown phone
3. Take out battery for 10s
4. Take out memory card (if you have one)
5. Reboot phone
6. tell phone to stay offline.
7. wait ~4 minutes (while phone settles (should still be offline mode))
8. open xterm
9. become root
10. start powertop, shut keyboard and press lock slide.
11. leave phone for 50s
12. paste output

From the little you pasted above somthing is keeping your CPU awake, just gotta see what it is.
............................................______ __
....................................,.-'"...................``~.,
.............................,.-"..................................."-.,
.........................,/...............................................":,
.....................,?........................... ...........................\,
.................../.................................................. .........,}
................./.................................................. ....,:`^`..}
.............../.................................................. .,:"........./
..............?.....__............................ .............:`.........../
............./__.(....."~-,_..............................,:`........../
.........../(_...."~,_........"~,_....................,:`..... ..._/
..........{.._$;_......"=,_......."-,_.......,.-~-,},.~";/....}
...........((.....*~_......."=-._......";,,./`..../"............../
...,,,___.\`~,......"~.,....................`..... }............../
............(....`=-,,.......`........................(......;_,,-"
............/.`~,......`-...............................\....../\
.............\`~.*-,.....................................|,./.....\,__
,,_..........}.>-._\...................................|........... ...`=~-,
.....`=~-,_\_......`\,.................................\
...................`=~-,,.\,...............................\
................................`:,,.............. .............`\..............__
.....................................`=-,...................,%`>--==``
........................................_\........ ..._,-%.......`\
...................................,<`.._|_,-&``................`


Disable all widgets and try again.

Make sure you wait 4 minutes this time.
__________________
N900: One of God's own prototypes. A high-powered mutant of some kind never even considered for mass production. Too weird to live, and too rare to die.
 
Posts: 306 | Thanked: 38 times | Joined on Dec 2009 @ Austria
#36
i did this without widgets no C2-C4 cpu usage
but i will post it in 5 minutes again...
 

The Following User Says Thank You to StocChr For This Useful Post:
Posts: 1,341 | Thanked: 708 times | Joined on Feb 2010
#37
Use -t option with powertop to get longer sample ;-)

I'd try to kill fmtxd from running processes. It seem to wake up CPU once every second.
 
Posts: 306 | Thanked: 38 times | Joined on Dec 2009 @ Austria
#38
once again, with clear desktops...

Code:
Nokia-N900:~# powertop
Powertop 1.13.3
status: Unknown job: pmtrackerdaemon
Sleeping for 6 seconds before sampling
Collecting data for 30 seconds
Sample interval was 00m 30s 13367us

C#      | Ratio  | Avg/dura | Frequency | Ratio
--------+--------+----------+-----------+--------+
     C0 |   8.8% |          |  1150 MHz |   nan% |
     C1 |  91.2% |  113.0ms | 
     C2 |   0.0% |          | 
     C3 |   0.0% |          | 
     C4 |   0.0% |          | 

IRQ#    | Activity   | Type           | Name
--------+------------+----------------+---------------------------
     37 |        467 |           INTC | gp
     56 |        322 |           INTC | i2c_omap
     86 |        130 |           INTC | mmc1
     12 |         89 |           INTC | DMA
     57 |         52 |           INTC | i2c_omap
     21 |         10 |           INTC | SGX
     61 |         10 |           INTC | i2c_omap
    225 |          4 |           GPIO | omap2-onenand
    341 |          1 |           GPIO | lis302dl

PID#    | Activity   | Name           | Function Entry (Expire)
--------+------------+----------------+---------------------------
      0 |        391 |  <kernel core> | tick_nohz_restart_sched_tick (tick_sched_timer)
    676 |         47D|  kernel-config | cpufreq_governor_dbs (delayed_work_timer_fn)
   3384 |         19 |     osso-xterm | schedule_hrtimeout_range (hrtimer_wakeup)
    651 |         17 |           dsme | __enqueue_rt_entity (sched_rt_period_timer)
      0 |         16 |  <kernel core> | hrtimer_start (tick_sched_timer)
    693 |         11 |      bme_RX-51 | sys_timer_settime (posix_timer_fn)
    568 |          9 |          mmcqd | queue_delayed_work (delayed_work_timer_fn)
    568 |          7 |          mmcqd | schedule_timeout (process_timeout)
    693 |          4 |      bme_RX-51 | schedule_timeout (process_timeout)
    999 |          3 |           Xorg | queue_delayed_work (delayed_work_timer_fn)
    568 |          3 |          mmcqd | cfq_completed_request (cfq_idle_slice_timer)
    999 |          2 |           Xorg | hrtimer_start (it_real_fn)
    970 |          2 |       sgx_perf | queue_delayed_work (delayed_work_timer_fn)
    925 |          2 |             sh | setup_wb_timer (wb_timer_fn)
    678 |          2D|<kernel module> | queue_delayed_work (delayed_work_timer_fn)
    651 |          2 |           dsme | do_nanosleep (hrtimer_wakeup)
      1 |          2D|  <kernel core> | queue_delayed_work (delayed_work_timer_fn)
    693 |          2 |      bme_RX-51 | do_nanosleep (hrtimer_wakeup)
   1274 |          2 |        signond | futex_wait (hrtimer_wakeup)
    942 |          1 |            mce | queue_delayed_work (delayed_work_timer_fn)
   3503 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   3517 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
    715 |          1 |    dbus-daemon | queue_delayed_work (delayed_work_timer_fn)
   3531 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   3545 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   3559 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   3573 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   3587 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   3601 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   1274 |          1 |        signond | schedule_hrtimeout_range (hrtimer_wakeup)
   3615 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   3629 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   3643 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   3657 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   1389 |          1 |  BatteryGraphd | schedule_hrtimeout_range (hrtimer_wakeup)
   3671 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   1389 |          1 |  BatteryGraphd | journal_get_write_access (commit_timeout)
   3685 |          1 |          sleep | do_nanosleep (hrtimer_wakeup)
   3459 |          1 |       powertop | do_nanosleep (hrtimer_wakeup)

Power domain activity breakdown
Domain  | % of time spent in states
--------+---------+---------+---------+---------+----------
usbhost |OFF:   0%|RET: 100%|INA:   0%| ON:   0%| now:(RET)
    sgx |OFF:  97%|RET:   0%|INA:   0%| ON:   2%| now:(OFF)
    per |OFF:   0%|RET:   0%|INA:   0%| ON: 100%| now:(ON)
    dss |OFF:   0%|RET: 100%|INA:   0%| ON:   0%| now:(RET)
    cam |OFF:   0%|RET: 100%|INA:   0%| ON:   0%| now:(RET)
   core |OFF:   0%|RET:   0%|INA:   0%| ON: 100%| now:(ON)
   neon |OFF:   0%|RET:   0%|INA:   0%| ON: 100%| now:(ON)
    mpu |OFF:   0%|RET:   0%|INA:   0%| ON: 100%| now:(ON)
   iva2 |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)

Clock activity breakdown at end of period
Domain  | Active clocks
--------+---------------+---------------+------------------
   core |          SDRC | HSOTGUSB_IDLE |      OMAPCTRL 
        |     MAILBOXES |
   wkup |          GPT1 |       32KSYNC |         GPIO1 
        |
  ckgen |          CORE |          PERI |           96M 
        |           48M |           12M |           54M 
        |      EMU_CORE |
    per |         GPIO2 |         GPIO3 |         GPIO4 
        |         GPIO5 |         GPIO6 |

Total wakeups  1648,  54.9/s | IRQ 1085,  36.2/s | Timers  563,  18.8/s
HW wakeups        0,   0.0/s |     Real gp_timers expired  211,   7.0/s
Attached Files
File Type: txt top_out.txt (12.3 KB, 101 views)
 
Posts: 1,680 | Thanked: 3,685 times | Joined on Jan 2011
#39
hmm, all those sleep | do_nanosleep (hrtimer_wakeup) are freaking me out.

ok try this, uninstall kernel-power-settings (dont worry you will still keep the power kernel).

Code:
apt-get remove kernel-power-settings
Then do the 'procedure again'. However this time don't worry about posting your output from top.

ok?
__________________
N900: One of God's own prototypes. A high-powered mutant of some kind never even considered for mass production. Too weird to live, and too rare to die.
 

The Following User Says Thank You to vi_ For This Useful Post:
Posts: 2,829 | Thanked: 1,459 times | Joined on Dec 2009 @ Finland
#40
 
Reply


 
Forum Jump


All times are GMT. The time now is 04:02.