First of all, this has been "feature" long time ago, but i noticed, that cpu usage changes dramatically after 3-5 days of continuous uptime. It seems to have EKG/Pulse (see picture)

What is causing that issue? Any brainiac/mythbuster want to help to solve this? I havent rebooted my phone after taking that screenshot.

I also can make this thing happen every time i let phone be on 3-5 days, so its not just "once in lifetime" thing.

You can also see Load been more than normal idle and CPU takes odd cycles.

Battery goes empty 2-3 times faster than normally.

I would advise to run powertop as root when this is happening to collect some data.

And when running powertop, leave it background and take screen off with key lock switch to minimize all the other processes.

I would advise to run powertop as root when this is happening to collect some data.

And when running powertop, leave it background and take screen off with key lock switch to minimize all the other processes.

Okay, i did that and here's the result:

/home/user # powertop
Powertop 1.13.3
status: Unknown job: pmtrackerdaemon
Sleeping for 11 seconds before sampling
Collecting data for 30 seconds
Sample interval was 00m 30s 19989us

C# | Ratio | Avg/dura | Frequency | Ratio
--------+--------+----------+-----------+--------+
C0 | 5.0% | | 600 MHz | 3.6% |
C1 | 0.6% | 2.4ms | 550 MHz | 0.0% |
C2 | 7.2% | 11.9ms | 500 MHz | 5.5% |
C3 | 33.9% | 231.6ms | 250 MHz | 90.9% |
C4 | 53.4% | 593.1ms |

IRQ# | Activity | Type | Name
--------+------------+----------------+---------------------------
12 | 908 | INTC | DMA
56 | 399 | INTC | i2c_omap
37 | 323 | INTC | gp
86 | 192 | INTC | mmc1
11 | 171 | INTC | prcm
57 | 121 | INTC | i2c_omap
21 | 82 | INTC | SGX
202 | 18 | GPIO | wl1251
67 | 6 | INTC | ssi_p1_mpu_irq0
376 | 4 | twl4030 | PwrButton
71 | 2 | INTC | ssi_gdd
311 | 2 | GPIO | ssi_p1_cawake_gpio
25 | 1 | INTC | OMAP

PID# | Activity | Name | Function Entry (Expire)
--------+------------+----------------+---------------------------
0 | 242 | <kernel core> | tick_nohz_restart_sched_tick (tick_sched_timer)
37 | 34D| awk | cpufreq_governor_dbs (delayed_work_timer_fn)
16666 | 28 | fmtxd | schedule_hrtimeout_range (hrtimer_wakeup)
0 | 23 | <kernel core> | hrtimer_start (tick_sched_timer)
1172 | 23 | wlancond | schedule_timeout (process_timeout)
536 | 10 | mmcqd | queue_delayed_work (delayed_work_timer_fn)
1113 | 7 |hildon-status-m | schedule_hrtimeout_range (hrtimer_wakeup)
536 | 7 | mmcqd | schedule_timeout (process_timeout)
717 | 7 | dbus-daemon | queue_delayed_work (delayed_work_timer_fn)
701 | 6 | bme_RX-51 | sys_timer_settime (posix_timer_fn)
691 | 6 | dsme | __enqueue_rt_entity (sched_rt_period_timer)
536 | 6 | mmcqd | cfq_completed_request (cfq_idle_slice_timer)
752 | 5 | mce | schedule_hrtimeout_range (hrtimer_wakeup)
802 | 5 | Xorg | OSEnableTimer (OSTimerCallbackWrapper)
1274 | 4 | icd2 | schedule_hrtimeout_range (hrtimer_wakeup)
780 | 4 | sgx_perf | queue_delayed_work (delayed_work_timer_fn)
16670 | 3 | osso-xterm | schedule_hrtimeout_range (hrtimer_wakeup)
1 | 3 | <kernel core> | queue_delayed_work (delayed_work_timer_fn)
691 | 3 | dsme | do_nanosleep (hrtimer_wakeup)
1172 | 3 | wlancond | schedule_timeout (process_timeout)
29 | 3 | mount | setup_wb_timer (wb_timer_fn)
701 | 2 | bme_RX-51 | sys_timer_settime (posix_timer_fn)
701 | 2 | bme_RX-51 | do_nanosleep (hrtimer_wakeup)
701 | 2 | bme_RX-51 | schedule_timeout (process_timeout)
752 | 2 | mce | schedule_timeout (process_timeout)
465 | 2 | wl12xx | schedule_timeout (process_timeout)
1620 | 2 | pdflush | blk_plug_device (blk_unplug_timeout)
752 | 1 | mce | tsc2005_start_scan (tsc2005_esd_timer_handler)
536 | 1 | mmcqd | schedule_timeout (process_timeout)
802 | 1 | Xorg | hrtimer_start (it_real_fn)
992 | 1 | systemui | schedule_hrtimeout_range (hrtimer_wakeup)
752 | 1 | mce | queue_delayed_work (delayed_work_timer_fn)
752 | 1 | mce | queue_delayed_work (delayed_work_timer_fn)
16676 | 1 | powertop | do_nanosleep (hrtimer_wakeup)

Power domain activity breakdown
Domain | % of time spent in states
--------+---------+---------+---------+---------+----------
usbhost |OFF: 100%|RET: 0%|INA: 0%| ON: 0%| nowOFF)
sgx |OFF: 96%|RET: 0%|INA: 0%| ON: 3%| nowOFF)
per |OFF: 85%|RET: 8%|INA: 0%| ON: 5%| nowON)
dss |OFF: 74%|RET: 0%|INA: 0%| ON: 25%| nowOFF)
cam |OFF: 100%|RET: 0%|INA: 0%| ON: 0%| nowOFF)
core |OFF: 48%|RET: 36%|INA: 3%| ON: 10%| nowON)
neon |OFF: 53%|RET: 33%|INA: 7%| ON: 5%| nowON)
mpu |OFF: 53%|RET: 33%|INA: 7%| ON: 5%| nowON)
iva2 |OFF: 100%|RET: 0%|INA: 0%| ON: 0%| nowOFF)

Clock activity breakdown at end of period
Domain | Active clocks
--------+---------------+---------------+------------------
core | SDRC | HSOTGUSB_IDLE | OMAPCTRL
| MAILBOXES |
wkup | GPT1 | 32KSYNC | GPIO1
| WDT1 |
ckgen | CORE | PERI | 96M
| 48M | 12M | 54M
| EMU_CORE |
per | GPIO2 | GPIO3 | GPIO4
| GPIO5 | GPIO6 |

Total wakeups 2680, 89.3/s | IRQ 2229, 74.3/s | Timers 451, 15.0/s
HW wakeups 171, 5.7/s | Real gp_timers expired 67, 2.2/s

I think time between those peaks in my screenshot is longer than this 30 second test.

Hi. I have been experiencing high cpu activity and system lags, mainly due to the "hildon-status-m" and "Hildon-home" processes, but also others.

This usually happens after 2-4 days of uptime.

I tend to use these lags as a cue to do a re-boot, which sort all that stuff out.

Jamil,
http://talk.maemo.org/showthread.php?t=42500

hmmm the power domain section seems to have a lot of unhappy smiley's perhaps this is the reason for the battery drain as it's clear the battery just is not happy... perhaps we need to replace with or something?

Also sorry, it's Monday and Im all wrong inside