[N900] issues with wlan (now after reboot device warm/poor battery) - Page 2 - maemo.org

Active Topics

Nokia Booklet 3G (RX-75) Drivers (60)
to Alternatives by Maemish - 3 days, 23 hrs ago
Is there a section to talk about Java ME? Apps, etc. (7)
to General by teroyk - 6 days, 19 hrs ago
more...

Page 2 of 5

< Prev

Next >

Last »

Thread Tools

	peterleinchen	2012-07-29 , 12:51
	Posts: 4,118 \| Thanked: 8,901 times \| Joined on Aug 2010 @ Ruhrgebiet, Germany	#11

Do you still have the bleeding edge drivers loaded by default?
Tr with:
~ $ lsmod | grep wl1251
If there is no output, forget my comment

If there is any, then this behaviour is well known.

Just a guess in the dark ...

Quote & Reply |

The Following 2 Users Say Thank You to peterleinchen For This Useful Post:
Estel, sixwheeledbeast

sixwheeledbeast	2012-07-29 , 12:59
Posts: 2,290 \| Thanked: 4,133 times \| Joined on Apr 2010 @ UK	#12

Originally Posted by peterleinchen

Do you still have the bleeding edge drivers loaded by default?
Try with:
~ $ lsmod | grep wl1251

No output.

Just had the battery/sim/sd removed for 10 minutes. Problem still there.

__________________

Wiki Admin
sixwheeledbeast's wiki
Testing Squad Subscriber
- mcallerx - tenminutecore - FlopSwap - Qnotted - zzztop - Bander - Fight2048 -

Before posting or starting a thread please try this.

Quote & Reply |

sixwheeledbeast	2012-07-29 , 17:21
Posts: 2,290 \| Thanked: 4,133 times \| Joined on Apr 2010 @ UK	#13

Just to follow up.
Uninstalled Kernel Power using the Nokia Kernel .desktop icon.
3 Hours later fairly normal service resumes (normally average 400-500 wakes) , device is cool again.
How odd.

Code:

$ sudo 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 91311us

C#      | Ratio  | Avg/dura | Frequency | Ratio
--------+--------+----------+-----------+--------+
     C0 |   1.1% |          |   600 MHz |   1.0% |
     C1 |   0.0% |    0.2ms |   550 MHz |   0.0% |
     C2 |   0.5% |    5.3ms |   500 MHz |   0.0% |
     C3 |   8.9% |  116.7ms |   250 MHz |  99.0% |
     C4 |  89.5% | 1495.9ms | 

IRQ#    | Activity   | Type           | Name
--------+------------+----------------+---------------------------
     37 |         93 |           INTC | gp
     56 |         86 |           INTC | i2c_omap
     12 |         74 |           INTC | DMA
     11 |         61 |           INTC | prcm
     57 |         22 |           INTC | i2c_omap
     21 |         10 |           INTC | SGX
     67 |         10 |           INTC | ssi_p1_mpu_irq0
    202 |          9 |           GPIO | wl1251
     71 |          8 |           INTC | ssi_gdd
    311 |          4 |           GPIO | ssi_p1_cawake_gpio

PID#    | Activity   | Name           | Function Entry (Expire)
--------+------------+----------------+---------------------------
      0 |         52 |  &lt;kernel core&gt; | tick_nohz_restart_sched_tick (tick_sched_timer)
     37 |         22D|            awk | cpufreq_governor_dbs (delayed_work_timer_fn)
   1850 |         18 |     osso-xterm | schedule_hrtimeout_range (hrtimer_wakeup)
    544 |          9 |         wl12xx | queue_delayed_work (delayed_work_timer_fn)
      0 |          7 |  &lt;kernel core&gt; | hrtimer_start (tick_sched_timer)
    881 |          4 |           Xorg | hrtimer_start (it_real_fn)
    728 |          4 |      bme_RX-51 | sys_timer_settime (posix_timer_fn)
      1 |          3 |  &lt;kernel core&gt; | queue_delayed_work (delayed_work_timer_fn)
     27 |          2 |    kondemand/0 | queue_delayed_work (delayed_work_timer_fn)
    831 |          2 |       sgx_perf | queue_delayed_work (delayed_work_timer_fn)
    728 |          2 |      bme_RX-51 | sys_timer_settime (posix_timer_fn)
    728 |          2 |      bme_RX-51 | do_nanosleep (hrtimer_wakeup)
    728 |          2 |      bme_RX-51 | schedule_timeout (process_timeout)
    689 |          2 |           dsme | do_nanosleep (hrtimer_wakeup)
    689 |          2 |           dsme | __enqueue_rt_entity (sched_rt_period_timer)
    817 |          1 |            mce | schedule_hrtimeout_range (hrtimer_wakeup)
   1213 |          1 |       wlancond | schedule_timeout (process_timeout)
   1213 |          1 |       wlancond | queue_delayed_work (delayed_work_timer_fn)
   1213 |          1 |       wlancond | ieee80211_ioctl_siwpower (ieee80211_dynamic_ps_timer)
   1652 |          1 |       browserd | futex_wait (hrtimer_wakeup)
      0 |          1 |  &lt;kernel core&gt; | queue_delayed_work (delayed_work_timer_fn)
   1855 |          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%| now:(OFF)
    sgx |OFF:  97%|RET:   0%|INA:   0%| ON:   2%| now:(OFF)
    per |OFF:  98%|RET:   0%|INA:   0%| ON:   1%| now:(ON)
    dss |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)
    cam |OFF: 100%|RET:   0%|INA:   0%| ON:   0%| now:(OFF)
   core |OFF:  89%|RET:   7%|INA:   0%| ON:   2%| now:(ON)
   neon |OFF:  89%|RET:   8%|INA:   0%| ON:   1%| now:(ON)
    mpu |OFF:  89%|RET:   8%|INA:   0%| ON:   1%| 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 
        |          WDT1 |
  ckgen |          CORE |          PERI |           96M 
        |           48M |           12M |           54M 
        |      EMU_CORE |
    per |         GPIO2 |         GPIO3 |         GPIO4 
        |         GPIO5 |         GPIO6 |

Total wakeups   517,  17.2/s | IRQ  377,  12.6/s | Timers  140,   4.7/s
HW wakeups       61,   2.0/s |     Real gp_timers expired   93,   3.1/s
~ $

Think I am going to flash rootfs just to be sure.

Quote & Reply |

The Following User Says Thank You to sixwheeledbeast For This Useful Post:
Estel

	Estel	2012-07-29 , 18:50
	Posts: 5,028 \| Thanked: 8,613 times \| Joined on Mar 2011	#14

wow, you're having some serious problems there. could you try with installing kp51r1 (or, in other words, any latest from repositories) on freshly flashed device and see, if problem ever happen again?

BTW, I would recommend using Advanced Interface Switcher in place of Wifi Switcher? the latter is very outdated, and I remember it had some part of code inferior (can't recall what it was).

sorry for not being more of a help, I've never saw something like that. I hope it isn't hardware problem with WiFi adaptor, that result in it being reset and firmware re-loaded.

/Estel

__________________
N900's aluminum backcover / body replacement
-
N900's HDMI-Out
-
Camera cover MOD
-
Measure battery's real capacity on-device
-
TrueCrypt 7.1 | ereswap | bnf
-
Hardware's mods research is costly. To support my work, please consider donating. Thank You!

Quote & Reply |

The Following User Says Thank You to Estel For This Useful Post:
sixwheeledbeast

Ungoliant	2012-07-29 , 19:00
Posts: 10 \| Thanked: 13 times \| Joined on Feb 2008	#15

I would not discard a malfunction due to the 1150 MHz OC ...

Quote & Reply |

sixwheeledbeast	2012-07-29 , 20:06
Posts: 2,290 \| Thanked: 4,133 times \| Joined on Apr 2010 @ UK	#16

Originally Posted by Ungoliant

I would not discard a malfunction due to the 1150 MHz OC ...

Sorry...BUT THIS DEVICE HAS NEVER BEEN OVERCLOCKED.
IIRC this is a Kernel issue that makes powertop show 1150 no matter what the clock freq is when using Kernel-Power.

[slightyofftopicrant]
I have KP and KPS but use KPS to underclock to 125-600.
I do this as I have tested 250-600 and 125-600 and find with my usage (music playing >6 hours a day) allowing 125 gives me better battery life.
This is why I also pasted the kernel-config show output!
[/slightyofftopicrant]

I don't believe in overclocking the N900 I wish to look after as much as possible. Even to the point of putting all personal files and swap on uSD to reduce eMMC wear. This is a concorde device that sadly may never happen again.

@Estel
I to hope it's not a hardware fault.
Any more info on this issue with wifi switcher would be useful.
Thanks, for the tip on AIS will try that out.
Will do some power searching...

I have reinstalled KP50 and KPS and will monitor for now.
I could really do without a reflash if I can help it, it will take me a while to rebuild everything.

All seems normal again ... for now.

Last edited by sixwheeledbeast; 2012-07-29 at 20:11. Reason: typos

Quote & Reply |

The Following User Says Thank You to sixwheeledbeast For This Useful Post:
reinob

	Estel	2012-07-29 , 23:51
	Posts: 5,028 \| Thanked: 8,613 times \| Joined on Mar 2011	#17

As for wifi switcher, it was something with not restarting wlancord, or restarting it too often... I really can't remember nor find it now.

As for KP, why kp50 not kp51r1? The latter is rox stable, while kp50 have some bugs left.

/Estel

Quote & Reply |

sixwheeledbeast	2012-07-30 , 14:55
Posts: 2,290 \| Thanked: 4,133 times \| Joined on Apr 2010 @ UK	#18

This is my "stable" use all the time device so only Extras is enabled.
I believe KP51 will be promoted soon anyway so I can wait.

I remember why I haven't tested AIS, it was due to the reports of poor battery life a while ago. Does it still poll the hardware all the time?

I can't find any reports of issues with wifi-switcher, if I knew the issue exactly I am hoping it maybe possible to modify wifi.sh. That's if AIS is a no go.

As for the original issue I have changed my wireless channel from 13 to 6, it maybe possible 11-13 are unstable as mentioned on your post in AIS thread.
I believe either 1/6/11 are the best channels to use as they don't overlap.
I used 13 originally due to no connections in my area use that channel (possibly due to others having AP's only capable of 1-11 global market)

I wonder if being connected with poor signal strength to channels between 11-13 causes the issue?
Or maybe an issue with my router? WAG200.
These theories seem to correlate with the timeline of events that happened to me.

Last edited by sixwheeledbeast; 2012-07-30 at 14:58. Reason: typos

Quote & Reply |

	Estel	2012-07-31 , 04:01
	Posts: 5,028 \| Thanked: 8,613 times \| Joined on Mar 2011	#19

AIS? I'm using it for ages, and it creates 0 battery impact (why should it?). Maybe You reffer to advanced-power-monitor, aka APM, from same author? BTW, the latter isn't fatal to battery, too (although, I don't use it since stopped using bq27x00_battery module, due to no additional benefits - it reads from BME, so regular battery applet does same thing, and for detailed info, I have bnf).

As for channel 13, I'm using it for my home network for ages, so it's not the issue. Where I've said about channel 13 instability? It must be misunderstanding, I remember only saying, that on some devices (depending on region used, in this case, EU and similar), without setting 'iw reg set JP', channel 13 works, but is crippled (no higher hop-range).

In any case, even crippled channel shouldn't result in things You're experiencing. BTW, Of course it's up to You, but I also have "main, everyday use" device of very high importance, yet, -devel is enabled 100% time - just be careful what you upgrade (no upgrade for things without changelog/TMO announcement, before checking who pushed higher version and why, as it may be not maintainer)

I know, such view on -devel isn't very popular, but our case is specific. We have plentora of packages with perfectly stable versions stuck in -devel, due to missing maintainers, alck of votes, or problems with repos, or... Summing it up, extras works a little like "legacy support", extras-testing is "old-stable", and -devel is... well, mix of things

/Estel

Quote & Reply |

The Following 2 Users Say Thank You to Estel For This Useful Post:
reinob, sixwheeledbeast

sixwheeledbeast	2012-07-31 , 16:45
Posts: 2,290 \| Thanked: 4,133 times \| Joined on Apr 2010 @ UK	#20

Originally Posted by Estel

AIS? I'm using it for ages, and it creates 0 battery impact (why should it?).

Advanced Interface Switcher was a pain to type out multiple times.
See below...

Originally Posted by vi_
unfortunatley I have to report AIS switcher blows. The idea is genius, I love it.

The problem is prior to installation I have around 450-600 hardware wakeups/second. After installation I see around 1400 hardware wakeups/second. I can consistantly do this.

This butt****s my batterylife.

Does this program poll the hardware or somthing?

Originally Posted by 412b
... every 5 seconds it makes d-bus call, when connected.

===

Originally Posted by Estel

As for channel 13, I'm using it for my home network for ages, so it's not the issue. Where I've said about channel 13 instability? It must be misunderstanding..

I mis-read your post. I will leave at "6" anyway.

Originally Posted by Estel

BTW, Of course it's up to You, but I also have "main, everyday use" device of very high importance, yet, -devel is enabled 100% time - just be careful ...

I do have devel and testing versions of packages on my stable device. This only happens strictly after checking changelogs (which I do with extras anyway), downloading the source to the desktop having a look through it (even tho I possibly don't understand most of it) and a full test for month or so on my testing device.

I have been bitten by the devil before and had to reflash at the most inconvenient time, so I try to look after my main devices filesystem as much as possible.

Originally Posted by Estel

I know, such view on -devel isn't very popular, but our case is specific. We have plentora of packages with perfectly stable versions stuck in -devel, due to missing maintainers, lack of votes, or problems with repos

Well, I am doing my best to help with that one.
Personally I'd like to see the return of QA and progressing through the repo's. It will reduce the number of newbies, enabling devel and coming here moaning when it's stopped working.