Active Topics
-
[ANNOUNCE] CSSU-thumb thread - stable Thumb2 on N900 (2,266)
to Maemo 5 / Fremantle by pali - 3 mins ago -
Installing CSSU Stable in year 2024 (0)
to Maemo 5 / Fremantle by teroyk - 2 hrs, 51 mins ago -
Firefox with Leste (7)
to Maemo 7 / Leste by teroyk - 4 hrs, 7 mins ago -
Porting apps to Leste (32)
to Maemo 7 / Leste by Arno_11 - 6 days, 7 hrs ago - more...
 |
|
2013-08-24
, 16:20
|
|
Posts: 5,028 |
Thanked: 8,613 times |
Joined on Mar 2011
|
#1492
|
Originally Posted by Android_808
Confirming, and it happens not only for browser. Could be related to hildon desktop?
another potential issue I'm experiencing another random, hard to reproduce issue ffor last week or so.
Whilst browsing web with microb, device loses track of focus. You can happily be browsing away and then it loads next page and nothing works. clicking/scrolling page results in haptic feedback but nothing else. text box is unresponsive. keyboard has no effect.
my initial reaction was unresponsive due to load but that usually slows haptic response as well. I then found that the task switcher button still works. you can get out to desktop but widgets don't work. you can get to menu and launch apps but they fail to respond to input. power button launched its menu but desaturated and unresponsive....
then i turned screen off with slider on the side, turn it on again using the slider and everything starts working again.
BTW, as a workaround, I found that locking device and bringing lockscreen by power-menu, then unlocking, fix the problem. Irritating, nevertheless.
No lockscreen replacements or other relevant things updated, so CSSU could be to be blamed.
/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!
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!
| The Following 4 Users Say Thank You to Estel For This Useful Post: | ||
|
|
2013-08-24
, 16:54
|
|
Posts: 1,203 |
Thanked: 3,027 times |
Joined on Dec 2010
|
#1493
|
I've only noticed it this last week or so, but I've been using device more for browsing than I have been lately.
No lockscreen replacements either. I am running thumb though but I wouldn't of thought it was connected to that.
No lockscreen replacements either. I am running thumb though but I wouldn't of thought it was connected to that.
| The Following User Says Thank You to Android_808 For This Useful Post: | ||
|
|
2013-08-24
, 16:59
|
|
Posts: 1,378 |
Thanked: 1,604 times |
Joined on Jun 2010
@ Göteborg, Sweden
|
#1494
|
tanago's device is peculiar in that all the tests, even where he reports low mA usage, the cpu spends much time in state C2. Further, the first powertop with stock kernel and low idle mA had those strange IRQs that fmg spotted.
My powertop with CSSU-T and thumb pretty well matches that of fmg re IRQ and C states.
Last edited by handaxe; 2013-08-24 at 17:02.
My powertop with CSSU-T and thumb pretty well matches that of fmg re IRQ and C states.
Last edited by handaxe; 2013-08-24 at 17:02.
| The Following 5 Users Say Thank You to handaxe For This Useful Post: | ||
 |
|
2013-08-24
, 17:23
|
|
Posts: 215 |
Thanked: 448 times |
Joined on Aug 2012
@ Burgas, Bulgaria
|
#1495
|
Originally Posted by handaxe
Yes...I'm going to run through further investigations - reflashing and installing everything one by one and running powertop after each installation
tanago's device is peculiar in that all the tests, even where he reports low mA usage, the cpu spends much time in state C2. Further, the first powertop with stock kernel and low idle mA had those strange IRQs that fmg spotted.
My powertop with CSSU-T and thumb pretty well matches that of fmg re IRQ and C states.
| The Following 2 Users Say Thank You to tanago For This Useful Post: | ||
|
|
2013-08-24
, 17:25
|
|
Posts: 3,074 |
Thanked: 12,960 times |
Joined on Mar 2010
@ Sofia,Bulgaria
|
#1496
|
Originally Posted by Estel
could be related to systemui-tklock replacement. can you try to find a reliable way to reproduce it? also, can you reproduce it with systemui-tklock that is in cssu-devel?
Confirming, and it happens not only for browser. Could be related to hildon desktop?
BTW, as a workaround, I found that locking device and bringing lockscreen by power-menu, then unlocking, fix the problem. Irritating, nevertheless.
No lockscreen replacements or other relevant things updated, so CSSU could be to be blamed.
/Estel
__________________
Never fear. I is here.
720p video support on N900,SmartReflex on N900,Keyboard and mouse support on N900
Nothing is impossible - Stable thumb2 on n900
Community SSU developer
kernel-power developer and maintainer
Never fear. I is here.
720p video support on N900,SmartReflex on N900,Keyboard and mouse support on N900
Nothing is impossible - Stable thumb2 on n900
Community SSU developer
kernel-power developer and maintainer
| The Following 4 Users Say Thank You to freemangordon For This Useful Post: | ||
|
|
2013-08-24
, 18:38
|
|
Posts: 5,028 |
Thanked: 8,613 times |
Joined on Mar 2011
|
#1497
|
Originally Posted by freemangordon
Will try the latter (Android_808, could you do the same, for consistent results? Also I may be of non-consistent and non-real time availability for some time), as for the former, sadly, it happens to be completely random.
can you try to find a reliable way to reproduce it? also, can you reproduce it with systemui-tklock that is in cssu-devel?
Well, maybe not *completely* - for me, it seems to appear most of the time, when I'm doing things that require "Scroll by drag" (like browser). But nothing more specific than that. It appeared on wide range of programs, from browser, to lonewolf4maemo, so I have no idea how to narrow it.
/Estel
// Edit
another thing that comes to my mind - bug is always related to screen becoming blurry for a second, like the topmost window become considered a background one. then, it returns to normal, but no input is detected, unless I do mentioned slide-lock & power-key-to-lockscreen & unlock combo.
__________________
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!
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!
Last edited by Estel; 2013-08-24 at 18:40.
| The Following 5 Users Say Thank You to Estel For This Useful Post: | ||
|
|
2013-08-24
, 19:04
|
|
Posts: 3,074 |
Thanked: 12,960 times |
Joined on Mar 2010
@ Sofia,Bulgaria
|
#1498
|
@Estel, Android_808 - another approach - could you install the stock systemui-tklock, just to verify if this is the package to blame. Though the more I am thinking the more it seems it is not, as I was using the replacement since January, had the lock you described only after the latest CSSU update.
@Android_808 - are you on CSSU-testing or on CSSU-thumb?
@Android_808 - are you on CSSU-testing or on CSSU-thumb?
__________________
Never fear. I is here.
720p video support on N900,SmartReflex on N900,Keyboard and mouse support on N900
Nothing is impossible - Stable thumb2 on n900
Community SSU developer
kernel-power developer and maintainer
Never fear. I is here.
720p video support on N900,SmartReflex on N900,Keyboard and mouse support on N900
Nothing is impossible - Stable thumb2 on n900
Community SSU developer
kernel-power developer and maintainer
| The Following 5 Users Say Thank You to freemangordon For This Useful Post: | ||
|
|
2013-08-24
, 23:36
|
|
Posts: 804 |
Thanked: 1,598 times |
Joined on Feb 2010
@ Gdynia, Poland
|
#1499
|
Originally Posted by tanago
Very wild guess from here, but (for completness) could you write which hw revision do you have? And if you're running stock kernel or power kernel (and if u-boot is installed)?
Yes...I'm going to run through further investigations - reflashing and installing everything one by one and running powertop after each installation
| The Following User Says Thank You to misiak For This Useful Post: | ||
|
|
2013-08-25
, 02:57
|
|
Posts: 5,028 |
Thanked: 8,613 times |
Joined on Mar 2011
|
#1500
|
Originally Posted by freemangordon
@Android_808 - are you on CSSU-testing or on CSSU-thumb?
Originally Posted by Android_808
I'm also running on thumb (All the time - I don't update to new CSSU-T, until new version of thumb appears. Not that it should matter for this particular case, but telling full picture, anyway).
I am running thumb though but I wouldn't of thought it was connected to that.
As for testing systemlock package, doing it now, one device with -devel, and will install stock on another one, tommorow. By stock, you mean full CSSU thumb, with just that package downgraded to stock, yep?
Anyway, I'm also inclined it is not related to tklock - all locking things seems to be "solution", instead of problem cause. + the fact that no one reported it before latest CSSU update (although, -devel isn't widely used on everyday devices, obviously, so random nature of this bug could prevent it from being detected).
Either way, due to - again - random nature of this bug, testing will take some time, and only answer I'll be able to give, is a) It happened (for sure), or b) I *think* it doesn't happen anymore (or I just was lucky to not run into it).
/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!
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!
| The Following 4 Users Say Thank You to Estel For This Useful Post: | ||
Powertop 1.13.3 Sleeping for 11 seconds before sampling Collecting data for 60 seconds Sample interval was 01m 00s 16876us C# | Ratio | Avg/dura | Frequency | Ratio --------+--------+----------+-----------+--------+ C0 | 0.4% | | 1150 MHz | nan% | C1 | 0.0% | 0.3ms | C2 | 25.8% | 201.1ms | C3 | 6.6% | 657.1ms | C4 | 67.2% | 2240.3ms | IRQ# | Activity | Type | Name --------+------------+----------------+--------------------------- 56 | 432 | INTC | i2c_omap 37 | 74 | INTC | gp 57 | 68 | INTC | i2c_omap 11 | 60 | INTC | prcm 86 | 9 | INTC | mmc1 12 | 1 | INTC | DMA PID# | Activity | Name | Function Entry (Expire) --------+------------+----------------+--------------------------- 0 | 41 | <kernel core> | tick_nohz_restart_sched_tick (tick_sched_timer) 38 | 15D| awk | cpufreq_governor_dbs (delayed_work_timer_fn) 714 | 13 | bme_RX-51 | sys_timer_settime (posix_timer_fn) 684 | 6 | dsme | __enqueue_rt_entity (sched_rt_period_timer) 684 | 5 | dsme | do_nanosleep (hrtimer_wakeup) 1 | 5D| <kernel core> | queue_delayed_work (delayed_work_timer_fn) 0 | 4 | <kernel core> | hrtimer_start (tick_sched_timer) 714 | 4 | bme_RX-51 | sys_timer_settime (posix_timer_fn) 714 | 4 | bme_RX-51 | do_nanosleep (hrtimer_wakeup) 714 | 4 | bme_RX-51 | schedule_timeout (process_timeout) 602 | 3 | mmcqd | queue_delayed_work (delayed_work_timer_fn) 757 | 2 | hald | schedule_hrtimeout_range (hrtimer_wakeup) 1205 | 2 | signond | futex_wait (hrtimer_wakeup) 30 | 1 | mount | setup_wb_timer (wb_timer_fn) 13 | 1 | pdflush | blk_plug_device (blk_unplug_timeout) 780 | 1 | mce | schedule_hrtimeout_range (hrtimer_wakeup) 1559 | 1 | browserd | futex_wait (hrtimer_wakeup) 1205 | 1 | signond | schedule_hrtimeout_range (hrtimer_wakeup) 901 | 1 | hald-addon-bme | schedule_hrtimeout_range (hrtimer_wakeup) 1055 | 1 | iphbd | schedule_hrtimeout_range (hrtimer_wakeup) 1630 | 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: 100%|RET: 0%|INA: 0%| ON: 0%| now:(OFF) per |OFF: 73%|RET: 25%|INA: 0%| ON: 0%| now:(ON) dss |OFF: 0%|RET: 0%|INA: 0%| ON: 100%| now:(ON) cam |OFF: 100%|RET: 0%|INA: 0%| ON: 0%| now:(OFF) core |OFF: 0%|RET: 0%|INA: 0%| ON: 100%| now:(ON) neon |OFF: 67%|RET: 6%|INA: 25%| ON: 0%| now:(ON) mpu |OFF: 67%|RET: 6%|INA: 25%| ON: 0%| 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 | dss | DSS_STBY | DSS_IDLE per | GPIO2 | GPIO3 | GPIO4 | GPIO5 | GPIO6 | Total wakeups 760, 12.7/s | IRQ 644, 10.7/s | Timers 116, 1.9/s HW wakeups 60, 1.0/s | Real gp_timers expired 74, 1.2/s