[DEV/DEBUG] CM11 debug thread for Falcon

Search This thread

matmutant

Senior Member
Mar 17, 2011
3,395
4,743
~/
andrux-and-me.blogspot.com
Moto G
Moto E
CM11 debug thread for Falcon

This thread is related to CM11 for Moto G (Falcon) maintained by Ethan Chen

for debugging purpose, i need this only thread to centralize the data
==> Report bugs (documented) i'll sort them by relevance (importance : low [L]; medium [M]; high [H]) and follow them to Ethan Chen.




Before Asking anything here, please refer to RC-FAQ & CM11 FAQ by @pinguijxy
refer to this post for general commands​

  • post 2 : known and solved
  • post 3 : pending
  • post 4: irrelevant

to improve CM11 for Falcon, you could :
  • post results after following protocols,
    [*]help me to set up test new protocols,
    [*]give advices about them
    [*]suggest things to investigate
    IF you can prove it worths to ;)

How To Report?
20140819-edit : Please use "Bug Report" button at the top of the thread instead of posting in thread.
give at least :

  • model : XT103* ?
  • rom build : cm-11-yyyymmdd-NIGHTLY-falcon ?
  • kernel build :
  • logcat?
  • way to reproduce ? (ok this is the best, but not always easy, or even possible to do)

XDA:DevDB Information
[DEV/DEBUG] Official CM11 debug and dev thread for Falcon, ROM for the Moto G

Contributors
matmutant, pinguijxy, CyanogenMod, Ethan Chen
ROM OS Version: 4.4.x KitKat
ROM Kernel: Linux 3.4.x
Based On: CyanogenMod

Version Information
Status: Testing

Created 2014-08-19
Last Updated 2015-01-10
 

Attachments

  • Screenshot_2014-08-19-21-07-47.png
    Screenshot_2014-08-19-21-07-47.png
    156.8 KB · Views: 1,781
Last edited:

matmutant

Senior Member
Mar 17, 2011
3,395
4,743
~/
andrux-and-me.blogspot.com
Moto G
Moto E
Known Bugs

Known and already reported:
  • OTG Mode does not charge [M]
  • LINE doesn't work in free call mode [M]
  • Hardware KB doesn't have layout settings in "language&input" [M]
  • Audio freeze and lags [H] [click for more]
  • Camera crashes after some time [H] [logcat - description in bugtracker - temporary workaround (XT1033): old libs]
  • [Wifi] Cannot connect to AP in some cases [H] (looking for more infos)
  • [Encryption] Encrypting 'bricks' device [H] (jira report)

Solved (kept for achive):
  • The flip case turns the screen on when opened, but not off when closed. (fixed since CM11-20140317-NIGHTLY)
  • Incall volume cannot be tuned [H] [click for more](fixed since cm-11-20140402-NIGHTLY)
  • Boot animation is lagging. [L] (fixed since cm11-20140611-NIGHTLY)
  • CPU minMAX stuck @1190MHz by default [H] (solved but frequency display is still somewhat broken)
  • In-Game freeze and Force-close [H] [click for more] (CM11-20140621~20140702 Nightlies)]
  • CM file manager crashes [M] [click for more]
  • GPS NO tracking at all issue [M] [click for more] ==> simply read this if you face a localization issue.
  • [Cosmetics] 2.4/5GHz switch available even though device is only capable of 2.4GHz Wifi [L]
  • Quiet Hours and Blacklist issue with XT1033 [M] [more info] 20141025-nightly : http://review.cyanogenmod.org/#/c/75556/
 
Last edited:

matmutant

Senior Member
Mar 17, 2011
3,395
4,743
~/
andrux-and-me.blogspot.com
Moto G
Moto E
Last edited:

myabc

Senior Member
Jan 27, 2006
1,232
64
in call receiving or outgoing the interlocutor listen echo of its own voice. this is very strong bug that keep this rom not good for using phone !
 
  • Like
Reactions: matmutant

matmutant

Senior Member
Mar 17, 2011
3,395
4,743
~/
andrux-and-me.blogspot.com
Moto G
Moto E
more info about the "ram bug"

first there is that story of "lost RAM" :
Code:
su
dumpsys meminfo

contrary to what i've been told, the "lost RAM" isn't increasing with time going on :
a few minutes ago i had more than 115MB of Lost RAM, and now it is 74209kB (74MB)
so... lost RAM can be somewhat turned into Free RAM ?

Note that i use CM11-2140317 on top of PinkKernel-exp-20140317, and greenify is installed.

Just to show it can actually decrease : see uptime as a piece of evidence ==> one time and a minute later


http://xdaforums.com/showpost.php?p=51202908&postcount=33
http://xdaforums.com/showpost.php?p=51214418&postcount=49
http://xdaforums.com/showpost.php?p=51228300&postcount=57


we need more data
 
Last edited:

matmutant

Senior Member
Mar 17, 2011
3,395
4,743
~/
andrux-and-me.blogspot.com
Moto G
Moto E
in call receiving or outgoing the interlocutor listen echo of its own voice. this is very strong bug that keep this rom not good for using phone !

ok ... i have an idea about this one, the top_of_device microphone is really near the speaker

hypothesis : top_of_device microphone catch both user voice and interlocutor's one and then send them to device creating the echo for interlocutor.
==> potential fix : disabling that microphone during call and only keep the other one.
 
Last edited:

matmutant

Senior Member
Mar 17, 2011
3,395
4,743
~/
andrux-and-me.blogspot.com
Moto G
Moto E
Screen turns off during call

bug: Screen randomly turns off after the action of emitting call (confirmed to exist on any nighlty with stock or pinkkernel)

workaround: force reboot or wait for interlocutor to end call.

potential fix: [?]

Would need more data from other users to confirm this bug on more devices.
 
Last edited:

ph4zrd

Senior Member
Nov 30, 2012
72
88
Dresden, Saxony
first there is that story of "lost RAM" :
Code:
su
dumpsys meminfo

contrary to what i've been told, the "lost RAM" isn't increasing with time going on :
a few minutes ago i had more than 115MB of Lost RAM, and now it is 74209kB (74MB)
so... lost RAM can be somewhat turned into Free RAM ?

Note that i use CM11-2140317 on top of PinkKernel-exp-20140317, and greenify is installed.

we need more data

For me the "Lost RAM" increased constantly since I rebooted yesterday. It was around 400,0 MB 10 minutes ago. I then tried to kill surfaceflinger: "Lost RAM" was at ~318 MB for a short time and began increasing again, now it's around 360 MB.

CM 20140317, CM default kernel, Dalvik btw...

Edit: It's a XT1032.
Tried killing one system process after another. "Lost RAM" value did not decrease. Induced soft reboots took very long. Any idea what I can try or which data could help locating the problem?

Edit 2: Phone slow as f*** now. Soft-rebooting itself, resetting the wallpaper, randomly killing apps, etc. Need to wait >30sec for it to open terminal emulator. I have a theory that the CM stock browser is somewhat involved cause after using it for ~10 min, my "Lost RAM" stat has risen by around 80 MB.
 
Last edited:

ph4zrd

Senior Member
Nov 30, 2012
72
88
Dresden, Saxony
More than enough info -> http://forum.cyanogenmod.com/topic/89863-cm11-on-moto-g-ram-full-after-24-hours/

Sent from my Moto G using XDA Premium 4 mobile app

Thank you for that link. Can confirm the information given there, my phone shows very similar results.
@matmutant: The "Lost RAM" does indeed decrease by some MB sometimes. I think the term is not 100% correct. The RAM is not lost for the phone, it is allocated/consumed by something dumpsys isn't aware about. Thus it can also be freed of course.
 

matmutant

Senior Member
Mar 17, 2011
3,395
4,743
~/
andrux-and-me.blogspot.com
Moto G
Moto E
Edit 2: Phone slow as f*** now. Soft-rebooting itself, resetting the wallpaper, randomly killing apps, etc. Need to wait >30sec for it to open terminal emulator. I have a theory that the CM stock browser is somewhat involved cause after using it for ~10 min, my "Lost RAM" stat has risen by around 80 MB.

More than enough info -> http://forum.cyanogenmod.com/topic/89863-cm11-on-moto-g-ram-full-after-24-hours/

Sent from my Moto G using XDA Premium 4 mobile app

Thank you for that link. Can confirm the information given there, my phone shows very similar results.

@matmutant: The "Lost RAM" does indeed decrease by some MB sometimes. I think the term is not 100% correct. The RAM is not lost for the phone, it is allocated/consumed by something dumpsys isn't aware about. Thus it can also be freed of course.


I made some investigations, your hypothesis about the implication of Stock Browser looks plausible, i've got a high raise of that "lost RAM" after launching that app ...
 

Gronnd

Senior Member
Nov 22, 2013
182
22
I made some investigations, your hypothesis about the implication of Stock Browser looks plausible, i've got a high raise of that "lost RAM" after launching that app ...

My highest raise of "lost RAM" was after using google+ and googl+ photos during 3 or 4 min. Before I had 107000mb(aprox), after I used that apps I had 180000mb
 

matmutant

Senior Member
Mar 17, 2011
3,395
4,743
~/
andrux-and-me.blogspot.com
Moto G
Moto E
A few small issues that no one else has reported so I might as well do it: When I set my volume panel to expandable it goes back to single after about 5 minutes, the only temporary fix is to change to single then back to expandable, but then it just goes back to single again. Also the animation when you go into standby doesn't work. Any fixes? And is it because of an app I'm running? Also I am running milestone 4 and I have the us version (xt1034)

No standby animation ==> did you change animations speed to 0 in Dev settings? or maybe disable "screen-off animation" in display and led settings :)
 
  • Like
Reactions: Jocker111

Jocker111

Member
Mar 17, 2014
25
23
Ok, I'm the original poster from the CM forum, where I've put already my analysis, but it seems here more ppl are active, so I try to find the solution here...
Shortly to my background:
- I'm quite familiar with linux itself, use it at work and privately
- I'm very new to android itself

Now here is what I do not get:
Coming from typical linux I am used to, that I can find ANY process running on my computer with the 'ps' command. The same I would expect from android. Now here for what I do not get:
- The linux kernel must be changed in android compared to original linux, in a way that NOT all memory appears in the ps command
- Can processes be hidden from the ps command (this basically means, they would not appear in /proc)
- If the answers to the above questions is: All processes are listed with ps and we can see their memory usage there, then in my oppinion there can only be a memory leak in the kernel itself.
This is just my oppinion in the problem, because that would be the only explenation for a plain linux system (again I'm new to Android and I have no idea how much google changed in the kernel code).

Now maybe also some more information how this happens on my phone:
- I'm rarely using the Browser on my phone, and the Lost RAM goes up to 500 in 2 days, without that I am ACTIVELY opening the browser (maybe something in the background is using the browser, but I'm not doing anything actively into that direction)

- Typically the Lost RAM does not increase much over night (maybe 50mb), but it seems that even when I'm not doing anything actively some ram gets lost.

- Running applications during the night are Xabber and the email client (also WhatsApp and Viber are running from time to time for checking new messages)

- It seems that most ram gets lost, when I actually move around with my phone, i.e. from one network setup at home, to another network setup at work, and again another network setup during lunch time.

So I am really interested in the answer to the question, how a process can have memory in usage, but this memory is not visible with the ps command or visible in /proc respectively
 

matmutant

Senior Member
Mar 17, 2011
3,395
4,743
~/
andrux-and-me.blogspot.com
Moto G
Moto E
Thank you @Jocker111, the main problem with this issue is that we didn't find (yet) the culprit...

That "Lost RAM" increase, then decrease a little, increase again (on my side at least) and that wouldn't be a real problem if it did not take this enormous amount of RAM as you describe (until now, i never got more than 160MB "lost")

We all need to investigate in different directions to remove false positive and silly hypothesis, and then brain storm a little and see what anyone could find.
 

Top Liked Posts

  • There are no posts matching your filters.
  • 20
    CM11 debug thread for Falcon

    This thread is related to CM11 for Moto G (Falcon) maintained by Ethan Chen

    for debugging purpose, i need this only thread to centralize the data
    ==> Report bugs (documented) i'll sort them by relevance (importance : low [L]; medium [M]; high [H]) and follow them to Ethan Chen.




    Before Asking anything here, please refer to RC-FAQ & CM11 FAQ by @pinguijxy
    refer to this post for general commands​

    • post 2 : known and solved
    • post 3 : pending
    • post 4: irrelevant

    to improve CM11 for Falcon, you could :
    • post results after following protocols,
      [*]help me to set up test new protocols,
      [*]give advices about them
      [*]suggest things to investigate
      IF you can prove it worths to ;)

    How To Report?
    20140819-edit : Please use "Bug Report" button at the top of the thread instead of posting in thread.
    give at least :

    • model : XT103* ?
    • rom build : cm-11-yyyymmdd-NIGHTLY-falcon ?
    • kernel build :
    • logcat?
    • way to reproduce ? (ok this is the best, but not always easy, or even possible to do)

    XDA:DevDB Information
    [DEV/DEBUG] Official CM11 debug and dev thread for Falcon, ROM for the Moto G

    Contributors
    matmutant, pinguijxy, CyanogenMod, Ethan Chen
    ROM OS Version: 4.4.x KitKat
    ROM Kernel: Linux 3.4.x
    Based On: CyanogenMod

    Version Information
    Status: Testing

    Created 2014-08-19
    Last Updated 2015-01-10
    10
    No major problems so far. RAM usage seems to increase over time like in official CM, but much slower and less predictable. "Lost RAM" drops by 20-30 MB from time to time. It feels like my voodoo "fixed" one of some more problems, but only longer testing will tell.

    So here you go. Standard disclaimer, no warranty yadda yadda yadda.

    Code:
    [...]
    
    Total PSS by category:
       108786 kB: Native
       101428 kB: Dalvik
        52426 kB: .dex mmap
        48599 kB: .so mmap
        47478 kB: Dalvik Other
        20583 kB: Other dev
        15317 kB: Ashmem
        13803 kB: .apk mmap
         6924 kB: Unknown
         4286 kB: Stack
         3092 kB: Other mmap
         1734 kB: .ttf mmap
          170 kB: .jar mmap
            8 kB: Cursor
            0 kB: code mmap
            0 kB: image mmap
            0 kB: Graphics
            0 kB: GL
            0 kB: Memtrack
    
    Total RAM: 903272 kB
     Free RAM: 469660 kB (110976 cached pss + 274788 cached + 83896 free)
     Used RAM: 373874 kB (313658 used pss + 3608 buffers + 11320 shmem + 45288 slab)
     Lost RAM: 59738 kB
         ZRAM: 4304 kB physical used for 15248 kB in swap (131068 kB total swap)
          KSM: 44048 kB saved from shared 8208 kB
               184956 kB unshared; 326004 kB volatile
       Tuning: 64 (large 256), oom 122880 kB, restore limit 40960 kB (high-end-gfx)

    Code:
    up time: 03:39:53, idle time: 01:56:57, sleep time: 01:35:43

    Code:
    $ md5sum cm-11-20140319-UNOFFICIAL-falcon-test2.zip 
    690636b7b10bd48bcec4fc9a8116fed6  cm-11-20140319-UNOFFICIAL-falcon-test2.zip

    Edit: Test build removed, use cm nightly + zip from here.
    8
    Im currently trying to reproduce the ram bug on a build I compiled 30min ago. So far I couldn't achieve more than 100 MB of "Lost RAM", currently it's at ~45 MB.

    Because someone said the bug happens with all versions since 20140224, I searched for commits in the device specific repositories that were pushed just after this date, especially in the kernel and qcom display repository. What I found were some commits from 26/02 in android_kernel_motorola_msm8226: https://github.com/CyanogenMod/android_kernel_motorola_msm8226/commits/cm-11.0?page=11
    I tried to revert those (cc12abf69bf608b1accffb15682bdeecd36aaaa3..6a8acccc9a65df6cd8782d8858003dea75cfa6e8), got an error during compilation and reverted 518351814c0f60c3ffad26bead05624b668c05ee in android_hardware_qcom_display-caf-new as well (git blame helped here).

    It seems that there are still some problems, "Lost RAM" does not go below 40 MB and rises sometimes to as high as 95 MB. But maybe that's just normal. I will test the build for some hours/days and report. If I don't get any problems, I may upload it for others to try so we can verify that the bug was introduced by said commits and track it down further.

    The only commit that needs to be reverted is the commit 1c8afaa0734547907076c45245aa4030a9049ccb with the funny commit message:
    Revert "Revert "msm: mdss: Detach mdss iommu on static screen"" commited and authored by dhacker29

    This raises for me 2 questions @dhacker29:
    - Why was the original commit reverted?
    - Why was the revert again reverted?
    Apparently when I revert the revert which is a revert (wow, that's a lot of reverts), then I get rid of the lost RAM bug...

    Just compiled the current git head with only 1c8afaa0734547907076c45245aa4030a9049ccb reverted, and everything seems to be fine. Let' see how the long-term usage will turn out.
    7
    Are you posting the fix for wifi to the official cm code review, or will you keep doing your own builds?

    Thanks for all the effort.

    Yeah I'll put it on the code review, I've just been lazy the last day or two there is one or two extra commits that need to be merged with the update for it to compile, ill get it up there tomorrow night probably.
    6
    Im currently trying to reproduce the ram bug on a build I compiled 30min ago. So far I couldn't achieve more than 100 MB of "Lost RAM", currently it's at ~45 MB.

    Because someone said the bug happens with all versions since 20140224, I searched for commits in the device specific repositories that were pushed just after this date, especially in the kernel and qcom display repository. What I found were some commits from 26/02 in android_kernel_motorola_msm8226: https://github.com/CyanogenMod/android_kernel_motorola_msm8226/commits/cm-11.0?page=11
    I tried to revert those (cc12abf69bf608b1accffb15682bdeecd36aaaa3..6a8acccc9a65df6cd8782d8858003dea75cfa6e8), got an error during compilation and reverted 518351814c0f60c3ffad26bead05624b668c05ee in android_hardware_qcom_display-caf-new as well (git blame helped here).

    It seems that there are still some problems, "Lost RAM" does not go below 40 MB and rises sometimes to as high as 95 MB. But maybe that's just normal. I will test the build for some hours/days and report. If I don't get any problems, I may upload it for others to try so we can verify that the bug was introduced by said commits and track it down further.