1096870 - [Flame][Device Storage]Device crash while copy file from device to itself on PC.

Status: RESOLVED WORKSFORME

(Firefox OS Graveyard :: Gaia::System, defect)

Description

[Alissa（leave from Mozilla）]

Attachment: logcat.txt

[1.Description]:
[Flame][v2.1][Device Storage]Device crash while copy file from device to itself on PC.
**Title
      ~B2G 34.0 Crash Report [@ pthread_kill ]
**Crash Report
    ~https://crash-stats.mozilla.com/report/index/f0f09bf7-ab82-4983-a1a7-cfe162141111
Occurence time:17:00
attachment:logcat.txt

[2.Testing Steps]: 
1.Enable USB Storage and connect with PC.
2.Copy some file to device from pc.
3.Copy some file on device internal storage and past on internal storage.(blank space is 600M while device crash)

[3.Expected Result]: 
3.Device should run normally.

[4.Actual Result]: 
3.Device crash report pop up.

[5.Reproduction build]: 
Gaia-Rev        0ec1925fc37b7c71d129ae44e42516a0cfb013c4
Gecko-Rev       https://hg.mozilla.org/releases/mozilla-b2g34_v2_1/rev/97487a2d1ee6
Build-ID        20141110001201
Version         34.0

[6.Reproduction Frequency]: 
Once,1/5

Comment 1

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

Crash stack : 
Frame 	Module 	Signature 	Source
0 	libc.so 	libc.so@0x220e8 	
1 	libc.so 	pthread_kill 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/bionic/libc/bionic/pthread_kill.cpp:49
2 	libc.so 	raise 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/bionic/libc/bionic/raise.cpp:32
3 	libc.so 	__libc_android_abort 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/bionic/libc/bionic/abort.cpp:55
4 	libc.so 	libc.so@0x2199e 	
5 	libc.so 	__libc_fatal 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/bionic/libc/bionic/libc_logging.cpp:514
6 	libc.so 	__assert2 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/bionic/libc/bionic/assert.cpp:41
7 	libxul.so 	play_callback 	media/libcubeb/src/cubeb_opensl.c
8 	libwilhelm.so 	audioTrack_handleMarker_lockPlay(CAudioPlayer_struct*) 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/frameworks/wilhelm/src/android/AudioPlayer_to_android.cpp:349
9 	libwilhelm.so 	audioTrack_callBack_pullFromBuffQueue 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/frameworks/wilhelm/src/android/AudioPlayer_to_android.cpp:1199
10 	libmedia.so 	android::AudioTrack::processAudioBuffer(android::sp<android::AudioTrack::AudioTrackThread> const&) 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/frameworks/av/media/libmedia/AudioTrack.cpp:1574
11 	libmedia.so 	android::AudioTrack::AudioTrackThread::threadLoop() 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/frameworks/av/media/libmedia/AudioTrack.cpp:1921
12 	libutils.so 	android::Thread::_threadLoop(void*) 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/system/core/libutils/Threads.cpp:770
13 	libutils.so 	thread_data_t::trampoline(thread_data_t const*) 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/system/core/libutils/Threads.cpp:95
14 	libc.so 	__thread_entry 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/bionic/libc/bionic/pthread_create.cpp:105
15 	libc.so 	pthread_create 	/builds/slave/b2g_m-cen_flm-kk_ntly-00000000/build/bionic/libc/bionic/pthread_create.cpp:224

Comment 2

[Norry.L.F(Leave from Mozilla)]

Hi Josh,

Could you help with this crash issue? Thanks.

Comment 3

[Mike Lien[:mlien]]

set blocking-b2g: 2.1? to see if this crash can be fixed in 2.1

Comment 4

[bhavana bajaj [:bajaj]]

NI :dhylands, to see if he can help or re-direct this someone who can investigate the issue.

Comment 5

[Dave Hylands [:dhylands]]

So I don't think that this has anything to do with copying a file to itself (at least not while using UMS).

When UMS is in play, the PC has total control over the storage area, and the phone has no access, and doesn't even know that the host is copying files.

It looks like the crash is happening when something tries to access the storage area while its being shared with the the PC.

It appears that this assert:
http://dxr.mozilla.org/mozilla-central/source/media/libcubeb/src/cubeb_opensl.c#79
or this one:
http://dxr.mozilla.org/mozilla-central/source/media/libcubeb/src/cubeb_opensl.c#83
is firing.

I think that this needs to be investigated by somebody who is familiar with the media codecs.

Comment 6

[Naoki Hirata :nhirata (please use needinfo instead of cc)]

similar crash : bug 1060912

Comment 7

[Tony Chung [:tchung]]

    qawanted for more investigation.

    Agree with Dave Dhylands here.  the crash stack in comment 1 seems to relate more to media codecs, rather than copying files from the device.   Reporter, can you please try and reproduce this again, and isolate any other apps running?   Did you have a music player running?   Please add more information, and check to see if you are logging the correct crash report.

Comment 8

[Pi Wei Cheng [:piwei] (inactive)]

I'm unable to reproduce the crash following the STR on reporter's build. Repeated steps 2 & 3 for 10 times with NO reproduction.

Tested on:
Device: Flame 2.1 (KK, full flash, 319MB)
BuildID: 20141110001201
Gaia: 0ec1925fc37b7c71d129ae44e42516a0cfb013c4
Gecko: 97487a2d1ee6
Gonk: 48835395daa6a49b281db62c50805bd6ca24077e
Version: 34.0 (2.1)
Firmware: V188-1
User Agent: Mozilla/5.0 (Mobile; rv:34.0) Gecko/34.0 Firefox/34.0

Leaving qawanted tag for others to attempt.

Comment 9

[Joshua Mitchell (Inactive)]

Should point out that we (Pi Wei) use Ubuntu OS laptops - just in case this is hardware dependent.

Comment 10

[Joshua Mitchell (Inactive)]

Maybe related to bug 1077642

Comment 11

[Alissa（leave from Mozilla）]

(In reply to Tony Chung [:tchung] from comment #7)
>     qawanted for more investigation.
> 
>     Agree with Dave Dhylands here.  the crash stack in comment 1 seems to
> relate more to media codecs, rather than copying files from the device.  
> Reporter, can you please try and reproduce this again, and isolate any other
> apps running?   Did you have a music player running?   Please add more
> information, and check to see if you are logging the correct crash report.

I'm not sure there is any app run in background,but my test file is (mp4),so if there is relationship with it.

Comment 12

[Mike Lien[:mlien]]

confirmed with reporter, this issue cannot be reproduced anymore

Comment 13

[Joshua Mitchell (Inactive)]

If this no longer reproduces we should close it as works-for-me

Comment 14

[Alissa（leave from Mozilla）]

(In reply to Joshua Mitchell [:Joshua_M] from comment #13)
> If this no longer reproduces we should close it as works-for-me

OK,I will track this issue.