893676 - [Bluetooth][A2DP] Quickly switch between devices would cause strange issues

Status: RESOLVED INVALID

(Firefox OS Graveyard :: Bluetooth, defect)

Description

[Walter Chen[:ypwalter][:wachen]]

*Environment:
Gaia:     6724eb7733dd425b65027d0c2fd414bc9b74d624
  B-D     2013-07-14 00:31:48
Gecko:    http://hg.mozilla.org/mozilla-central/rev/18467a85acf6
BuildID   20130714030201
Version   25.0a1

pvt master build with manually pushed audio.conf
    adb remount
    adb push B2G/system/bluetooth/data/audio.conf /etc/bluetooth/audio.conf

*How to reproduce:
1. Launch "Settings" app
2. Go to "bluetooth" subsection
3. Turn on bluetooth
4. Pair with 2 A2DP-supported bluetooth earphones
5. Launch "Music" app
6. Play any music
7. You can hear it from earphone A
8. Go back to settings app in bluetooth subsection, connect to earphone B
9. connect to earphone A
10. connect to earphone B
11. repeat 9&10 for lots of times (like 10 times)

*Expected Result:
You can hear the music from earphone B

*Actual Result:
a. Music app might crash
b. you might hear the music from earphone A

*Note:
Sorry, these are hard to reproduce. However, I do meet them twice for each.
However, I didn't have the logcat on that comment.
I will try to attach a logcat after a long time :P

Comment 1

[Walter Chen[:ypwalter][:wachen]]

Attachment: logcat

Comment 2

[Eric Chou [:ericchou] [:echou]]

This doesn't look like a Bluetooth bug.

E/ProcessKiller(  104): Process /system/b2g/plugin-container (2711) has open file /mnt/secure/staging/downloads/bluetooth/01.çŽ‹è² - 百年孤寂.mp3
W/ProcessKiller(  104): Sending SIGHUP to process 2711
E/OMXNodeInstance( 2250): !!! Observer died. Quickly, do something, ... anything...
F/OMXNodeInstance( 2250): frameworks/base/media/libstagefright/omx/OMXNodeInstance.cpp:203 !"should not be here, unknown state."
F/libc    ( 2250): Fatal signal 11 (SIGSEGV) at 0xdeadbaad (code=1)
I/Gecko   (  109): 
I/Gecko   (  109): ###!!! [Parent][AsyncChannel] Error: Channel error: cannot send/recv
I/Gecko   (  109): 
I/GeckoDump(  109): Crash reporter : Not online, postponing.
I/DEBUG   (  114): *** *** *** *** *** *** *** *** *** *** *** *** *** *** *** ***
I/DEBUG   (  114): Build fingerprint: 'toro/full_unagi/unagi:4.0.4.0.4.0.4/OPENMASTER/eng.cltbld.20130714.063019:eng/test-keys'
I/DEBUG   (  114): pid: 2250, tid: 2250  >>> /system/bin/mediaserver <<<
I/DEBUG   (  114): signal 11 (SIGSEGV), code 1 (SEGV_MAPERR), fault addr deadbaad
I/DEBUG   (  114):  r0 deadbaad  r1 00bebdf0  r2 a0000000  r3 00000000
I/DEBUG   (  114):  r4 00000000  r5 00000027  r6 00be6614  r7 00be6648
I/DEBUG   (  114):  r8 00be6620  r9 00bbb6bc  10 00bbb6a8  fp 00000000
I/DEBUG   (  114):  ip 4009612c  sp beb83610  lr 400b2191  pc 400ae8f4  cpsr 60000030
I/DEBUG   (  114):  d0  6f6e20646c756f68  d1  7265682065622074
I/DEBUG   (  114):  d2  6f6e6b6e75202c65  d3  6574617473206e77
I/DEBUG   (  114):  d4  0000000000000000  d5  0000000000000000
I/DEBUG   (  114):  d6  4220000041300000  d7  3f8000003debc8c1
I/DEBUG   (  114):  d8  0000000000000000  d9  0000000000000000
I/DEBUG   (  114):  d10 0000000000000000  d11 0000000000000000
I/DEBUG   (  114):  d12 0000000000000000  d13 0000000000000000
I/DEBUG   (  114):  d14 0000000000000000  d15 0000000000000000
I/DEBUG   (  114):  d16 3ff0000000000000  d17 3ff0000000000000
I/DEBUG   (  114):  d18 7e37e43c8800759c  d19 3f636ee70bc10671
I/DEBUG   (  114):  d20 3f11566a8631d7d6  d21 bebbbd418439103e
I/DEBUG   (  114):  d22 3ff0000000000000  d23 3ff009bb63fc01c8
I/DEBUG   (  114):  d24 3e66376972bea4d0  d25 0000000000000000
I/DEBUG   (  114):  d26 0000000000000000  d27 0000000000000000
I/DEBUG   (  114):  d28 0000000000000000  d29 0000000000000000
I/DEBUG   (  114):  d30 0000000000000000  d31 0000000000000000
I/DEBUG   (  114):  scr 60000010
I/DEBUG   (  114): 
I/DEBUG   (  114):          #00  pc 000178f4  /system/lib/libc.so
I/DEBUG   (  114):          #01  pc 0000144a  /system/lib/liblog.so (__android_log_assert)
I/DEBUG   (  114):          #02  pc 0000d6b8  /system/lib/libstagefright_omx.so (_ZN7android15OMXNodeInstance8freeNodeEPNS_9OMXMasterE)
I/DEBUG   (  114):          #03  pc 0000d7a0  /system/lib/libstagefright_omx.so (_ZN7android15OMXNodeInstance14onObserverDiedEPNS_9OMXMasterE)
I/DEBUG   (  114):          #04  pc 0000b9ba  /system/lib/libstagefright_omx.so (_ZN7android3OMX10binderDiedERKNS_2wpINS_7IBinderEEE)
I/DEBUG   (  114):          #05  pc 00018116  /system/lib/libbinder.so (_ZN7android8BpBinder14reportOneDeathERKNS0_8ObituaryE)
I/DEBUG   (  114):          #06  pc 00018186  /system/lib/libbinder.so (_ZN7android8BpBinder12sendObituaryEv)
I/DEBUG   (  114):          #07  pc 0001b200  /system/lib/libbinder.so (_ZN7android14IPCThreadState14executeCommandEi)
I/DEBUG   (  114):          #08  pc 0001b36e  /system/lib/libbinder.so (_ZN7android14IPCThreadState14joinThreadPoolEb)
I/DEBUG   (  114):          #09  pc 00008c24  /system/bin/mediaserver
I/DEBUG   (  114):          #10  pc 00016778  /system/lib/libc.so (__libc_init)

Comment 3

[Eric Chou [:ericchou] [:echou]]

The log I attached to comment 2 is about bug 894530 and bug 864188. We should focus on why Music player was dead.

Comment 4

[Joe Cheng [:jcheng] (please needinfo)]

retriage: corner case. minus

Comment 5

[Gina Yeh [:gyeh] [:ginayeh]]

I'd like to steal the bug, Eric.

Comment 6

[Gina Yeh [:gyeh] [:ginayeh]]

The UX behaviour has been changed for a while. After disconnecting A2DP, Music App should pause the music in case of the audio is routed to speaker at a high volume and scare the user away.

I tried with the latest build of both v1.2 and m-c. The routing path seems great and no crash. Walter, can you or other QA help to verify it again?

Comment 7

[Walter Chen[:ypwalter][:wachen]]

The behavior is changed as Gina described. I think currently this behavior is hard to reproduce.

Comment 8

[Gina Yeh [:gyeh] [:ginayeh]]

That's great! Thanks, Walter.