Closed Bug 1121612 Opened 9 years ago Closed 9 years ago

Intermittent test_dynamicsCompressorNode.html | application crashed [@ quant_all_bands]

Categories

(Core :: Audio/Video, defect)

Product:
Core
Component:
Audio/Video
Platform:
ARM
Android
Type:
defect
Priority:
Not set
Severity:
normal

Tracking

()

Status:
RESOLVED WORKSFORME

People

(Reporter: RyanVM, Unassigned)

References

Details

(Keywords: crash, intermittent-failure) 

19:22:10 WARNING - PROCESS-CRASH | dom/media/webaudio/test/test_dynamicsCompressorNode.html | application crashed [@ quant_all_bands]
19:22:10 INFO - Crash dump filename: /tmp/tmpv6mwqY/633b12ff-5abb-4b90-00490ac0-628dfef1.dmp
19:22:10 INFO - Operating system: Android
19:22:10 INFO - 0.0.0 Linux 3.2.0+ #2 SMP PREEMPT Thu Nov 29 08:06:57 EST 2012 armv7l pandaboard/pandaboard/pandaboard:4.0.4/IMM76I/5:eng/test-keys
19:22:10 INFO - CPU: arm
19:22:10 INFO - 2 CPUs
19:22:10 INFO - Crash reason: SIGILL
19:22:10 INFO - Crash address: 0x6343c3e8
19:22:10 INFO - Thread 66 (crashed)
19:22:10 INFO - 0 libxul.so!quant_all_bands [bands.c:b2d1d57c25b9 : 439 + 0x0]
19:22:10 INFO - r4 = 0x0000000e r5 = 0x0000000e r6 = 0x00002000 r7 = 0x6e984828
19:22:10 INFO - r8 = 0x00004002 r9 = 0x00004002 r10 = 0x6e9855d0 fp = 0xfdda6ed4
19:22:10 INFO - sp = 0x6e984340 lr = 0x6343c38f pc = 0x6343c3e8
19:22:10 INFO - Found by: given as instruction pointer in context
19:22:10 INFO - 1 libxul.so!celt_decode_with_ec [celt_decoder.c:b2d1d57c25b9 : 898 + 0x3]
19:22:10 INFO - r4 = 0x00000000 r5 = 0x716077d0 r6 = 0x0000002a r7 = 0x6e985af0
19:22:10 INFO - r8 = 0x00000000 r9 = 0x6e986b78 r10 = 0x6e9858b0 fp = 0x6e985970
19:22:10 INFO - sp = 0x6e984930 pc = 0x6343d9fb
19:22:10 INFO - Found by: call frame info
19:22:10 INFO - 2 libxul.so!opus_decode_frame [opus_decoder.c:b2d1d57c25b9 : 479 + 0x1b]
19:22:10 INFO - r4 = 0x00000078 r5 = 0x000003c0 r6 = 0x716032b8 r7 = 0x6e986b20
19:22:10 INFO - r8 = 0x00000001 r9 = 0x00000000 r10 = 0x00000001 fp = 0x7160110c
19:22:10 INFO - sp = 0x6e985c00 pc = 0x6345f3b1
19:22:10 INFO - Found by: call frame info
19:22:10 INFO - 3 libxul.so!opus_decode_native [opus_decoder.c:b2d1d57c25b9 : 678 + 0x1b]
19:22:10 INFO - r4 = 0x7160110c r5 = 0x00000000 r6 = 0x6e986c0a r7 = 0x000003c0
19:22:10 INFO - r8 = 0x71b7ac02 r9 = 0x6e986ca8 r10 = 0x6e986c0a fp = 0x000003c0
19:22:10 INFO - sp = 0x6e986bd0 pc = 0x6345fe5b
19:22:10 INFO - Found by: call frame info
19:22:10 INFO - 4 libxul.so!opus_multistream_decode_native [opus_multistream_decoder.c:b2d1d57c25b9 : 251 + 0x3]
19:22:10 INFO - r4 = 0x00000000 r5 = 0x6e986ca8 r6 = 0x00000000 r7 = 0x6e987bb0
19:22:10 INFO - r8 = 0x6adeb000 r9 = 0x000003c0 r10 = 0x000003bd fp = 0x71601000
19:22:10 INFO - sp = 0x6e986c90 pc = 0x634634cd
19:22:10 INFO - Found by: call frame info
19:22:10 INFO - 5 libxul.so!mozilla::OggReader::DecodeOpus(ogg_packet*) [OggReader.cpp:b2d1d57c25b9 : 586 + 0xb]
19:22:10 INFO - r4 = 0x6b5d8e40 r5 = 0x00000001 r6 = 0x6eb26370 r7 = 0x6e987d24
19:22:10 INFO - r8 = 0x6e987d84 r9 = 0x000003c0 r10 = 0x70d4bc00 fp = 0x00000008
19:22:10 INFO - sp = 0x6e987c78 pc = 0x62d0ab31
19:22:10 INFO - Found by: call frame info
19:22:10 INFO - 6 libxul.so!mozilla::OggReader::DecodeAudioData() [OggReader.cpp:b2d1d57c25b9 : 703 + 0x7]
19:22:10 INFO - r4 = 0x70d4bc00 r5 = 0x6b5d8e40 r6 = 0x6eb26370 r7 = 0x6e987d24
19:22:10 INFO - r8 = 0x6e987d84 r9 = 0x00000000 r10 = 0x6adae5d4 fp = 0x001205bc
19:22:10 INFO - sp = 0x6e987cc8 pc = 0x62d107cb
19:22:10 INFO - Found by: call frame info
19:22:10 INFO - 7 libxul.so!mozilla::MediaDecoderReader::RequestAudioData() [MediaDecoderReader.cpp:b2d1d57c25b9 : 244 + 0x9]
19:22:10 INFO - r4 = 0x70d4bc00 r5 = 0x6e987cf0 r6 = 0x6f7df420 r7 = 0x6e987d24
19:22:10 INFO - r8 = 0x6e987d84 r9 = 0x00000000 r10 = 0x6adae5d4 fp = 0x001205bc
19:22:10 INFO - sp = 0x6e987ce8 pc = 0x62cdf511
19:22:10 INFO - Found by: call frame info
19:22:10 INFO - 8 libxul.so!mozilla::MediaDecoderStateMachine::DecodeAudio() [MediaDecoderStateMachine.cpp:b2d1d57c25b9 : 712 + 0x5]
19:22:10 INFO - r4 = 0x71b78400 r5 = 0x6e987d30 r6 = 0x70337ce0 r7 = 0x6e987d8c
19:22:10 INFO - r8 = 0x6e987d84 r9 = 0x00000000 r10 = 0x6adae5d4 fp = 0x001205bc
19:22:10 INFO - sp = 0x6e987d10 pc = 0x62cdf017
19:22:10 INFO - Found by: call frame info
19:22:10 INFO - 9 libxul.so!nsRunnableMethodImpl<tag_nsresult (mozilla::MediaDecoderReader::*)(), void, true>::Run() + 0x23
19:22:10 INFO - r4 = 0x6b5c5740 r5 = 0x6b5c5740 r6 = 0x70337ce0 r7 = 0x6e987d8c
19:22:10 INFO - r8 = 0x6e987d84 r9 = 0x00000000 r10 = 0x6adae5d4 fp = 0x001205bc
19:22:10 INFO - sp = 0x6e987d50 pc = 0x62ce261f
19:22:10 INFO - Found by: call frame info
19:22:10 INFO - 10 libxul.so!mozilla::MediaTaskQueue::Runner::Run() [MediaTaskQueue.cpp:b2d1d57c25b9 : 230 + 0x7]
19:22:10 INFO - r4 = 0x6b1ee5c0 r5 = 0x6b5c5740 r6 = 0x70337ce0 r7 = 0x6e987d8c
19:22:10 INFO - r8 = 0x6e987d84 r9 = 0x00000000 r10 = 0x6adae5d4 fp = 0x001205bc
19:22:10 INFO - sp = 0x6e987d58 pc = 0x62cd5655
19:22:10 INFO - Found by: call frame info
See Also: → 1057067
Any ideas what may be going on here, Matt? Doesn't seem to be a slave-specific issue like I'd hoped. I think this may be related to bug 1057067 as well based on the stack.
Flags: needinfo?(kinetik)
I'm not an Opus internals or ARM expert, so not really, sorry.  What's changed recently, are these new slaves or have we switched compilers?

Ralph might be a better person to start with, so I'll bounce the ni? to him.
Flags: needinfo?(kinetik) → needinfo?(giles)
(In reply to Matthew Gregan [:kinetik] from comment #5)
> I'm not an Opus internals or ARM expert, so not really, sorry.  What's
> changed recently, are these new slaves or have we switched compilers?

Nothing new in Android land from an infra standpoint.
Toolchain changes are the only thing I can think of which could raise new crashes in libopus. The code hasn't changed since November.
Flags: needinfo?(giles)
Any ideas, Jean-Marc? Is this something we've addressed upstream?
Flags: needinfo?(jmvalin)
I'm not aware of any crash in libopus in quite some time. Most likely explanations I can think of (in no particular order):
1) A bug in the ARM assembly, which has received less testing than the C code
2) Some corruption that's totally unrelated to opus and is just causing a crash there
3) A bug in the toolchain

Of course, I can't totally rule out a bug in the libopus C code, but given the size, simplicity and the number of testers, I consider this pretty unlikely (bugs in Opus tend to cause bad audio but very rarely crashes).
Flags: needinfo?(jmvalin)
Seems to have gone away on its own.
Status: NEW → RESOLVED
Closed: 9 years ago
Resolution: --- → WORKSFORME
You need to log in before you can comment on or make changes to this bug.