Closed Bug 845146 Opened 8 years ago Closed 8 years ago

Intermittent test_bug610212.html, test_bug606817.html, test_bug613979.html | application crashed [@ libGLESv2_POWERVR_SGX540_120.so + 0x19904]

Categories

(Core :: DOM: Core & HTML, defect)

ARM
Android
defect
Not set
critical

Tracking

()

RESOLVED WORKSFORME

People

(Reporter: RyanVM, Unassigned)

References

(Depends on 1 open bug)

Details

(Keywords: crash, intermittent-failure, Whiteboard: [likely a real bug, unlikely a test flakyness])

Crash Data

https://tbpl.mozilla.org/php/getParsedLog.php?id=20079389&tree=Mozilla-Inbound

Android 4.0 Panda mozilla-inbound opt test mochitest-2 on 2013-02-25 13:58:14 PST for push 99f36af07e3b
slave: panda-0612

75986 INFO TEST-END | /tests/content/html/content/test/test_bug607145.html | finished in 595ms
75987 INFO TEST-START | /tests/content/html/content/test/test_bug610212.html
INFO | automation.py | Application ran for: 0:03:15.845754
INFO | automation.py | Reading PID log: /tmp/tmp7LA8K5pidlog
getting files in '/mnt/sdcard/tests/profile/minidumps/'
Downloading symbols from: http://ftp.mozilla.org/pub/mozilla.org/mobile/tinderbox-builds/mozilla-inbound-android/1361827474/fennec-22.0a1.en-US.android-arm.crashreporter-symbols.zip
PROCESS-CRASH | /tests/content/html/content/test/test_bug610212.html | application crashed [@ libGLESv2_POWERVR_SGX540_120.so + 0x19904]
Crash dump filename: /tmp/tmpetEQr_/2888228b-001c-8d4d-61215398-36209f1c.dmp
Operating system: Android
                  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
CPU: arm
     0 CPUs

Crash reason:  SIGSEGV
Crash address: 0x9

Thread 0 (crashed)
 0  libGLESv2_POWERVR_SGX540_120.so + 0x19904
     r4 = 0x0000002d    r5 = 0x0190ae00    r6 = 0x01b16b58    r7 = 0x00000005
     r8 = 0x00000000    r9 = 0x01d42910   r10 = 0x0190b450    fp = 0x01d72ec0
     sp = 0xbebd8ef8    lr = 0x5bcef5d8    pc = 0x5be67904
    Found by: given as instruction pointer in context
 1  libGLESv2_POWERVR_SGX540_120.so + 0x1a862
     sp = 0xbebd8f08    pc = 0x5be68864
    Found by: stack scanning
 2  libc.so + 0x137c1
     sp = 0xbebd8f28    pc = 0x400e77c3
    Found by: stack scanning
 3  libGLESv2_POWERVR_SGX540_120.so + 0x10c8e
     sp = 0xbebd8f58    pc = 0x5be5ec90
    Found by: stack scanning
 4  libGLESv2_POWERVR_SGX540_120.so + 0x1e42e
     sp = 0xbebd8f60    pc = 0x5be6c430
    Found by: stack scanning
 5  libGLESv2_POWERVR_SGX540_120.so + 0x22532
     sp = 0xbebd8f70    pc = 0x5be70534
    Found by: stack scanning
https://tbpl.mozilla.org/php/getParsedLog.php?id=20091807&tree=Mozilla-Inbound
Severity: normal → critical
Summary: Intermittent content/html/content/test/test_bug610212.html | application crashed [@ libGLESv2_POWERVR_SGX540_120.so + 0x19904] → Intermittent content/html/content/test/test_bug610212.html, test_bug606817.html | application crashed [@ libGLESv2_POWERVR_SGX540_120.so + 0x19904]
Whiteboard: [likely a real bug, unlikely a test flakyness]
Maybe someone from the Android team should have a look at this?
https://tbpl.mozilla.org/php/getParsedLog.php?id=20238004&tree=Firefox
Summary: Intermittent content/html/content/test/test_bug610212.html, test_bug606817.html | application crashed [@ libGLESv2_POWERVR_SGX540_120.so + 0x19904] → Intermittent test_bug610212.html, test_bug606817.html, test_bug613979.html | application crashed [@ libGLESv2_POWERVR_SGX540_120.so + 0x19904]
(In reply to Mounir Lamouri (:mounir) from comment #6)
> Maybe someone from the Android team should have a look at this?

Please can you find an owner for this intermittent-failure - the current overall tree intermittent failure rate is spiralling out of control & the majority of bugs are unowned (see dev.platform thread).
Flags: needinfo?(blassey.bugs)
assigning to ted since I presume getting symbols for libc and libdv will shed some light on this.
Assignee: nobody → ted
Flags: needinfo?(blassey.bugs)
(In reply to Brad Lassey [:blassey] from comment #12)
> assigning to ted since I presume getting symbols for libc and libdv will
> shed some light on this.

Thank you :-)
I've got nothing here. I grabbed the minidump from the crash in comment 10, got the binaries for libGLESv2.. and libc, dumped what I could get out of them, and still can't get a useful stack. This is basically everything that's on the stack:
0xbec4af04: libGLESv2_POWERVR_SGX540_120.so!glScissor + 0x1cbc
0xbec4af24: libc.so!dlmalloc + 0xae
0xbec4af54: libGLESv2_POWERVR_SGX540_120.so!glGenRenderbuffers + 0xfc
0xbec4af5c: libGLESv2_POWERVR_SGX540_120.so!glBindAttribLocation + 0x1c4
0xbec4af6c: libGLESv2_POWERVR_SGX540_120.so!glCopyTexSubImage2D + 0xc00
0xbec4b120: libGLESv2_POWERVR_SGX540_120.so!glVertexAttribPointer + 0x226c8
0xbec4b134: libGLESv2_POWERVR_SGX540_120.so!glDrawArrays + 0x328
0xbec4b14c: libGLESv2_POWERVR_SGX540_120.so!glFinish + 0xa54
0xbec4b1b4: libGLESv2_POWERVR_SGX540_120.so!glFinish + 0x944
0xbec4b1c4: libGLESv2_POWERVR_SGX540_120.so!glGenTextures + 0x190
0xbec4bb1c: libc.so!dlmalloc + 0xae
0xbec4bc14: libc.so!__libc_init + 0x2a
0xbec4bf40: libxul.so!set_rgb_colorants [transform.c:e02eec70a603 : 261 + 0x33]
0xbec4bff4: libxul.so!bits_image_fetch_separable_convolution_affine_normal_a8 [pixman-bits-image.c:e02eec70a603 : 798 + 0x0]

The stuff at the bottom looks totally bogus, so as best I can tell this is just a bunch of crap in the GL driver.
Assignee: ted → nobody
How do you get minidumps from tinderbox runs? The uuid in the log doesn't seem to be a socorro id.
(In reply to Jim Chen [:jchen :nchen] from comment #15)
> How do you get minidumps from tinderbox runs? The uuid in the log doesn't
> seem to be a socorro id.

Crash reporter is disabled on our testing infrastructure; you need to ask #releng to fetch the minidump from the test machine (as soon as possible after the job ran, to ensure it hasn't been purged).
I wasn't able to get anything either from the comment 16 minidump

The crashing thread is not the Gecko thread, so the stack doesn't contain information about Gecko. The Gecko thread is running JS that's AFAIK not related to graphics.

Either a core dump (bug 824760) or full Java stacks (bug 849220) will help the diagnosis.
Depends on: 824760, 849220
Haven't hit this in ages. I'm assuming that whatever caused this has been fixed or morphed.
Status: NEW → RESOLVED
Closed: 8 years ago
Resolution: --- → WORKSFORME
Blocks: 872858
Blocks: 903110
You need to log in before you can comment on or make changes to this bug.