Closed GoogleCodeExporter closed 9 years ago
We will likely need https://bugzilla.mozilla.org/show_bug.cgi?id=997336
Original comment by classi...@floodgap.com
on 13 May 2014 at 2:57
JavaScript passes tests with generational GC (whew!).
Original comment by classi...@floodgap.com
on 22 May 2014 at 3:30
Canvas and thumbnails work.
Original comment by classi...@floodgap.com
on 25 May 2014 at 4:48
Google Maps does paint correctly, but crashes when trying to interact with the
map. Does so even in safe mode, so it's not the JIT.
Assertion failure: contents, at
/Volumes/BruceDeuce/src/mozilla-31a/js/src/vm/ArrayBufferObject.cpp:1055
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_PROTECTION_FAILURE at address: 0x00000000
0x0acab7f0 in JS_NewArrayBufferWithContents (cx=<value temporarily unavailable,
due to optimizations>, nbytes=<value temporarily unavailable, due to
optimizations>, contents=0x0) at
/Volumes/BruceDeuce/src/mozilla-31a/js/src/vm/ArrayBufferObject.cpp:1055
1055 JS_ASSERT(contents);
(gdb) bt 15
#0 0x0acab7f0 in JS_NewArrayBufferWithContents (cx=<value temporarily
unavailable, due to optimizations>, nbytes=<value temporarily unavailable, due
to optimizations>, contents=0x0) at
/Volumes/BruceDeuce/src/mozilla-31a/js/src/vm/ArrayBufferObject.cpp:1055
#1 0x0ad96ba0 in JSStructuredCloneReader::readTransferMap (this=0xefffce54) at
/Volumes/BruceDeuce/src/mozilla-31a/js/src/vm/StructuredClone.cpp:1559
#2 0x0ad98214 in JSStructuredCloneReader::read (this=0xefffce54,
vp=0xefffd068) at
/Volumes/BruceDeuce/src/mozilla-31a/js/src/vm/StructuredClone.cpp:1605
#3 0x0ad98774 in ReadStructuredClone (cx=<value temporarily unavailable, due
to optimizations>, data=<value temporarily unavailable, due to optimizations>,
nbytes=<value temporarily unavailable, due to optimizations>,
vp={<js::MutableHandleBase<JS::Value>> =
{<js::MutableValueOperations<JS::MutableHandle<JS::Value> >> =
{<js::ValueOperations<JS::MutableHandle<JS::Value> >> = {<No data fields>}, <No
data fields>}, <No data fields>}, ptr = 0xefffd068}, cb=0xb7f0f2c,
cbClosure=0x0) at
/Volumes/BruceDeuce/src/mozilla-31a/js/src/vm/StructuredClone.cpp:350
#4 0x0ad98ba0 in JSAutoStructuredCloneBuffer::read (this=<value temporarily
unavailable, due to optimizations>, cx=0xefffcee4,
vp={<js::MutableHandleBase<JS::Value>> =
{<js::MutableValueOperations<JS::MutableHandle<JS::Value> >> =
{<js::ValueOperations<JS::MutableHandle<JS::Value> >> = {<No data fields>}, <No
data fields>}, <No data fields>}, ptr = 0xefffd068}, optionalCallbacks=<value
temporarily unavailable, due to optimizations>, closure=0x0) at
/Volumes/BruceDeuce/src/mozilla-31a/js/src/vm/StructuredClone.cpp:1651
#5 0x08ecb25c in DispatchDOMEvent (this=0x29971ad0, aCx=0x27621300,
aTarget=0x283be800, aIsMainThread=<value temporarily unavailable, due to
optimizations>) at
/Volumes/BruceDeuce/src/mozilla-31a/dom/workers/WorkerPrivate.cpp:986
#6 0x08ecee48 in WorkerRun (this=0xefffd044, aCx=0x27621300,
aWorkerPrivate=0x283be800) at
/Volumes/BruceDeuce/src/mozilla-31a/dom/workers/WorkerPrivate.cpp:1043
#7 0x08edaabc in mozilla::dom::workers::WorkerRunnable::Run (this=0x29971ad0)
at /Volumes/BruceDeuce/src/mozilla-31a/dom/workers/WorkerRunnable.cpp:312
#8 0x075e164c in nsThread::ProcessNextEvent (this=0x41e4f0, mayWait=<value
temporarily unavailable, due to optimizations>, result=0xefffd2b8) at
/Volumes/BruceDeuce/src/mozilla-31a/xpcom/threads/nsThread.cpp:715
#9 0x075146d0 in NS_ProcessPendingEvents (thread=0x41e4f0, timeout=20) at
/Volumes/BruceDeuce/src/mozilla-31a/xpcom/glue/nsThreadUtils.cpp:210
#10 0x08927f6c in nsBaseAppShell::NativeEventCallback (this=0x48e6c0) at
/Volumes/BruceDeuce/src/mozilla-31a/widget/xpwidgets/nsBaseAppShell.cpp:98
#11 0x088ba028 in _ZN10nsAppShell18ProcessGeckoEventsEPv (aInfo=0x48e6c0) at
/Volumes/BruceDeuce/src/mozilla-31a/widget/cocoa/nsAppShell.mm:409
#12 0x907df300 in __CFRunLoopDoSources0 ()
#13 0x907de830 in __CFRunLoopRun ()
#14 0x907de2b0 in CFRunLoopRunSpecific ()
(More stack frames follow...)
Original comment by classi...@floodgap.com
on 25 May 2014 at 5:09
Keystrokes documented in
https://support.mozilla.org/en-US/kb/keyboard-shortcuts-perform-firefox-tasks-qu
ickly work.
Original comment by classi...@floodgap.com
on 25 May 2014 at 5:21
Audio still busted on web cam.
Original comment by classi...@floodgap.com
on 25 May 2014 at 5:24
The MTE works fine. The QTE hangs and does not display the Going to dialogue
box.
Original comment by classi...@floodgap.com
on 25 May 2014 at 5:32
Printing works.
Editing keys have regressed from 24 definitely.
Other than that, most of the browser seems to work well.
So, worklist:
- Google Maps crash (endian???)
- Fix keystrokes
- Fix webcam
- Fix QTE, either with new QTE version or figure out problem (latter preferred)
Original comment by classi...@floodgap.com
on 25 May 2014 at 5:34
The jetpack version in QTE 119 is too old. Updating it to the one we're using
for the MTE works. After 31 sticks, we'll update both of them, but this
suffices for the transition.
Now: Google Maps crash
Original comment by classi...@floodgap.com
on 26 May 2014 at 12:40
Oh hai another endian bug in Firefox
https://bugzilla.mozilla.org/show_bug.cgi?id=1015824
Fixed!
Original comment by classi...@floodgap.com
on 26 May 2014 at 3:52
The browser is even faster with OMTC disabled. Need to see if that's the WebM
regression.
Original comment by classi...@floodgap.com
on 26 May 2014 at 7:58
Accumulated settings changes to make: roll back timeslice to 30ms, disable
OMTC, what else?
Original comment by classi...@floodgap.com
on 26 May 2014 at 8:07
[deleted comment]
Working on the keys problem. NSPR logging of left, right, Cmd-Left, Cmd-Right
in a textbox:
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=123,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView keyUp:
keycode=123,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=124,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView keyUp:
keycode=124,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=123,modifiers=b00108,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=124,modifiers=b00108,chars=.,charsIgnoringModifiers=.
Not sure if the loss of keyUps is the problem, but the modifiers and keycodes
look basically correct ... ?
Original comment by classi...@floodgap.com
on 26 May 2014 at 11:49
Some other data points:
Cmd-A
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=0,modifiers=100108,chars=a,charsIgnoringModifiers=a
Notice the modifiers!
Cmd-C:
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=8,modifiers=100108,chars=c,charsIgnoringModifiers=c
-1610551928[409b00]: writing out clipboard data of type text/unicode (0)
Cmd-Opt-T:
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=17,modifiers=180128,chars=.,charsIgnoringModifiers=t
Delete, Opt-Delete:
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=51,modifiers=100,chars=,charsIgnoringModifiers=
-1610551928[409b00]: ChildView keyUp:
keycode=51,modifiers=100,chars=,charsIgnoringModifiers=
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=51,modifiers=80120,chars=,charsIgnoringModifiers=
-1610551928[409b00]: ChildView keyUp:
keycode=51,modifiers=80120,chars=,charsIgnoringModifiers=
Original comment by classi...@floodgap.com
on 26 May 2014 at 11:58
Cmd-Up/Down, Up/Down:
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=126,modifiers=b00108,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=125,modifiers=b00108,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=126,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView keyUp:
keycode=126,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=125,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView keyUp:
keycode=125,modifiers=a00100,chars=.,charsIgnoringModifiers=.
Original comment by classi...@floodgap.com
on 26 May 2014 at 11:59
Cmd-1,2,3 (these work);
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=18,modifiers=100108,chars=1,charsIgnoringModifiers=1
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=19,modifiers=100108,chars=2,charsIgnoringModifiers=2
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=20,modifiers=100108,chars=3,charsIgnoringModifiers=3
Original comment by classi...@floodgap.com
on 27 May 2014 at 12:00
Option-left/right DOES WORK:
(l/r, then o-l/r)
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=123,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView keyUp:
keycode=123,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=124,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView keyUp:
keycode=124,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=123,modifiers=a80120,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView keyUp:
keycode=123,modifiers=a80120,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=124,modifiers=a80120,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView keyUp:
keycode=124,modifiers=a80120,chars=.,charsIgnoringModifiers=.
Original comment by classi...@floodgap.com
on 27 May 2014 at 12:06
I think the problem is `[nsChildView convertCocoaKeyEvent]`.
Original comment by classi...@floodgap.com
on 27 May 2014 at 12:08
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=123,modifiers=a00100,chars=.,charsIgnoringModifiers=.
editor commanded = cmd_charPrevious
-1610551928[409b00]: ChildView keyUp:
keycode=123,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=124,modifiers=a00100,chars=.,charsIgnoringModifiers=.
editor commanded = cmd_charNext
-1610551928[409b00]: ChildView keyUp:
keycode=124,modifiers=a00100,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=123,modifiers=b00108,chars=.,charsIgnoringModifiers=.
-1610551928[409b00]: ChildView processKeyDownEvent:
keycode=124,modifiers=b00108,chars=.,charsIgnoringModifiers=.
Original comment by classi...@floodgap.com
on 27 May 2014 at 12:49
It is, but indirectly. Bug 282097 introduced Cocoa keybindings and
decommissioned the old XBL ones. Our 10.4-compatible code didn't know about
this and did not give the WidgetKeyboardEvent a copy of the NSEvent key down
event, so NativeKeyBindings.mm didn't get an event to process, and only the
default residual XBL bindings took effect. Added to convertCocoaKeyEvent and
now they work.
This might be the solution for the double keys in Classilla, btw (the XBL
bindings).
Original comment by classi...@floodgap.com
on 27 May 2014 at 2:56
Last problem, the webcam.
When checking RecordingIsAvailable,
MEMORY ; (21: 7:44:218 | 0) AUDIO DEVICE: 1 99; 52231;
AudioDeviceBuffer::SetRecordingSampleRate(fsHz=48000)
MEMORY ; (21: 7:44:218 | 0) AUDIO DEVICE: 1 99; 52231;
AudioDeviceBuffer::SetRecordingChannels(channels=2)
STATEINFO ; (21: 7:44:219 | 1) UTILITY: -1; 61383; Thread
with name:CaptureWorkerThread started
DEBUG ; (21: 7:44:241 | 22) AUDIO DEVICE: 1 99; 87555;
Recording device stopped
DEBUG ; (21: 7:44:241 | 0) AUDIO DEVICE: 1 99; 52231;
Recording stopped
STATEINFO ; (21: 7:44:259 | 18) UTILITY: -1; 61383; Thread
with name:CaptureWorkerThread stopped
STATEINFO ; (21: 7:44:261 | 2) AUDIO DEVICE: 1 99; 52231; output:
available=1
STATEINFO ; (21: 7:44:261 | 0) VOICE: 1 99; 52231;
Output: isAvailable = 1)
the CaptureWorkerThread does not seem to get restarted.
Original comment by classi...@floodgap.com
on 27 May 2014 at 3:02
Actually, it does ...
MEMORY ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799;
InitRecording
DEBUGINFO ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799; Input
device: Apple iSight
DEBUGINFO ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799;
AudioMixerManagerMac::OpenMicrophone(id=258)
DEBUGINFO ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799; No
process has hogged the input device
MEMORY ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799;
SpeakerIsInitialized
DEBUGINFO ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799; Input
stream format:
DEBUGINFO ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799;
mSampleRate = 48000.000000, mChannelsPerFrame = 1
DEBUGINFO ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799;
mBytesPerPacket = 4, mFramesPerPacket = 1
DEBUGINFO ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799;
mBytesPerFrame = 4, mBitsPerChannel = 32
DEBUGINFO ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799;
mFormatFlags = 11
DEBUGINFO ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799;
mFormatID: lpcm
DEBUGINFO ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799; Stereo
recording unavailable on this device
MEMORY ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799;
AudioDeviceBuffer::SetRecordingSampleRate(fsHz=48000)
MEMORY ; (21: 8:31:597 | 0) AUDIO DEVICE: 1 99; 50799;
AudioDeviceBuffer::SetRecordingChannels(channels=1)
STATEINFO ; (21: 8:31:598 | 1) UTILITY: -1; 98343; Thread
with name:CaptureWorkerThread started
DEBUGINFO ; (21: 8:31:601 | 3) VOICE: 1 1; 50799;
Channel::StartSend()
DEBUGINFO ; (21: 8:31:601 | 0) VOICE: 1 1; 50799;
Channel::SetInitSequenceNumber()
MODULECALL; (21: 8:31:601 | 0) RTP/RTCP: 1 1; 50799;
SetSequenceNumber(4222)
MODULECALL; (21: 8:31:601 | 0) RTP/RTCP: 1 1; 50799;
SetSendingStatus(sending)
MODULECALL; (21: 8:31:601 | 0) RTP/RTCP: 1 1; 50799;
CurrentSendFrequencyHz()
MODULECALL; (21: 8:31:601 | 0) RTP/RTCP: 1 1; 50799;
PacketCountSent()
MODULECALL; (21: 8:31:601 | 0) RTP/RTCP: 1 1; 50799;
ByteCountSent()
APICALL ; (21: 8:31:601 | 4) VOICE: 1 99; 50799;
RegisterExternalMediaProcessing(channel=1, type=2, processObject=0x260cc1e8)
DEBUGINFO ; (21: 8:31:601 | 0) VOICE: 1 1; 50799;
Channel::RegisterExternalMediaProcessing()
APICALL ; (21: 8:31:601 | 0) VOICE: 1 99; 50799;
SetAgcStatus(enable=0, mode=1)
APICALL ; (21: 8:31:601 | 0) VOICE: 1 99; 50799;
SetNsStatus(enable=0, mode=1)
STREAM ; (21: 8:31:608 | 7) VOICE: 1 99; 53359;
TransmitMixer::OnPeriodicProcess()
DEBUG ; (21: 8:31:614 | 6) AUDIO DEVICE: 1 99; 98623;
AudioDeviceMac::implObjectListenerProc()
DEBUG ; (21: 8:31:680 | 66) AUDIO DEVICE: 1 99; 98623;
AudioDeviceMac::implObjectListenerProc()
MODULECALL; (21: 8:31:702 | 22) RTP/RTCP: 1 1; 53359;
CurrentSendFrequencyHz()
MODULECALL; (21: 8:31:702 | 0) RTP/RTCP: 1 1; 53359;
PacketCountSent()
MODULECALL; (21: 8:31:702 | 0) RTP/RTCP: 1 1; 53359;
ByteCountSent()
STREAM ; (21: 8:31:702 | 0) VOICE: 1 1; 53359;
Channel::SendRTCPPacket(channel=1, len=40)
MODULECALL; (21: 8:32:295 | 0) RTP/RTCP: 1 0; 53359;
CurrentSendFrequencyHz()
MODULECALL; (21: 8:32:295 | 0) RTP/RTCP: 1 0; 53359;
PacketCountSent()
MODULECALL; (21: 8:32:295 | 0) RTP/RTCP: 1 0; 53359;
ByteCountSent()
STREAM ; (21: 8:32:295 | 0) VOICE: 1 0; 53359;
Channel::SendRTCPPacket(channel=0, len=8)
STREAM ; (21: 8:32:608 | 313) VOICE: 1 99; 53359;
TransmitMixer::OnPeriodicProcess()
STREAM ; (21: 8:33:608 | 0) VOICE: 1 99; 53359;
TransmitMixer::OnPeriodicProcess()
MODULECALL; (21: 8:34:464 | 0) RTP/RTCP: 1 2; 53359;
CurrentSendFrequencyHz()
MODULECALL; (21: 8:34:465 | 1) RTP/RTCP: 1 2; 53359;
PacketCountSent()
MODULECALL; (21: 8:34:465 | 0) RTP/RTCP: 1 2; 53359;
ByteCountSent()
STREAM ; (21: 8:34:465 | 0) VOICE: 1 2; 53359;
Channel::SendRTCPPacket(channel=2, len=8)
STREAM ; (21: 8:34:608 | 143) VOICE: 1 99; 53359;
TransmitMixer::OnPeriodicProcess()
MODULECALL; (21: 8:35:540 | 0) RTP/RTCP: 1 0; 53359;
CurrentSendFrequencyHz()
MODULECALL; (21: 8:35:540 | 0) RTP/RTCP: 1 0; 53359;
PacketCountSent()
MODULECALL; (21: 8:35:540 | 0) RTP/RTCP: 1 0; 53359;
ByteCountSent()
STREAM ; (21: 8:35:540 | 0) VOICE: 1 0; 53359;
Channel::SendRTCPPacket(channel=0, len=8)
STREAM ; (21: 8:35:608 | 68) VOICE: 1 99; 53359;
TransmitMixer::OnPeriodicProcess()
APICALL ; (21: 8:36:209 | 0) VOICE: 1 99; 50799;
DeRegisterExternalMediaProcessing(channel=1)
DEBUGINFO ; (21: 8:36:209 | 0) VOICE: 1 1; 50799;
Channel::DeRegisterExternalMediaProcessing()
APICALL ; (21: 8:36:209 | 0) VOICE: 1 99; 50799;
StopSend(channel=1)
DEBUGINFO ; (21: 8:36:209 | 0) VOICE: 1 1; 50799;
Channel::StopSend()
MODULECALL; (21: 8:36:210 | 1) RTP/RTCP: 1 1; 50799;
SequenceNumber()
MODULECALL; (21: 8:36:210 | 0) RTP/RTCP: 1 1; 50799;
SetSendingStatus(stopped)
MODULECALL; (21: 8:36:210 | 0) RTP/RTCP: 1 1; 50799;
CurrentSendFrequencyHz()
MODULECALL; (21: 8:36:210 | 0) RTP/RTCP: 1 1; 50799;
PacketCountSent()
MODULECALL; (21: 8:36:210 | 0) RTP/RTCP: 1 1; 50799;
ByteCountSent()
STREAM ; (21: 8:36:210 | 0) VOICE: 1 1; 50799;
Channel::SendRTCPPacket(channel=1, len=16)
MODULECALL; (21: 8:36:210 | 0) RTP/RTCP: 1 1; 50799;
ResetSendDataCountersRTP()
DEBUGINFO ; (21: 8:36:210 | 0) VOICE: 1 99; 50799;
VoEBaseImpl::StopSend()
DEBUG ; (21: 8:36:210 | 0) AUDIO DEVICE: 1 99; 98623;
Recording device stopped
DEBUG ; (21: 8:36:210 | 0) AUDIO DEVICE: 1 99; 50799;
Recording stopped
Original comment by classi...@floodgap.com
on 28 May 2014 at 3:31
RecordedDataIsAvailable never gets called. That implies the worker thread is
not calling it. I suspect the worker thread isn't calling
DeliverRecordedData(). Did it fail this in audio_device_mac.cc?
// TODO(xians): what if the returned size is incorrect?
if (size == ENGINE_REC_BUF_SIZE_IN_SAMPLES)
{
Original comment by classi...@floodgap.com
on 28 May 2014 at 3:54
Key check routine was the cause.
Graphics really have regressed overall. OMTC eventually does more harm than
good. Disabled multithreaded image decoding (probably wasn't even running) and
OMTC, tab animate, fade in animate.
We will need https://bugzilla.mozilla.org/show_bug.cgi?id=1008793
Original comment by classi...@floodgap.com
on 13 Jun 2014 at 5:07
WFM
Original comment by classi...@floodgap.com
on 20 Jun 2014 at 9:53
Original issue reported on code.google.com by
classi...@floodgap.com
on 11 May 2014 at 12:25