a6f59e760b
Bug 1062661 - Part 1: Discard decoders that don't have any data in them. r=karlt
2014-11-04 11:16:34 +13:00
0f909f5b5b
Bug 1062661 - Part 1: Fix ResourceQueue::Evict so that it tries to remove enough data to get us under the threshold. r=kinetik
2014-11-04 11:16:31 +13:00
a3fba57a19
Bug 1085186. Don't readahead more than 30s of media data per resource on mobile/B2G. r=cpearce
...
--HG--
extra : rebase_source : 6fce6b5390d0eb76463db04f7dd72f79b97d39c5
2014-10-21 10:32:11 +13:00
6f28d2bb7c
Bug 1021682 - Fix and re-enable test_streams_gc.html on B2G. r=cpearce
2014-11-04 18:11:00 -05:00
ab3a404727
Backed out 10 changesets (bug 1091008, bug 1093020, bug 1063323) for windows m2 permanent test failures on a CLOSED TREE
...
Backed out changeset 21ddb8a58fea (bug 1093020)
Backed out changeset fe9e11333c7d (bug 1093020)
Backed out changeset bba774c54652 (bug 1063323)
Backed out changeset 16f58d7e1e17 (bug 1091008)
Backed out changeset 649bfc6dad4d (bug 1091008)
Backed out changeset 6f270b2d90f4 (bug 1091008)
Backed out changeset 966093bbc26a (bug 1091008)
Backed out changeset 9de4746aa59a (bug 1091008)
Backed out changeset 856016c0118a (bug 1091008)
Backed out changeset 8aaa10a8d956 (bug 1091008)
2014-11-05 12:57:43 +01:00
b163c632f5
Bug 1088130 - Fixed mingw cross compilation.
2014-11-05 11:29:47 +01:00
420dd07ad2
Bug 1093020 - Tests. r=karlt
2014-11-05 10:09:00 +01:00
28d2d55247
Bug 1093020 - Don't use a buffering wait for MSE. r=cpearce
2014-11-05 10:09:00 +01:00
a170ee7065
Bug 1063323 - Tests. r=cajbir
2014-11-05 10:08:59 +01:00
d89289ac96
Bug 1091008 - Reimplement HasLowUndecodedData in terms of GetBuffered. r=cpearce
...
Since GetBuffered now has a sane implementation for MSE, this should
make this function sane for MSE as well.
2014-11-05 10:08:59 +01:00
51b2c702bb
Bug 1091008 - Remove special handling for MSE in HTMLMediaElement::Buffered. r=cajbir
...
The same code will now be reached by invoking this method on the decoder.
2014-11-05 10:08:59 +01:00
1cd2b803c6
Bug 1091008 - Implement a sensible GetBuffered override for MediaSourceReader using the existing GetBuffered on MediaSource. r=cajbir
2014-11-05 10:08:59 +01:00
5a2f54ca85
Bug 1091008 - Hoist stream pinning into the MediaDecoderReaders and make MediaDecoderStateMachine::GetBuffered just forward to them. r=cpearce
...
Whether or not we ping a stream depends a lot on what kind of decoder we're
dealing with. In particular, it doesn't really make sense for MSE.
2014-11-05 10:08:58 +01:00
461d13b4d5
Bug 1091008 - Remove the aStartTime argument from MediaDecoderReader::GetBuffered. r=cpearce
...
We now have this stashed on the superclass.
2014-11-05 10:08:58 +01:00
ccff922922
Bug 1091008 - Teach MediaDecoderReader about its start time. r=cpearce
...
Duplicating state is never great, but this lets the reader make calculations
using this immutable state variable without involving the state machine. We
could alternatively punch a hole from MediaDecoderReader to
MediaDecoderStateMachine and access it there, but that would create tighter
coupling, and weird relationships for MSE.
2014-11-05 10:08:58 +01:00
9939d4bbd6
Bug 1091008 - Factor out IsDataCachedToEndOfResource() into a higher-level question and answer it sensibly for MSE. r=cpearce
2014-11-05 10:08:57 +01:00
af0f511f43
Bug 1094049 - Make GonkDecoderModule use currect display height; r=bwu
2014-11-05 16:18:32 +08:00
4cc292fee5
Bug 1090523 - Add more logging to help debug infrequent EME/MSE failure. r=edwin
2014-11-05 15:11:44 +13:00
8cbaa94377
Bug 1093399. Fire UpdateReadyStateForData every time we get new data. r=cpearce
...
--HG--
extra : rebase_source : e243ecb8a72615cdf697228f30e631144cde129d
2014-11-04 14:19:14 +13:00
9d4b88f360
Bug 1078017 - Deregister track callbacks before unlinking tracks. r=jesup,jib
2014-11-02 00:56:10 +01:00
bbae2a8da1
bug 1092932 detach MediaSource before initiating MDStateMachine shutdown r=bholley
...
to avoid races between the detach process and decoder task queue shutdown.
--HG--
extra : rebase_source : 91761d9594c53a7ccb974caf4931bcac90533e5a
2014-11-04 17:24:52 +13:00
4ef94165c9
Bug 966247 - Make MediaElementAudioSourceNode take the HTMLMediaElement volume into account. r=padenot
...
--HG--
extra : rebase_source : 6ba3d01780be44c4164bd8ed1b69e6171eaa66dd
2014-06-22 14:12:57 +02:00
688c64b71b
Backout 58e2cab9572e, bad patch.
2014-11-04 12:35:02 +01:00
baa4f7ed4e
Bug 1089049 - crash in nsContentUtils::CanCallerAccess(nsINode*), Browser crashes if contextNode is null of document.evaluate. r=bz.
...
--HG--
extra : rebase_source : 00d2454f8de660a6432d30324529df02b52dd618
2014-11-04 10:20:08 +01:00
a369313d4b
Backed out changeset 221ca54f4df0 (bug 1092096) for m2 test failures
2014-11-04 10:32:32 +01:00
b8014abf5b
Bug 1088130 - Change for loop in GMP OP gtest so that it builds with VisualStudio 2010. r=bobowen
2014-11-03 22:40:00 +00:00
005b05de5c
Bug 1092096 - Separate MSE prefs for MP4 and WebM; r=cajbir
2014-11-04 15:22:26 +08:00
9ac0482856
Bug 1083347 - Removing entries from android23.json and annotating manifests. r=gbrown
2014-11-03 12:01:51 -08:00
de12f8c051
Bug 1088130: Add to TestGMPCrossOrigin.cpp gtest to test that the APIs required for Output Protection on Windows work within the sandbox. r=cpearce
2014-11-03 10:02:30 +00:00
1fb284960f
Bug 1092130 - Fixed build errors found by mingw after bug 1076698 landed (-Werror=missing-braces errors). r=cpearce
2014-11-03 10:28:52 +01:00
efd90d884a
Bug 1090991 - Introduce a new NotDecodedReason WAITING_FOR_DATA and use it for MSE. r=cpearce,r=cajbir
2014-11-03 09:20:15 +01:00
a757b63f7e
Bug 1090991 - Unify MediaDecoderStateMachine::On{DecodeError,AudioEOS,VideoEOS} and eliminate duplicated logic. r=cpearce
...
We take this opportunity to align the behavior of Finish() calls between audio
and video EOS, invoking them unconditionally for both cases. Currently both
cases always call Finish() immediately, with the exception of:
(A) Video in seeking mode, where we may push mFirstVideoFrameAfterSeek before
doing so, and
(B) Video in the |default:| case.
Push() and Finish() seem like orthogonal operations on MediaQueue, but we
nonetheless preserve the old order just in case. There doesn't seem to be a good
reason for (B).
2014-11-03 09:20:14 +01:00
a215c61b6a
Bug 1090991 - Refactor RequestSampleCallback to use a single callback for all "not decoded" message. r=cpearce,r=cajbir
...
I'm going to add another one, and want this API to scale better than it does.
2014-11-03 09:20:14 +01:00
a6e295e57c
Bug 1090991 - Modify MediaData::Type so that it may serve as a general-purpose enum for distinguishing audio and video. r=cpearce
...
We don't seem to have something like this already, and this seemed better than
introducing yet another new enum.
2014-11-03 09:20:14 +01:00
eaff88ccea
Bug 1079695 - Part 3 - fix const-correctness of MediaDecoderStateMachine. r=kinetik
2014-10-30 21:12:00 +01:00
4d39082ebf
Bug 1079695 - Part 2 - remove audio clock wordaround from MediaDecoderStateMachine. r=kinetik.
2014-11-02 18:06:00 +01:00
b170447565
Bug 1079695 - Part 1 - ensure AudioSink::GetPosition() is mono-increasing in spite of AudioStream error. r=kinetik
2014-10-30 21:10:00 +01:00
ffcd9af8d5
Bug 1021794 - Fix and re-enable test_played.html. r=cpearce
2014-10-30 20:51:00 +01:00
f67d2db3b3
Bug 1073805 - Apply ADTS header for Gonk. r=cpearce.
2014-10-30 17:01:00 -07:00
e4de5b19f4
Bug 1073805 - Add ADTS header for FFmpeg. r=cpearce.
...
Keep a reference to the AudioDecoderConfig so we can pass the
appropriate values to the Adts method from our Input method.
2014-10-30 16:26:00 -07:00
ba2b4afbf2
Bug 1073805 - Apply ADTS header for Android. r=cpearce.
2014-10-30 17:07:00 -07:00
f16eb3b79f
Bug 1073805 - Apply ADTS headers to AAC on MacOS. r=cpearce
...
Our AudioFileStream parser needs the headers to find the sample
boundaries.
2014-10-30 15:47:00 -07:00
6a90558a75
Bug 1073805 - Pass raw AAC to Windows PlatformDecoder. r=kinetik
...
This avoids problems creating an ADTS header for newer HE-AAC profiles.
2014-10-30 15:45:00 -07:00
7ab137b8fd
Bug 1092860 - Fix GMPVideoFrameType and GMPSessionType serialization. r=jesup
2014-11-03 13:53:01 +13:00
bc6d9cfe16
Bug 1091436 - Enable test_chaining.html on B2G. r=cpearce
2014-10-30 06:37:00 +01:00
91aaa3e40b
Bug 1089159 - Correctly use MediaCodec's audio output format r=cpearce
...
--HG--
extra : rebase_source : 2b3b3a43e54cdcdac87e303a2705e8be7c60d3dd
2014-10-29 18:40:06 -05:00
d010e3f711
Bug 1090900 - Issue SimpleTest.finish() even if test failed. r=padenot
2014-10-30 12:15:58 +11:00
053e9ca254
Bug 1090983 - Invoke UpdateReadyState() after invoking SetState(). r=cpearce
...
Otherwise, UpdateReadyState() will think that nothing has changed.
2014-10-30 09:16:02 +01:00
47c853314f
Bug 1077148 part 4 - Add and use new moz.build templates for Gecko programs and libraries. r=gps
...
There are, sadly, many combinations of linkage in use throughout the tree.
The main differentiator, though, is between program/libraries related to
Gecko or not. Kind of. Some need mozglue, some don't. Some need dependent
linkage, some standalone.
Anyways, these new templates remove the need to manually define the
right dependencies against xpcomglue, nspr, mozalloc and mozglue
in most cases.
Places that build programs and were resetting MOZ_GLUE_PROGRAM_LDFLAGS
or that build libraries and were resetting MOZ_GLUE_LDFLAGS can now
just not use those Gecko-specific templates.
2014-10-30 13:06:12 +09:00
6440005c47
Bug 1071482 - Add more logging to help debug rare intermittent failures. r=edwin
2014-10-30 15:05:49 +13:00
Matt Woodrow
Robert O'Callahan
JW Wang
Carsten "Tomcat" Book
Jacek Caban
Bobby Holley
Anthony Jones
Chris Pearce
Andreas Pehrson
Karl Tomlinson
Cgg
Peter Van der Beken
Vaibhav Agrawal
Bob Owen
Ralph Giles
James Willcox
Jean-Yves Avenard
Mike Hommey