external/webrtc.git
5 hours agoMake bands vector in SplittingFilter Analysis const master
aluebs@webrtc.org [Fri, 28 Nov 2014 00:26:27 +0000 (00:26 +0000)]
Make bands vector in SplittingFilter Analysis const

BUG=webrtc:3146
R=kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30229004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7761 4adac7df-926f-26a2-2b94-8c16560cd09d

5 hours agoMove ChannelBuffer class to channel_buffer file
aluebs@webrtc.org [Thu, 27 Nov 2014 23:40:25 +0000 (23:40 +0000)]
Move ChannelBuffer class to channel_buffer file

No change in functionallity.

BUG=webrtc:3146
R=andrew@webrtc.org, bjornv@webrtc.org, kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/28109004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7760 4adac7df-926f-26a2-2b94-8c16560cd09d

15 hours agoRemove unused RtpStatistics struct.
pbos@webrtc.org [Thu, 27 Nov 2014 13:48:35 +0000 (13:48 +0000)]
Remove unused RtpStatistics struct.

This unused struct is basically a copy of RtcpStatistics in
webrtc/common_types.h. I expect this wasn't properly removed when that
one was added.

R=tommi@webrtc.org
BUG=

Review URL: https://webrtc-codereview.appspot.com/25239004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7758 4adac7df-926f-26a2-2b94-8c16560cd09d

18 hours agoRoll chromium_revision d8c9041..309cf65
kjellander@webrtc.org [Thu, 27 Nov 2014 10:41:04 +0000 (10:41 +0000)]
Roll chromium_revision d8c9041..309cf65

Relevant changes:
* testing/gtest 4650552..8245545
* testing/gmock 896ba0e..2976396
* third_party/boringssl 2f3ba91..69a0160
* third_party/icu: 6242e2f..dd72764
* third_party/libyuv: 5a09c3e..d204db6
* tools/gyp: b13d8f2..0a381c0

Details: https://chromium.googlesource.com/chromium/src/+/d8c9041..309cf65/DEPS

Clang version was not updated in this roll.

R=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/25219004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7757 4adac7df-926f-26a2-2b94-8c16560cd09d

21 hours agoAdd receive bitrates to histogram stats:
asapersson@webrtc.org [Thu, 27 Nov 2014 07:38:56 +0000 (07:38 +0000)]
Add receive bitrates to histogram stats:
- total bitrate ("WebRTC.Video.BitrateReceivedInKbps")
- media bitrate ("WebRTC.Video.MediaBitrateReceivedInKbps")
- rtx bitrate ("WebRTC.Video.RtxBitrateReceivedInKbps")
- padding bitrate ("WebRTC.Video.PaddingBitrateReceivedInKbps")

BUG=crbug/419657
R=mflodman@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/27189005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7756 4adac7df-926f-26a2-2b94-8c16560cd09d

28 hours agoRefactor iOS AppRTC parsing code.
tkchin@webrtc.org [Thu, 27 Nov 2014 00:52:38 +0000 (00:52 +0000)]
Refactor iOS AppRTC parsing code.

Moved parsing code to JSON categories for the relevant objects.
No longer prefer ISAC as audio codec.

BUG=
R=glaznev@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/31989005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7755 4adac7df-926f-26a2-2b94-8c16560cd09d

33 hours agoImplement 3 band splitting filter bank by upsampling and splitting twice into 2 bands
aluebs@webrtc.org [Wed, 26 Nov 2014 20:21:38 +0000 (20:21 +0000)]
Implement 3 band splitting filter bank by upsampling and splitting twice into 2 bands

Implemented the 3 bands splitting filter bank by:
1. Upsample by 4/3.
2. Split twice into 2 bands.
3. Discard upper most band, because it is empty anyway.

A unittest was also implemented:
1. Generate a signal from presence or absence of sine waves of different frequencies.
2. Split into 3 bands and check their presence or absence.
3. Recombine the bands.
4. Calculate delay (as it is an IIR it depends on frequency).
5. Check that the cross correlation of input and output is high enough at that delay.

BUG=webrtc:3146
R=andrew@webrtc.org, bjornv@webrtc.org, kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/31029004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7754 4adac7df-926f-26a2-2b94-8c16560cd09d

33 hours agoFix an ASSERT that fires in a browser test for renegotiation.
jiayl@webrtc.org [Wed, 26 Nov 2014 19:58:50 +0000 (19:58 +0000)]
Fix an ASSERT that fires in a browser test for renegotiation.
See https://code.google.com/p/chromium/issues/detail?id=293125#c33

BUG=crbug/293125
R=juberti@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/32429004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7753 4adac7df-926f-26a2-2b94-8c16560cd09d

34 hours agoRevert 7750 "Don't reset sequence number for a stream on deactiv..."
sprang@webrtc.org [Wed, 26 Nov 2014 19:33:15 +0000 (19:33 +0000)]
Revert 7750 "Don't reset sequence number for a stream on deactiv..."

> Don't reset sequence number for a stream on deactivate/reactivate.
>
> BUG=chromium:431908
> R=pbos@webrtc.org
>
> Review URL: https://webrtc-codereview.appspot.com/32199004

TBR=sprang@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/29099004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7752 4adac7df-926f-26a2-2b94-8c16560cd09d

36 hours agoEnabling building with NEON on ARM64
andrew@webrtc.org [Wed, 26 Nov 2014 17:01:40 +0000 (17:01 +0000)]
Enabling building with NEON on ARM64

This patch enables NEON on ARM64 platform. Passed building both on
Android ARMv7 and Android ARM64.

BUG=3580
R=andrew@webrtc.org, jridges@masque.com

Review URL: https://webrtc-codereview.appspot.com/25069004

Patch from Zhongwei Yao <zhongwei.yao@arm.com>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7751 4adac7df-926f-26a2-2b94-8c16560cd09d

36 hours agoDon't reset sequence number for a stream on deactivate/reactivate.
sprang@webrtc.org [Wed, 26 Nov 2014 16:55:52 +0000 (16:55 +0000)]
Don't reset sequence number for a stream on deactivate/reactivate.

BUG=chromium:431908
R=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/32199004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7750 4adac7df-926f-26a2-2b94-8c16560cd09d

37 hours agoRename RtpFileReader::Packet to RtpPacket and move out of RtpFileReader
henrik.lundin@webrtc.org [Wed, 26 Nov 2014 15:50:30 +0000 (15:50 +0000)]
Rename RtpFileReader::Packet to RtpPacket and move out of RtpFileReader

This is in preparation for creating a new class RtpFileWriter which
will use the same RtpPacket struct.

R=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/33429004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7749 4adac7df-926f-26a2-2b94-8c16560cd09d

46 hours agoRevert "Revert "This adds an Android apk for running tests on the Java layer of...
perkj@webrtc.org [Wed, 26 Nov 2014 07:35:37 +0000 (07:35 +0000)]
Revert "Revert "This adds an Android apk for running tests on  the Java layer of PeerConnection.""

This reverts commit 308e7ff61327d64ba5c7761ce6b58cd1fbc4847e.

Original cl description:

This adds an Android apk for running tests on the Java layer of PeerConnection.

The only testcase is currently the same test we run on Java standalone.
To run the test adb shell am instrument -w org.webrtc.test/android.test.InstrumentationTestRunner

BUG=4031
R=kjellander@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/32529004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7748 4adac7df-926f-26a2-2b94-8c16560cd09d

2 days agoAdd video encoder fps and bitrate statistics to
glaznev@webrtc.org [Wed, 26 Nov 2014 00:39:42 +0000 (00:39 +0000)]
Add video encoder fps and bitrate statistics to
Android AppRTCDemo UI.

BUG=4045
R=jiayl@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/25229004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7747 4adac7df-926f-26a2-2b94-8c16560cd09d

2 days agoImplement settable min/start/max bitrates in Call.
pbos@webrtc.org [Tue, 25 Nov 2014 14:03:34 +0000 (14:03 +0000)]
Implement settable min/start/max bitrates in Call.

These parameters are set by the x-google-*-bitrate SDP parameters. This
is implemented on a Call level instead of per-stream like the currently
underlying VideoEngine implementation to allow this refactoring to not
reconfigure the VideoCodec at all but rather adjust bandwidth-estimator
parameters.
Also implements SetMaxSendBandwidth in WebRtcVideoEngine2 as it's a SDP
parameter and allowing it to be dynamically readjusted in Call.

R=mflodman@webrtc.org, stefan@webrtc.org
BUG=1788

Review URL: https://webrtc-codereview.appspot.com/26199004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7746 4adac7df-926f-26a2-2b94-8c16560cd09d

2 days agoAdd back EXPECT_TRUEs.
pbos@webrtc.org [Tue, 25 Nov 2014 11:13:28 +0000 (11:13 +0000)]
Add back EXPECT_TRUEs.

These shouldn't fail, but EXPECT_TRUE gives nicer error messages that
work in Release. These changes got through unreviewed in r7726.

R=stefan@webrtc.org
BUG=

Review URL: https://webrtc-codereview.appspot.com/26249004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7745 4adac7df-926f-26a2-2b94-8c16560cd09d

2 days agoReenable GetStats test.
pbos@webrtc.org [Tue, 25 Nov 2014 09:39:04 +0000 (09:39 +0000)]
Reenable GetStats test.

Also increasing start bitrate to have the test go significantly faster
on average. Hopefully an assert hit in the jitter buffer while running
this test was fixed in r7735.

R=stefan@webrtc.org
BUG=4014

Review URL: https://webrtc-codereview.appspot.com/26239004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7744 4adac7df-926f-26a2-2b94-8c16560cd09d

3 days agoUse mirror image for Android AppRTCDemo local preview.
glaznev@webrtc.org [Mon, 24 Nov 2014 17:31:01 +0000 (17:31 +0000)]
Use mirror image for Android AppRTCDemo local preview.

Similar to Chrome apprtc using mirror image for camera
local preview provides better experience when device
is rotated.

R=jiayl@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/25209004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7741 4adac7df-926f-26a2-2b94-8c16560cd09d

3 days agoAdd wav output capability to neteq_rtpplay
henrik.lundin@webrtc.org [Mon, 24 Nov 2014 14:50:53 +0000 (14:50 +0000)]
Add wav output capability to neteq_rtpplay

This CL makes neteq_rtpplay capable of writing to wav files as well as
pcm files. This is done through the new class OutputWavFile, which
wraps a WavWriter object in an AudioSink interface.

BUG=2692
R=kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/32139004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7740 4adac7df-926f-26a2-2b94-8c16560cd09d

3 days agoAdd new test for VP8 packetizer to test tight partitions
henrik.lundin@webrtc.org [Mon, 24 Nov 2014 12:36:58 +0000 (12:36 +0000)]
Add new test for VP8 packetizer to test tight partitions

It was discovered that if remaining_bytes is an exact multiple of
max_payload_len in RtpPacketizerVp8::CalcNextSize, then the packetizer
will produce too many packets (i.e., split the payload into more
packets than needed).

This CL adds a test to trigger the problem.

BUG=4019
R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/24289004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7739 4adac7df-926f-26a2-2b94-8c16560cd09d

3 days agosync_chromium.py: Check for chromium/src
kjellander@webrtc.org [Mon, 24 Nov 2014 10:08:03 +0000 (10:08 +0000)]
sync_chromium.py: Check for chromium/src

Make sure the script alwyas downloads Chromium
if there's no current download. This case can happen
if a user is removing the 'src' folder but doesn't know
to remove the .last_sync_chromium file.

BUG=
TESTED=Renamed chromium/src and ran a sync keeping the .last_sync_chromium file, verified it started downloading.
TBR=iannucci@chromium.org

Review URL: https://webrtc-codereview.appspot.com/27099004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7738 4adac7df-926f-26a2-2b94-8c16560cd09d

3 days agoPRESUBMIT: Only notify GN changes for GYP files in webrtc/*
kjellander@webrtc.org [Mon, 24 Nov 2014 10:05:37 +0000 (10:05 +0000)]
PRESUBMIT: Only notify GN changes for GYP files in webrtc/*

We don't maintain a BUILD.gn file for talk/ since it's a part
of Chromium:
https://code.google.com/p/chromium/codesearch#chromium/src/third_party/libjingle/BUILD.gn
Because of this, it's confusing to get warnings about updating
a GYP file in talk/ from the PRESUBMIT check.

TESTED=Successsfully ran git cl presubmit with this change
applied on top of a CL containing changes in .gyp files.

R=niklas.enbom@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/27189004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7737 4adac7df-926f-26a2-2b94-8c16560cd09d

3 days agoOWNERS: Remove tomasl@ and mallinath@
kjellander@webrtc.org [Mon, 24 Nov 2014 10:05:05 +0000 (10:05 +0000)]
OWNERS: Remove tomasl@ and mallinath@

mallinath@ has left the team and tomasl@ says he doesn't
know why he's owner in webrtc/test/channel_transport

R=henrika@webrtc.org, perkj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30119004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7736 4adac7df-926f-26a2-2b94-8c16560cd09d

3 days agoSimplifying VideoReceiver and JitterBuffer.
pbos@webrtc.org [Mon, 24 Nov 2014 09:06:48 +0000 (09:06 +0000)]
Simplifying VideoReceiver and JitterBuffer.

Removing frame_buffers_ array and dual-receiver mechanism. Also adding
some thread annotations to VCMJitterBuffer.

R=stefan@webrtc.org
BUG=4014

Review URL: https://webrtc-codereview.appspot.com/27239004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7735 4adac7df-926f-26a2-2b94-8c16560cd09d

3 days agoUse vector of CSRCs for DeliverFrame & SetCSRCs.
pbos@webrtc.org [Mon, 24 Nov 2014 08:25:50 +0000 (08:25 +0000)]
Use vector of CSRCs for DeliverFrame & SetCSRCs.

BUG=
R=pbos@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/28029004

Patch from Changbin Shao <changbin.shao@intel.com>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7734 4adac7df-926f-26a2-2b94-8c16560cd09d

4 days agoRevert "This adds an Android apk for running tests on the Java layer of PeerConnection."
kjellander@webrtc.org [Sun, 23 Nov 2014 21:23:00 +0000 (21:23 +0000)]
Revert "This adds an Android apk for running tests on  the Java layer of PeerConnection."

This reverts r7732

Reason: Breaks compilation on Linux:
[813/818] LINK libjingle_media_unittest
FAILED: cd ../../talk; build/build_jar.sh /usr/lib/jvm/java-7-openjdk-amd64 ../out/Debug/libjingle_peerconnection_test.jar ../out/Debug/obj/talk/libjingle_peerconnection_test_jar.gen app/webrtc/javatests/src:../out/Debug/libjingle_peerconnection.jar:../third_party/junit/junit-4.11.jar app/webrtc/java/testcommon/src/org/webrtc/PeerConnectionTest.java app/webrtc/javatests/src/org/webrtc/PeerConnectionTestJava.java
build/build_jar.sh: Entering directory `/mnt/data/b/build/slave/linux64/build/src/talk'
app/webrtc/java/testcommon/src/org/webrtc/PeerConnectionTest.java:46:warning: [deprecation] Assert in junit.framework has been deprecated
import static junit.framework.Assert.*;
                             ^
app/webrtc/javatests/src/org/webrtc/PeerConnectionTestJava.java:36:error: cannot find symbol
  @Test
   ^
  symbol:   class Test
  location: class PeerConnectionTestJava
app/webrtc/javatests/src/org/webrtc/PeerConnectionTestJava.java:43:error: cannot find symbol
  @Test
   ^
  symbol:   class Test
  location: class PeerConnectionTestJava
2 errors
1 warning
ninja: build stopped: subcommand failed.

TBR=perkj@webrtc.org
BUG=

Review URL: https://webrtc-codereview.appspot.com/32169004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7733 4adac7df-926f-26a2-2b94-8c16560cd09d

4 days agoThis adds an Android apk for running tests on the Java layer of PeerConnection.
perkj@webrtc.org [Sun, 23 Nov 2014 16:00:57 +0000 (16:00 +0000)]
This adds an Android apk for running tests on  the Java layer of PeerConnection.

The only testcase is currently the same test we run on Java standalone.
To run the test adb shell am instrument -w org.webrtc.test/android.test.InstrumentationTestRunner

R=kjellander@webrtc.org, phoglund@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/26219004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7732 4adac7df-926f-26a2-2b94-8c16560cd09d

6 days agoRemove expensive and unnecessary memory alloc for sending black frames on video
thorcarpenter@google.com [Sat, 22 Nov 2014 01:04:26 +0000 (01:04 +0000)]
Remove expensive and unnecessary memory alloc for sending black frames on video
mute.

Remove old crusty is_black_ member var in webrtcvideoengine which was not adding value.

R=henrike@webrtc.org, tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/26229004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7731 4adac7df-926f-26a2-2b94-8c16560cd09d

6 days agoBuild fix for MIPS Android Webview build.
andrew@webrtc.org [Fri, 21 Nov 2014 16:28:32 +0000 (16:28 +0000)]
Build fix for MIPS Android Webview build.

Excluding optimizations to support MIPS32R6 platform for Android Webview build (see also https://code.google.com/p/webrtc/source/detail?r=7580).

R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/32459004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7729 4adac7df-926f-26a2-2b94-8c16560cd09d

6 days agocricket::VideoFrame: Refactor ConvertToRgbBuffer into base class
magjed@webrtc.org [Fri, 21 Nov 2014 10:53:00 +0000 (10:53 +0000)]
cricket::VideoFrame: Refactor ConvertToRgbBuffer into base class

There is also an implementation in Chromium that can be removed if/when this lands:
content/renderer/media/webrtc/webrtc_video_capturer_adapter.cc

R=fbarchard@google.com, pbos@webrtc.org, perkj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/32059004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7728 4adac7df-926f-26a2-2b94-8c16560cd09d

6 days agoUpdate mock_frame_dropper.h to use size_t
kjellander@webrtc.org [Fri, 21 Nov 2014 09:40:57 +0000 (09:40 +0000)]
Update mock_frame_dropper.h to use size_t

This mock was missed in the work of
https://webrtc-codereview.appspot.com/23129004 since the file
is not currently used by any test in this repo.

BUG=chromium:81439
R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/25199004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7727 4adac7df-926f-26a2-2b94-8c16560cd09d

7 days agoUse size_t more consistently for packet/payload lengths.
pkasting@chromium.org [Thu, 20 Nov 2014 22:28:14 +0000 (22:28 +0000)]
Use size_t more consistently for packet/payload lengths.

See design doc at https://docs.google.com/a/chromium.org/document/d/1I6nmE9D_BmCY-IoV6MDPY2V6WYpEI-dg2apWXTfZyUI/edit?usp=sharing for more information.

This CL was reviewed and approved in pieces in the following CLs:
https://webrtc-codereview.appspot.com/24209004/
https://webrtc-codereview.appspot.com/24229004/
https://webrtc-codereview.appspot.com/24259004/
https://webrtc-codereview.appspot.com/25109004/
https://webrtc-codereview.appspot.com/26099004/
https://webrtc-codereview.appspot.com/27069004/
https://webrtc-codereview.appspot.com/27969004/
https://webrtc-codereview.appspot.com/27989004/
https://webrtc-codereview.appspot.com/29009004/
https://webrtc-codereview.appspot.com/30929004/
https://webrtc-codereview.appspot.com/30939004/
https://webrtc-codereview.appspot.com/31999004/
Committing as TBR to the original reviewers.

BUG=chromium:81439
TEST=none
TBR=pthatcher,henrik.lundin,tina.legrand,stefan,tkchin,glaznev,kjellander,perkj,mflodman,henrika,asapersson,niklas.enbom

Review URL: https://webrtc-codereview.appspot.com/23129004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7726 4adac7df-926f-26a2-2b94-8c16560cd09d

7 days agoSupport loopback mode and command line execution
glaznev@webrtc.org [Thu, 20 Nov 2014 21:16:12 +0000 (21:16 +0000)]
Support loopback mode and command line execution
for Android AppRTCDemo when using WebSocket signaling.

- Add loopback support for new signaling. In loopback mode
only room connection is established, WebSocket connection is
not opened and all candidate/sdp messages are automatically
routed back.
- Fix command line support both for channek and new signaling.
Exit from application when room connection is closed and add
an option to run application for certain time period and exit.
- Plus some fixes for WebSocket signaling - support
POST (not used for now) and DELETE request to WebSocket server
and making sure that all available TURN server are used by
peer connection client.

BUG=3995,3937
R=jiayl@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/24339004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7725 4adac7df-926f-26a2-2b94-8c16560cd09d

7 days agoFix problems if first packet into NetEq is rejected
henrik.lundin@webrtc.org [Thu, 20 Nov 2014 14:14:49 +0000 (14:14 +0000)]
Fix problems if first packet into NetEq is rejected

This CL fixes the problem described in issue 4021. In summary, of the
very first packet coming in to NetEq gets rejected because the RTP
payload type is unknown, subsequent GetAudio calls would trigger asserts
(in debug builds). The problem was that the first_packet_ variable was
reset and new_codec_ was set, even though the packet was discarded
further down the line. Now, these variables are modified after the
packet has been verified.

BUG=4021
R=tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/29089004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7724 4adac7df-926f-26a2-2b94-8c16560cd09d

7 days agoCreate a NetEq test for when the first incoming payload type is unknown
henrik.lundin@webrtc.org [Thu, 20 Nov 2014 11:01:02 +0000 (11:01 +0000)]
Create a NetEq test for when the first incoming payload type is unknown

This test is currently disabled. It triggers an issue where the NetEq
will trigger asserts on subsequent GetAudio calls if the first inserted
packet is rejected, for instance since the payload type is unknown to
NetEq.

BUG=4021
R=tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/28069004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7723 4adac7df-926f-26a2-2b94-8c16560cd09d

7 days agoChange default values for CpuOveruseOptions.
asapersson@webrtc.org [Thu, 20 Nov 2014 10:19:46 +0000 (10:19 +0000)]
Change default values for CpuOveruseOptions.
Enabled method based on encode time and modified values for the low (60->55) and high threshold (90->85).

Moved DelayedEncoder to fake_encoder.h and added configuration for the delay.

R=mflodman@webrtc.org, pbos@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30969004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7722 4adac7df-926f-26a2-2b94-8c16560cd09d

8 days agocricket::VideoAdapter: Drop frames before spending time converting/scaling, not after.
magjed@webrtc.org [Wed, 19 Nov 2014 18:09:14 +0000 (18:09 +0000)]
cricket::VideoAdapter: Drop frames before spending time converting/scaling, not after.

In VideoCapturer::OnFrameCaptured, we currently convert cricket::CapturedFrame to cricket::VideoFrame and then send that to VideoAdapter::AdaptFrame. AdaptFrame may then decide to drop the frame. It would be faster to drop the frame before converting to cricket::VideoFrame.

This CL refactors VideoAdapter with a new function AdaptFrameResolution that takes captured resolution as input and output adapted resolution, or 0x0 if the frame should be dropped. Using that function, frames can be dropped before any conversion takes place.

R=fbarchard@google.com, perkj@webrtc.org, tommi@webrtc.org

Committed: https://code.google.com/p/webrtc/source/detail?r=7702

Committed: https://code.google.com/p/webrtc/source/detail?r=7707

Review URL: https://webrtc-codereview.appspot.com/29949004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7721 4adac7df-926f-26a2-2b94-8c16560cd09d

8 days agoRevert "Add DCHECK to ensure that NetEq's packet buffer is not empty"
henrik.lundin@webrtc.org [Wed, 19 Nov 2014 13:46:52 +0000 (13:46 +0000)]
Revert "Add DCHECK to ensure that NetEq's packet buffer is not empty"

This reverts r7719. It failed to compile in Chromium.

TBR=kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30179004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7720 4adac7df-926f-26a2-2b94-8c16560cd09d

8 days agoAdd DCHECK to ensure that NetEq's packet buffer is not empty
henrik.lundin@webrtc.org [Wed, 19 Nov 2014 13:02:24 +0000 (13:02 +0000)]
Add DCHECK to ensure that NetEq's packet buffer is not empty

This DCHECK ensures that one packet was inserted after the buffer was
flushed.

R=kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30169004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7719 4adac7df-926f-26a2-2b94-8c16560cd09d

9 days agoAppRTCDemoActivity: Add a config CheckBox for enabling/disabling CPU overuse adaptati...
henrika@webrtc.org [Tue, 18 Nov 2014 13:22:28 +0000 (13:22 +0000)]
AppRTCDemoActivity: Add a config CheckBox for enabling/disabling CPU overuse adaptation. (re-land)

This CL was incorrectly reverted in r7647 by the libjingle sync bot.

TBR=kjellander@webrtc.org
BUG=

Review URL: https://webrtc-codereview.appspot.com/32489004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7717 4adac7df-926f-26a2-2b94-8c16560cd09d

9 days agoRoll chromium_revision 91f1781..d8c9041
kjellander@webrtc.org [Tue, 18 Nov 2014 10:25:04 +0000 (10:25 +0000)]
Roll chromium_revision 91f1781..d8c9041

Relevant changes:
* buildtools: c27f95b..6ea835d
* third_party/icu: d8b2a9d..6242e2f
* tools/gyp: 487c0b6..b13d8f2
* tools/swarming_client: 1f8ba35..5b827c9
Details: https://chromium.googlesource.com/chromium/src/+/91f1781..d8c9041/DEPS

Clang version was not updated in this roll, although the
-Wunused-local-typedef warning was enabled by default.

R=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/26189004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7716 4adac7df-926f-26a2-2b94-8c16560cd09d

10 days agoAdd empty 3 band splitting filter API
aluebs@webrtc.org [Mon, 17 Nov 2014 23:01:23 +0000 (23:01 +0000)]
Add empty 3 band splitting filter API

This is only an empty API that will never be used. For now is 48kHz not supported in AudioProcessing. For that it needs to be added in InitializeLocked. But before the 3 band filter bank needs to be populated.

BUG=webrtc:3146
R=bjornv@webrtc.org, kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30139004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7715 4adac7df-926f-26a2-2b94-8c16560cd09d

10 days agoFix ExpectedQueueTimeMs() to avoid truncation or overflow.
pkasting@chromium.org [Mon, 17 Nov 2014 22:21:14 +0000 (22:21 +0000)]
Fix ExpectedQueueTimeMs() to avoid truncation or overflow.

BUG=none
TEST=none
R=asapersson@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30129004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7714 4adac7df-926f-26a2-2b94-8c16560cd09d

10 days agoAdd jmi field for packets discarded due to network error
guoweis@webrtc.org [Mon, 17 Nov 2014 19:42:14 +0000 (19:42 +0000)]
Add jmi field for packets discarded due to network error

Also included the total packets attempted to send.

BUG=427555

Copied from https://webrtc-codereview.appspot.com/25959004/

R=harryjin@google.com, juberti@webrtc.org

Committed: https://code.google.com/p/webrtc/source/detail?r=7693

Review URL: https://webrtc-codereview.appspot.com/32039004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7713 4adac7df-926f-26a2-2b94-8c16560cd09d

10 days agoAdd preliminary empty file videoframefactory.cc
magjed@webrtc.org [Mon, 17 Nov 2014 16:34:00 +0000 (16:34 +0000)]
Add preliminary empty file videoframefactory.cc

The purpose of this CL is to add a new file in libjingle without breaking Chromium in the process. The plan is to do the following:
1. Land a no-op videoframefactory.cc in webrtc (this file).
2. Wait for it to roll into Chromium.
3. Modify libjingle.gyp in Chromium to include this file.
4. Make the real change in webrtc with the real implementation of this file.
5. Wait for the change to roll into Chromium.

R=tommi@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/28049004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7712 4adac7df-926f-26a2-2b94-8c16560cd09d

10 days agoAnnotate COMPILE_ASSERT with __attribute__((unused)).
pbos@webrtc.org [Mon, 17 Nov 2014 13:47:38 +0000 (13:47 +0000)]
Annotate COMPILE_ASSERT with __attribute__((unused)).

Also renames UNUSED -> ATTRIBUTE_UNUSED to be able to use this when
building peerconnection_jni.cc which apparently has this defined to
something else.

R=kjellander@webrtc.org
TBR=mflodman@webrtc.org
BUG=4018

Review URL: https://webrtc-codereview.appspot.com/28039005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7711 4adac7df-926f-26a2-2b94-8c16560cd09d

10 days agoSetting Opus FEC as default
minyue@webrtc.org [Mon, 17 Nov 2014 09:26:39 +0000 (09:26 +0000)]
Setting Opus FEC as default

BUG=3986
R=mflodman@webrtc.org, tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/26899004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7710 4adac7df-926f-26a2-2b94-8c16560cd09d

10 days agoUse RtpFileSource in NetEqDecodingTest
henrik.lundin@webrtc.org [Mon, 17 Nov 2014 09:08:38 +0000 (09:08 +0000)]
Use RtpFileSource in NetEqDecodingTest

This CL removes the dependency on the old NETEQTEST_RTPpacket class
from the NetEqDecodingTest code, and also removes the dependency
from the modules_unittests target to neteq_test_tools.

BUG=2692
R=tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/24269004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7709 4adac7df-926f-26a2-2b94-8c16560cd09d

11 days agoRevert 7707 "cricket::VideoAdapter: Drop frames before spending ..."
tommi@webrtc.org [Sun, 16 Nov 2014 22:58:11 +0000 (22:58 +0000)]
Revert 7707 "cricket::VideoAdapter: Drop frames before spending ..."

This didn't compile on the FYI bots.  Example error:

FAILED: E:\b\depot_tools\python276_bin\python.exe gyp-win-tool link-with-manifests environment.x86 True chrome_child.dll "E:\b\depot_tools\python276_bin\python.exe gyp-win-tool link-wrapper environment.x86 False link.exe /nologo /IMPLIB:chrome_child.dll.lib /DLL /OUT:chrome_child.dll @chrome_child.dll.rsp" 2 mt.exe rc.exe "obj\chrome\chrome_child_dll.chrome_child.dll.intermediate.manifest" obj\chrome\chrome_child_dll.chrome_child.dll.generated.manifest
content_renderer.lib(content_renderer.webrtc_video_capturer_adapter.obj) : error LNK2001: unresolved external symbol "public: virtual class cricket::VideoFrame * __thiscall cricket::VideoFrameFactory::CreateAliasedFrame(struct cricket::CapturedFrame const *,int,int,int,int)const " (?CreateAliasedFrame@VideoFrameFactory@cricket@@UBEPAVVideoFrame@2@PBUCapturedFrame@2@HHHH@Z)

libjingle_webrtc_common.lib(libjingle_webrtc_common.peerconnectionfactory.obj) : error LNK2001: unresolved external symbol "public: virtual class cricket::VideoFrame * __thiscall cricket::VideoFrameFactory::CreateAliasedFrame(struct cricket::CapturedFrame const *,int,int,int,int)const " (?CreateAliasedFrame@VideoFrameFactory@cricket@@UBEPAVVideoFrame@2@PBUCapturedFrame@2@HHHH@Z)

libjingle_webrtc_common.lib(libjingle_webrtc_common.videocapturer.obj) : error LNK2001: unresolved external symbol "public: virtual class cricket::VideoFrame * __thiscall cricket::VideoFrameFactory::CreateAliasedFrame(struct cricket::CapturedFrame const *,int,int,int,int)const " (?CreateAliasedFrame@VideoFrameFactory@cricket@@UBEPAVVideoFrame@2@PBUCapturedFrame@2@HHHH@Z)

libjingle_webrtc_common.lib(libjingle_webrtc_common.dummydevicemanager.obj) : error LNK2001: unresolved external symbol "public: virtual class cricket::VideoFrame * __thiscall cricket::VideoFrameFactory::CreateAliasedFrame(struct cricket::CapturedFrame const *,int,int,int,int)const " (?CreateAliasedFrame@VideoFrameFactory@cricket@@UBEPAVVideoFrame@2@PBUCapturedFrame@2@HHHH@Z)

chrome_child.dll : fatal error LNK1120: 1 unresolved externals

> cricket::VideoAdapter: Drop frames before spending time converting/scaling, not after.
>
> In VideoCapturer::OnFrameCaptured, we currently convert cricket::CapturedFrame to cricket::VideoFrame and then send that to VideoAdapter::AdaptFrame. AdaptFrame may then decide to drop the frame. It would be faster to drop the frame before converting to cricket::VideoFrame.
>
> This CL refactors VideoAdapter with a new function AdaptFrameResolution that takes captured resolution as input and output adapted resolution, or 0x0 if the frame should be dropped. Using that function, frames can be dropped before any conversion takes place.
>
> R=fbarchard@google.com, perkj@webrtc.org, tommi@webrtc.org
>
> Committed: https://code.google.com/p/webrtc/source/detail?r=7702
>
> Review URL: https://webrtc-codereview.appspot.com/29949004

TBR=magjed@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/28019004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7708 4adac7df-926f-26a2-2b94-8c16560cd09d

11 days agocricket::VideoAdapter: Drop frames before spending time converting/scaling, not after.
magjed@webrtc.org [Sun, 16 Nov 2014 18:21:51 +0000 (18:21 +0000)]
cricket::VideoAdapter: Drop frames before spending time converting/scaling, not after.

In VideoCapturer::OnFrameCaptured, we currently convert cricket::CapturedFrame to cricket::VideoFrame and then send that to VideoAdapter::AdaptFrame. AdaptFrame may then decide to drop the frame. It would be faster to drop the frame before converting to cricket::VideoFrame.

This CL refactors VideoAdapter with a new function AdaptFrameResolution that takes captured resolution as input and output adapted resolution, or 0x0 if the frame should be dropped. Using that function, frames can be dropped before any conversion takes place.

R=fbarchard@google.com, perkj@webrtc.org, tommi@webrtc.org

Committed: https://code.google.com/p/webrtc/source/detail?r=7702

Review URL: https://webrtc-codereview.appspot.com/29949004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7707 4adac7df-926f-26a2-2b94-8c16560cd09d

13 days agoRevert 7693 "Add jmi field for packets discarded due to network error" breaks chromiu...
henrike@webrtc.org [Fri, 14 Nov 2014 22:33:13 +0000 (22:33 +0000)]
Revert 7693 "Add jmi field for packets discarded due to network error" breaks chromium's webrtc_cases.

TBR=guoweis@webrtc.org

BUG=N/A

Review URL: https://webrtc-codereview.appspot.com/25179004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7706 4adac7df-926f-26a2-2b94-8c16560cd09d

13 days agoWrap the splitting filter in its own class
aluebs@webrtc.org [Fri, 14 Nov 2014 22:18:10 +0000 (22:18 +0000)]
Wrap the splitting filter in its own class

This doesn't change the behavior at all.
The logic behind this is having one class which manages all the splitting filters, because in the future we plan to add a 3 band one for 48kHz support.
It also breaks the dependency of the AudioBuffer with the filter states of these filters (which are going to be different for the 3 band one). The AudioBuffer is complicated enough and is going to need changes to support 3 bands in the future, so any simplification is a good idea.
On top of that it eliminates repeated code in the APM (now only iterating over channels, but then also deciding in how many bands to split). This should be managed by the AudioBuffer directly.

BUG=webrtc:3146
R=bjornv@webrtc.org, kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/32469004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7705 4adac7df-926f-26a2-2b94-8c16560cd09d

13 days agoDisable EndToEnd.GetStats test.
pbos@webrtc.org [Fri, 14 Nov 2014 17:42:51 +0000 (17:42 +0000)]
Disable EndToEnd.GetStats test.

Looks like this test exposes a bug in jitter buffer after enabling
multiple streams. Will disable to be able to debug it in peace and not
have to revert.

TBR=stefan@webrtc.org
BUG=4014

Review URL: https://webrtc-codereview.appspot.com/31009004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7704 4adac7df-926f-26a2-2b94-8c16560cd09d

13 days agoRevert 7702 "cricket::VideoAdapter: Drop frames before spending ..."
magjed@webrtc.org [Fri, 14 Nov 2014 13:25:25 +0000 (13:25 +0000)]
Revert 7702 "cricket::VideoAdapter: Drop frames before spending ..."

Rease for revert: failed internal test cases

> cricket::VideoAdapter: Drop frames before spending time converting/scaling, not after.
>
> In VideoCapturer::OnFrameCaptured, we currently convert cricket::CapturedFrame to cricket::VideoFrame and then send that to VideoAdapter::AdaptFrame. AdaptFrame may then decide to drop the frame. It would be faster to drop the frame before converting to cricket::VideoFrame.
>
> This CL refactors VideoAdapter with a new function AdaptFrameResolution that takes captured resolution as input and output adapted resolution, or 0x0 if the frame should be dropped. Using that function, frames can be dropped before any conversion takes place.
>
> R=fbarchard@google.com, perkj@webrtc.org, tommi@webrtc.org
>
> Review URL: https://webrtc-codereview.appspot.com/29949004

TBR=magjed@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/24279004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7703 4adac7df-926f-26a2-2b94-8c16560cd09d

13 days agocricket::VideoAdapter: Drop frames before spending time converting/scaling, not after.
magjed@webrtc.org [Fri, 14 Nov 2014 12:10:46 +0000 (12:10 +0000)]
cricket::VideoAdapter: Drop frames before spending time converting/scaling, not after.

In VideoCapturer::OnFrameCaptured, we currently convert cricket::CapturedFrame to cricket::VideoFrame and then send that to VideoAdapter::AdaptFrame. AdaptFrame may then decide to drop the frame. It would be faster to drop the frame before converting to cricket::VideoFrame.

This CL refactors VideoAdapter with a new function AdaptFrameResolution that takes captured resolution as input and output adapted resolution, or 0x0 if the frame should be dropped. Using that function, frames can be dropped before any conversion takes place.

R=fbarchard@google.com, perkj@webrtc.org, tommi@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/29949004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7702 4adac7df-926f-26a2-2b94-8c16560cd09d

13 days agoReport total bitrate for all streams in GetStats.
pbos@webrtc.org [Fri, 14 Nov 2014 11:52:04 +0000 (11:52 +0000)]
Report total bitrate for all streams in GetStats.

This regression wasn't caught because I accidentally disabled multiple
streams for EndToEndTest.GetStats in a refactoring.

R=stefan@webrtc.org, xians@webrtc.org
BUG=1667

Review URL: https://webrtc-codereview.appspot.com/27179004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7701 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoRevert 7698 "WebRtcVideoMediaChannel::SetSendParams: Don't cap r..."
magjed@webrtc.org [Thu, 13 Nov 2014 16:21:49 +0000 (16:21 +0000)]
Revert 7698 "WebRtcVideoMediaChannel::SetSendParams: Don't cap r..."

Reason for revert is failed testcases:
WebRtcVideoEngineExtendedTestFake.ResetSimulcastSendCodecOnNewFrameSize
WebRtcVideoEngineExtendedTestFake.MultipleSendStreamsDifferentFormats

> WebRtcVideoMediaChannel::SetSendParams: Don't cap resolution
>
> BUG=3936
> R=pthatcher@webrtc.org
>
> Review URL: https://webrtc-codereview.appspot.com/30039004

TBR=magjed@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/28009004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7700 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoRemove unnecessary copying of libjingle resource files.
kjellander@webrtc.org [Thu, 13 Nov 2014 15:53:08 +0000 (15:53 +0000)]
Remove unnecessary copying of libjingle resource files.

This copying has probably not been needed since
https://code.google.com/p/webrtc/source/detail?r=7088

BUG=398
TESTED=Removed the top-level talk directory and ran
libjingle_media_unittest from the following working directories:
* checkout-root/src/out/Debug
* checkout-root/src
* checkout-root

R=phoglund@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/26149004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7699 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoWebRtcVideoMediaChannel::SetSendParams: Don't cap resolution
magjed@webrtc.org [Thu, 13 Nov 2014 15:43:11 +0000 (15:43 +0000)]
WebRtcVideoMediaChannel::SetSendParams: Don't cap resolution

BUG=3936
R=pthatcher@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30039004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7698 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoMake SetREMBData accept vector of SSRCs.
pbos@webrtc.org [Thu, 13 Nov 2014 14:42:37 +0000 (14:42 +0000)]
Make SetREMBData accept vector of SSRCs.

BUG=
R=pbos@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/32049004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7697 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoFix and enable CanReceiveFec test.
pbos@webrtc.org [Thu, 13 Nov 2014 14:40:15 +0000 (14:40 +0000)]
Fix and enable CanReceiveFec test.

Test relied on the first protected media packet that was dropped to
actually be rendered, while rendering it could have been skipped on slow
systems due to newer frames being decoded before rendering happens.

R=stefan@webrtc.org
BUG=3269

Review URL: https://webrtc-codereview.appspot.com/25159004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7696 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoSet correct sample rate in far_frame in audioproc tool.
bjornv@webrtc.org [Thu, 13 Nov 2014 11:00:10 +0000 (11:00 +0000)]
Set correct sample rate in far_frame in audioproc tool.

One debug recording with non matching sample rates between render and capture revealed a bug in modules/audio_processing/test/process_test.cc
The far_frame (render audio frame) used was loaded with the capture rate instead of the render rate with a data length mismatch error as result.

BUG=N/A
TESTED=manually on linux
R=kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/27169004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7695 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoUpdate isolate files for Android APK tests.
kjellander@webrtc.org [Thu, 13 Nov 2014 08:35:05 +0000 (08:35 +0000)]
Update isolate files for Android APK tests.

This should speed up test execution on Android since only
the files needed by the test will be processed (instead
of the whole data + resources directories).

A few files for modules_unittests had to be explicitly added
for Android, since they were previously a part of the
add-whole-directories entries for the resources and data
directories.

BUG=webrtc:3741
TEST=Passing android+android_rel trybots.
R=phoglund@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/22559004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7694 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoAdd jmi field for packets discarded due to network error
guoweis@webrtc.org [Thu, 13 Nov 2014 03:38:05 +0000 (03:38 +0000)]
Add jmi field for packets discarded due to network error

Also included the total packets attempted to send.

BUG=427555

Copied from https://webrtc-codereview.appspot.com/25959004/

R=harryjin@google.com, juberti@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/32039004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7693 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoFix a platform check to use WEBRTC_WIN instead of OS_WIN.
jiayl@webrtc.org [Wed, 12 Nov 2014 20:53:00 +0000 (20:53 +0000)]
Fix a platform check to use WEBRTC_WIN instead of OS_WIN.

BUG=4006
R=wez@chromium.org

Review URL: https://webrtc-codereview.appspot.com/25169004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7691 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoFix a SCTP message reordering issue in datachannel.cc.
jiayl@webrtc.org [Wed, 12 Nov 2014 17:28:40 +0000 (17:28 +0000)]
Fix a SCTP message reordering issue in datachannel.cc.
Previously DataChannel::SendQueuedDataMessages continues the loop of sending queued messages if the channel is blocked, which will cause message reordering if the channel becomes unblocked during the loop, i.e. messages attempted after the unblocking will be sent earlier than the older messages attempted before the unblocking.

BUG=3979
R=pthatcher@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/25149004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7690 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agowebrtc::Scaler: Preserve aspect ratio
magjed@webrtc.org [Wed, 12 Nov 2014 09:52:03 +0000 (09:52 +0000)]
webrtc::Scaler: Preserve aspect ratio

BUG=3936
R=glaznev@webrtc.org, stefan@webrtc.org

Committed: https://code.google.com/p/webrtc/source/detail?r=7679

Review URL: https://webrtc-codereview.appspot.com/28969004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7689 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoVideoSendStreamTest.SwapsI420VideoFrames: Initialize frame memory to avoid drmemory...
magjed@webrtc.org [Wed, 12 Nov 2014 08:58:49 +0000 (08:58 +0000)]
VideoSendStreamTest.SwapsI420VideoFrames: Initialize frame memory to avoid drmemory errors

R=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/27149004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7688 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoChange the static_library("webrtc") to a source set in the GN build.
kjellander@webrtc.org [Wed, 12 Nov 2014 07:56:21 +0000 (07:56 +0000)]
Change the static_library("webrtc") to a source set in the GN build.

Static libraries cannot have only headers as sources (libtool complains
that there's nothing to actually link).

TBR=brettw@chromium.org

Review URL: https://webrtc-codereview.appspot.com/27159004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7687 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoreplace inline assembly WebRtcAecm_CalcLinearEnergiesNeon by intrinsics.
andrew@webrtc.org [Tue, 11 Nov 2014 19:34:14 +0000 (19:34 +0000)]
replace inline assembly WebRtcAecm_CalcLinearEnergiesNeon by intrinsics.

The modification only uses the unique part of the CalcLinearEnergies
 function. Pass byte to byte conformance test both on ARMv7 and ARM64,
 and the single function performance is similar with original assembly
 version on different platforms. If not specified, the code is compiled
 by GCC 4.6. The result is the "X version / C version" ratio, and the
 less is better.

| run 100k times             | cortex-a7 | cortex-a9 | cortex-a15 |
| use C as the base on each  |  (1.2Ghz) |  (1.0Ghz) |   (1.7Ghz) |
| CPU target                 |           |           |            |
|----------------------------+-----------+-----------+------------|
| Neon asm                   |    19.48% |    19.26% |     13.68% |
| Neon inline                |    27.90% |    28.87% |     17.79% |
| Neon intrinsics (GCC 4.8)  |    18.69% |    20.18% |     14.69% |
| Neon intrinsics (LLVM 3.4) |    18.52% |    21.15% |     13.56% |

BUG=3580
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/23349004

Patch from Zhongwei Yao <zhongwei.yao@arm.com>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7686 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoreplace inline assembly WebRtcAecm_StoreAdaptiveChannelNeon by intrinsics.
andrew@webrtc.org [Tue, 11 Nov 2014 19:32:33 +0000 (19:32 +0000)]
replace inline assembly WebRtcAecm_StoreAdaptiveChannelNeon  by intrinsics.

The modification only uses the unique part of the StoreAdaptiveChannel
 function. Pass byte to byte conformance test both on ARM32 and ARM64,
 and the single function performance is similar with original assembly
 version on different platforms. If not specified, the code is compiled
 by GCC 4.6.  The result is the "X version / C version" ratio, and the
 less is better.

| run 100k times             | cortex-a7 | cortex-a9 | cortex-a15 |
| use C as the base on each  |  (1.2Ghz) |  (1.0Ghz) |   (1.7Ghz) |
| CPU target                 |           |           |            |
|----------------------------+-----------+-----------+------------|
| Neon asm                   |    20.97% |    37.70% |     25.41% |
| Neon inline                |    36.93% |    51.80% |     38.14% |
| Neon intrinsics (GCC 4.6)  |    27.78% |    43.71% |     26.50% |
| Neon intrinsics (GCC 4.8)  |    27.16% |    38.22% |     26.87% |
| Neon intrinsics (LLVM 3.4) |    27.82% |    39.90% |     26.69% |

Change-Id: Ia55d8a268a70164b50676c604ae40b68fc183106

BUG=3580
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30029004

Patch from Zhongwei Yao <zhongwei.yao@arm.com>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7685 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoUse ScreenCapturer to capture the whole and clip to the window rect when the shared...
jiayl@webrtc.org [Tue, 11 Nov 2014 18:15:55 +0000 (18:15 +0000)]
Use ScreenCapturer to capture the whole and clip to the window rect when the shared window is on the top.

BUG=crbug/403703, crbug/316603
R=wez@chromium.org

Review URL: https://webrtc-codereview.appspot.com/22759004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7684 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoBump to version 40
tnakamura@webrtc.org [Tue, 11 Nov 2014 16:23:15 +0000 (16:23 +0000)]
Bump to version 40

TBR=niklas.enbom

Review URL: https://webrtc-codereview.appspot.com/26109004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7683 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoRevert 7679 "webrtc::Scaler: Preserve aspect ratio"
magjed@webrtc.org [Tue, 11 Nov 2014 13:12:09 +0000 (13:12 +0000)]
Revert 7679 "webrtc::Scaler: Preserve aspect ratio"

> webrtc::Scaler: Preserve aspect ratio
>
> BUG=3936
> R=glaznev@webrtc.org, stefan@webrtc.org
>
> Review URL: https://webrtc-codereview.appspot.com/28969004

TBR=magjed@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30989004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7682 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoAdd PROJECT to codereview.settings
kjellander@webrtc.org [Tue, 11 Nov 2014 10:00:47 +0000 (10:00 +0000)]
Add PROJECT to codereview.settings

This is needed once we move over to Chromium's
Rietveld instance at codereview.chromium.org.

BUG=3884
R=niklas.enbom@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/25129004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7681 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoRoll chromium_revision 375f736..91f1781
kjellander@webrtc.org [Tue, 11 Nov 2014 09:57:19 +0000 (09:57 +0000)]
Roll chromium_revision 375f736..91f1781

Relevant changes:
* buildtools: 51ca1f2..c27f95b
* tools/gyp: b13d8f2..487c0b6
* tools/swarming_client: 41036ec..1f8ba35
Details: https://chromium.googlesource.com/chromium/src/+/375f736..91f1781/DEPS

Clang version was not updated in this roll.

BUG=
R=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30049004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7680 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agowebrtc::Scaler: Preserve aspect ratio
magjed@webrtc.org [Tue, 11 Nov 2014 09:51:30 +0000 (09:51 +0000)]
webrtc::Scaler: Preserve aspect ratio

BUG=3936
R=glaznev@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/28969004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7679 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoAdd thread annotations to overuse_frame_detector class.
asapersson@webrtc.org [Tue, 11 Nov 2014 09:40:19 +0000 (09:40 +0000)]
Add thread annotations to overuse_frame_detector class.

R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/29029004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7678 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoFollow-up fixes for G722
henrik.lundin@webrtc.org [Tue, 11 Nov 2014 08:38:24 +0000 (08:38 +0000)]
Follow-up fixes for G722

This CL addresses post-commit comments on r7662. See
https://webrtc-codereview.appspot.com/27089004/#ps40001.

BUG=3951
R=pthatcher@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/30979004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7677 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoRevert 7675 "Make an AudioEncoder subclass for iSAC"
turaj@webrtc.org [Tue, 11 Nov 2014 01:44:13 +0000 (01:44 +0000)]
Revert 7675 "Make an AudioEncoder subclass for iSAC"

Above CL did not compile on Android. Followings are links to Android builds

http://chromegw.corp.google.com/i/internal.client.webrtc/builders/Android%20Builder%20%28dbg%29/builds/2648

http://chromegw.corp.google.com/i/internal.client.webrtc/builders/Android%20Clang%20%28dbg%29/builds/2369

http://chromegw.corp.google.com/i/internal.client.webrtc/builders/Android%20ARM64%20%28dbg%29/builds/1320

> Make an AudioEncoder subclass for iSAC
>
> BUG=3926
> R=henrik.lundin@webrtc.org, kjellander@webrtc.org
>
> Review URL: https://webrtc-codereview.appspot.com/25019004

TBR=kwiberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/32439004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7676 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoMake an AudioEncoder subclass for iSAC
kwiberg@webrtc.org [Mon, 10 Nov 2014 23:53:08 +0000 (23:53 +0000)]
Make an AudioEncoder subclass for iSAC

BUG=3926
R=henrik.lundin@webrtc.org, kjellander@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/25019004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7675 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoChange from talk/p2p (r7664) "(Auto)update libjingle 79414100-> 79428003".
henrike@webrtc.org [Mon, 10 Nov 2014 19:43:11 +0000 (19:43 +0000)]
Change from talk/p2p (r7664) "(Auto)update libjingle 79414100-> 79428003".

BUG=3379
R=tpsiaki@google.com

Review URL: https://webrtc-codereview.appspot.com/27119004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7674 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoChange from talk/p2p (r7572): "Improve the logging when a TCP connection is deleted."
henrike@webrtc.org [Mon, 10 Nov 2014 19:40:29 +0000 (19:40 +0000)]
Change from talk/p2p (r7572): "Improve the logging when a TCP connection is deleted."

BUG=3379
R=jiayl@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/27109004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7673 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoreplace inline assembly WebRtcAecm_ResetAdaptiveChannelNeon by intrinsics.
andrew@webrtc.org [Mon, 10 Nov 2014 17:27:53 +0000 (17:27 +0000)]
replace inline assembly WebRtcAecm_ResetAdaptiveChannelNeon  by intrinsics.

The modification only uses the unique part of the ResetAdaptiveChannel
 function. Pass byte to byte conformance test both on ARM32 and ARM64,
 and the single function performance is similar with original assembly
 version on different platforms. If not specified, the code is compiled
 by GCC 4.6. The result is the "X version / C version" ratio, and the
 less is better.

| run 100k times             | cortex-a7 | cortex-a9 | cortex-a15 |
| use C as the base on each  |  (1.2Ghz) |  (1.0Ghz) |   (1.7Ghz) |
| CPU target                 |           |           |            |
|----------------------------+-----------+-----------+------------|
| Neon asm                   |       15% |       30% |        12% |
| Neon inline                |       21% |       30% |        12% |
| Neon intrinsics (GCC 4.6)  |       19% |       32% |        12% |
| Neon intrinsics (GCC 4.8)  |       20% |       32% |        12% |
| Neon intrinsics (LLVM 3.4) |       19% |       30% |        12% |

BUG=3580
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/29019004

Patch from Zhongwei Yao <zhongwei.yao@arm.com>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7672 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoclear asm code and unused functions in audio processing module
andrew@webrtc.org [Mon, 10 Nov 2014 17:19:57 +0000 (17:19 +0000)]
clear asm code and unused functions in audio processing module

BUG=
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/25119004

Patch from Zhongwei Yao <zhongwei.yao@arm.com>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7671 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoRemoves talk/xmllite, talk/xmpp and talk/p2p as they are no longer used by gyp/gn...
henrike@webrtc.org [Mon, 10 Nov 2014 15:31:24 +0000 (15:31 +0000)]
Removes talk/xmllite, talk/xmpp and talk/p2p as they are no longer used by gyp/gn builds.

TBR=niklas.enbom@webrtc.org
BUG=3379

Review URL: https://webrtc-codereview.appspot.com/30959004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7670 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoWire up DSCP support in WebRtcVideoEngine2.
pbos@webrtc.org [Mon, 10 Nov 2014 14:41:43 +0000 (14:41 +0000)]
Wire up DSCP support in WebRtcVideoEngine2.

R=stefan@webrtc.org
BUG=1788

Review URL: https://webrtc-codereview.appspot.com/24249004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7669 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoPut send-side bwe probing under finch experiment.
stefan@webrtc.org [Mon, 10 Nov 2014 13:55:16 +0000 (13:55 +0000)]
Put send-side bwe probing under finch experiment.

BUG=crbug/425925
R=mflodman@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/26079004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7668 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoRefactor SetDefaultEncoderConfig to work on existing codecs.
pbos@webrtc.org [Mon, 10 Nov 2014 12:36:11 +0000 (12:36 +0000)]
Refactor SetDefaultEncoderConfig to work on existing codecs.

Addresses issue where SetDefaultEncoderConfig modifies the codec list
rather than just the targeted codec. This was previously done just to
pass more unit tests rather than be done properly. This incidentally
addresses a TODO causing this to work with external codecs as well.

R=stefan@webrtc.org
BUG=1788

Review URL: https://webrtc-codereview.appspot.com/32009004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7667 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoAdd unit to dropped frames.
pbos@webrtc.org [Mon, 10 Nov 2014 09:54:19 +0000 (09:54 +0000)]
Add unit to dropped frames.

Missing unit causes less dropped frames to be reported as a regression
and not an improvement.

R=stefan@webrtc.org
BUG=chromium:429206

Review URL: https://webrtc-codereview.appspot.com/25139004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7666 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks ago.gitignore updates
kjellander@webrtc.org [Mon, 10 Nov 2014 06:51:34 +0000 (06:51 +0000)]
.gitignore updates

Update after Chromium roll in https://review.webrtc.org/24179004/
and Android project updates in https://review.webrtc.org/25029004/

BUG=
TBR=phoglund@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/24239004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7665 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks ago(Auto)update libjingle 79414100-> 79428003
buildbot@webrtc.org [Fri, 7 Nov 2014 17:58:41 +0000 (17:58 +0000)]
(Auto)update libjingle 79414100-> 79428003

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7664 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoEnable VP9 video codec support on webrtcvideoengine behind a field trial.
andresp@webrtc.org [Fri, 7 Nov 2014 13:21:04 +0000 (13:21 +0000)]
Enable VP9 video codec support on webrtcvideoengine behind a field trial.

BUG=chromium:431285
R=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/27929004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7663 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoReapply "Advertise G722 as 8 kHz rather than 16 kHz""
henrik.lundin@webrtc.org [Fri, 7 Nov 2014 12:25:00 +0000 (12:25 +0000)]
Reapply "Advertise G722 as 8 kHz rather than 16 kHz""

This reverts r7653 and relands r7645. The reason for the original revert was that G722 disappeared from the SDP offer. This is now fixed. Also, a unit test was updated compared with the original change.

BUG=3951
TBR=pthatcher@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/27089004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7662 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoChange dummy address to use 0.0.0.0 instead of ::
perkj@webrtc.org [Fri, 7 Nov 2014 11:22:06 +0000 (11:22 +0000)]
Change dummy address to use 0.0.0.0 instead of ::
This is to not break compatiblity with FF.

https://code.google.com/p/chromium/issues/detail?id=430333

TBR=pthatcher@webrtc.org, juberti@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/31979004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7661 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoRemove partially defined WebRtcRTPHeader from Parse().
pbos@webrtc.org [Fri, 7 Nov 2014 11:02:12 +0000 (11:02 +0000)]
Remove partially defined WebRtcRTPHeader from Parse().

It' bit ugly that RtpDepacketizer::ParsedPayload partially defines WebRtcRTPHeader, and then sent to Parse() function for internal change.
To make it clearer, the CL gets rid of using partially-defined WebRtcRTPHeader.

BUG=
R=pbos@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/28919004

Patch from Changbin Shao <changbin.shao@intel.com>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7660 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoPrevent a lot of VideoSendStream reconfigures.
pbos@webrtc.org [Fri, 7 Nov 2014 10:54:43 +0000 (10:54 +0000)]
Prevent a lot of VideoSendStream reconfigures.

Checking whether we're setting the same configuration or not.
Experimentally this brings down underlying reconfigures from ~20 to
about 4-5.

R=stefan@webrtc.org
BUG=1788

Review URL: https://webrtc-codereview.appspot.com/30909004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7659 4adac7df-926f-26a2-2b94-8c16560cd09d

2 weeks agoRefactor webrtcvideoengines to have the default list of supported codecs being genera...
andresp@webrtc.org [Fri, 7 Nov 2014 09:37:54 +0000 (09:37 +0000)]
Refactor webrtcvideoengines to have the default list of supported codecs being generated in runtime.
This will allow to plugin VP9 based on a field trial.

R=pbos@webrtc.org, pbos, pthatcher

Review URL: https://webrtc-codereview.appspot.com/27949004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7658 4adac7df-926f-26a2-2b94-8c16560cd09d

3 weeks agoReland Volume buttons in AppRTCDemo should affect output audio volume (part I).
henrika@webrtc.org [Thu, 6 Nov 2014 20:35:13 +0000 (20:35 +0000)]
Reland Volume buttons in AppRTCDemo should affect output audio volume (part I).

Second attempt to land https://webrtc-codereview.appspot.com/32399004/

TBR=perkj@webrtc.org
BUG=

Review URL: https://webrtc-codereview.appspot.com/30919004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7657 4adac7df-926f-26a2-2b94-8c16560cd09d

3 weeks agoUse uint16s for port numbers in webrtc/p2p/base.
pkasting@chromium.org [Thu, 6 Nov 2014 20:19:22 +0000 (20:19 +0000)]
Use uint16s for port numbers in webrtc/p2p/base.

This is a necessary precursor to using uint16s for port numbers more
consistently in Chromium code.

This also makes some minor formatting changes in surrounding code (function declaration wrapping, virtual -> override).

BUG=chromium:81439
TEST=none
R=henrike@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/32379004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@7656 4adac7df-926f-26a2-2b94-8c16560cd09d