admin/webrtc_m130
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
5,681 Commits 1 Branch 0 Tags
Commit Graph

1309 Commits

Author SHA1 Message Date
fischman@webrtc.org
eb7def234e Fix compilation errors on Fedora 20. 
peerconnection_jni.cc: syscall() comes from <unistd.h>
RTPtimeshift.cc: char* being compared to 0 instead of the atoi() of it
rtp_payload_registry_unittest.cc: avoid narrowing int to uint32.

BUG=2700
R=andrew@webrtc.org, fischman@webrtc.org, henrik.lundin@webrtc.org, henrike@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/5019004

Patch from Victor Costan <costan@gmail.com>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5248 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-09 21:34:30 +00:00
andrew@webrtc.org
de7c9e8884 Ensure WEBRTC_MODULE_UTILITY_VIDEO is undefined for enable_video==0. 
Move the logic to common.gypi to reduce the chance of the define being
unprotected in the future.

BUG=b/12018143
TESTED=git try, and local Linux build with -Denable_video=0
R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/5309004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5244 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-09 16:23:00 +00:00
sergeyu@chromium.org
5e13ac967b Add shape in DesktopFrame. 
The shape will be used for Me2App mode in chromoting.

R=wez@chromium.org

Review URL: https://webrtc-codereview.appspot.com/4369005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5243 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-07 01:03:28 +00:00
andrew@webrtc.org
8d0ca7f5d2 Add new method to MockAudioProcessing. 
TBR=henrikg

Review URL: https://webrtc-codereview.appspot.com/5279004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5241 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-06 17:52:27 +00:00
henrikg@webrtc.org
863b536100 Allow opening an AEC dump from an existing file handle. 
This is necessary for Chromium to be able enable the dump with the sanbox enabled. It will open the file in the browser process and pass the handle to the render process.

This changes FileWrapper to deal with the case were the file handle is not managed by the wrapper.

BUG=2567
R=andrew@webrtc.org, henrika@webrtc.org, perkj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4649004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5239 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-06 16:05:17 +00:00
sprang@webrtc.org
88615f0659 Fix use of uninitialized memory in RtpSenderTest::StreamDataCountersCallbacks 
BUG=
R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/5229004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5236 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-06 13:16:44 +00:00
asapersson@webrtc.org
96a9b2dcdc Use the RTT from RtcpRttStats class if provided when sending/receiving NACK. 
R=holmer@google.com

Review URL: https://webrtc-codereview.appspot.com/5049004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5228 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-05 15:06:56 +00:00
sprang@webrtc.org
ebad765ee0 Add callbacks for send channel rtp statistics 
BUG=2235
R=mflodman@webrtc.org, pbos@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4449004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5227 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-05 14:29:02 +00:00
stefan@webrtc.org
0a3c1471b8 Add API to query video engine for the send-side delay. 
R=mflodman@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4559005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5225 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-05 14:05:07 +00:00
sprang@webrtc.org
a6ad6e5b58 Add callbacks for send channel rtcp statistics 
BUG=2235
R=mflodman@webrtc.org, pbos@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4429004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5220 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-05 09:48:44 +00:00
stefan@webrtc.org
c4726d06fa Make RTPSender::SendPadData public. 
R=andresp@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/5029004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5219 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-05 09:16:33 +00:00
andrew@webrtc.org
3d9981d58a Remove unused ThreadData struct. 
TBR=bjornv

Review URL: https://webrtc-codereview.appspot.com/4949004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5216 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-04 17:13:47 +00:00
sprang@webrtc.org
71f055fb41 Add send frame rate statistics callback 
BUG=2235
R=mflodman@webrtc.org, pbos@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4479005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5213 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-04 15:09:27 +00:00
stefan@webrtc.org
79b63206b9 Fixes a crash in fullstack tests introduced with r5209. 
TBR=mflodman@webrtc.org
BUG=1812

Review URL: https://webrtc-codereview.appspot.com/4689005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5211 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-04 13:34:28 +00:00
henrik.lundin@webrtc.org
b477fa6d21 Small fixes to plot_neteq_delay.m 
Fixing problems with wrap-arounds and other small things. Adding an
extra output value.

Review URL: https://webrtc-codereview.appspot.com/4929004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5210 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-04 12:28:47 +00:00
stefan@webrtc.org
7e9315b42e Adds support for sending redundant payloads over RTX. 
TEST=trybots
BUG=1812
R=mflodman@webrtc.org, pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4169004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5209 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-04 10:24:26 +00:00
henrik.lundin@webrtc.org
9523b55826 Fix a typo in neteq.gypi 
This CL is for NetEq3. The #define for iSAC-fb was wrong on one
line. It did not affect the defualt use case, but resulted in
errors if 48 kHz mode was enabled.

TBR=tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4909004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5208 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-04 08:24:49 +00:00
andrew@webrtc.org
d7696c4ed1 Compile-out functions only used by the bit-exact test. 
Causes errors on platforms where the test is unused.

TBR=bjornv

Review URL: https://webrtc-codereview.appspot.com/4869004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5207 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-03 23:39:16 +00:00
fischman@webrtc.org
d3865e9124 Don't HANDLE_EINTR(close). Use IGNORE_EINTR(close). 
It is incorrect to wrap close in HANDLE_EINTR on Linux.

BUG=chromium:269623
R=fischman@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4759004

Patch from Mark Mentovai <mark@chromium.org>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5206 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-03 19:10:20 +00:00
solenberg@webrtc.org
812dd11f8c Add baseline generation/verification to BWE test framework. 
Updating resource file separately, once LGTM. Generates ~628k of files for current tests, highly compressable, once/if we need that.

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

Review URL: https://webrtc-codereview.appspot.com/4639004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5204 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-03 15:11:14 +00:00
sprang@webrtc.org
499631c1e4 Utility class for reading/writing network-byte-ordered integers. 
BUG=
R=holmer@google.com, mflodman@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/2151008

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5203 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-03 13:22:48 +00:00
sprang@webrtc.org
37968a9be7 Change BitrateStats to more generalized RateStatistics 
BUG=2656
R=holmer@google.com, mflodman@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4559004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5202 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-03 10:31:59 +00:00
henrik.lundin@webrtc.org
5ecdef11cc Do not use recursive calling in NetEq test tools 
This CL removes recursive calling in:
- NETEQTEST_DummyRTPpacket::readFromFile,
- NETEQTEST_RTPpacket::readFromFile.

The files currently exist for both NetEq3 and NetEq4, and all are
changed with this CL.

R=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4699004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5200 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-12-03 08:26:49 +00:00
tina.legrand@webrtc.org
8418e9696b Fixing NetEq tests for new Opus version 
The new version of Opus doesn't generate the same number of bytes encoding the test vectors in audio_decoder_unittest. Therefore the test was updated not to check the length of the encoded packet, to prepare for the coming roll of Opus. Same change was applied to iSAC, which can also generate different number of bytes on different platforms.

BUG=1459
R=henrik.lundin@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4609004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5195 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-29 09:30:43 +00:00
bjornv@webrtc.org
bd41a84694 This CL adds an API to enable robust validation of delay estimates. 
Added is
- a member variable for turning robust validation on and off.
- API to enable/disable feature.
- API to check if enabled.
- unit tests for these APIs.

Not added is
- the actual functionality (separate CL), hence turning feature on/off has no impact currently.
- calls in AEC and AEC, where the delay estimator is used. This is also done in a separate CL when we know if it should be turned on in both components.

TESTED=trybots, module_unittest
BUG=
R=aluebs@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4609005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5191 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-28 14:58:35 +00:00
stefan@webrtc.org
b627f676b3 Fixes a crash in the pacer where it fails to find a normal prio packet if there are no high prio packets, given that the queue has grown too large. 
BUG=2682
R=mflodman@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4599005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5190 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-28 14:00:09 +00:00
bjornv@webrtc.org
d1a1c353ac Recommit CL5184 
TBR=aluebs@webrtc.org
BUG=

Review URL: https://webrtc-codereview.appspot.com/4599004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5187 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-28 11:45:05 +00:00
solenberg@webrtc.org
c8f76ddc19 Refactor Remote Estimators Test into a more reusable form. 
BUG=
R=andresp@webrtc.org, mflodman@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4359004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5186 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-28 10:11:35 +00:00
bjornv@webrtc.org
82eb3a690e Revert 5184 "Small refactoring change in delay_estimator." 
> Small refactoring change in delay_estimator.
> 
> This CL produce the bit exact output and is a preparing step for an upcoming robust validation scheme.
> 
> TESTED=trybots, module_unittest
> BUG=None
> R=aluebs@webrtc.org
> 
> Review URL: https://webrtc-codereview.appspot.com/4549004

TBR=bjornv@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4589004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5185 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-28 09:44:47 +00:00
bjornv@webrtc.org
eea079a376 Small refactoring change in delay_estimator. 
This CL produce the bit exact output and is a preparing step for an upcoming robust validation scheme.

TESTED=trybots, module_unittest
BUG=None
R=aluebs@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4549004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5184 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-28 07:59:04 +00:00
stefan@webrtc.org
19a40ff05b Ensure that no packet stays in the pacer queue for longer than 2 seconds. 
BUG=2682
TEST=trybots
R=mflodman@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4519004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5182 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-27 14:16:20 +00:00
sprang@webrtc.org
4070935f4f Implement and test EncodedImageCallback in new ViE API. 
R=mflodman@webrtc.org, pbos@webrtc.org, stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4059004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5179 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-26 11:41:59 +00:00
mikhal@webrtc.org
d89b52af80 Faster implementation of BitRateStats. 
Landing cl for hguihot.

At high bitrate, EraseOld() could account for a significant part of
the total CPU usage on certain platforms. The new implementation
eliminates per-packet memory allocations and records the number of
bytes in buckets (one bucket per millisecond in window).

R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4269004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5175 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-25 17:49:28 +00:00
stefan@webrtc.org
47fadba750 Add include stdlib.h to files using abs. 
abs function is declared in stdlib.h

Committing for alextaran@chromium.org.
Reviewed here: http://review.webrtc.org/4239004/

TBR=mflodman@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4349004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5170 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-25 12:03:56 +00:00
fbarchard@google.com
b5bc098e20 Clear empty video frames in unittest so DrMemory will allow them to be read without an uninitialized read error. 
BUG=libyuv:263
TESTED=drmemory out\Debug\modules_unittests.exe --gtest_filter=*PreprocessorLogic
R=kjellander@webrtc.org, mflodman@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4319004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5168 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-25 09:06:33 +00:00
turaj@webrtc.org
5272eb8d83 Don't register iSAC-swb and iSAC-fb in NetEqDecodingTest. 
Android bots break due to r5164. This CL patches that issue.

BUG=
TEST=modules_unittests on local device.
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4309004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5166 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-23 00:11:32 +00:00
sergeyu@chromium.org
e839da02c1 Fix MouseCursor to MouseCursorShape conversion in ScreenCapturerWin. 
BUG=crbug.com/322596
R=dcaiafa@chromium.org, wez@chromium.org

Review URL: https://webrtc-codereview.appspot.com/4279005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5165 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-22 20:39:16 +00:00
turaj@webrtc.org
78b41a09bd Fix issues with sequence number wrap-around in jitter statistics. 
Related CL for NetEq 3 is https://code.google.com/p/webrtc/source/detail?r=5150

Jitter statistics was not very sensitive to timestamp warp-around, and NetEqDecodingTest.TimestampWrap *DID NOT* fail before fixes applied. However, we still keep the test.

The criteria for the tests are not satisfied for first few packets, before any wrap-around happens. We could either relax the bound or ignore the first few packets. We chose the latter.

BUG=2662
TEST=modules_unittests,trybots
R=tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4219004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5164 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-22 20:27:07 +00:00
turaj@webrtc.org
1e8c93c953 Distinguish instances of ACM1 from ACM2 by a version string. This is fpr testing purposes and will be removed when the experiment is done and ACM1 is fade out. 
BUG=
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4069006

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5161 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-22 17:04:49 +00:00
pbos@webrtc.org
2ffb149c2c Replace VideoFrameI420 with I420VideoFrame. 
Gives one less struct/class for I420 video frames.

BUG=2657
R=mflodman@webrtc.org, wu@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4149004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5160 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-22 13:10:13 +00:00
andrew@webrtc.org
b0ed8f8a08 Don't reset the AEC filter in extended mode. 
I don't believe I've witnessed this "feature" ever provide a benefit,
and have now collected some evidence of its harm when using the
extended filter mode. It can cause erroneous resets in two cases:

1. Some preprocessing noise suppression is enabled in the system (i.e.
"audio enhancements") that push the noise floor very low, possibly to
zero. If the filter is non-zero this condition can be triggered very
easily, and erroneously.

2. Non-zero energy in the filter before the peak impulse response can
cause a slight (and harmless) "pre-echo" in the error signal. This
becomes more significant as the peak is set further back in the filter.
This effect can cause needless resets during echo onsets.

In short, this isn't a great criterion for filter reset and has the
potential to cause serious harm. Ideally we would remove it entirely,
but in the interests of safety, can start with the extended mode.

BUG=1261
R=aluebs@webrtc.org, bjornv@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/3959004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5159 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-22 06:39:42 +00:00
stefan@webrtc.org
ef2d55461b Increase size of pacer window to 500 ms as that better matches the encoder. 
BUG=1812
R=andresp@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4129006

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5154 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-21 14:37:11 +00:00
pbos@webrtc.org
ffe1b17b57 Lock access to ModuleRtpRtcpImpl::simulcast_. 
Fixes race between RegisterSendPayload and SendOutgoingData.

BUG=
R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4099006

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5152 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-21 13:53:13 +00:00
henrik.lundin@webrtc.org
6f6ba6edee Fix issues with sequence number wrap-around in jitter statistics 
Wrap-arounds in sequence numbers (and in timestamps) were not always
treated correctly. This is fixed by introducing two helper functions
for correct comparisons, and by casting to the right word size.

Also added a new member variable to AutomodeInst_t. The new member keeps
track of when the first packet has been registered in the automode code.
This was previously done implicitly (and not very good) using the fact
that the lastSeqNo and lastTimestamp members were initialized to zero.

Two new unit tests were added as part of this CL.
NetEqDecodingTest.SequenceNumberWrap was failing before the fixes were
made; now it is ok.

BUG=2654
R=tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/4139004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5150 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-21 13:17:29 +00:00
asapersson@webrtc.org
8d02f5dc71 Added API for enabling/disabling RTCP Receiver Reference Time extension. 
R=mflodman@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/3419005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5147 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-21 08:57:04 +00:00
henrike@webrtc.org
a750044396 Fixes a crash in VoE when unregistering JNI hooks. 
BUG=11695087
R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/3939004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5144 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-20 22:32:12 +00:00
asapersson@webrtc.org
1ae1d0c471 Add possibility to get the last processed RTT from the call stats class (to be used by RTP/RTCP module). 
R=mflodman@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/2383004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5139 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-20 12:46:11 +00:00
aluebs@webrtc.org
0b72f5863b Add experimental noise suppression dummy API. 
Add this flag to the voe_cmd_test.

R=andrew@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/3879004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5134 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-19 15:17:51 +00:00
sergeyu@chromium.org
5d85819dd2 Fix DesktopAndCursorComposer to restore frames to the original state. 
Screen capturers may reuse frame buffers and they expect that the
frame content isn't changed by the frame consumer.
DesktopAndCursorComposer draws mouse cursor on generated frames and
it was releasing the frames with the mouse cursor on them. Fixed
it to restore frame content erasing mouse cursor before returning
desktop frames.

BUG=crbug.com/316297
R=wez@chromium.org

Review URL: https://webrtc-codereview.appspot.com/3899004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5133 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-19 02:15:47 +00:00
turaj@webrtc.org
7a05ae5c69 Adding back main() to the test. Now it is possible to choose between ACM1 and ACM2, furthermore, the test can simulate a channel with packet loss and FEC can be activated. Packet loss pattern is based on channel implementation in Channel{.cc,.h}, which currently is a determenistic pattern with 1 every 3rd packet is discarded. 
The main() was deleted in r4731.

BUG=
R=andrew@webrtc.org, minyue@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/2370004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@5132 4adac7df-926f-26a2-2b94-8c16560cd09d
 2013-11-18 18:16:53 +00:00