1309 Commits

Author SHA1 Message Date
xians@webrtc.org
233c58de47 Landing 1399004, Minor clean up on the un-used _measureDelay code
Those code is/will never used, removing it makes the code better.



git-svn-id: http://webrtc.googlecode.com/svn/trunk@3961 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-06 11:52:47 +00:00
mikhal@webrtc.org
45f2da0920 VCM/JB: Porting jitter_buffer_test to gtest.
Tests were not modified, but ported as is.

R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1391004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3956 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-03 22:22:46 +00:00
andrew@webrtc.org
a31c428307 Remove 44.1 kHz workaround from AudioDevice on PulseAudio.
We currently inform VoE that 44.1 kHz audio is 44 kHz. We now have arbitrary
resampling in VoE, allowing us to pass in the native 44.1 kHz.

Our ALSA interface always requires 48 kHz, allowing ALSA to handle resampling.

This also removes WEBRTC_PA_GTALK which was not defined anywhere.

BUG=webrtc:1395
TESTED=Using 44.1 for capture and render in loopback, ran through all codec channel/rate combinations. Quality is good. Testing AEC was difficult as I can't find a way to change the sample rate of an individual device in PulseAudio. Using a webcam at 32 kHz, other problems were the overriding contribution to quality degradation (delay issues, possible clock drift from the camera). At least I verified that the quality got no worse with this patch.
R=xians@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1384004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3955 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-03 19:01:46 +00:00
andrew@webrtc.org
7cb766b016 Remove 44.1 kHz workaround from AudioDevice on WASAPI.
We currently inform VoE that 44.1 kHz audio is 44 kHz. We now have arbitrary
resampling in VoE, allowing us to pass in the native 44.1 kHz.

BUG=webrtc:1395
TESTED=Set capture device to 44.1 and render device to 48 and vice versa and observed good AEC. The quality is considerably worse before this change. Using 44.1 for capture and render in loopback, ran through all codec channel/rate combinations. Quality is good.
R=xians@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1383004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3954 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-03 18:56:38 +00:00
sergeyu@chromium.org
bd4a2feddb Fix off-by-one buffer overflow in WebRtcNetEQ_PacketBufferInsert().
BUG=1725
R=tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1395004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3953 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-03 18:11:36 +00:00
mikhal@webrtc.org
d3cd565ecf VCM: Updating receiver logic
R=stefan@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1363005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3952 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-03 17:54:18 +00:00
pbos@webrtc.org
77f6b2175e Revert 3934 "Revert 3933 "Remove traces of deprecated WebRtc_Wor..."
> Revert 3933 "Remove traces of deprecated WebRtc_Word types."
> 
> > Remove traces of deprecated WebRtc_Word types.
> > 
> > BUG=314
> > R=tommi@webrtc.org
> > 
> > Review URL: https://webrtc-codereview.appspot.com/1385004
> 
> TBR=pbos@webrtc.org
> 
> Review URL: https://webrtc-codereview.appspot.com/1386004

TBR=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1397004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3948 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-03 12:02:11 +00:00
tina.legrand@webrtc.org
d5726a1286 Formatting ACM tests
Pure formatting of all files located in /webrtc/modules/audio_coding/main/test/

Smaller manual modifications done after using Eclipse formatting tool, like wrapping long lines (mostly comments).

BUG=issue1024

Review URL: https://webrtc-codereview.appspot.com/1342004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3946 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-03 07:34:12 +00:00
pwestin@webrtc.org
03efc89151 Fix when SetMinimumPlayoutDelay is configured to 0
BUG=1720
R=turaj@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1386005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3942 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-02 21:19:12 +00:00
pwestin@webrtc.org
52b4e8871a Adding trace and changing pacing constants
BUG=1721,1722
R=mikhal@webrtc.org, niklas.enbom@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1380005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3940 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-02 19:02:17 +00:00
pbos@webrtc.org
68e5a68f07 Revert 3933 "Remove traces of deprecated WebRtc_Word types."
> Remove traces of deprecated WebRtc_Word types.
> 
> BUG=314
> R=tommi@webrtc.org
> 
> Review URL: https://webrtc-codereview.appspot.com/1385004

TBR=pbos@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1386004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3934 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-02 09:30:12 +00:00
pbos@webrtc.org
265a5d298a Remove traces of deprecated WebRtc_Word types.
BUG=314
R=tommi@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1385004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3933 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-05-02 09:11:20 +00:00
andrew@webrtc.org
342353780d Consolidate common_audio into a single target.
In principle should reduce gyp processing time, but the difference was not measurable. In any case, it's a good simplification that aligns with having a single common_video target.

R=bjornv@webrtc.org, kma@webrtc.org, tina.legrand@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1375004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3928 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-30 23:43:26 +00:00
andrew@webrtc.org
dff69c56b0 Add AEC suppression level option to audioproc.
TBR=bjornv

Review URL: https://webrtc-codereview.appspot.com/1368007

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3927 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-30 23:01:09 +00:00
stefan@webrtc.org
4980679d35 Fixes two bugs in receive statistics.
- Reported bitrate wasn't reset correctly when no frames had been received.
- Internal framerate estimate wasn't reset when no frames had been received.

BUG=1713

Review URL: https://webrtc-codereview.appspot.com/1377004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3924 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-30 22:05:07 +00:00
pwestin@webrtc.org
d35964a1ce Fixing AV sync.
Increased 2 const to allow for a bigger difference in AV sync.

BUG=1711

Re-wrote the ComputeDelays to be readable and remove the possibilities of returning values lower than base_target_delay_ms

R=mflodman@webrtc.org, mikhal@webrtc.org, niklas.enbom@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1367004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3922 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-30 16:06:10 +00:00
mikhal@webrtc.org
6faba6edc9 VCM: Setting buffering delay in timing
Review URL: https://webrtc-codereview.appspot.com/1338004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3921 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-30 15:39:34 +00:00
solenberg@webrtc.org
8efc623fc2 Apply Chromium C++ style to RemoteRateControl.
BUG=

Review URL: https://webrtc-codereview.appspot.com/1329004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3919 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-30 08:33:46 +00:00
sergeyu@chromium.org
15e32ccd30 Add DesktopCapturer interface for desktop capturers.
The new DesktopCapturer interface will be used for screen and window
captures. Beside DesktopCapturer itself also added classes/interfaces
that it depends on.

R=alexeypa@chromium.org, tommi@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1322007

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3917 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-29 20:10:57 +00:00
mikhal@webrtc.org
865ada3a52 Don't reset the last je value and mode
Review URL: https://webrtc-codereview.appspot.com/1369004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3916 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-29 19:09:41 +00:00
stefan@webrtc.org
5b7120c81b Fix two issues where we might end up busy looping in decoder_render mode.
This happens if
- Next frame is far into the future (> 200 ms).
- Next frame is ready for decode/render but incomplete.

BUG=1696
TESTS=trybots

Review URL: https://webrtc-codereview.appspot.com/1354005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3914 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-29 16:41:30 +00:00
pwestin@webrtc.org
b0061f94b2 Enable Nack pacing.
Review URL: https://webrtc-codereview.appspot.com/1357004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3912 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-27 00:41:08 +00:00
andrew@webrtc.org
1acb3b33bc Add comfort noise disabling and routing mode selection to audioproc.
TBR=bjornv

Review URL: https://webrtc-codereview.appspot.com/1358004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3907 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-26 00:39:27 +00:00
mikhal@webrtc.org
381da4be9c VCM: Adding API for the size(duration) of the jitter buffer.
Refers to the duration in time of the frames which are ready to be sent to the decoder.

Review URL: https://webrtc-codereview.appspot.com/1319004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3903 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-25 21:45:29 +00:00
mikhal@webrtc.org
8392cd9edd VCM/JB: Using last decoded state for waiting for key
relanding 1323006

BUG=

Review URL: https://webrtc-codereview.appspot.com/1354004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3902 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-25 21:30:50 +00:00
mikhal@webrtc.org
dc3cd217b2 VCM/JB: FrameForDecoding->IncompleteFrameForDecoding
- Update complete frame for decoding
- Remove FrameForDecodingNack

This CL should only be committed after issue http://webrtc-codereview.appspot.com/1313007/

Review URL: https://webrtc-codereview.appspot.com/1316007

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3901 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-25 20:27:04 +00:00
pwestin@webrtc.org
52aa019e98 Avoid adding duplicates in pacer lists.
Review URL: https://webrtc-codereview.appspot.com/1329007

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3899 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-25 17:35:56 +00:00
stefan@webrtc.org
cb60fb2e6c Make sure timestamps are monotonically increasing.
BUG=1563

Review URL: https://webrtc-codereview.appspot.com/1325008

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3898 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-24 23:54:31 +00:00
andrew@webrtc.org
df9c0e5ec9 Revert 3892 "VCM/JB: Using last decoded state for waiting for key"
> VCM/JB: Using last decoded state for waiting for key
> 
> Review URL: https://webrtc-codereview.appspot.com/1323006

Although I have no idea why, it appears this might be causing failures in ViEStandardIntegrationTest.RunsFileTestWithoutErrors. I was unable to reproduce locally. This is a trial revert to verify. If the errors continue to happen, I will restore this change.

TBR=mikhal@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1321010

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3896 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-24 02:13:18 +00:00
mikhal@webrtc.org
1248d4effc VCM/JB: Using last decoded state for waiting for key
Review URL: https://webrtc-codereview.appspot.com/1323006

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3892 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-23 20:57:06 +00:00
stefan@webrtc.org
8ca8a71de2 Revert "Add a default RTT to CallStats and use different values for buffered/real-time mode."
This reverts commit aae26db1da5803482b094357c546b8454ab1c26d.

BUG=1613
TBR=mflodman@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1327008

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3890 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-23 16:48:32 +00:00
turaj@webrtc.org
a942692725 Buf fix for r3883.
Review URL: https://webrtc-codereview.appspot.com/1319012

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3889 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-23 16:08:29 +00:00
stefan@webrtc.org
ccd4b2aec8 Add a default RTT to CallStats and use different values for buffered/real-time mode.
BUG=1613

Review URL: https://webrtc-codereview.appspot.com/1326007

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3888 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-23 15:58:23 +00:00
pbos@webrtc.org
d25b602dc0 VP8: Avoid copying the codec struct on Reset().
BUG=

Review URL: https://webrtc-codereview.appspot.com/1319013

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3887 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-23 13:08:04 +00:00
mflodman@webrtc.org
efdf778d3f BUG=1351
Propose of this CL: Close the camera properly on MacOS in order to allow other apps to use it.

Changes in this CL:
1. video_capture_qtkit_info_objc.mm _captureDevicesInfo is never released. I have found this memory leak using Instruments from XCode. The patch is releasing it in dealloc.

2. In video_capture_qtkit_objc.h:
a) _captureDeviceName is not needed. Is allocated in the class but never used.
b) I don't see the role of the  NSAutoreleasePool. also if you use it you have to release it when the class is destroyed. Otherwise you will leak memory. Libjingle has for each thread a pool on mac os.

3. In video_capture_qtkit_objc.mm
a) the camera is not stopped properly . See the changes from dealloc. NOTE : If you don't call [[_captureVideoDeviceInput device] close] other apps will not be able to use the camera since you are not closing your app

b) Removed QTCaptureDevice* videoDevice = (QTCaptureDevice*)[_captureDevices objectAtIndex:0]; I don't know why this because the desired camera is opened in setCaptureDeviceById and can be different than position 0 in the camera array. At this moment if you have two cameras and user want to pick the one on index 1 the app also locks the one on 0 .

Other changes I have done to improve (and are not in this CL):
a) I have set the FPS properly to the desired. I have succeeded to reduce the CPU with 3 % doing this. The current code for setting FPS is commented in webrtc
b) I have removed _rLock from the equation. I don't know if it's good or not but I hadn't understood what exactly we are trying to protect with this. Anyway in the current implementation is never released.

Review URL: https://webrtc-codereview.appspot.com/1097014

Patch from Silviu Caragea <silviu.cpp@gmail.com>.

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3886 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-23 11:57:56 +00:00
mikhal@webrtc.org
c1f243f8e7 VCM/JB: Skip to the next complete key frame
Review URL: https://webrtc-codereview.appspot.com/1317006

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3885 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-22 22:24:38 +00:00
turaj@webrtc.org
28d54ab18f Improve AV-sync when initial delay is set and NetEq has long buffer.
Review URL: https://webrtc-codereview.appspot.com/1324006

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3883 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-22 18:53:35 +00:00
kjellander@webrtc.org
1b427719dc emove desktop_capture.gypi from modules.gyp
When adding this in
we started getting linking problems on the mac_asan bot due to
the empty list of source files for the library target.

Please re-add it into modules.gyp when the library has source files
to compile.

BUG=none
TEST=Passing mac_asan trybot.

Review URL: https://webrtc-codereview.appspot.com/1313009

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3882 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-22 13:13:49 +00:00
mikhal@webrtc.org
e0e029e8cb Revert 3876
Review URL: https://webrtc-codereview.appspot.com/1341005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3877 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-19 21:11:41 +00:00
mikhal@webrtc.org
ee184b9520 VCM/Receiver: Only update render time when decoding
Review URL: https://webrtc-codereview.appspot.com/1336004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3876 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-19 19:15:47 +00:00
mikhal@webrtc.org
a73d52ca52 revert r3871
TBR= solenberg@webrtc.org

Review URL: https://webrtc-codereview.appspot.com/1331004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3872 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-18 20:26:28 +00:00
solenberg@webrtc.org
9756017717 - Replace the BWE_MIN and BWE_MAX macros with std::min and std::max
- Add 'virtual' to a bunch of overridden methods of RemoteBitrateEstimatorMultiStream and RemoteBitrateEstimatorSingleStream.

BUG=

Review URL: https://webrtc-codereview.appspot.com/1324005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3871 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-18 19:12:42 +00:00
solenberg@webrtc.org
d26457fdeb Apply Chromium C++ style to BitRateStats.
BUG=

Review URL: https://webrtc-codereview.appspot.com/1325006

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3870 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-18 12:25:32 +00:00
braveyao@webrtc.org
c14b728b71 Add lock to prevent possible rare race condition in Win coreAudio capture implementation.
BUG = 
TEST = voe_auto_test

Review URL: https://webrtc-codereview.appspot.com/1320011

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3868 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-18 09:32:07 +00:00
sergeyu@chromium.org
a0cd9182aa Add desktop_capture directory for screen and window capturers.
The screen captures will be moved from chromium to WebRTC to make it easy
to share this code with other projects. This CL adds a new directory where
the current screen capturer code will be moved to.

Review URL: https://webrtc-codereview.appspot.com/1297005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3866 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-17 18:21:42 +00:00
mikhal@webrtc.org
dbd6a6d653 Updating delay for first value
BUG=

Review URL: https://webrtc-codereview.appspot.com/1327005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3865 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-17 16:23:22 +00:00
andresp@webrtc.org
48c5882f2a Remove libvpx pre-processor conditions and conditional compile of default temporal layers files.
R=stefan@webrtc.org,marpan@webrtc.org
BUG=201

Review URL: https://webrtc-codereview.appspot.com/1323005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3864 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-17 15:31:40 +00:00
tina.legrand@webrtc.org
db11fab49e Adding Opus unit test
This CL adds a unit test for Opus, as well as new APIs for true stereo decoding (skipping master/slave approach).

BUG=

Review URL: https://webrtc-codereview.appspot.com/1222006

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3860 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-17 10:39:41 +00:00
turaj@webrtc.org
f1a3b4bc0c Issue 1647. Avoid unsequenced modification.
issue=1647
test=trybots,manual

Review URL: https://webrtc-codereview.appspot.com/1327004

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3858 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-16 17:01:35 +00:00
pbos@webrtc.org
6e788df19e Remove vim/emacs modelines from .gypi files
BUG=1655

Review URL: https://webrtc-codereview.appspot.com/1326005

git-svn-id: http://webrtc.googlecode.com/svn/trunk@3857 4adac7df-926f-26a2-2b94-8c16560cd09d
2013-04-16 12:40:34 +00:00