Always use the PF_SRTP_BYPASS when sending RTP/RTCP packets.
Change the condition in BaseChannel::GetSrtpOverhead.
Get the SRTP overhead when using either SDES or DTLS-SRTP.
Bug: None
Change-Id: I44aeff8b75e56b12acefd73299a95a3e38cd401b
Reviewed-on: https://webrtc-review.googlesource.com/6580
Reviewed-by: Taylor Brandstetter <deadbeef@webrtc.org>
Commit-Queue: Zhi Huang <zhihuang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20152}
Some applications enable logging before using PeerConnectionFactory.
Not loading native library in Logging.java broke these applications.
TBR=magjed@webrtc.org
Bug: webrtc:7474, b/67419105
Change-Id: I5984d2241cfb76e0edb5b5da0974c8693bf50603
Reviewed-on: https://webrtc-review.googlesource.com/6600
Reviewed-by: Sami Kalliomäki <sakal@webrtc.org>
Commit-Queue: Sami Kalliomäki <sakal@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20151}
The comment was updated here:
https://webrtc-review.googlesource.com/c/src/+/6200
But it should really refer to PeerConnectionFactory.initialize, which
is the new approved way of initializing webrtc in Java.
NOTRY=True
TBR=sakal@webrtc.org
Bug: None
Change-Id: I7b57f5abf926a1437bd61d6592ba236eb30ee9c5
Reviewed-on: https://webrtc-review.googlesource.com/6560
Commit-Queue: Taylor Brandstetter <deadbeef@webrtc.org>
Reviewed-by: Taylor Brandstetter <deadbeef@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20150}
It was not producing any results on presubmit for 2 reasons:
- The list of modified GN files contained relative paths but the list
of all GN files contains absolute paths.
- The root directory was not passed to the script and the path of the
first file substituted it.
Fix potential problem with using unescaped names in a regex.
Blacklist testdata directory with intentionally bad BUILD.gn files.
NOTRY=True
Bug: webrtc:6954
Change-Id: Ib0b845b9440b594960bc8a656e8a84d2ccb4a684
Reviewed-on: https://webrtc-review.googlesource.com/5981
Reviewed-by: Edward Lemur <ehmaldonado@webrtc.org>
Reviewed-by: Henrik Kjellander <kjellander@webrtc.org>
Commit-Queue: Oleh Prypin <oprypin@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20149}
Needed because swarming adds --isolated-script-test-output and --isolated-script-test-perf-output
See for example:
https://chromium-swarm.appspot.com/task?id=39006c763bebf710
No-Try: true
Bug: chromium:755660
Change-Id: Iff9fb3441200f760c511a67211fbc4a1272717b4
Reviewed-on: https://webrtc-review.googlesource.com/6362
Reviewed-by: Henrik Kjellander <kjellander@webrtc.org>
Reviewed-by: Oleh Prypin <oprypin@webrtc.org>
Commit-Queue: Edward Lemur <ehmaldonado@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20146}
This renames I420BufferImpl to JavaI420Buffer and moves it as part of
the API.
Bug: webrtc:7749
Change-Id: I70726f248ba4601b4922996712bdfdafbfa4a1e1
Reviewed-on: https://webrtc-review.googlesource.com/5381
Commit-Queue: Magnus Jedvert <magjed@webrtc.org>
Commit-Queue: Sami Kalliomäki <sakal@webrtc.org>
Reviewed-by: Magnus Jedvert <magjed@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20145}
This is a reland of f4898a650954691d79bbc146d5b454fb5e67ec47
Original change's description:
> Don't download PESQ and POLQA in the low_bandwidth_audio_test.py script.
>
> They should've been downloaded already.
>
> NOTRY=True
>
> Bug: chromium:755660
> Change-Id: I8ecb355f06026a38bd9377633e2be6c55d7c6452
> Reviewed-on: https://webrtc-review.googlesource.com/5620
> Commit-Queue: Edward Lemur <ehmaldonado@webrtc.org>
> Reviewed-by: Oleh Prypin <oprypin@webrtc.org>
> Reviewed-by: Henrik Kjellander <kjellander@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#20109}
No-Try: true
Bug: chromium:755660
Change-Id: I391130545eee5d4928101f87ac4a4e0945d665a1
Reviewed-on: https://webrtc-review.googlesource.com/6380
Commit-Queue: Edward Lemur <ehmaldonado@webrtc.org>
Reviewed-by: Henrik Kjellander <kjellander@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20140}
A script for producing boxplots by parsing data generated by the
apm_quality_assessment.py tool.
The script groups data by the values of one or several audioproc_f
parameters. For every such subgroup it draws a boxplot. All boxplots
are shown next to each other with the parameter value as the x axis.
It is similar to this matplotlib example:
https://matplotlib.org/mpl_examples/pylab_examples/boxplot_demo_06.png
The script
1. reads config file names from the pandas dataframe generated by
quality_assurance.collect_data
2. parses the (JSON) config files to read the parameter values
3. groups data with matching param values together
4. draws a boxplot for each group using matplotlib
TBR=alessiob@webrtc.org # reviewed already in old gerrit https://chromium-review.googlesource.com/c/external/webrtc/+/660559
BUG: webrtc:7218
Change-Id: I380a1363d26721feb975fad1506835c622e9d926
Reviewed-on: https://webrtc-review.googlesource.com/6340
Reviewed-by: Alex Loiko <aleloi@webrtc.org>
Commit-Queue: Alex Loiko <aleloi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20139}
This is a reland of 34cdd2d402b08aee4e17a6fd38c87e0e5cd7aa30
Original change's description:
> Remove AudioDeviceObserver and make ADM not inherit from the Module interface.
>
> (Re-upload of https://codereview.webrtc.org/3020493002/)
>
> Bug: webrtc:4690, webrtc:7306
> Change-Id: I67fb9ebca1296aabc08eae8a292a5c69832dc35e
> Reviewed-on: https://webrtc-review.googlesource.com/5360
> Commit-Queue: Fredrik Solenberg <solenberg@webrtc.org>
> Reviewed-by: Henrik Andreassson <henrika@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#20083}
Bug: webrtc:4690, webrtc:7306
Change-Id: Ib019439fe6ab0e6b759819e1e9bd320ba1d983bd
Reviewed-on: https://webrtc-review.googlesource.com/6300
Commit-Queue: Fredrik Solenberg <solenberg@webrtc.org>
Reviewed-by: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20137}
This is needed in the general case, now that we aim to support codecs
other than those built-in to WebRTC.
BUG=webrtc:8159
Change-Id: I40a41252bf69ad5d4d0208e3c1e8918da7394706
Reviewed-on: https://webrtc-review.googlesource.com/5380
Commit-Queue: Karl Wiberg <kwiberg@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20136}
We need to support two modes of writing to the output:
1. Current way - the application lets lets WebRTC know which file to write to, and WebRTC is then in charge of the writing.
2. New way - the application would receive indications from WebRTC about (encoded) RTC events, and would itself be in charge of processing them (be it writing it to a file, uploading it somewhere, etc.).
We achieve this by creating an interface for output - RtcEventLogOutput. By providing an instance of the subclass, RtcEventLogOutputFile, the old behavior is achieved. The subclass of the new behavior is to be added by a later CL.
TBR=stefan@webrtc.org
Bug: webrtc:8111
Change-Id: I9c50521a7f7144d86d8353a65995795862e19c44
Reviewed-on: https://webrtc-review.googlesource.com/2686
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Reviewed-by: Elad Alon <eladalon@webrtc.org>
Reviewed-by: Björn Terelius <terelius@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20135}
The first argument of FindMatchingCodec is supposed to be the list that
contains the codec to be found, specifically to handle RTX codecs that
point to other codecs. But this wasn't being done everywhere, and wasn't
noticed because *most of the time* it just results in adding the RTX
codec in a different location, which isn't an issue.
But, it's still not standards-compliant. And it sometimes is an issue
when talking to older endpoints.
Adding a regression test, and DCHECK in FindMatchingCodec to ensure this
doesn't happen by accident again.
Bug: webrtc:8332
Change-Id: I5def056b245c6d00a49a59d429f1dee303fb7cef
Reviewed-on: https://webrtc-review.googlesource.com/6240
Reviewed-by: Justin Uberti <juberti@webrtc.org>
Commit-Queue: Taylor Brandstetter <deadbeef@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20130}
This method allows the client to get details about the SSL
certificate sent by the remote side in the DTLS handshake.
This functionality in this new method has been standardized in the
RTCDtlsTransport, but until we have that implemented we wish to
expose this functionality so clients do not need to depend on
WebRtcSession.
Bug: webrtc:8323
Change-Id: Ic964266dd7e734cec07289a147fd8d090d74ce6b
Reviewed-on: https://webrtc-review.googlesource.com/5641
Reviewed-by: Taylor Brandstetter <deadbeef@webrtc.org>
Commit-Queue: Steve Anton <steveanton@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20129}
This CL changed the behavior of the logging APIs:
https://webrtc-review.googlesource.com/c/src/+/4160
They now no longer automatically load the native library. So the
comment needs to be updated. This change broke at least one application
that was trying to enable logging before anything else.
NOTRY=True
TBR=sakal@webrtc.org
Bug: None
Change-Id: I845aa8de2eebf3687884fbf69668f56e11a3d93c
Reviewed-on: https://webrtc-review.googlesource.com/6200
Reviewed-by: Taylor Brandstetter <deadbeef@webrtc.org>
Commit-Queue: Taylor Brandstetter <deadbeef@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20127}
SignalProcessingUtils.MixSignals() now allows different padding options.
This CL also adds more unit tests for SignalProcessingUtils.MixSignals().
Bug: webrtc:7494
Change-Id: Id62fe9998e512c275cb6399e0aedf11f23a9f36e
Reviewed-on: https://webrtc-review.googlesource.com/5780
Commit-Queue: Alessio Bazzica <alessiob@webrtc.org>
Reviewed-by: Alex Loiko <aleloi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20122}
RegisterPacketFeedbackObserver signals congestion controller object is
ready to process incoming packet, thus call it as last statement in the constructor
Bug: webrtc:8325
Change-Id: I31d8ab04c568e639db12c97b649c2d50a489ce24
Reviewed-on: https://webrtc-review.googlesource.com/5860
Reviewed-by: Niels Moller <nisse@webrtc.org>
Reviewed-by: Fredrik Solenberg <solenberg@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20121}
The two timestamps, rtp time and corresponding system time, are always
used together, for audio/video sync. The new method reads both
timestamps, without releasing a lock in between. Ensures that the
caller gets values corresponding to the same packet.
Bug: webrtc:7135
Change-Id: I25bdcbe9ad620016bfad39841b339c266efade14
Reviewed-on: https://webrtc-review.googlesource.com/4062
Commit-Queue: Niels Moller <nisse@webrtc.org>
Commit-Queue: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Fredrik Solenberg <solenberg@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20120}
RtcEventLogImpl no longer hard-codes the way encoding is done. It now relies on RtcEventEncoder for it. This gives two benefits:
1. We can decide between the current encoding and the new encoding (which is still WIP) without code duplication (no need for RtcEventLogImplNew).
2. Encoding is done only when the event needs to be written to a file. This both avoids unnecessary encoding of events which don't end up getting written to a file, as well as is useful for the new, delta-based encoding, which is stateful.
BUG=webrtc:8111
Change-Id: I9517132e5f96b8059002a66fde8d42d3a678c3bb
Reviewed-on: https://webrtc-review.googlesource.com/1365
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Björn Terelius <terelius@webrtc.org>
Commit-Queue: Stefan Holmer <stefan@webrtc.org>
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20118}
We're moving to an RtcEventLog interface that accepts std::unique_ptr<EventLog> and stores the event for encoding when encoding becomes necessary, rather than before. This will be useful while we maintain the legacy (current) encoding alongside the new encoding on which we're working.
This CL introduces RtcEventLogEncoderLegacy, which takes provides the encoding currently done by RtcEventLogImpl. After this, we can modify RtcEventLogImpl to use a dynamically chosen encoding, allowing us to easily choose between the current encoding and the new one on which we're working.
BUG=webrtc:8111
TBR=stefan@webrtc.org
Change-Id: I3dde7e222a40a117549a094a59b04219467f490a
Reviewed-on: https://webrtc-review.googlesource.com/1364
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Reviewed-by: Elad Alon <eladalon@webrtc.org>
Reviewed-by: Björn Terelius <terelius@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20116}
We're moving to an RtcEventLog interface that accepts std::unique_ptr<EventLog> and stores the event for encoding when encoding becomes necessary, rather than before. This will be useful while we maintain the legacy (current) encoding alongside the new encoding on which we're working.
This CL adds the internals of RtcEvent's subclasses - the actual data that they keep. (Work on this was broken down into several CLs in order to make reviewing easier.)
BUG=webrtc:8111
Change-Id: I402c9c64bffef6a5a6d227bde5da0fd3152daba1
Reviewed-on: https://webrtc-review.googlesource.com/1362
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Reviewed-by: Björn Terelius <terelius@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20108}
This reverts commit b7239a9dc825ddb25dbc800aed3a065163b9a10e.
Reason for revert: Broke chromium mac build, compilation failures on content/renderer/media/webrtc/webrtc_video_frame_adapter.h.
Original change's description:
> Make rtc_base/refcount.h self contained, not including refcountedobject.h.
>
> The refcount.h file doesn't depend on anything from
> refcountedobject.h. The motivation of this change to make it possible
> to add additional declarations to refcount.h, and include it from
> refcountedobject.h.
>
> Bug: webrtc:8270
> Change-Id: I24f6131f471e675570968d00065ff9b1f55e3373
> Reviewed-on: https://webrtc-review.googlesource.com/5760
> Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
> Commit-Queue: Niels Moller <nisse@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#20106}
TBR=kwiberg@webrtc.org,nisse@webrtc.org
Change-Id: I7334597cc8979ba9cfaff526967084349ef27f3c
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:8270
Reviewed-on: https://webrtc-review.googlesource.com/5800
Reviewed-by: Niels Moller <nisse@webrtc.org>
Commit-Queue: Niels Moller <nisse@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20107}
The refcount.h file doesn't depend on anything from
refcountedobject.h. The motivation of this change to make it possible
to add additional declarations to refcount.h, and include it from
refcountedobject.h.
Bug: webrtc:8270
Change-Id: I24f6131f471e675570968d00065ff9b1f55e3373
Reviewed-on: https://webrtc-review.googlesource.com/5760
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Commit-Queue: Niels Moller <nisse@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20106}
This CL is a clean-up to prepare for adding more supported codecs for the internal H264 SW codec.
Bug: webrtc:8317
Change-Id: If483d05c01c40bbc81cbd1a6aad89961689714ef
Reviewed-on: https://webrtc-review.googlesource.com/4940
Reviewed-by: Sami Kalliomäki <sakal@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Commit-Queue: Magnus Jedvert <magjed@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20105}
