This CL removes the constraint that freezes the filter adaptation
whenever the estimated echo or the prediction error is saturated. This
allows for much more rapid filter recovery in cases where the echo path
gain for some reason changes, such as when the analog AGC gain is
adjusted or the loudspeaker volume is changed.
TBR: devicentepena@webrtc.org
Bug: webrtc:9466,chromium:857426
Change-Id: Ic0b3b03f41f12e9a607aaadd2ee91cbaa16cac52
Reviewed-on: https://webrtc-review.googlesource.com/86124
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23775}
The Copy() function previously did not copy the logging timestamp.
To be able to use Copy() in this test, we add private copy
constructors for RtcEvents which the Copy() can use to copy
everything including the timestamp.
Also adds missing test for RtcEventAlrState,
RtcEventIceCandidatePairConfig and RtcEventIceCandidatePair.
Bug: webrtc:8111
Change-Id: I3901231735baa4e671173c921eada0a4be6de7c9
Reviewed-on: https://webrtc-review.googlesource.com/86042
Commit-Queue: Björn Terelius <terelius@webrtc.org>
Reviewed-by: Elad Alon <eladalon@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23774}
This CL adds a field trial that disables the feature that the pacer will
ignore the pacing rate and send extra fast to drain the queues if the
pacer queue starts to fill up. BBR assumes that the pacing rate will be
respected and sending more increase the risk of overestimating the
bandwidth.
Bug: webrtc:8415
Change-Id: Ibba315360dafef1c317d14a83199172f9f8cc6aa
Reviewed-on: https://webrtc-review.googlesource.com/80964
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Björn Terelius <terelius@webrtc.org>
Commit-Queue: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23773}
This CL changes the behavior when the main filter diverges.
Instead of entering non-linear mode, the AEC continues to operate in
linear mode but estimates the residual echo differently. R2 is S2
scaled by a factor of 10.
Bug: chromium:857018,webrtc:9462
Change-Id: I41212efe164ad319cf38a163cdf9d3ea151e0997
Reviewed-on: https://webrtc-review.googlesource.com/85981
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Reviewed-by: Jesus de Vicente Pena <devicentepena@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23772}
Also adds tests, and adds a bit of logging in ParseIceServers.
Bug: chromium:718508
Change-Id: Id41ccb7cccbdab5af76e380b32b4d8ba0c4a0a72
Reviewed-on: https://webrtc-review.googlesource.com/86121
Reviewed-by: Henrik Boström <hbos@webrtc.org>
Commit-Queue: Harald Alvestrand <hta@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23769}
Those are static functions in the spec, so implemented as member functions
of the PeerConnectionFactory instead.
Bug: webrtc:7577, webrtc:9441
Change-Id: Iccb24180e096e713d24e7e25ecfd5d7bbd7638f9
Reviewed-on: https://webrtc-review.googlesource.com/85341
Commit-Queue: Florent Castelli <orphis@webrtc.org>
Reviewed-by: Taylor Brandstetter <deadbeef@webrtc.org>
Reviewed-by: Steve Anton <steveanton@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23768}
The behavior of division-by-zero is undefined, so the DivisionByZeroFails test isn't correct. As we don't need any specific behavior on division-by-zero we leave the current code untouched.
Additionally, since the DivisionFailsOnLargeSize EXPECT_DEATH checks rely on DCHECKs, we only run those when DCHECKs are enabled.
Bug: webrtc:9443
Change-Id: I0fdd7be55a7bc76b4203b2f6d5cd0ed8ac5cc688
Reviewed-on: https://webrtc-review.googlesource.com/85362
Commit-Queue: Jonas Olsson <jonasolsson@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23767}
Both incoming and outgoing datachannels should cause
the DATA_ADDED flag to be set.
This CL also moves all tests into their own file, and
improves scaffolding.
Bug: chromium:718508
Change-Id: I5c4c257ccb6f26799f7593bce8b27ebf59015b1e
Reviewed-on: https://webrtc-review.googlesource.com/85348
Commit-Queue: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Henrik Boström <hbos@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23766}
this function is now only used in combination with StreamStatistician::IsRetransmitOfOldPacket
but IsRetransmitOfOldPacket internally checks if packet is in_order, thus making extra check unnecessary
In addition to making code simpler, removing this checks avoids
taking two extra CritSection on common code path of incoming rtp packet.
Bug: webrtc:8016
Change-Id: I050004e256b5698ce700e3416aa86b55f446a270
Reviewed-on: https://webrtc-review.googlesource.com/85361
Reviewed-by: Niels Moller <nisse@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Fredrik Solenberg <solenberg@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23762}
This will allow us to add unstandardized stats for the benefit of
native applications, and easily filter them out in chromium (without
having to maintain a whitelist that lists out every member
individually).
Unstandardized stats are declared as "RTCNonStandardStatsMember",
to make it clear in the declaration (in rtcstats_objects.h) whether
something is standardized or not.
Bug: webrtc:9410
Change-Id: I7c9804c261b7af96738e94dadeaa4b8a56b9ef2c
Reviewed-on: https://webrtc-review.googlesource.com/83743
Reviewed-by: Henrik Boström <hbos@webrtc.org>
Commit-Queue: Taylor Brandstetter <deadbeef@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23760}
This test creates a one way audio and video call, allows for bandwidth
estimation to ramp up and then runs the call for 10 seconds. The
average bandwidth estimate over this time is recorded as a perf metric.
This is done at the PeerConnection level with the intention to catch
regressions related to ICE configurations. Stats are taken from
PeerConnection for BWE, and the network simulation is done with a
VirtualSocketServer.
Bug: webrtc:7668
Change-Id: Ib8a449da80fc74be1e505ac34c0c6b7479cb58db
Reviewed-on: https://webrtc-review.googlesource.com/78361
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Taylor Brandstetter <deadbeef@webrtc.org>
Commit-Queue: Seth Hampson <shampson@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23758}
This goes back to using a subtree mirror of Chromium's third_party directory (managed by gclient).
The related scripts for syncing the files are also deleted.
The plan is to solve the conflict by creating third_party directories in subdirectories of WebRTC rather than the repo root.
Bug: webrtc:8366
Change-Id: I0b9f6a86c6d4075e2fa12c2db19aa54682ddb11f
Reviewed-on: https://webrtc-review.googlesource.com/85300
Reviewed-by: Oleh Prypin <oprypin@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23757}
The variable, num_active_spatial_layers, is used to construct ssData.
This CL reverts two instances of num_active_spatial_layers not
related to ssData construction.
Bug: None
Change-Id: I4d90d4578684dfdf8bd5a39c7a2fe778fce4414c
Reviewed-on: https://webrtc-review.googlesource.com/85643
Commit-Queue: Michael Horowitz <mhoro@webrtc.org>
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23756}
This will allow experimenting with audio bitrate allocation in video calls without increasing transport overhead.
Bug: webrtc:8243
Change-Id: If961780921d53bdce95b68c26641df6875509c1f
Reviewed-on: https://webrtc-review.googlesource.com/84501
Commit-Queue: Alex Narest <alexnarest@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23755}
fft.c is third party library and have to be moved to proper third_party
directory. So this CL will extract it to separate gn target to be able
then to move it to proper location.
Bug: webrtc:8366
Change-Id: I228ebab3c821aa7095f7aa460c23c2ea0fb98f01
Reviewed-on: https://webrtc-review.googlesource.com/85640
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Commit-Queue: Artem Titov <titovartem@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23753}
Removed the need to create a custom parser function and reuses some of
the code to reduce binary overhead of enums.
Bug: webrtc:9346
Change-Id: I51c9da713ed5456a86a2afbcf0991477bb83b894
Reviewed-on: https://webrtc-review.googlesource.com/83623
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Commit-Queue: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23752}
We now always enable any address ports, only using them if they end up
using interfaces that weren't otherwise accessible. This flag is no
longer used by downstream projects.
TBR=deadbeef@webrtc.org
Bug: None
Change-Id: I6e4e93958cbc4300811bafb103f1a2e8732274ed
Reviewed-on: https://webrtc-review.googlesource.com/85860
Commit-Queue: Qingsi Wang <qingsi@google.com>
Reviewed-by: Qingsi Wang <qingsi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23751}
A stub of sparse histogram factory getter is added so that Chromium can
provide an implementation using base::SparseHistogram for the metrics
macro RTC_HISTOGRAM_ENUMERATION_SPARSE. The default implementation in
WebRTC reuses the non-sparse version.
Bug: None
Change-Id: Ia091ca7aaacb6baa92027cd99d821bbc8da8d780
Reviewed-on: https://webrtc-review.googlesource.com/85740
Reviewed-by: Tommi <tommi@webrtc.org>
Commit-Queue: Qingsi Wang <qingsi@google.com>
Cr-Commit-Position: refs/heads/master@{#23750}
For HDR codecs, we expect to receive input that has 10-bit color depth. But
currently, WebRTC assumes only 8-bit input and output. This CL adds k010
format that represent this input.
Bug: webrtc:9376
Change-Id: Ie7df64b0eddb0752b493e0457a49083a1e87ba51
Reviewed-on: https://webrtc-review.googlesource.com/81920
Commit-Queue: Emircan Uysaler <emircan@webrtc.org>
Reviewed-by: Niklas Enbom <niklas.enbom@webrtc.org>
Reviewed-by: Niels Moller <nisse@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23749}
This can happen with the following sequence of API calls:
1) AddTrack(track) + offer/answer
2) RemoveTrack(track's sender) + offer/answer
3) AddTrack(same track)
Since the first transceiver had already been used to send, it will
not get re-used by the second call to AddTrack. Another RtpSender
will be created with its ID = the track ID. But the code hits a
DCHECK when CreateOffer is later called since both m= sections will
offer the same track ID component of the MSID.
The fix implemented here is to randomly generate a sender ID if
there is already an RtpSender with the track's ID.
Bug: webrtc:8734
Change-Id: Ic2dda23d66e364e77ff7505e1c37e53105a17dae
Reviewed-on: https://webrtc-review.googlesource.com/84249
Commit-Queue: Steve Anton <steveanton@webrtc.org>
Reviewed-by: Taylor Brandstetter <deadbeef@webrtc.org>
Reviewed-by: Henrik Boström <hbos@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23748}
The frequency shape of the echo path has been included in the reverberation model.
Bug: webrtc:9454,chromium:856636
Change-Id: Id2bc3096df31e29328936f94fe965ed1883d70f7
Reviewed-on: https://webrtc-review.googlesource.com/85370
Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
Commit-Queue: Jesus de Vicente Pena <devicentepena@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23746}
This removes call of av_register_all(), which is deprecated, and
related code.
Bug: webrtc:9352
Change-Id: Ib7de5931c900eaf1023ecf3046f560feaaeec8ef
Reviewed-on: https://webrtc-review.googlesource.com/85347
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23743}
common_audio/fft4g.c is third party codem that have to be moved into
third_party folder, so to be able to do it at first we have to extract
it into separate target. It is extracted with corresponding header file
and will be moved in futher CL.
Bug: webrtc:8366
Change-Id: I586ca94d4e9242c23163b987fa334dfa705020ed
Reviewed-on: https://webrtc-review.googlesource.com/85372
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Commit-Queue: Artem Titov <titovartem@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23742}
When GetSvcConfig returned fewer spatial layers than the number
statically configured from the test, we would crash on a SIGFPE.
This is not a problem in the production code, since there we
reset the encoder with the correct number of spatial layers
whenever the resolution changes.
Bug: None
Change-Id: I339e4a3c0fa993c7c649533c0eae71e1314194e7
Reviewed-on: https://webrtc-review.googlesource.com/85374
Reviewed-by: Michael Horowitz <mhoro@webrtc.org>
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Commit-Queue: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23741}
This slightly increases fuzzer coverage of the APM.
(.25 % points more line coverage.)
Bug: webrtc:9413
Change-Id: Ic992423f1dcf34fa0aa9649c8035a8e48b0ccdb2
Reviewed-on: https://webrtc-review.googlesource.com/85342
Reviewed-by: Alex Loiko <aleloi@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23732}
Before this CL, the packetsLost and jitter stats (as returned by
GetStats, at the API level) were only being updated when an RTCP SR or
RR is generated. According to the stats spec, "local" stats like this
should be updated any time a packet is received.
This CL also fixes some minor issues with the calculation of packetsLost
(and fractionLost):
* Packets weren't being count as lost if lost over a sequence number
rollover.
* Temporary periods of "negative" loss (caused by duplicate or out of
order packets) weren't being accumulated into the cumulative loss
counter. Example:
Period 1: Received packets 1, 2, 4
Loss over that period: 1 (expected 4 packets, got 3)
Reported cumulative loss: 1
Period 2: Received packets 3, 5
Loss over that period: -1 (expected 1 packet, got 2)
Reported cumulative loss: 1 (should be 0!)
Landing with NOTRY because Android compile bots are broken for an
unrelated reason.
NOTRY=True
Bug: webrtc:8804
Change-Id: I840ba34de8957b1276f6bdaf93718f805629f5c8
Reviewed-on: https://webrtc-review.googlesource.com/50020
Commit-Queue: Taylor Brandstetter <deadbeef@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Oskar Sundbom <ossu@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23731}
This change also standardizes the RtpSender to a single constructor
and moves the |track| and |stream_ids| arguments to setter methods.
Bug: webrtc:8734
Change-Id: I227a84868a80797f6cc2a1af6eec6d76da8ea159
Reviewed-on: https://webrtc-review.googlesource.com/84248
Commit-Queue: Steve Anton <steveanton@webrtc.org>
Reviewed-by: Taylor Brandstetter <deadbeef@webrtc.org>
Reviewed-by: Seth Hampson <shampson@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23730}
