When bitrate is allocated to streams that does not have packet feedback,
the allocated bitrate should be included in the estimate. This was
previously only implemented for the old congestion controller and not
for the new task queue based version.
To make the behavior more robust, the responsibility for tracking this
is moved to BitrateAllocator where it's handled consistently for
multiple streams without feedback.
Bug: webrtc:9586, webrtc:8243
Change-Id: I8af7fec23e1bdc08cc61cf1b4ff10461c3711fb0
Reviewed-on: https://webrtc-review.googlesource.com/102681
Commit-Queue: Sebastian Jansson <srte@webrtc.org>
Reviewed-by: Christoffer Rodbro <crodbro@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24905}
This features is not needed anymore, with this CL it is also possible
to address two issues:
- The need to pick a default implementation.
- The need to use -Wno-global-constructors.
Bug: webrtc:9631, webrtc:9693
Change-Id: Id3daf34179fbc8db26969fc701ccbfa7182c6a9b
Reviewed-on: https://webrtc-review.googlesource.com/102543
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24904}
This reverts commit 2fbb83b16b4c2c1712cbe898ca3ba42d6da3e96f.
Reason for revert: Speculative revert over failing Chromium bot:
https://ci.chromium.org/p/chromium/builders/luci.chromium.webrtc.fyi/WebRTC%20Chromium%20FYI%20Android%20Tests%20%28dbg%29%20%28M%20Nexus5X%29/117
Original change's description:
> Remove APM-internal usage of EchoControlMobile
>
> This is a sibling CL to a similar one for EchoCancellation:
> https://webrtc-review.googlesource.com/c/src/+/97603
>
> - EchoControlMobileImpl will no longer inherit EchoControlMobile.
> - Removes usage of AudioProcessing::echo_control_mobile() inside most of
> the audio processing module and unit tests.
>
> The CL breaks audioproc_f backwards compatibility: It can no longer
> use all recorded settings (comfort noise, routing mode), but prints an
> error message when unsupported settings are encountered.
>
> Tested: audioproc_f with .wav and aecdump inputs.
> Bug: webrtc:9535
> Change-Id: I63c3c81bcaf44021315978e1a0f3e42173b988ce
> Reviewed-on: https://webrtc-review.googlesource.com/101621
> Reviewed-by: Alex Loiko <aleloi@webrtc.org>
> Commit-Queue: Sam Zackrisson <saza@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#24888}
TBR=saza@webrtc.org,aleloi@webrtc.org
Change-Id: I1f8a27ac291f2cdc16c8daa32e399b74d489dbb9
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:9535
Reviewed-on: https://webrtc-review.googlesource.com/102642
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24895}
These submodules implicitly rely on low cut filtering being enabled.
This CL clarifies a distinction:
High pass filtering is a feature that users can enable, according to the WebRTC standard.
Low cut filtering is a processing effect that is applied when any of the following is active:
- high pass filter
- noise suppression
- builtin echo cancellation
Bug: webrtc:9535
Change-Id: I9474276fb11354ea3b01e65a0699f6c29263770b
Reviewed-on: https://webrtc-review.googlesource.com/102600
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24892}
Adds SimulatedTimeClient, a class that simulates time so congestion
controllers can be tested using the Scenario test framework without
running in real time.
This allows using simplified scenario tests as unit tests, narrowing
the gap between end to end tests and unit tests.
Bug: webrtc:9510
Change-Id: I61ab388bd610f636b926675b1f14b8d85e3c1114
Reviewed-on: https://webrtc-review.googlesource.com/99801
Commit-Queue: Sebastian Jansson <srte@webrtc.org>
Reviewed-by: Christoffer Rodbro <crodbro@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24890}
Add new method OnRtpPacket, but leave
ReceiveStatisticsImpl::IncomingPacket and most of the implementation
unchanged. Deleting the old method and converting implementation from
RTPHeader to RtpPacketreceived is planned for a followup, after
downstream code is updated.
Bug: webrtc:7135, webrtc:8016
Change-Id: I697ec12804618859f8d69415622d1b957e1d0847
Reviewed-on: https://webrtc-review.googlesource.com/100104
Reviewed-by: Fredrik Solenberg <solenberg@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Niels Moller <nisse@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24889}
This is a sibling CL to a similar one for EchoCancellation:
https://webrtc-review.googlesource.com/c/src/+/97603
- EchoControlMobileImpl will no longer inherit EchoControlMobile.
- Removes usage of AudioProcessing::echo_control_mobile() inside most of
the audio processing module and unit tests.
The CL breaks audioproc_f backwards compatibility: It can no longer
use all recorded settings (comfort noise, routing mode), but prints an
error message when unsupported settings are encountered.
Tested: audioproc_f with .wav and aecdump inputs.
Bug: webrtc:9535
Change-Id: I63c3c81bcaf44021315978e1a0f3e42173b988ce
Reviewed-on: https://webrtc-review.googlesource.com/101621
Reviewed-by: Alex Loiko <aleloi@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24888}
These changes are based on finding when using Tsan v2. More changes are
needed before usage of the THREAD_SANITIZER build flag can be removed.
Hence, all tests are still ignored when this flag is set. The changes
are still improvements.
See https://bugs.chromium.org/p/webrtc/issues/detail?id=9778#c10
for more details.
Bug: webrtc:9778
Change-Id: I1266cec48165046dcffc16f104ec5b88b41500b2
Reviewed-on: https://webrtc-review.googlesource.com/102440
Reviewed-by: Sami Kalliomäki <sakal@webrtc.org>
Commit-Queue: Henrik Andreassson <henrika@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24880}
This CL removes some deprecated build targets (and their headers)
from system_wrappers:
- field_trial_api
- field_trial_default
- metrics_api
- metrics_default
It also refreshes all the dependencies on field_trial.h and metrics.h.
A nice side effect is that it is finally possible to remove 'nogncheck'
from the following files (when it was used with field_trial_default
and metrics_default):
- sdk/objc/api/peerconnection/RTCMetricsSampleInfo+Private.h
- sdk/android/src/jni/pc/peerconnectionfactory.cc
- sdk/objc/api/peerconnection/RTCFieldTrials.mm
Bug: webrtc:9631
Change-Id: Ib621f41ef8ad0aba4fe1c1d7e749c044afc956c3
No-Try: True
Reviewed-on: https://webrtc-review.googlesource.com/100524
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24878}
This CL removes killswitches for code that has been properly tested in
experiments and is to be considered to be permanent.
The changes have been tested for bitexactness.
Bug: webrtc:8671
Change-Id: I0f9db16f377390d9dd3779096da91f3abc0fb4a5
Reviewed-on: https://webrtc-review.googlesource.com/102360
Reviewed-by: Gustaf Ullberg <gustaf@webrtc.org>
Commit-Queue: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24877}
This is the first step in order to remove runtime_enabled_features
code from WebRTC.
Bug: webrtc:9693
Change-Id: Ic67f770c2166755ea45c782efb3e4184433ac15e
Reviewed-on: https://webrtc-review.googlesource.com/102361
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24868}
The RTC_DCHECK is hit sometimes. This happens when there is no overlap
between the nack_list and frames in keyframes. The existing code
correctly handles this situation.
Bug: webrtc:9629
Change-Id: I7e3eed1b04781cd69974c5d3eb86e382e9587268
Reviewed-on: https://webrtc-review.googlesource.com/102340
Reviewed-by: Philip Eliasson <philipel@webrtc.org>
Commit-Queue: Johannes Kron <kron@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24860}
Specially for devices with high echo path gain, even low render signal can allow the linear filter of the AEC3 to converge. However, the conditions that were used for updating the ERLE avoided to update that estimation. In this commit, we allow adapting the ERLE estimator using even low render signal but the update of the ERLE is constraint in a way that decreases are not allowed.
Bug: webrtc:9776
Change-Id: Ic4331efcc47a0b05f394cdea9a88f336292de5a1
Reviewed-on: https://webrtc-review.googlesource.com/101641
Commit-Queue: Jesus de Vicente Pena <devicentepena@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24859}
The tests only use the automatic mode, and I'd rather not maintain
(and test!) the rest.
Bug: webrtc:8396
Change-Id: I4cd1096e088d2ea8807a605b8448bd44ff9e88ed
Reviewed-on: https://webrtc-review.googlesource.com/102060
Reviewed-by: Ivo Creusen <ivoc@webrtc.org>
Commit-Queue: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24852}
Mark all low spatial layer frames as references (not just frames of
active low spatial layers) if inter-layer prediction is enabled since
these frames are indirect references of high spatial layer, which can
later be enabled without key frame.
Bug: webrtc:9782
Change-Id: Iffa5039fab2673a5582e7cdc9be4a36d9e8deb63
Reviewed-on: https://webrtc-review.googlesource.com/102063
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24849}
This reverts commit 8b7bc5d7010c84ac57459518fe18309ef5fee1dd.
Reason for revert: Slow RTC_DCHECK has been removed.
Original change's description:
> Revert "Second reland of "Optimize execution time of RTPSender::UpdateDelayStatistics""
>
> This reverts commit 9def3b45ef06de9e068e8f4d1644e9d508baa913.
>
> Reason for revert: webrtc_perf_tests fails on Mac-10.12.
>
> Original change's description:
> > Second reland of "Optimize execution time of RTPSender::UpdateDelayStatistics"
> >
> > The reland has a lot of additional DCHECKS for easier debugging,
> > so in debug builds it will actually be a ~2x slowdown compared to the old code.
> > The excessive DCHECKS should be removed in a followup CL.
> >
> > Bug: webrtc:9439
> > Change-Id: I8389cd84f1ca12c29cc6993f0d2cf7e6d7dd8379
> > Reviewed-on: https://webrtc-review.googlesource.com/101761
> > Reviewed-by: Åsa Persson <asapersson@webrtc.org>
> > Reviewed-by: Björn Terelius <terelius@webrtc.org>
> > Commit-Queue: Johannes Kron <kron@webrtc.org>
> > Cr-Commit-Position: refs/heads/master@{#24821}
>
> TBR=terelius@webrtc.org,asapersson@webrtc.org,kron@webrtc.org
>
> Change-Id: I98c4c96d552858d0299d49993e9b9be6a6204dfe
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: webrtc:9439
> Reviewed-on: https://webrtc-review.googlesource.com/101860
> Reviewed-by: Johannes Kron <kron@webrtc.org>
> Commit-Queue: Johannes Kron <kron@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#24825}
TBR=terelius@webrtc.org,asapersson@webrtc.org,kron@webrtc.org
Change-Id: I260c56932710d26f9d7201c07279fef8d2150bd9
Bug: webrtc:9439
Reviewed-on: https://webrtc-review.googlesource.com/102000
Reviewed-by: Björn Terelius <terelius@webrtc.org>
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Reviewed-by: Johannes Kron <kron@webrtc.org>
Commit-Queue: Johannes Kron <kron@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24843}
This reverts commit 3f4a4fad8cd661309ff5d9a631e89518f32e7c5e.
Reason for revert: Breaking internal tests
Original change's description:
> Added field trial WebRTC-GenericDescriptor for the new generic descriptor.
>
> Also parameterized tests to test the new generic descriptor and
> added --generic_descriptor flag to loopback tests.
>
> Bug: webrtc:9361
> Change-Id: I2b76889606fe2e81249ecdebb0b7b61151682485
> Reviewed-on: https://webrtc-review.googlesource.com/101900
> Commit-Queue: Philip Eliasson <philipel@webrtc.org>
> Reviewed-by: Erik Språng <sprang@webrtc.org>
> Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
> Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#24835}
TBR=danilchap@webrtc.org,sprang@webrtc.org,stefan@webrtc.org,philipel@webrtc.org
Change-Id: I4d4714a9f4ab0e95adf0f4130bc1a932efc448fa
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:9361
Reviewed-on: https://webrtc-review.googlesource.com/101940
Reviewed-by: Lu Liu <lliuu@webrtc.org>
Commit-Queue: Lu Liu <lliuu@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24839}
Also parameterized tests to test the new generic descriptor and
added --generic_descriptor flag to loopback tests.
Bug: webrtc:9361
Change-Id: I2b76889606fe2e81249ecdebb0b7b61151682485
Reviewed-on: https://webrtc-review.googlesource.com/101900
Commit-Queue: Philip Eliasson <philipel@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24835}
The flaky crash is happening in the constructor.
Bug: webrtc:9778
Change-Id: I9ac9a89e033e17de690e594ef263ff83d14fcc5a
Reviewed-on: https://webrtc-review.googlesource.com/101880
Reviewed-by: Henrik Andreassson <henrika@webrtc.org>
Commit-Queue: Sami Kalliomäki <sakal@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24829}
This reverts commit 9def3b45ef06de9e068e8f4d1644e9d508baa913.
Reason for revert: webrtc_perf_tests fails on Mac-10.12.
Original change's description:
> Second reland of "Optimize execution time of RTPSender::UpdateDelayStatistics"
>
> The reland has a lot of additional DCHECKS for easier debugging,
> so in debug builds it will actually be a ~2x slowdown compared to the old code.
> The excessive DCHECKS should be removed in a followup CL.
>
> Bug: webrtc:9439
> Change-Id: I8389cd84f1ca12c29cc6993f0d2cf7e6d7dd8379
> Reviewed-on: https://webrtc-review.googlesource.com/101761
> Reviewed-by: Åsa Persson <asapersson@webrtc.org>
> Reviewed-by: Björn Terelius <terelius@webrtc.org>
> Commit-Queue: Johannes Kron <kron@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#24821}
TBR=terelius@webrtc.org,asapersson@webrtc.org,kron@webrtc.org
Change-Id: I98c4c96d552858d0299d49993e9b9be6a6204dfe
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:9439
Reviewed-on: https://webrtc-review.googlesource.com/101860
Reviewed-by: Johannes Kron <kron@webrtc.org>
Commit-Queue: Johannes Kron <kron@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24825}
The reland has a lot of additional DCHECKS for easier debugging,
so in debug builds it will actually be a ~2x slowdown compared to the old code.
The excessive DCHECKS should be removed in a followup CL.
Bug: webrtc:9439
Change-Id: I8389cd84f1ca12c29cc6993f0d2cf7e6d7dd8379
Reviewed-on: https://webrtc-review.googlesource.com/101761
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Reviewed-by: Björn Terelius <terelius@webrtc.org>
Commit-Queue: Johannes Kron <kron@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24821}
It hasn't been changed in any meaningful way since 2013, the same year
it was created.
Bug: webrtc:8396
Change-Id: I5633188134f71f24311fbd3098d046632fc4ee3a
Reviewed-on: https://webrtc-review.googlesource.com/101563
Reviewed-by: Henrik Lundin <henrik.lundin@webrtc.org>
Commit-Queue: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24816}
A potential bug was introduced in "Refactor to remove direct memory
dependency on kMaxId" due to a memory restructuring,
commit c5744b8b21b627213286f1b6f2c65da5df9ce8d0
Bug: webrtc:7990
Change-Id: I0dcaf47e1c1e361d65220c278a2326d6f2686af7
Reviewed-on: https://webrtc-review.googlesource.com/101642
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Johannes Kron <kron@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24814}
RtpPacket::UpdateDelayStatistics was previously optimized with several
sanity checks added. These sanity checks caused many of the unit tests
in peerconnection_integration_unittests to fail and the CL was therefore
reverted. This CL contains the sanity checks along with patches so that
the unit tests pass.
Bug: webrtc:9439
Change-Id: Ia5f5e8b125e5f3f4b79d433e2282901143530a25
Reviewed-on: https://webrtc-review.googlesource.com/99802
Reviewed-by: Björn Terelius <terelius@webrtc.org>
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Steve Anton <steveanton@webrtc.org>
Commit-Queue: Johannes Kron <kron@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24813}
This avoids crashing if an application resets the starting bitrate
before adding streams to the call.
Bug: webrtc:9586
Change-Id: I8d31aba1f4fee40c67c8930f5a32d17700ccadc3
Reviewed-on: https://webrtc-review.googlesource.com/101680
Reviewed-by: Björn Terelius <terelius@webrtc.org>
Commit-Queue: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24808}
It will soon lose the ability to do so.
Also, the ACM no longer creates comfort noise encoders for us, so
don't bother testing that.
Bug: webrtc:8396
Change-Id: I24a12e26bef142f9f8e7532b764f28572e0c6ace
Reviewed-on: https://webrtc-review.googlesource.com/101640
Reviewed-by: Ivo Creusen <ivoc@webrtc.org>
Commit-Queue: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24803}
Speeds up adaptation of the matched filter of the delay estimator by
allowing the estimated echo and the error signal (microphone minus
estimated echo) to be saturated. Only microphone saturation pauses
the filter adaptation.
Bug: webrtc:9773
Change-Id: I8b8400539fde3ee821f36a95818bece02ddd626b
Reviewed-on: https://webrtc-review.googlesource.com/101341
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Gustaf Ullberg <gustaf@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24802}
When two-byte header extensions are enabled, kMaxId will change from 15
to 255. This CL is a refactor to remove the direct dependency between
memory allocation and kMaxId.
Bug: webrtc:7990
Change-Id: I38974a9c705eb6a0fdba9038a7d909861587d98d
Reviewed-on: https://webrtc-review.googlesource.com/101580
Commit-Queue: Johannes Kron <kron@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24801}
A (actually several weeks) while ago, we noticed an error with the
WebRTC.Audio.Agc2.EstimatedNoiseLevel histogram. It always reported
the value 0. Here is why:
The histogram bins go from 0 to 100. But the value logged is dBFS. It is
always less than or equal to 0. This CL changes the bins.
Bug: webrtc:7494
Change-Id: I45fd122e98f9396f9871bc965a708987bd1815f6
Reviewed-on: https://webrtc-review.googlesource.com/101340
Commit-Queue: Alex Loiko <aleloi@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24800}
Right after a volume decrease, the echo path estimate is overestimated and, as a side effect, the nearend signal is also overestimated. Due to that, the suppression gains are kept high avoiding the suppression of echoes. In this CL the neared power spectrum estimation is limited to a level given by the power spectrum or the microphone input signal. Additionally, the minimum gain that is computed inside the suppressor is also modified. Instead of using the nearend power spectrum that is now bounded, the power spectrum of the signal after the linear echo canceler is used.
Bug: webrtc:9762
Change-Id: Ia24cd2ce248f2c2ba124711b75acff3b8c5cfa9f
Reviewed-on: https://webrtc-review.googlesource.com/100720
Commit-Queue: Jesus de Vicente Pena <devicentepena@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24796}
