This ensure BWE works as intended with transport sequence numbers on
audio.
Tested with webrtc_perf_tests --gtest_filter=CallPerfTest.Min_Bitrate_VideoAndAudio
and --gtest_filter=Rampup*
Bug: webrtc:14854, webrtc:7135, b/266786240
Change-Id: I3b7a743149c22035e582a2157b5f0a93747857cf
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291523
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Reviewed-by: Andrey Logvin <landrey@webrtc.org>
Auto-Submit: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39208}
Instead, ensure extensions are registered so that both transport and send streams are aware.
Bug: webrtc:7135,webrtc:14795,b/266658815
Change-Id: I7710113893e2c5e23c1365de6aa3b761e3408308
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291333
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39193}
This reverts commit 897ea04db5db2e591e28bd884191be58d9bcdc63.
Reason for revert: Speculative revert as it could be the reason why perf tests started failing: https://ci.chromium.org/p/webrtc/g/perf/console?limit=200
Original change's description:
> Delete PacketReceiver::DeliverPacket from all implementations
>
> And fix tests that still depend on extensions to be known by the receiver.
>
> Change-Id: I62227829af81af07769189e547f1cdb8ed4d06b3
>
> Bug: webrtc:7135,webrtc:14795
> Change-Id: I62227829af81af07769189e547f1cdb8ed4d06b3
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290996
> Commit-Queue: Per Kjellander <perkj@webrtc.org>
> Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
> Reviewed-by: Erik Språng <sprang@webrtc.org>
> Cr-Commit-Position: refs/heads/main@{#39184}
Bug: webrtc:7135,webrtc:14795,b/266658815
Change-Id: I9d03f4952938d176ffee110a707acadc1846457c
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291400
Commit-Queue: Andrey Logvin <landrey@webrtc.org>
Bot-Commit: rubber-stamper@appspot.gserviceaccount.com <rubber-stamper@appspot.gserviceaccount.com>
Owners-Override: Andrey Logvin <landrey@webrtc.org>
Reviewed-by: Jeremy Leconte <jleconte@google.com>
Cr-Commit-Position: refs/heads/main@{#39189}
And fix tests that still depend on extensions to be known by the receiver.
Change-Id: I62227829af81af07769189e547f1cdb8ed4d06b3
Bug: webrtc:7135,webrtc:14795
Change-Id: I62227829af81af07769189e547f1cdb8ed4d06b3
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290996
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39184}
This reverts commit 3e61f881cd2ba9040a07371e0ba6dda902aa60ae.
Reason for revert: Issue fixed in https://webrtc-review.googlesource.com/c/src/+/291104
Original change's description:
> Revert "Change CallTests to use new PacketReceiver::DeliverRtp and PacketReceiver::DeliverRtcp"
>
> This reverts commit 3b96f2c770df7691df90c2cc1be40509a76ae425.
>
> Reason for revert: Seems to cause test failures and perf regressions in tests: webrtc:14833, and CallPerfTest.Min_Bitrate_VideoAndAudio
>
>
> Original change's description:
> > Change CallTests to use new PacketReceiver::DeliverRtp and PacketReceiver::DeliverRtcp
> >
> > PacketReceiver::DeliverRtp requires delivered packets to have extensions already mapped.
> > Therefore DirectTransport is provided with the extension mapping.
> >
> > CallTests and tests derived from CallTest create transports in different ways, this cl change CallTest to create tests in only one way to simplify how extensions are provided to the transport but at the same time still allows different network behaviour.
> >
> >
> > Change-Id: Ie8b3ad947c170be61e62c02dadf4adedbb3841f1
> > Bug: webrtc:7135, webrtc:14795
> > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290980
> > Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
> > Commit-Queue: Per Kjellander <perkj@webrtc.org>
> > Reviewed-by: Erik Språng <sprang@webrtc.org>
> > Cr-Commit-Position: refs/heads/main@{#39137}
>
> Bug: webrtc:7135, webrtc:14795, webrtc:14833
> Change-Id: Ib6180a47cf7611ed2bc648acc3b9e5cfeec4d9cf
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291220
> Owners-Override: Björn Terelius <terelius@webrtc.org>
> Auto-Submit: Per Kjellander <perkj@webrtc.org>
> Reviewed-by: Björn Terelius <terelius@webrtc.org>
> Commit-Queue: Björn Terelius <terelius@webrtc.org>
> Bot-Commit: rubber-stamper@appspot.gserviceaccount.com <rubber-stamper@appspot.gserviceaccount.com>
> Cr-Commit-Position: refs/heads/main@{#39146}
Bug: webrtc:7135, webrtc:14795, webrtc:14833
Change-Id: I3fb0210d7a33c600ead5719ce2acb8cc68ec20bd
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291222
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Bot-Commit: rubber-stamper@appspot.gserviceaccount.com <rubber-stamper@appspot.gserviceaccount.com>
Owners-Override: Per Kjellander <perkj@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39157}
This reverts commit 3b96f2c770df7691df90c2cc1be40509a76ae425.
Reason for revert: Seems to cause test failures and perf regressions in tests: webrtc:14833, and CallPerfTest.Min_Bitrate_VideoAndAudio
Original change's description:
> Change CallTests to use new PacketReceiver::DeliverRtp and PacketReceiver::DeliverRtcp
>
> PacketReceiver::DeliverRtp requires delivered packets to have extensions already mapped.
> Therefore DirectTransport is provided with the extension mapping.
>
> CallTests and tests derived from CallTest create transports in different ways, this cl change CallTest to create tests in only one way to simplify how extensions are provided to the transport but at the same time still allows different network behaviour.
>
>
> Change-Id: Ie8b3ad947c170be61e62c02dadf4adedbb3841f1
> Bug: webrtc:7135, webrtc:14795
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290980
> Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
> Commit-Queue: Per Kjellander <perkj@webrtc.org>
> Reviewed-by: Erik Språng <sprang@webrtc.org>
> Cr-Commit-Position: refs/heads/main@{#39137}
Bug: webrtc:7135, webrtc:14795, webrtc:14833
Change-Id: Ib6180a47cf7611ed2bc648acc3b9e5cfeec4d9cf
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291220
Owners-Override: Björn Terelius <terelius@webrtc.org>
Auto-Submit: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Björn Terelius <terelius@webrtc.org>
Commit-Queue: Björn Terelius <terelius@webrtc.org>
Bot-Commit: rubber-stamper@appspot.gserviceaccount.com <rubber-stamper@appspot.gserviceaccount.com>
Cr-Commit-Position: refs/heads/main@{#39146}
PacketReceiver::DeliverRtp requires delivered packets to have extensions already mapped.
Therefore DirectTransport is provided with the extension mapping.
CallTests and tests derived from CallTest create transports in different ways, this cl change CallTest to create tests in only one way to simplify how extensions are provided to the transport but at the same time still allows different network behaviour.
Change-Id: Ie8b3ad947c170be61e62c02dadf4adedbb3841f1
Bug: webrtc:7135, webrtc:14795
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290980
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39137}
Perf tests upload its results to CPD.
With the current design, an assertion failure in one test prevents the upload for all the tests.
https://ci.chromium.org/ui/p/webrtc/builders/perf/Perf%20Mac%20M1%20Arm64%2012/1719/overview
The "quick" perf test mode is made to run on regular CQ/CI bots without any metrics upload so it's fine to have an assertion failure there.
Bug: b/264502081
Change-Id: I22e8e8b7ce317f43297cb8837694e420cd80613d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290571
Commit-Queue: Jeremy Leconte <jleconte@google.com>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Christoffer Jansson <jansson@google.com>
Cr-Commit-Position: refs/heads/main@{#39063}
With this cl, a packet is only parsed once in RtpTransport::DemuxPacket and the metadata is reused.
Extensions are still identified twice- one for demuxing based on mid. The second time in Channel::OnReceivedPacket in order to use extensions specific to that mid.
Bug: webrtc:7135, webrtc:14795
Change-Id: I50e3814af92ca4378f148876b20a54bcfac1e146
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290540
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39058}
and in DegradedCall. In DegradedCall - ThreadPacketReceiver is no longer needed.
Implementation of DeliverRtpPacket is done in preparation of https://webrtc-review.googlesource.com/c/src/+/290540, where the parsed packet will be propagated to Call without extra parsing.
Bug: webrtc:7135, webrtc:14795
Change-Id: Ic068105d6d1f337afc6b4539b0e7184e736e7ee0
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290704
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39048}
Keeping the headers to allow compatibility with current users
that expect the headers to be in that target before they are
also updated.
Bug: webrtc:9838
Change-Id: I8b1e88850958e92c043686587a37791f01860220
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290569
Commit-Queue: Florent Castelli <orphis@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Auto-Submit: Florent Castelli <orphis@webrtc.org>
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39031}
DeliverRtpPacket use a parsed RTP packet as argument where the RTP extensions are supposed to be known.
This method is implemented in webrt::Call and temporary used by the exising method Call::DeliverRtp, but the idea is to instead avoid extra packet parsing by forwarding a RtpPacketReceived from RtpTransport::DemuxRtpPacket via WebrtcVideoChannel::OnPacketReceived and WebrtcVoiceChannel.
DeliverRtcpPacket is also implemented in Call and is directly used in PeerConnection::InitializeRtcpCallback.
Bug: webrtc:14795, webrtc:7135
Change-Id: Ib6ffe8e1229ac07fa459ee2fc9a0af8455a23bac
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290401
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39015}
In order for a packet to be parseable and include a transport sequence number, it has to be negotiated. Thus, there is no need to check again.
Bug: webrtc:14795
Change-Id: I1fa76abdbad11d15ecae80fbaa227bd12a8035bc
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290565
Auto-Submit: Per Kjellander <perkj@webrtc.org>
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39011}
This is a reland of commit 97ba853295578975a04fc504315cccd465f9f0bd
This cl did not cause the regression in Chrome rolls https://chromium-review.googlesource.com/c/chromium/src/+/4132644?tab=checks. Real culprit reverted in https://webrtc-review.googlesource.com/c/src/+/290502.
Original change's description:
> Remove use of ReceiveStreamRtpConfig:transport_cc
>
> With this change, webrtc will send RTCP transport feedback for all received packets that have a transport sequence number, if the header extension
> http://www.ietf.org/id/draft-holmer-rmcat-transport-wide-cc-extensions is negotiated.
> I.e the SDP attribute a=rtcp-fb:96 transport-cc is ignored.
>
>
> Change-Id: I95d8d4405dc86a2f872f7883b7bafd623d5f7841
>
> Bug: webrtc:14802
> Change-Id: I95d8d4405dc86a2f872f7883b7bafd623d5f7841
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290403
> Reviewed-by: Harald Alvestrand <hta@webrtc.org>
> Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
> Commit-Queue: Per Kjellander <perkj@webrtc.org>
> Cr-Commit-Position: refs/heads/main@{#38980}
Bug: webrtc:14802
Change-Id: Ib98e61413161d462da60144942cdb0140e12bc42
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290503
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#38997}
With this change, webrtc will send RTCP transport feedback for all received packets that have a transport sequence number, if the header extension
http://www.ietf.org/id/draft-holmer-rmcat-transport-wide-cc-extensions is negotiated.
I.e the SDP attribute a=rtcp-fb:96 transport-cc is ignored.
Change-Id: I95d8d4405dc86a2f872f7883b7bafd623d5f7841
Bug: webrtc:14802
Change-Id: I95d8d4405dc86a2f872f7883b7bafd623d5f7841
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290403
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#38980}
Instead of getting header extension mapping from a receiver object, get the mapping from the received packet.
The purpose is to be able to remove extension information from webrtc/call/receive_stream.h.
Header extensions are negotiated per mid, not per receive stream.
The goal is to reduce the number of places where packets are parsed and demuxed.
Bug: webrtc:7135, webrtc:14795
Change-Id: I8944bc06a11dc572d9e14e7d7ee446a841096295
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/288968
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#38944}
This is a reland of commit d49d49ad89e67d1a3c63fbc638af445af5648875
Fixed seconds to milliseconds conversion in VideoAnalyzer.
Original change's description:
> Report total and squared inter frame delays measured in OnRenderedFrame
>
> After https://webrtc-review.googlesource.com/c/src/+/160042 we ended up with two sets of metrics representing total and total squared inter frame delays: old is measured in OnDecodedFrame and new in OnRenderedFrame. Reporting of old metrics was unshipped in https://webrtc-review.googlesource.com/c/src/+/278100. The metrics are used for calculation of harmonic frame rate and are desired to be measured as close as possible to rendering. This CL removes calculation of inter frame delay metrics from OnDecodedFrame and reports the metrics calculated in OnRenderedFrame to the stats.
>
> Bug: webrtc:11108, b/261512902
> Change-Id: Ia21b321aab3a1ac0b6136dc0df7d95f2f0fd24c6
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/286842
> Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
> Reviewed-by: Erik Språng <sprang@webrtc.org>
> Reviewed-by: Harald Alvestrand <hta@webrtc.org>
> Reviewed-by: Henrik Boström <hbos@webrtc.org>
> Cr-Commit-Position: refs/heads/main@{#38909}
Bug: webrtc:11108, webrtc:14792, b/261512902
Change-Id: Ic5d0bc4622ee0cb46b6c225cdddccc217200e794
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/288641
Reviewed-by: Henrik Boström <hbos@webrtc.org>
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Johannes Kron <kron@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#38929}
This reverts commit d49d49ad89e67d1a3c63fbc638af445af5648875.
Reason for revert:
# Check failed: total_freezes_duration_ms_double <= total_frames_duration_ms_double (196 vs. 0.044783)
https://ci.chromium.org/p/webrtc/builders/perf/Perf%20Mac%20M1%20Arm64%2012
it also breaks the metric 'freeze_duration_ratio':
https://chromeperf.appspot.com/report?sid=6e919d271ff5885c3fa6363dd255b9793d5e79332a9f202b725c33cc7d3da31a
Original change's description:
> Report total and squared inter frame delays measured in OnRenderedFrame
>
> After https://webrtc-review.googlesource.com/c/src/+/160042 we ended up with two sets of metrics representing total and total squared inter frame delays: old is measured in OnDecodedFrame and new in OnRenderedFrame. Reporting of old metrics was unshipped in https://webrtc-review.googlesource.com/c/src/+/278100. The metrics are used for calculation of harmonic frame rate and are desired to be measured as close as possible to rendering. This CL removes calculation of inter frame delay metrics from OnDecodedFrame and reports the metrics calculated in OnRenderedFrame to the stats.
>
> Bug: webrtc:11108, b/261512902
> Change-Id: Ia21b321aab3a1ac0b6136dc0df7d95f2f0fd24c6
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/286842
> Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
> Reviewed-by: Erik Språng <sprang@webrtc.org>
> Reviewed-by: Harald Alvestrand <hta@webrtc.org>
> Reviewed-by: Henrik Boström <hbos@webrtc.org>
> Cr-Commit-Position: refs/heads/main@{#38909}
Bug: webrtc:11108, b/261512902, webrtc:14789
Change-Id: Ie0da33c1071c48c50bff6608830c9e2a5a928fb4
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/288402
Commit-Queue: Jeremy Leconte <jleconte@google.com>
Bot-Commit: rubber-stamper@appspot.gserviceaccount.com <rubber-stamper@appspot.gserviceaccount.com>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#38922}
After https://webrtc-review.googlesource.com/c/src/+/160042 we ended up with two sets of metrics representing total and total squared inter frame delays: old is measured in OnDecodedFrame and new in OnRenderedFrame. Reporting of old metrics was unshipped in https://webrtc-review.googlesource.com/c/src/+/278100. The metrics are used for calculation of harmonic frame rate and are desired to be measured as close as possible to rendering. This CL removes calculation of inter frame delay metrics from OnDecodedFrame and reports the metrics calculated in OnRenderedFrame to the stats.
Bug: webrtc:11108, b/261512902
Change-Id: Ia21b321aab3a1ac0b6136dc0df7d95f2f0fd24c6
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/286842
Commit-Queue: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Henrik Boström <hbos@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#38909}
This CL adds functionality to remove packets matching a given SSRC from
the pacer queue, and calls that with any SSRCs used by an RTP module
when that module is removed.
Bug: chromium:1395081
Change-Id: I13c0285ddca600e784ad04a806727a508ede6dcc
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/287124
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Philip Eliasson <philipel@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#38880}
This is in the webrtc-stats spec at
https://www.w3.org/TR/webrtc-stats/#dom-rtcoutboundrtpstreamstats-scalabilitymode.
This adds the scalability mode to CodecSpecificInfo which is used to
plumb the modes for each simulcast layer.
TBR=orphis@webrtc.org
Tested: Compiled into Chrome and confirmed the scalability mode set for AV1, VP9, VP8 and H264 software encoders in chrome://webrtc-internals.
Bug: webrtc:14730
Change-Id: I71ceba8f6485a4f4a73e0856031b8d5f16f913f2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/285085
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Commit-Queue: Evan Shrubsole <eshr@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#38847}
