Checking in sending classes avoids using global field trial string in favor of the injected one.
In addition to that RateLimiter looks wrong layer for check that field trial:
checking inside RateLimiter class might be surprising if it is used for limiting something else than RTX bitrate.
evaluating field trial for each retransmitting packet might be expensive
Bug: webrtc:15184, webrtc:10335
Change-Id: I87bae3522bbd9692629d4f9b6caa119be03f2bd6
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/322720
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Ying Wang <yinwa@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#40908}
Combine all parameters into single struct so that it is easier to add and remove optional parameters
Use Timestamp type instad of plain int to represent capture time
Use rtc::ArrayView instead of pointer+size to represent payload
Merge passing audio level into send function.
Bug: webrtc:13757, webrtc:14870
Change-Id: I0386b710eb99b864334d61235add9abcde9bc69d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/317442
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#40688}
Keep the logic managing whether audio RTP timestamps have the random
start offset added or not inside ChannelSend, so that the
ChannelSendFrameTransformerDelegate doesn't need to worry about it.
Crucially, this means that frames moved between senders by encoded
transforms clients will always use the correct offset for the channel
where we actually get sent.
Also rename TS variables throughout both classes to be explicit over
whether the offset has been added or not.
Bug: chromium:1464847
Change-Id: I19955ec4c1cb834161b00dd74622725a070b713a
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/317900
Commit-Queue: Tony Herre <herre@google.com>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#40655}
NetworkLinkRtcpObserver is similar to RtcpBandwidthObserver but pass
time variables using unit types instead of raw integers.
Bug: webrtc:13757
Change-Id: Iaa0bbe0b108620b3a24013c40e7d9004032e904d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/305022
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#40087}
Delete VoERtcpObserver proxy:
pass BWE related message directly to transport controller
pass ReportBlock directly to ChannelSend, assuming there will be single report block per source ssrc
Bug: None
Change-Id: I8378326bff1dc3c2736960166fc782ee822a9c12
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/305224
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#40081}
ReportBlockData class is better documented and has wider usage.
Bug: webrtc:13757
Change-Id: Ie5f2275f2f0236267172e6dd1ce5c2dfb2193ba0
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/304101
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39980}
This change makes AudioCodingModule a pure sender and AcmReceiver a pure
receiver.
The Config struct is in practice no longer used by AudioCodingModule,
so a new definition is included in AcmReceiver. The old definition
remains in AudioCodingModule while downstream clients are being
updated.
Bug: webrtc:14867
Change-Id: If0d0b4214c5aa278cf6c85c5b62c6da644de20e0
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291533
Reviewed-by: Tomas Lundqvist <tomasl@google.com>
Commit-Queue: Henrik Lundin <henrik.lundin@webrtc.org>
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39244}
The lowest level and some of the highest levels of this function are
already using ArrayView. Make this consistent throughout.
Use deprecation for the old API rather than deleting it, since upstream
may be using it.
Bug: webrtc:14870
Change-Id: If5e1a6e9802ecf7e8e3ec27befb5167ca9985517
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291706
Commit-Queue: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Evan Shrubsole <eshr@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39241}
This removes the previous approach where we continued to update the timestamp when the capturer is running but the send stream is stopped in favor of a more general approach that also works when the capturer is paused.
Some assumptions for this change to be correct: input sample rate and frame size will be the same before/after the stream is paused.
Bug: webrtc:12397
Change-Id: I3b03741cd6d3285cbc9aee3893800729852e6cfa
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291526
Commit-Queue: Jakob Ivarsson <jakobi@webrtc.org>
Reviewed-by: Henrik Lundin <henrik.lundin@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39213}
This is to clear any remaining buffers and other state such as the next packet timestamp.
Bug: webrtc:12397
Change-Id: I2ef9a6f7254d82a69a2896ec8aa619ced2694fb8
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291327
Reviewed-by: Henrik Lundin <henrik.lundin@webrtc.org>
Commit-Queue: Jakob Ivarsson <jakobi@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39206}
Remove the default enabled "WebRTC-Audio-FixTimestampStall" field trial which was rolled out 2 years ago without any issues.
Also change the include audio level indication member to be atomic since it is accessed on multiple threads.
Bug: webrtc:14804
Change-Id: I4c5145e1fb03351154162b4293a3bd870e4793cb
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/290886
Reviewed-by: Markus Handell <handellm@webrtc.org>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Commit-Queue: Jakob Ivarsson <jakobi@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#39125}
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 information is now readily available. Let's expose it.
In practise we don't pace audio by default and the delay is ~0, however
we can tell that this metric is working as intended by setting
PacingController's pace_audio_ to true via the "WebRTC-Pacer-BlockAudio"
field trial. In this case chrome://webrtc-internals/ plots neats graphs
for audio send delay.
Bug: webrtc:10635
Change-Id: Iecfd93bb84ec61e5d54232769a9e7a500601b199
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/280523
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Henrik Boström <hbos@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#38483}
It defined RTC_PRIuS, which was needed for compatibility with MSVC
prior to version 2015.
Bug: webrtc:6424
Change-Id: I5668d473376201cad3e8da65927c967fc397804b
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/261314
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Commit-Queue: Niels Moller <nisse@webrtc.org>
Reviewed-by: Tomas Gunnarsson <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36814}
To mitigate race between ~ChannelSend and task created in ProcessAndEncodeAudio.
as described in the comment next to the task queue member.
Bug: b/228933184
Change-Id: Ia0efd050c76a4539dc2525ef8efc065fab96861c
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/258983
Auto-Submit: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Commit-Queue: Jakob Ivarsson <jakobi@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#36553}
Currently, the RtpRtcp module of AudioSendStream is (de)registered in
the packet router on calls to
(Register|Reset)SenderCongestionControlObjects.
This CL changes that to happen on Start/Stop instead, which allows us
to safely call (Get|Set)RtpState on suspend/resume without the need
for extra locking in the rtp module.
See also https://webrtc-review.googlesource.com/c/src/+/228430
Bug: webrtc:11340
Change-Id: I54243a9ace8a7659924269418468b49b967b9465
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/228433
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#34738}
This change achieves an Idle Wakeup savings of 200 Hz.
ModuleRtcRtcpImpl2 had Process() logic only active if TMMBR() is
enabled in RtcpSender, which it never is. Hence the Module
inheritance could be removed. The change removes all known
dependencies of the module inheritance, and any related mentions
of ProcessThread.
Fixed: webrtc:11581
Change-Id: I440942f07187fdb9ac18186dab088633969b340e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/222604
Reviewed-by: Tommi <tommi@webrtc.org>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Commit-Queue: Markus Handell <handellm@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#34358}
This reverts commit f23e2144e8.
Reason for revert: Need further discussion on appropriate thread/tq requirements.
Original change's description:
> Add task queue to RtpRtcpInterface::Configuration.
>
> Let ModuleRtpRtcpImpl2 use the configured value instead of
> TaskQueueBase::Current().
>
> Intention is to allow construction of RtpRtcpImpl2 on any thread.
> If a task queue is provided (required for periodic rtt updates), the
> destruction of the object must be done on that same task queue.
>
> Also, delete ModuleRtpRtcpImpl2::Create, callers updated to use std::make_unique.
>
> Bug: None
> Change-Id: I412b7b1e1ce24722ffd23d16aa6c48a7214c9bcd
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/199968
> Reviewed-by: Sebastian Jansson <srte@webrtc.org>
> Reviewed-by: Sam Zackrisson <saza@webrtc.org>
> Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
> Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Commit-Queue: Niels Moller <nisse@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#32949}
TBR=danilchap@webrtc.org,ilnik@webrtc.org,saza@webrtc.org,nisse@webrtc.org,srte@webrtc.org
Change-Id: I7e5007f524a39a6552973ec9744cd04c13162432
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: None
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/201420
Reviewed-by: Niels Moller <nisse@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Reviewed-by: Tommi <tommi@webrtc.org>
Commit-Queue: Tommi <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32953}
Let ModuleRtpRtcpImpl2 use the configured value instead of
TaskQueueBase::Current().
Intention is to allow construction of RtpRtcpImpl2 on any thread.
If a task queue is provided (required for periodic rtt updates), the
destruction of the object must be done on that same task queue.
Also, delete ModuleRtpRtcpImpl2::Create, callers updated to use std::make_unique.
Bug: None
Change-Id: I412b7b1e1ce24722ffd23d16aa6c48a7214c9bcd
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/199968
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Niels Moller <nisse@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32949}
The lock in TransportFeedbackProxy could cause a dead-lock if audio is
included in transport feedback messages, and necessitated a revert:
https://webrtc-review.googlesource.com/c/src/+/178100
This CL removes that lock and in fact the entire TransportFeedbackProxy
class, and instead sets the observer at construction time.
We therefore don't need to guard the observer pointer anymore.
For further context, see also internal bug b/153893626
Bug: webrtc:10809
Change-Id: I79b08d8d0e587a59736b383c3596a26836c33d2e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/178207
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31583}
The 'Module' part of the implementation must not be
called via the RtpRtcp interface, but is rather a part of
the contract with ProcessThread. That in turn is an
implementation detail for how timers are currently implemented
in the default implementation.
Along the way I'm deprecating away the factory function which
was inside the interface and tied it to one specific implementation.
Instead, I'm moving that to the implementation itself and down the
line, we don't have to go through it if we just want to create an
instance of the class.
The key change is in rtp_rtcp.h and the new rtp_rtcp_interface.h
header file (things moved from rtp_rtcp.h), the rest falls from that.
Change-Id: I294f13e947b9e3e4e649400ee94a11a81e8071ce
Bug: webrtc:11581
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/176419
Reviewed-by: Magnus Flodman <mflodman@webrtc.org>
Commit-Queue: Tommi <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31440}
The method is being used externally to create instances
of the deprecated internal implementation.
Instead, I'm moving how we instantiate the internal implementation into
the implementation itself and move towards keeping the interface
separate from a single implementation.
Change-Id: I743aa86dc4c812b545699c546c253c104719260e
Bug: webrtc:11581
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/176404
Commit-Queue: Tommi <tommi@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31420}
This interface has a couple of issues. Primarily for me, it makes it
difficult work with the paced sender as we need to either temporarily
release a lock or force a thread-handover in order to avoid a cyclic
lock order.
For video in particular, its behavior is also falky since header sizes
can vary not only form frame to frame, but from packet to packet within
a frame (e.g. TimingInfo extension is only on the last packet, if set).
On bitrate allocation, the last reported value is picked, leading to
timing issues affecting the bitrate set.
This CL removes the callback interface and instead we simply poll the
RTP module for a packet overhead. This consists of an expected overhead
based on which non-volatile header extensions are registered (so for
instance AbsoluteCaptureTime is disregarded since it's only populated
once per second). The overhead estimation is a little less accurate but
instead simpler and deterministic.
Bug: webrtc:10809
Change-Id: I2c3d3fcca6ad35704c4c1b6b9e0a39227aada1ea
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173704
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Ali Tofigh <alito@webrtc.org>
Reviewed-by: Jakob Ivarsson <jakobi@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31185}
includes the (random) rtp start offset in the timestamp passed to the frame transformer callback
Bug: chromium:1069278
Change-Id: I7d10130404d93df7cee3b8f87a0b780801a51415
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173329
Commit-Queue: Marina Ciocea <marinaciocea@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Reviewed-by: Marina Ciocea <marinaciocea@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31067}
The frame transformer is passed from RTPSenderInterface through the
library to be eventually set in ChannelSend, where the frame
transformation will occur in the follow-up CL.
Insertable Streams Web API explainer:
https://github.com/alvestrand/webrtc-media-streams/blob/master/explainer.md
Design doc for WebRTC library changes:
http://doc/1eiLkjNUkRy2FssCPLUp6eH08BZuXXoHfbbBP1ZN7EVk
Bug: webrtc:11380
Change-Id: I01b2adc3c96b948d182d5401a9a4fe14cf5960a2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/171870
Commit-Queue: Tommi <tommi@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Reviewed-by: Tommi <tommi@webrtc.org>
Reviewed-by: Magnus Flodman <mflodman@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30946}
This reduces indirection and makes it easier to follow code. It also
fits into a long term strategy of reducing the scope of ChannelSend.
Bug: webrtc:9883
Change-Id: I2661c4aa6c561f7691beaaa289636254f7a58b72
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/166042
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30273}
