Implements a two-pass approach to packetization which creates
packets of an even size similar to RtpPacketizer::SplitAboutEqually.
This improves the bandwidth estimation.
The algorithm does a first pass with the existing packetizer, then
iterates through the resulting packet sizes and sums up the bytes left unused in each packet.
It then calculates a new maximum packet length as
configured_max_packet_len - ((unused_bytes - packets + 1) / packets)
adjusts for the overhead and re-runs the packetization algorithm.
For example, a list of OBUs with sizes
{1206, 1476, 1431}
currently gets packetized greedily as payload sizes
{1200, 1200, 1200, 523}
With this change, it gets packetized as
{1032, 1032, 1032, 1028}
This change is guarded by the field trial
WebRTC-Video-AV1EvenPayloadSizes
which is acting as a rollout flag.
BUG=webrtc:15927
Co-authored-by: Shyam Sadhwani <shyamsadhwani@meta.com>
Change-Id: I4f0b3c27de6f06104908dd769c4dd1f34115712c
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/348100
Commit-Queue: Philipp Hancke <phancke@meta.com>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Per Kjellander <perkj@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#42203}
Removed thread safety: for a low level helper it adds overhead that users may not need. In particular RtpSenderVideo doesn't need it because calls to SendVideo are already synchronized.
Added a feature to force producing extension as requested by downstream.
Cleanup and document api:
Changed rtp_frequency type to int as it has no reason to use uint32_t per style guide
Changed absolute_capture_time to NtpTime to clarify both units and offset of the time. NtpTime has trivial conversion to/from uint64_t
Documented all the parameters.
Cleanup tests.
Bug: b/307553606
Change-Id: I0922ca4d3c89f124eeb561742dca79ed5c2327fd
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/325022
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Chen Xing <chxg@google.com>
Cr-Commit-Position: refs/heads/main@{#41023}
Remove support for setting one limit without another limit
because related rtp header extension doesn't support such values.
Start morphing VideoPlayouDelay into a class and stricter type: add accessors returning TimeDelta
Bug: webrtc:13756
Change-Id: If0dd02620528dc870b015beeff3a8103e04022ee
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/315921
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Philip Eliasson <philipel@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#40570}
Currently FecController knows about network conditions, these information can be used to control RTX settings in-call.
Change-Id: I8f84164aeac48ea13b7f1cf82fd7424431f98ada
Bug: webrtc:15167
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/304800
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Per Kjellander <perkj@webrtc.org>
Commit-Queue: Ying Wang <yinwa@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#40192}
Allow CSRCs to be modified per-frame in an Encoded Insertable Streams
transform, to support a web API which allows per-frame CSRC
modifications to signal when a JS application has changed the source
of the video which is written into an encoded frame.
Initially only for Video, with Audio support likely to follow later.
Bug: webrtc:14709
Change-Id: Ib34f35faa9cee56216b30eaae42d7e65c78bb9f2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/291324
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Tove Petersson <tovep@google.com>
Reviewed-by: Henrik Boström <hbos@webrtc.org>
Commit-Queue: Tony Herre <herre@google.com>
Cr-Commit-Position: refs/heads/main@{#39214}
Instead of re-using the sender task queue, a new task queue will
suffice.
Bug: webrtc:14445
Change-Id: Ia7395ace2f0bb66bf9e76e3783b208f2cd0385dc
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/275771
Commit-Queue: Evan Shrubsole <eshr@webrtc.org>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#38332}
Sends a VideoLayersAllocation header extension if frame rate change more than 5fps since the last time it was sent with valid frame rate and resolution.
Bug: webrtc:12000
Change-Id: I2572c966025cc2c22743bbe2187cec7cceb86d01
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/234752
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Cr-Commit-Position: refs/heads/main@{#35180}
This allows callers to use timestamps generated from their own clocks
without worrying about converting to webrtc time.
No-Try because of lack of infra lack of capacity on macs.
No-Try: True
Bug: webrtc:11327
Change-Id: I7b1935654a2b23cf844c7b3622ed68763ced9da5
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/219785
Commit-Queue: Paul Hallak <phallak@google.com>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Minyue Li <minyue@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#34076}
increase.
VP9 and other codecs can in theory add spatial layers without a key
frame.
Bug: webrtc:12000
Change-Id: I27461af2e34c855203a130e400a6aa01144d3cf7
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/198781
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32883}
This adds a method to allow VideoLayersAllocation to be sent using the header extension RtpVideoLayersAllocationExtension.
Bug: webrtc:12000
Change-Id: Iafdc1e16911c57ca55d7cc0559a0b45774211e92
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/187495
Commit-Queue: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32397}
Video playout delay is used to give a hint to the receiver
how the video should be played out.
Add the field trial WebRTC-ForceSendPlayoutDelay to set the video playout
delay of outgoing RTP packets to enable experimentation with this feature.
Bug: webrtc:11896
Change-Id: Ie6123b5967763bde6a830f4c5e5a963e73fb0acb
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/185042
Commit-Queue: Johannes Kron <kron@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32169}
We can then finally delete the top-level common_types.h, and the
corresponding build target webrtc_common.
Bug: webrtc:7660
Change-Id: I1c1096541477586d90774c7a3405b9d36edec14a
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/182800
Commit-Queue: Niels Moller <nisse@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32044}
It is not longer needed by the rtp_rtcp module.
Bug: webrtc:6471
Change-Id: I89a4374a50c54a02e9f20a5ce789eac308aaffeb
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/179523
Reviewed-by: Philip Eliasson <philipel@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31773}
worker_queue is used in many places and it can be confusing. This queue
is the send transport's worker queue. Rename to send_transport_queue to
reflect that.
Bug: none
Change-Id: I43c5c4cbddaee3dae1ff75aa38dc3ddee6585902
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/176362
Commit-Queue: Tommi <tommi@webrtc.org>
Reviewed-by: Tommi <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31396}
When video frame encoding is done on an external thread (for example in
the case of hardware encoders), the WebRTC TaskQueueBase::Current() is
null; in this case use the worker queue instead to send transformed
frames.
Bug: chromium:1086373
Change-Id: I903ddc52ad6832557fc5b5f76396fe26cf5a88f3
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/176303
Reviewed-by: Magnus Flodman <mflodman@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Marina Ciocea <marinaciocea@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31388}
RtpVideoSender now stores fec type and overhead instead of querying the
generator all the time. Setting of protection parameters and asking for
current bitrate is also now handled just by the VideoFecGenerator
instance, instead of going via RtpVideoSender.
Finally, adds method to query for RtpState in VideoFecGenerator
interface. This avoids an ugly cast that would have been even more
trouble after moving fec generation.
For context, see https://webrtc-review.googlesource.com/c/src/+/173708
Bug: webrtc:11340
Change-Id: Ia5e6cd919e71850c9cc5ed5a4f4417338d577162
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/174203
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31166}
make it a member function which allows to reduce number of parameters
and simplify accessing more state in the future.
Bug: None
Change-Id: Iba35125c0c2cf1d6bb67b106c1f73a33ecb8e44e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/170366
Reviewed-by: Johannes Kron <kron@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30797}
This is a reland of 49734dc0fa
Patchset 2 contains a fix for the fuzzer set up. Since we now parse
an RtpPacket out of the fuzzer data, the header needs to be correct,
otherwise we fail before even reaching the FEC code that we actually
want to test.
Bug: webrtc:11340, chromium:1052323, chromium:1055974
TBR=stefan@webrtc.org
Original change's description:
> Reland "Refactors UlpFec and FlexFec to use a common interface."
>
> This is a reland of 11af1d7444
>
> Original change's description:
> > Refactors UlpFec and FlexFec to use a common interface.
> >
> > The new VideoFecGenerator is now injected into RtpSenderVideo,
> > and generalizes the usage.
> > This also prepares for being able to genera FEC in the RTP egress
> > module.
> >
> > Bug: webrtc:11340
> > Change-Id: I8aa873129b2fb4131eb3399ee88f6ea2747155a3
> > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168347
> > Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> > Reviewed-by: Sebastian Jansson <srte@webrtc.org>
> > Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
> > Commit-Queue: Erik Språng <sprang@webrtc.org>
> > Cr-Commit-Position: refs/heads/master@{#30515}
>
> Bug: webrtc:11340, chromium:1052323
> Change-Id: Id646047365f1c46cca9e6f3e8eefa5151207b4a0
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168608
> Commit-Queue: Erik Språng <sprang@webrtc.org>
> Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#30593}
Bug: webrtc:11340, chromium:1052323
Change-Id: Ib8925f44e2edfcfeadc95c845c3bfc23822604ed
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/169222
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30724}
Add a new API in RTPSenderInterface, to be called from the browser side
to insert a frame transformer between the Encoded and the Packetizer.
The frame transformer is passed from RTPSenderInterface through the
library to be eventually set in RTPSenderVideo, where the frame
transformation will occur in the follow-up CL
https://webrtc-review.googlesource.com/c/src/+/169128.
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: I46cd0d8a798c2736c837e90cbf90d8901c7d27fb
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/169127
Commit-Queue: Marina Ciocea <marinaciocea@webrtc.org>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30642}
This reverts commit 49734dc0fa.
Reason for revert: Still something wrong with ulpfec fuzzer setup.
Original change's description:
> Reland "Refactors UlpFec and FlexFec to use a common interface."
>
> This is a reland of 11af1d7444
>
> Original change's description:
> > Refactors UlpFec and FlexFec to use a common interface.
> >
> > The new VideoFecGenerator is now injected into RtpSenderVideo,
> > and generalizes the usage.
> > This also prepares for being able to genera FEC in the RTP egress
> > module.
> >
> > Bug: webrtc:11340
> > Change-Id: I8aa873129b2fb4131eb3399ee88f6ea2747155a3
> > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168347
> > Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> > Reviewed-by: Sebastian Jansson <srte@webrtc.org>
> > Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
> > Commit-Queue: Erik Språng <sprang@webrtc.org>
> > Cr-Commit-Position: refs/heads/master@{#30515}
>
> Bug: webrtc:11340, chromium:1052323
> Change-Id: Id646047365f1c46cca9e6f3e8eefa5151207b4a0
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168608
> Commit-Queue: Erik Språng <sprang@webrtc.org>
> Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#30593}
TBR=sprang@webrtc.org,stefan@webrtc.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: webrtc:11340, chromium:1052323
Change-Id: I920ce0a48a08768d7a98a563e2b66bd6eb8602b7
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/169121
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30616}
This is a reland of 11af1d7444
Original change's description:
> Refactors UlpFec and FlexFec to use a common interface.
>
> The new VideoFecGenerator is now injected into RtpSenderVideo,
> and generalizes the usage.
> This also prepares for being able to genera FEC in the RTP egress
> module.
>
> Bug: webrtc:11340
> Change-Id: I8aa873129b2fb4131eb3399ee88f6ea2747155a3
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168347
> Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> Reviewed-by: Sebastian Jansson <srte@webrtc.org>
> Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
> Commit-Queue: Erik Språng <sprang@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#30515}
Bug: webrtc:11340, chromium:1052323
Change-Id: Id646047365f1c46cca9e6f3e8eefa5151207b4a0
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168608
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30593}
This reverts commit 11af1d7444.
Reason for revert: Possible crash
Original change's description:
> Refactors UlpFec and FlexFec to use a common interface.
>
> The new VideoFecGenerator is now injected into RtpSenderVideo,
> and generalizes the usage.
> This also prepares for being able to genera FEC in the RTP egress
> module.
>
> Bug: webrtc:11340
> Change-Id: I8aa873129b2fb4131eb3399ee88f6ea2747155a3
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168347
> Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> Reviewed-by: Sebastian Jansson <srte@webrtc.org>
> Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
> Commit-Queue: Erik Språng <sprang@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#30515}
TBR=brandtr@webrtc.org,sprang@webrtc.org,stefan@webrtc.org,srte@webrtc.org
Change-Id: Iddf112d801621c8a4370b853cee3fa42bf2c7fba
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:11340
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168603
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30524}
The new VideoFecGenerator is now injected into RtpSenderVideo,
and generalizes the usage.
This also prepares for being able to genera FEC in the RTP egress
module.
Bug: webrtc:11340
Change-Id: I8aa873129b2fb4131eb3399ee88f6ea2747155a3
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168347
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30515}
References to PlayoutDelayOracle and the deprecated RtpSenderVideo
constructor have been removed in downstream code, we can now clean the
unused code away.
Bug: webrtc:10809, webrtc:11340
Change-Id: I789274be2079ad4ddd7e83a5fa249b06a32a4e82
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168400
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30499}
This is a reland of 4f68f5398d
Original change's description:
> Remove PlayoutDelayOracle and make RtpSenderVideo guarantee delivery
>
> The PlayoutDelayOracle was responsible for making sure the PlayoutDelay
> header extension was successfully propagated to the receiving side. Once
> it was determined that the receiver had received a frame with the new
> delay tag, it's no longer necessary to propagate.
>
> The issue with this implementation is that it is based on max
> extended sequence number reported via RTCP, which makes it often slow
> to react, could theoretically fail to produce desired outcome (max
> received > X does not guarantee X was fully received and decoded), and
> added a lot of code complexity.
>
> The guarantee of delivery can in fact be accomplished more reliably and
> with less code by making sure to tag each frame until an undiscardable
> frame is sent.
>
> This allows containing the logic fully within RTPSenderVideo.
>
> Bug: webrtc:11340
> Change-Id: I2d1d2b6b67f4f07b8b33336f8fcfcde724243eef
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168221
> Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> Reviewed-by: Sebastian Jansson <srte@webrtc.org>
> Commit-Queue: Erik Språng <sprang@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#30473}
TBR=stefan@webrtc.org
Bug: webrtc:11340
Change-Id: I2fdd0004121b13b96497b21e052359e31d0c477a
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168305
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30479}
This reverts commit 4f68f5398d.
Reason for revert: Breaks downstream project
Original change's description:
> Remove PlayoutDelayOracle and make RtpSenderVideo guarantee delivery
>
> The PlayoutDelayOracle was responsible for making sure the PlayoutDelay
> header extension was successfully propagated to the receiving side. Once
> it was determined that the receiver had received a frame with the new
> delay tag, it's no longer necessary to propagate.
>
> The issue with this implementation is that it is based on max
> extended sequence number reported via RTCP, which makes it often slow
> to react, could theoretically fail to produce desired outcome (max
> received > X does not guarantee X was fully received and decoded), and
> added a lot of code complexity.
>
> The guarantee of delivery can in fact be accomplished more reliably and
> with less code by making sure to tag each frame until an undiscardable
> frame is sent.
>
> This allows containing the logic fully within RTPSenderVideo.
>
> Bug: webrtc:11340
> Change-Id: I2d1d2b6b67f4f07b8b33336f8fcfcde724243eef
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168221
> Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> Reviewed-by: Sebastian Jansson <srte@webrtc.org>
> Commit-Queue: Erik Språng <sprang@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#30473}
TBR=sprang@webrtc.org,stefan@webrtc.org,srte@webrtc.org
Change-Id: Ide922e680ae36bb69b95e58002482cf5ed57e254
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:11340
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168304
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30475}
The PlayoutDelayOracle was responsible for making sure the PlayoutDelay
header extension was successfully propagated to the receiving side. Once
it was determined that the receiver had received a frame with the new
delay tag, it's no longer necessary to propagate.
The issue with this implementation is that it is based on max
extended sequence number reported via RTCP, which makes it often slow
to react, could theoretically fail to produce desired outcome (max
received > X does not guarantee X was fully received and decoded), and
added a lot of code complexity.
The guarantee of delivery can in fact be accomplished more reliably and
with less code by making sure to tag each frame until an undiscardable
frame is sent.
This allows containing the logic fully within RTPSenderVideo.
Bug: webrtc:11340
Change-Id: I2d1d2b6b67f4f07b8b33336f8fcfcde724243eef
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/168221
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30473}
