In several places VideoFrame::Builder is used to create a new VideoFrame
when intent is to change only one or two fields of a const VideoFrame&.
This approach is bad because each and every metadata field have to be
added to all the places.
Instead, this CL adds missing setters and refactors the code to use
full copy of a VideoFrame and update required fields only.
Along the way few actual bugs are fixed, e.g. when ColorSpace isn't copied
when frame rotation or buffer is cropped or converted.
Bug: webrtc:10460
Change-Id: I2895a473ca938b150eed2916c689060bdf58cb25
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/140102
Reviewed-by: Niels Moller <nisse@webrtc.org>
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28170}
Some of the TODOs associated with webrtc:10336 which are
currently in the codebase have recently been resolved,
but not all relevant TODOs have been removed.
TBR=kwiberg@webrtc.org
Bug: webrtc:10336
Change-Id: Iff1d0fc94dee5bf49226f6ea3d9127fea77e9d68
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/139902
Reviewed-by: Elad Alon <eladalon@webrtc.org>
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28167}
Replaced by separate methods
SendPictureLossIndication and SendFullIntraRequest.
The split SetKeyFrameRequestMethod/RequestKeyFrame implicitly
requires that the two methods are called on the same thread, to avoid a
data race. After downstream code is updated, both deprecated
methods and the member |ModuleRtpRtcpImpl::key_frame_req_method_| can
be deleted.
Bug: None
Change-Id: I454f6d16b667f2306cba0dec467ddc183ad449c8
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/140043
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Niels Moller <nisse@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28163}
Reset() is called each time the encoder is reconfigured, but then it
happens the target bitrate isn't reset in encoder. So it might produce a
frame before next bitrate estimate is propagated to the metadata writer.
The incorrect zero bitrate would be treated as a paused encoder and would
cause metadata to be dropped.
Also, added unittest for that scenario at VideoStreamEncoder level.
Bug: webrtc:10460
Change-Id: I28024a527f1fb8474b172e2c5c2394fd38d69a07
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/140101
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28159}
Loss notifications may either be sent immediately, or wait until another
RTCP feedback message is sent.
Bug: webrtc:10336
Change-Id: I40601d9fa1dec6c17b2ce905cb0c8cd2dcff7893
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/139242
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28142}
Currently, if LNTF and NACK messages are both created, they will
be sent out in separate RTCP messages. This is wasteful.
This CL is the first of in a series of CLs that will ensure that
these feedback messages can be buffered together, without introducing
more of a delay than the CPU time required to process both messages.
Bug: webrtc:10336
Change-Id: I950324112ee346695a12a17d025483ea5e99c732
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/139112
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28136}
This is a partial revert of
https://webrtc-review.googlesource.com/c/src/+/130101.
The KeyFrameRequestSender argument is added back to the constructor of
RtpVideoStreamReceiver. It is optional; if a null pointer is passed,
key frame requests are sent via the internal RtpRtcp module, and this is
how the class is used by VideoReceiveStream. An injectable
KeyFrameRequestSender is useful for tests, for downstream applications
that want to route key frame requests elsewhere, and should also aid
later migration to RtcpTransciever.
Bug: None
Change-Id: Idf9baeed21570625ad74e9afbe38f7ea5bf79feb
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/139107
Commit-Queue: Niels Moller <nisse@webrtc.org>
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28102}
Currently used ':' is bad because it prevents from specifying absolute
path on windows (e.g. "C:\directory").
Now to specify path on windows, it can be passed unchanged in field trial:
"WebRTC-DecoderDataDumpDirectory/C:\path\on\windows/"
On linux ';' has to be used instead of '/':
"WebRTC-DecoderDataDumpDirectory/;path;on;linux/"
Bug: none
Change-Id: Ia46c94bdfab95385618dde4fd3f2857e9ddf2d1c
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/138832
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Elad Alon <eladalon@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28085}
LNTF (loss notifications) are no longer mutually exclusive with
NACK and key frames; a receiver may send both, and the sender would
be allowed to choose which to regard.
Bug: webrtc:10336
Change-Id: I1ae7d972f9f47b07fe2f493bb31c1916456d6a3a
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/138828
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28081}
This is part of implementing RTCRemoteInboundRtpStreamStats. The CL was
split up into smaller pieces for reviewability. Spec:
https://w3c.github.io/webrtc-stats/#dom-rtcremoteinboundrtpstreamstats
In [1], ReportBlockData was implemented and tested.
This CL adds the plumbing to make it available in MediaSenderInfo for
video streams, but the code is not wired up to make use of these stats.
In follow-up CL [2], ReportBlockData will be used to implement
RTCRemoteInboundRtpStreamStats. The follow-up CL will add integration
tests exercising the full code path.
[1] https://webrtc-review.googlesource.com/c/src/+/136584
[2] https://webrtc-review.googlesource.com/c/src/+/138067
Bug: webrtc:10456
Change-Id: Icd20452cb4b4908203b28ae9d9f52c25693cf91d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/138065
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Niels Moller <nisse@webrtc.org>
Commit-Queue: Henrik Boström <hbos@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28071}
When the LossNotifications field trial is in effect, LNTF should
be offered/accepted in the SDP message, not assumed to be configured
on both sides equally.
Bug: webrtc:10662
Change-Id: Ibd827779bd301821cbb4196857f6baebfc9e7dc2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/138079
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Reviewed-by: Niels Moller <nisse@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28056}
Reland with fixes.
If input framerate is a little unstable, using it to cap layers will
make output framerate even smaller for longer periods of time.
Also, fix screenshare_loopback test for low-fps vp9 testing.
Bug: webrtc:10257
Change-Id: Id40a780d461e6b51cb44d275b8aa5d7b348d3586
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/138215
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28054}
This reverts commit eb1754c5750dfcad23ac62b47aa3aa2176ae7be2.
Reason for revert: breaks downstream projects
Original change's description:
> VP9 screenshare: Don't base layers frame-rate on input frame-rate
>
> If input framerate is a little unstable, using it to cap layers will
> make output framerate even smaller for longer periods of time.
>
> Also, fix screenshare_loopback test for low-fps vp9 testing.
>
> Bug: webrtc:10257
> Change-Id: I64aa087e859ab4ab8e484c9ab7f5ac0fb18bd37d
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/138204
> Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#28050}
TBR=ilnik@webrtc.org,ssilkin@webrtc.org
Change-Id: I82bfbac58249cfe0da5ff565aa97a4745fd078ff
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:10257
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/138213
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28051}
If input framerate is a little unstable, using it to cap layers will
make output framerate even smaller for longer periods of time.
Also, fix screenshare_loopback test for low-fps vp9 testing.
Bug: webrtc:10257
Change-Id: I64aa087e859ab4ab8e484c9ab7f5ac0fb18bd37d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/138204
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28050}
This is a standardized metric:
https://w3c.github.io/webrtc-stats/#dom-rtcoutboundrtpstreamstats-totalencodedbytestarget
We estimate the target frame size in bytes from the current encoder
target bitrate and encoder framerate.
We would expect that the average bytes produced by the encoder would
over time match the average target, which is calculated by polling
getStats() twice and dividing the delta totalEncodedBytesTarget with
the delta framesEncoded. This is meant to make googTargetEncBitrate
obsolete.
Bug: webrtc:10446
Change-Id: Ib10ce236476a2f965582d5c536f419952926d4e6
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/137200
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Commit-Queue: Henrik Boström <hbos@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28022}
RTP timestamp was recently added to contributing sources in the WebRTC
specification. This CL implements that change in WebRTC.
Bug: webrtc:10650
Change-Id: Ic0ccfbea7049a5b66063fa6cf60d01d5bd713132
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/137515
Reviewed-by: Fredrik Solenberg <solenberg@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Commit-Queue: Johannes Kron <kron@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28020}
Currently we pass media_transport from PeerConnection to media layers. The goal of this change is to replace media_transport with struct MediaTransportCondif, which will enable adding different transports (i.e. we plan to add DatagramTransport) as well as other media-transport related settings without changing 100s of files.
TODO: In the future we should consider also adding rtp_transport in the same config, but it will require a bit more work, so I did not include it in the same change.
Bug: webrtc:9719
Change-Id: Ie31e1faa3ed9e6beefe30a3da208130509ce00cd
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/137181
Commit-Queue: Anton Sukhanov <sukhanov@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Fredrik Solenberg <solenberg@webrtc.org>
Reviewed-by: Steve Anton <steveanton@webrtc.org>
Reviewed-by: Bjorn Mellem <mellem@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#28016}
RtpVideoStreamReceiver used to pass the PacketRouter when creating its
RtpRtcp module, but it's not needed for a receive-only module. Make the
PacketRouter optional to the constructor; it's used only for registering
the created RtpRtcp module as a candidate for sending rtcp feedback.
Bug: None
Change-Id: I371a0bdb9d68ac48b16f52e1d7939f8c177dc528
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/137429
Commit-Queue: Niels Moller <nisse@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27984}
This is a standardized metric. Spec:
https://w3c.github.io/webrtc-stats/#dom-rtcoutboundrtpstreamstats-totalpacketsenddelay
It is meant to replace the legacy googBucketDelay. The average
packet delay over any interval can be calculated as the delta
totalPacketSendDelay divided by the delta packetsSent between two
calls to getStats().
Bug: webrtc:10506
Change-Id: I3d6c6d66e5a06937d0ea8d182a82cd255084ad19
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/137044
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Commit-Queue: Henrik Boström <hbos@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27979}
Reland with fixes.
Currently some video frames metadata like rotation or ntp timestamps are
copied in every encoder and decoder separately. This CL makes copying to
happen at a single place for send or receive side. This will make it
easier to add new metadata in the future.
Also, added some missing tests.
Original Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/133346
Bug: webrtc:10460
Change-Id: I98629589fa55ca1d74056033cf86faccfdf848cd
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/136582
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Niels Moller <nisse@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27930}
Use the C++-style stdlib headers, add `std::` prefix, in order to avoid implicit casts to double.
Bug: None
Change-Id: I78d9caaee715be341d2480c6d5e769068966d577
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/133625
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Commit-Queue: Oleh Prypin <oprypin@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27905}
This reverts commit 4fb12b0caec9faa57cfbceb0f86b0e10c32a0cc2.
Reason for revert: Breaks some asan chromium bots
Original change's description:
> Reland "Copy video frames metadata between encoded and plain frames in one place"
>
> Reland with fixes.
>
> Currently some video frames metadata like rotation or ntp timestamps are
> copied in every encoder and decoder separately. This CL makes copying to
> happen at a single place for send or receive side. This will make it
> easier to add new metadata in the future.
>
> Also, added some missing tests.
>
> Original Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/133346
>
> Bug: webrtc:10460
> Change-Id: Ia71198685de7fbd990704b575231cdce94dc0645
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/134961
> Reviewed-by: Johannes Kron <kron@webrtc.org>
> Reviewed-by: Niels Moller <nisse@webrtc.org>
> Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#27828}
TBR=ilnik@webrtc.org,nisse@webrtc.org,sprang@webrtc.org,kron@webrtc.org,artit@webrtc.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: webrtc:10460
Change-Id: I9c87a43a716622b389974cb8377f973573fc29a5
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/135747
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27895}
It's more convenient to have it next to other freeze-related metrics
when sorted alphabetically.
Bug: None
Change-Id: I8f73e32a9cdb8166e139c193ffcd1dcc1fa18533
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/135563
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27874}
Make Vp8FrameBufferController::UpdateConfiguration return a set
of desired overrides. These overrides are cumulative with
previously returned override sets.
Bug: webrtc:10382
Change-Id: I1aa9544ae0cf6c57115e80963b3bbcdc3101db5e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/134649
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27835}
Reland with fixes.
Currently some video frames metadata like rotation or ntp timestamps are
copied in every encoder and decoder separately. This CL makes copying to
happen at a single place for send or receive side. This will make it
easier to add new metadata in the future.
Also, added some missing tests.
Original Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/133346
Bug: webrtc:10460
Change-Id: Ia71198685de7fbd990704b575231cdce94dc0645
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/134961
Reviewed-by: Johannes Kron <kron@webrtc.org>
Reviewed-by: Niels Moller <nisse@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27828}
