Vp8FrameBufferController is currently just a renamed Vp8TemporalLayers,
but subsequent CLs will modify Vp8FrameBufferController in ways that are
not relevant for Vp8TemporalLayers. Namely:
1. Loss notifications will be added.
2. Packet-loss rate will be tracked.
3. RTT will be tracked.
4. Vp8FrameBufferController will be made injectable.
Vp8TemporalLayers is retained in order to:
1. Avoid needlessly changing api/.
2. Place for code shared between DefaultTemporalLayers and ScreenshareLayers.
We can remove it in the future (with a proper public announcement).
Bug: webrtc:10382
Change-Id: I49ad1b9bc1954d51bb0b5e60361985f1eb12ae9f
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/126045
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27009}
This CL plumbs an additional signal from VideoSendStream down to
VideoStreamEncoder, namely the amount of headroom that's left between
the encoder max bitrate and the current bitrate allocation for the
media track.
This will be used in follow-up CLs to tune encoder rate adjustment
and some codec specific paramaters a bit differently, based on the
knowledge if we are network constrained or not.
Bug: webrtc:10155
Change-Id: Ic6ccc79be5c6845468bab65b4ca9918b56923fa4
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125981
Reviewed-by: Niels Moller <nisse@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27008}
This change disables DTLS 1.0, TLS 1.0 and TLS 1.1 in WebRTC by default. This
is part of a larger effort at Google to remove old TLS protocols:
https://security.googleblog.com/2018/10/modernizing-transport-security.html
For the M74 timeline I have added a disabled by default field trial
WebRTC-LegacyTlsProtocols which can be enabled to support these cipher suites
as consumers move away from these legacy cipher protocols but it will be off
in Chrome.
This is compliant with the webrtc-security-arch specification which states:
All Implementations MUST implement DTLS 1.2 with the
TLS_ECDHE_ECDSA_WITH_AES_128_GCM_SHA256 cipher suite and the P-256
curve [FIPS186]. Earlier drafts of this specification required DTLS
1.0 with the cipher suite TLS_ECDHE_ECDSA_WITH_AES_128_CBC_SHA, and
at the time of this writing some implementations do not support DTLS
1.2; endpoints which support only DTLS 1.2 might encounter
interoperability issues. The DTLS-SRTP protection profile
SRTP_AES128_CM_HMAC_SHA1_80 MUST be supported for SRTP.
Implementations MUST favor cipher suites which support (Perfect
Forward Secrecy) PFS over non-PFS cipher suites and SHOULD favor AEAD
over non-AEAD cipher suites.
Bug: webrtc:10261
Change-Id: I847c567592911cc437f095376ad67585b4355fc0
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125141
Commit-Queue: Benjamin Wright <benwright@webrtc.org>
Reviewed-by: David Benjamin <davidben@webrtc.org>
Reviewed-by: Qingsi Wang <qingsi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27006}
If minQP is reached and encoder undershoot consistently, we consider the
quality good enough and throttle encode frame rate.
Bug: webrtc:10310
Change-Id: Ifd07280040dd67ef6e544efdd4619d47bff951e8
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125461
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27003}
If minQP is reached and encoder undershoot consistently, we consider the
quality good enough and throttle encode frame rate.
This CL also adds perf tests for high fps vp9 screenshare.
Bug: webrtc:10310
Change-Id: I49fc7d31f9f596a9ecb5f85fe9e0c7861d4915f9
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125761
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26997}
to clearly signal passed ownership.
Drop support for accepting nullptr clock to avoid copying the Configuration structure.
Update all calls in webrtc to the new factory function
Bug: None
Change-Id: Ic5a78da8e59ba3988a757a9d9634fa31499ce0db
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125901
Reviewed-by: Niels Moller <nisse@webrtc.org>
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Reviewed-by: Oskar Sundbom <ossu@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26994}
I will deprecate deps in proto_library for improved build throughput.
We can use link_deps here instead.
Bug: chromium:938011
Change-Id: Iafa83000c3f7f9ffdc0c376a2297b4a9380b7594
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125820
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Commit-Queue: Takuto Ikuta <tikuta@google.com>
Cr-Commit-Position: refs/heads/master@{#26989}
Subclasses of FakeEncoder need to fill out the CodecSpecificInfo and
RTPFragmentationHeader, and they also write to the encoded data of the
EncodedImage. This used to be done by subclasses chaining onto the
parent's OnEncodedImage callback, but that's not so nice, since the
EncodedImage argument is passed as a const ref there.
This change introduces a protected method EncodeHook for this purpose.
FakeEncoder calls this prior to calling OnEncodedImage, with non-const
pointers.
In addition, change FakeEncoder to use EncodedImage::Allocate, rather
than explicit new and delete.
Bug: webrtc:9378
Change-Id: Ie8182d1d5224aa3b7f15905612f6dbcebef0a555
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125880
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Commit-Queue: Niels Moller <nisse@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26988}
In this CL:
- Updated Vp8TemporalLayers::OnEncodeDone to take a CodecSpecificInfo
instead of a CodecSpecificInfoVP8, so that both the VP8 specific and
generic information can be populated.
- Added structs to represent the GFD template structure.
- Added code to generate templates for video/screensharing.
Bug: webrtc:10342
Change-Id: I978f9d708597a6f86bbdc494e62acf7a7b400db3
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/123422
Commit-Queue: Philip Eliasson <philipel@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26987}
The initial implementation forced the sender to use different sizes
of the RTP header extension depending on if a feedback request is
included or not. This can be a problem if the RTP header is pre-
allocated.
This CL changes this so that a static size of 4 bytes can be used
for the TransportSequenceNumberV2 RTP header extension. The change
in the protocol to get this to work is that
FeedbackRequest::sequence_count == 0 means that no feedback is
requested, and FeedbackRequest::sequence_count == 1 means that
feedback is requested for the current packet only.
Bug: webrtc:10262
Change-Id: Ia5134b3daf49f8a5b89f6c717894f6e055f39c8e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125420
Commit-Queue: Johannes Kron <kron@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26985}
Metrics are added to measure:
1. The number of send encodings in calls to AddTransceiver.
2. The number of times that simulcast is disabled because there is no
support from remote peer.
3. The number of times simulcast is indicated in ApplyLocal and
ApplyRemote and with which API surface (no simulcast, legacy munging,
spec-compliant).
Bug: webrtc:10372
Change-Id: I84717a1911efdf8aaf43cd6c04c7f09fcf2c58f0
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125482
Commit-Queue: Amit Hilbuch <amithi@webrtc.org>
Reviewed-by: Steve Anton <steveanton@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26979}
This is a reland of d9f798a6b368024513b0dee5456703849608827d
Original change's description:
> Remove field trial include from decision logic.
>
> Bug: webrtc:9289
> Change-Id: I2e465bf9eddda8bde50daeb14cfd51405e536ff4
> Reviewed-on: https://webrtc-review.googlesource.com/c/125097
> Reviewed-by: Minyue Li <minyue@webrtc.org>
> Commit-Queue: Jakob Ivarsson <jakobi@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#26925}
Bug: webrtc:9289
Change-Id: I40fbd999fc8495beaeb46799c333f91d72b5be37
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125720
Reviewed-by: Minyue Li <minyue@webrtc.org>
Commit-Queue: Minyue Li <minyue@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26978}
Ignore rtc_link_task_queue_impl flag,
instead use build_with_chromium for custom chromium implementation injection
This changes TaskQueue implementation used in webrtc fuzzers in chromium:
from own webrtc implementation to chromium's.
Bug: webrtc:10191
Change-Id: I63be28b680ae8ea8ee1dbf0c699263c392ce29d3
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125196
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26977}
Currently, tests that verify metrics use a combination of
metrics::NumSamples and metrics::NumEvents to assert which samples
were recorded and how many times they were recorded. This means
that a comprehensive tests has n + 1 assertions for n distinct
samples.
The new metrics::Samples function returns a map of sample --> num
events which can be asserted against using gmock matchers,
achieving better coverage and better test failure messages in just
one line.
Bug: None
Change-Id: I07d4a766654cfc04e414b77b6de02927683a361f
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125486
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Commit-Queue: Steve Anton <steveanton@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26974}
In QualityAnalyzingVideoEncoder all encoded images that belongs to
unrelated simulcast streams will be marked as to be discarded. So
to support simulcast streams QualityAnalyzingVideoDecoder have to return
black frames when all encoded images in received concatenated encoded
image are marked as to be discarded. Also QualityAnalyzingVideoDecoder
shouldn't pass such encoded image into VideoQualityAnalyzerInterface.
Bug: webrtc:10138
Change-Id: I0f793a7dc04b5d6b10949479bd074b2db86c5c6f
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125460
Commit-Queue: Artem Titov <titovartem@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Peter Slatala <psla@webrtc.org>
Reviewed-by: Yves Gerey <yvesg@google.com>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26973}
This reverts commit 9b9344742b186b14d87e827e71a1757f4c94b30e.
Reason for revert: Caused test flakiness.
Original change's description:
> Removes lock from ChannelSend.
>
> The lock isn't really needed as encoder_queue_is_active_ can be checked
> on the task queue to provide synchronization.
>
> There is one behavioral change due to this: We will not cancel any currently
> pending encoding tasks when we stop sending, they will be allowed to finish.
>
> Bug: webrtc:10365
> Change-Id: I2b4897dde8d49bc7ee5d2d69694616aee8aaea38
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125096
> Reviewed-by: Oskar Sundbom <ossu@webrtc.org>
> Commit-Queue: Sebastian Jansson <srte@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#26963}
TBR=ossu@webrtc.org,srte@webrtc.org
Change-Id: I30409414d3dc7b0be75b14a70dfc4457f5682a8c
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:10365
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125726
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Commit-Queue: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26971}
For a single layer vp9, the target bitrate was not set correctly. This
may cause a problem for screenshare case, since target bitrate is
respected in that case. If it were less than a min bitrate, the only
spatial layer was permanently disabled.
Bug: webrtc:10257
Change-Id: I0980349adfc2970f810acc51a3e2a31ecbb2bbd2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125681
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26970}
The lock isn't really needed as encoder_queue_is_active_ can be checked
on the task queue to provide synchronization.
There is one behavioral change due to this: We will not cancel any currently
pending encoding tasks when we stop sending, they will be allowed to finish.
Bug: webrtc:10365
Change-Id: I2b4897dde8d49bc7ee5d2d69694616aee8aaea38
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/125096
Reviewed-by: Oskar Sundbom <ossu@webrtc.org>
Commit-Queue: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26963}
