This CL paves the way to making FrameBufferController injectable.
LibvpxVp8Encoder can manage multiple streams. Prior to this CL,
each stream had its own frame buffer controller, all of them held
in a vector by LibvpxVp8Encoder. This complicated the code and
produced some code duplication (cf. SetupTemporalLayers).
This CL:
1. Replaces CreateVp8TemporalLayers() by a factory. (Later CLs
will make this factory injectable.)
2. Makes LibvpxVp8Encoder use a single controller. This single
controller will, in the case of multiple streams, delegate
its work to multiple controllers, but that fact is not visible
to LibvpxVp8Encoder.
This CL also squashes CL #126046 (Send notifications of RTT and
PLR changes to Vp8FrameBufferController) into it.
Bug: webrtc:10382
Change-Id: Id9b55734bebb457acc276f34a7a9e52cc19c8eb9
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/126483
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#27206}
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}
there's no easy way to inject the Clock in ScreenshareLayers under
normal use. To allow faking the clock, rtc::TimeMillis is used instead.
Bug: webrtc:10365
Change-Id: I46c7f76514672190a0f0f5816a2c858bc6c76fa4
Reviewed-on: https://webrtc-review.googlesource.com/c/125189
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Commit-Queue: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26946}
This CL continues the work began by CL #119958, extending it
to ScreenshareLayers.
Bug: webrtc:10249
Change-Id: I59d0c062a93b288007977e00aa3a2e0929509e0c
Reviewed-on: https://webrtc-review.googlesource.com/c/120042
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26526}
FrameConfig is not specific to temporal layers. Anything that
can control referenced/updated buffers could potentially use it.
Bug: webrtc:10259
Change-Id: I04ed177ee884693798c3b69e35fd4255ce1e9062
Reviewed-on: https://webrtc-review.googlesource.com/c/120355
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Elad Alon <eladalon@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#26448}
Also renaming it Vp8TemporalLayers to show that it is codec specific.
Bug: webrtc:9012
Change-Id: I18187538b8142cdd7538f1a4ed1bada09d040f1f
Reviewed-on: https://webrtc-review.googlesource.com/c/104643
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Reviewed-by: Niels Moller <nisse@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#25137}
This refactoring merged PopulateCodecSpecific and FrameEncoded into a
single callback method. It also removes the FrameConfig parameter and
instead relies on the temporal layer to remember it internally.
Bug: webrtc:9012
Change-Id: I489b76821b534398ad452643f1322f411d3455b1
Reviewed-on: https://webrtc-review.googlesource.com/95681
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24957}
Today, the internal frame dropper in libvpx vp8 encoder is enabled or
disabled based on video or screen content. This is then expected to
match up with screenshare vs default temporal layers implementation.
This cl makes libvpx query the temporal layers implementation as well,
breaking this implicit dependency and allows frames to be dropped if
default temporal layers is used with screen content.
Bug: webrtc:9734
Change-Id: If2523a211f4929f16e65a02fa7a6b4edf7328571
Reviewed-on: https://webrtc-review.googlesource.com/99062
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24645}
This CL is a minimum effort/low risk fix.
Later CLs take a more thorough approach.
Bug: webrtc:9634
Change-Id: I728a061a4e71b38a559ee438646de83cd2cb3517
Reviewed-on: https://webrtc-review.googlesource.com/94760
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24326}
This CL introduces a few changes to the default VP8 temporal layers:
* The pattern is now reset on keyframes
* The sync flag is inferred rather than hard-coded
* Support is added for buffer search order
Bug: webrtc:9012
Change-Id: Ice19d32413d20982368a01a7d2540d155e185ad4
Reviewed-on: https://webrtc-review.googlesource.com/91863
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#24288}
* Move SimulcastEncoderAdapter out under modules/video_coding
* Move SimulcastRateAllocator back out to modules/video_coding/utility
* Move TemporalLayers and ScreenshareLayers to modules/video_coding/utility
* Move any VP8 specific code - such as temporal layer bitrate budgeting -
under codec type dependent conditionals.
* Plumb the simulcast index for H264 in the codec specific and RTP format data structures.
TBR=sprang@webrtc.org,stefan@webrtc.org,titovartem@webrtc.org
Bug: webrtc:5840
Change-Id: I2d3b130622dd7ceec5528f3ab6c46f109e6bafb8
Reviewed-on: https://webrtc-review.googlesource.com/84743
Commit-Queue: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23715}
This reverts commit 07efe436c9.
Reason for revert: Breaks downstream project.
cricket::GetSimulcastConfig method signature has been updated.
I think you can get away with a default value for temporal_layers_supported (and then you can remove it after a few days when projects will be updated).
Original change's description:
> Implement H264 simulcast support and generalize SimulcastEncoderAdapter use for H264 & VP8.
>
> * Move SimulcastEncoderAdapter out under modules/video_coding
> * Move SimulcastRateAllocator back out to modules/video_coding/utility
> * Move TemporalLayers and ScreenshareLayers to modules/video_coding/utility
> * Move any VP8 specific code - such as temporal layer bitrate budgeting -
> under codec type dependent conditionals.
> * Plumb the simulcast index for H264 in the codec specific and RTP format data structures.
>
> Bug: webrtc:5840
> Change-Id: Ieced8a00e38f273c1a6cfd0f5431a87d07b8f44e
> Reviewed-on: https://webrtc-review.googlesource.com/64100
> Commit-Queue: Harald Alvestrand <hta@webrtc.org>
> Reviewed-by: Stefan Holmer <stefan@webrtc.org>
> Reviewed-by: Erik Språng <sprang@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#23705}
TBR=sprang@webrtc.org,stefan@webrtc.org,mflodman@webrtc.org,hta@webrtc.org,sergio.garcia.murillo@gmail.com,titovartem@webrtc.org,agouaillard@gmail.com
Change-Id: Ic9d3b1eeaf195bb5ec2063954421f5e77866d663
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:5840
Reviewed-on: https://webrtc-review.googlesource.com/84760
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23710}
* Move SimulcastEncoderAdapter out under modules/video_coding
* Move SimulcastRateAllocator back out to modules/video_coding/utility
* Move TemporalLayers and ScreenshareLayers to modules/video_coding/utility
* Move any VP8 specific code - such as temporal layer bitrate budgeting -
under codec type dependent conditionals.
* Plumb the simulcast index for H264 in the codec specific and RTP format data structures.
Bug: webrtc:5840
Change-Id: Ieced8a00e38f273c1a6cfd0f5431a87d07b8f44e
Reviewed-on: https://webrtc-review.googlesource.com/64100
Commit-Queue: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23705}
Running clang-format with chromium's style guide.
The goal is n-fold:
* providing consistency and readability (that's what code guidelines are for)
* preventing noise with presubmit checks and git cl format
* building on the previous point: making it easier to automatically fix format issues
* you name it
Please consider using git-hyper-blame to ignore this commit.
Bug: webrtc:9340
Change-Id: I694567c4cdf8cee2860958cfe82bfaf25848bb87
Reviewed-on: https://webrtc-review.googlesource.com/81185
Reviewed-by: Patrik Höglund <phoglund@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#23660}
OnRatesUpdated() is called every time the bitrate estimate, or once per
second. However, since we don't want to reconfigure libvpx too often,
just in case it interferes with the rate controller, so
ScreenshareLayers contains a boolean |bitrate_update_| which indicate
if the configuration should be updated on a call to
UpdateConfiguration().
However, it two rate updates happened between two frames, the first of
which changes the rates and second one does not, |bitrate_update_| will
be reset to false and the encoder won't get the desired config.
This CL makes sure we update the configuration iff the rate has changed
at any time since the last call to UpdateConfiguration().
Bug: webrtc:9012
Change-Id: I62af36cffe20ecb7d3f403b3eb11f23a9692d719
Reviewed-on: https://webrtc-review.googlesource.com/69040
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Niels Moller <nisse@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#22826}
It would be nice to also delete the fields from CodecSpecificInfo,
but those fields are used on the receive side.
Bug: webrtc:8830
Change-Id: I1a3f13ea2c024cbd73b33fd9dd58e531d3576a55
Reviewed-on: https://webrtc-review.googlesource.com/64780
Commit-Queue: Niels Moller <nisse@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#22625}
As the rate allocation has been moved into entirely into
SimulcastRateAllocator, and the listeners are thus no longer needed,
this class doesn't fill any other purpose than to determine if
ScreenshareLayers or TemporalLayers should be created for a given
simulcast stream. This can however be done just from looking at the
VideoCodec instance, so changing this into a static factory method.
Due to dependencies from upstream projects, keep the class name and
field in VideoCodec around for now.
Bug: webrtc:9012
Change-Id: I028fe6b2a19e0d16b35956cc2df01dcf5bfa7979
Reviewed-on: https://webrtc-review.googlesource.com/63264
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#22529}
This CL moves all temporal layer rate allocation from
DefaultTemporalLayers and ScreenshareLayers into SimulcastRateAllocator.
This means we don't need an extra call-out to the TemporalLayers
interface to get the last allocation, which simplifies the code path a
lot.
It also paves the wave for removing the TemporalLayersFactory interface
(in a separate cl), which will further simplify the ownership model.
Bug: webrtc:9012
Change-Id: I6540b1848efa1a136dce449f13902ad479d5ee37
Reviewed-on: https://webrtc-review.googlesource.com/62420
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#22502}
This is in preparation for
https://webrtc-review.googlesource.com/c/src/+/36340
With these changes we can avoid some strange #ifdefs in the code
that uses temporal layers.
Bug: webrtc:7925
Change-Id: I472210738ccc9f73812b8863951befeabec56f15
Reviewed-on: https://webrtc-review.googlesource.com/41280
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Anders Carlsson <andersc@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21759}
Try to use frame timestamps first if they look reasonable, otherwise
use realtime clock.
Also, lower limit from 90% of target to 85%.
Bug: webrtc:4172, chromium:802290
Change-Id: Iad489be7c7cf637345be4795e5089936ab9fab07
Reviewed-on: https://webrtc-review.googlesource.com/41041
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21729}
This is a reland of https://webrtc-review.googlesource.com/34380
The main problem with that CL was that we used frame timestamps as basis
for frame dropping, but those might not be continuous or even populated
in some circumstances.
Additionally, I found that the bitrate was off since the encoder does
not not take the dropped frames into account, so if we drop every other
frame continiusoly, the bitrate sent will be around half of the target.
Patch set 1 is the original CL, subsequent patch sets cotains fixes.
Bug: webrtc:4172
Change-Id: I8ec8dddcebf4ce44f28dd9055cf9c46bbd68e4a6
Reviewed-on: https://webrtc-review.googlesource.com/39201
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21601}
This reverts commit 28a06b16cc.
Reason for revert: Causes some unexpected perf changes.
Original change's description:
> Smoother frame dropping when screenshare_layers limits fps
>
> Currently, when input fps is higher than the configured target fps in
> screenshare_layers, we drop frames with the help of a rate tracker using
> a one second sliding window. This is not optimal.
> For instance, given a 5fps limit and a 30fps capturer, the window will
> not be saturated until we have added the first 5 frames. Then we will
> drop all frames until the oldest one drops out, at which point we can
> immediately fill it's place. This results in quick 5 frame bursts every
> second.
>
> In addition to rate tracker, also set a limit on minimum interval
> required between input frames, based on target frame rate.
>
> Bug: webrtc:4172
> Change-Id: I49f0abf4d549673cc6b3fafe580b529ea3feaf57
> Reviewed-on: https://webrtc-review.googlesource.com/34380
> Commit-Queue: Erik Språng <sprang@webrtc.org>
> Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#21325}
TBR=ilnik@webrtc.org,sprang@webrtc.org
Change-Id: I7ca5b0c847310dbb11dce28773082eac946c0ba4
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:4172
Reviewed-on: https://webrtc-review.googlesource.com/34780
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21354}
Currently, when input fps is higher than the configured target fps in
screenshare_layers, we drop frames with the help of a rate tracker using
a one second sliding window. This is not optimal.
For instance, given a 5fps limit and a 30fps capturer, the window will
not be saturated until we have added the first 5 frames. Then we will
drop all frames until the oldest one drops out, at which point we can
immediately fill it's place. This results in quick 5 frame bursts every
second.
In addition to rate tracker, also set a limit on minimum interval
required between input frames, based on target frame rate.
Bug: webrtc:4172
Change-Id: I49f0abf4d549673cc6b3fafe580b529ea3feaf57
Reviewed-on: https://webrtc-review.googlesource.com/34380
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#21325}
This is expected to result in a slight loss of overall quality, but
should offset by quicker switching between temporal layers with flaky
connections.
Bug: webrtc:7694
Change-Id: Ib605802bb59f12773652324ac66cdf4774ae6bb6
Reviewed-on: https://webrtc-review.googlesource.com/6881
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20160}
All frames are checked against hard-coded dependency graph
using new helper class. It's invoked in RTC_DCHECK(). Only
DefaultTemporalLayers are fully implemented in this CL, checker
for ScreenshareLayers is not doing anything for now.
Bug: none
Change-Id: I3ec017176d8c25f7572c8f161e52f2ebfac8220f
Reviewed-on: https://webrtc-review.googlesource.com/3740
Reviewed-by: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#20066}
In https://webrtc-review.googlesource.com/c/src/+/1560 we moved WebRTC
from src/webrtc to src/ (in order to preserve an healthy git history).
This CL takes care of fixing header guards, #include paths, etc...
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
TBR=tommi@webrtc.org
Bug: chromium:611808
Change-Id: Iea91618212bee0af16aa3f05071eab8f93706578
Reviewed-on: https://webrtc-review.googlesource.com/1561
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Henrik Kjellander <kjellander@webrtc.org>
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#19846}
In order to eliminate the WebRTC Subtree mirror in Chromium,
WebRTC is moving the content of the src/webrtc directory up
to the src/ directory.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
TBR=tommi@webrtc.org
Bug: chromium:611808
Change-Id: Iac59c5b51b950f174119565bac87955a7994bc38
Reviewed-on: https://webrtc-review.googlesource.com/1560
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Henrik Kjellander <kjellander@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#19845}
2017-09-15 04:25:06 +00:00
Renamed from webrtc/modules/video_coding/codecs/vp8/screenshare_layers.cc (Browse further)
