This CL is functionally a noop but may reduce thread wakupes in some
cases.
In particular, consider a send task scheduled for time T. While waiting
for that, a higher-priority packet than the top of the current queue is
added (e.g. an audio packet), and a send is executed immediately.
After sending, it resets the field indicating that a scheduled task is
expected at time T. It then polls NextSendTime() and schedules a new
task, likely at or very close to T. Causing unnecessary task queue
churn and behavior that is more difficult to reason about.
Bug: webrtc:10809
Change-Id: Ic5706f2cc06df3f27cc3e7b473d4de29a669473b
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173700
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31116}
Today when the pacing debt is cleared, we blindly ask for 50 bytes of
padding, which is above a static magic number for RTX payload padding.
Instead, we should adjust the target size based on the current padding
rate. The old pacer sort-of does this, it allows the budget to grow up
to one process interval (usually 5ms).
This CL makes the dynamic pacer also use a duration as target, by
default 5ms to match old pacer but with a trial to allow tweaking it.
This will be important for good behavior due to
https://bugs.chromium.org/p/webrtc/issues/detail?id=11508
Bug: webrtc:10809
Change-Id: I9c14acc5730c6e2e0d7821adf5fb058b8d5487c6
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173687
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31091}
This was introduced on trial but turned out to perform badly for WebRTC
purposes and never used in production.
Bug: webrtc:9883
Change-Id: Ib72acddf4d90fc9ac042084dddf526c04661f290
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173680
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31085}
https://webrtc-review.googlesource.com/c/src/+/172847
------------ original description --------------
Preparation for ReceiveStatisticsProxy lock reduction.
Update tests to call VideoReceiveStream::GetStats() in the same or at
least similar way it gets called in production (construction thread,
same TQ/thread).
Mapped out threads and context for ReceiveStatisticsProxy,
VideoQualityObserver and VideoReceiveStream. Added
follow-up TODOs for webrtc:11489.
One functional change in ReceiveStatisticsProxy is that when sender
side RtcpPacketTypesCounterUpdated calls are made, the counter is
updated asynchronously since the sender calls the method on a different
thread than the receiver.
Make CallClient::SendTask public to allow tests to run tasks in the
right context. CallClient already does this internally for GetStats.
Remove 10 sec sleep in StopSendingKeyframeRequestsForInactiveStream.
Bug: webrtc:11489
Change-Id: I491e13344b9fa714de0741dd927d907de7e39e83
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173583
Commit-Queue: Tommi <tommi@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31077}
This is a reland of 12e2d4ddb235da6ec7a5c1c3a83ac33d394920b0
Original change's description:
> APM: Remove the usage of AudioFrame in the AudioProcessing interface
>
> This CL removes the AudioFrame-based APIs from the AudioProcessing
> interface.
>
> Bug: webrtc:5298
> Change-Id: Iab470b26b10e06dcf29c543851ae0085bc5b66f0
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172939
> Reviewed-by: Sam Zackrisson <saza@webrtc.org>
> Commit-Queue: Per Åhgren <peah@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31016}
Bug: webrtc:5298
Change-Id: I70e6d59afc3716ee6109d8b9dc384abc71c93624
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173476
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31066}
This is more logical way to remove inactive lower layers.
Current way is to notify the encoder that the layer is inactive,
then renumber layers at the packatization level.
This Cl will allow to simplify libvpx vp9 encoder, svcRateAllocator and
vp9 packetizer.
Bug: webrtc:11319
Change-Id: Idf0bb30b729f5ecc97e31454b32934546b681aa2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173182
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Henrik Boström <hbos@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31058}
The transformer was previously moved into the config of the first stream
which resulted in incorrect behavior for simulcast. Use the transformer
in all the streams.
Pass the sender's ssrc on registring the transformed frame callback, to
associate separate transformer sinks for each sender.
Bug: chromium:1065838
Change-Id: I5c52dacb241c68268681b85f875257b24987849e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173332
Commit-Queue: Marina Ciocea <marinaciocea@webrtc.org>
Reviewed-by: Tommi <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31050}
This reverts commit 24eed2735b2135227bcfefbabf34a89f9a5fec99.
Reason for revert: Speculative revert: breaks downstream project
Original change's description:
> Preparation for ReceiveStatisticsProxy lock reduction.
>
> Update tests to call VideoReceiveStream::GetStats() in the same or at
> least similar way it gets called in production (construction thread,
> same TQ/thread).
>
> Mapped out threads and context for ReceiveStatisticsProxy,
> VideoQualityObserver and VideoReceiveStream. Added
> follow-up TODOs for webrtc:11489.
>
> One functional change in ReceiveStatisticsProxy is that when sender
> side RtcpPacketTypesCounterUpdated calls are made, the counter is
> updated asynchronously since the sender calls the method on a different
> thread than the receiver.
>
> Make CallClient::SendTask public to allow tests to run tasks in the
> right context. CallClient already does this internally for GetStats.
>
> Remove 10 sec sleep in StopSendingKeyframeRequestsForInactiveStream.
>
> Bug: webrtc:11489
> Change-Id: Ib45bfc59d8472e9c5ea556e6ecf38298b8f14921
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172847
> Commit-Queue: Tommi <tommi@webrtc.org>
> Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
> Reviewed-by: Magnus Flodman <mflodman@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31008}
TBR=mbonadei@webrtc.org,henrika@webrtc.org,kwiberg@webrtc.org,tommi@webrtc.org,juberti@webrtc.org,mflodman@webrtc.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: webrtc:11489
Change-Id: I48b8359cdb791bf22b1a2c2c43d46263b01e0d65
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173082
Reviewed-by: Artem Titov <titovartem@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31023}
This CL removes the AudioFrame-based APIs from the AudioProcessing
interface.
Bug: webrtc:5298
Change-Id: Iab470b26b10e06dcf29c543851ae0085bc5b66f0
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172939
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31016}
The TaskQueuePacedSender today has some inefficiencies:
* Enqueuing a packet will trigger a MaybeProcessPackets() call, but it
won't actually run immediately even if it should - instead it will
schedule a new call in at least 1ms. This incurs delays and extra
CPU overhead.
* Sometimes thread wakeups are scheduled simply in order to do
book-keeping: ProcessPackets() will be called when the media debt has
gone down to 0 even if there is no packet in the queue, in order to
check if we should send padding.
This CL fixes that by called ProcessPackets() immediately if it is
actually time to do so, and by immediately determining when padding
should be sent without having a separate call to drain media debt.
Bug: webrtc:10809
Change-Id: I4870e86e6de2ce4197463fd5b788ad4717fc7177
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172842
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31010}
Update tests to call VideoReceiveStream::GetStats() in the same or at
least similar way it gets called in production (construction thread,
same TQ/thread).
Mapped out threads and context for ReceiveStatisticsProxy,
VideoQualityObserver and VideoReceiveStream. Added
follow-up TODOs for webrtc:11489.
One functional change in ReceiveStatisticsProxy is that when sender
side RtcpPacketTypesCounterUpdated calls are made, the counter is
updated asynchronously since the sender calls the method on a different
thread than the receiver.
Make CallClient::SendTask public to allow tests to run tasks in the
right context. CallClient already does this internally for GetStats.
Remove 10 sec sleep in StopSendingKeyframeRequestsForInactiveStream.
Bug: webrtc:11489
Change-Id: Ib45bfc59d8472e9c5ea556e6ecf38298b8f14921
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172847
Commit-Queue: Tommi <tommi@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Magnus Flodman <mflodman@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31008}
This CL adds several field-trial-based overrides for parameters related
to AEC transparency.
The changes have been shown to be bitexact for a test dataset.
Bug: webrtc:11487,chromium:1067597
Change-Id: Ica9613db74240687fc85efe059874ef8c20aa7d5
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172844
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31007}
Non-paced audio should be sent "immediately", but in several places that
was determined by looking at the current time - which can lead to
inconsistencies.
E.g. if a packet is enqueued and ProcessPackets() is called 1ms later,
the pacer should see NextSendTime() as 1ms ago, so that buffer levels
are cleared at the right pace.
Bug: webrtc:10809
Change-Id: I04a169f3df3e28a5c8ef7fa8a042b9c482c307ce
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172845
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31002}
Especially with the TaskQueuePacerSender, it is very common that a
single packet is added to the queue and then immediately removed as
it gets sent to the network.
This CL adds a fast-path for that case, that avoid creating book-
keeping in the form of stream-priorities and timestamp sets etc.
Functionally, it should be a noop, but hopefully it can save a few
CPU cycles.
Bug: webrtc:10809
Change-Id: Idaa06b4f8d1da444fce78cc742e2ab52f9efe815
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172090
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31001}
This CL adds two kill-switches to the AEC3 code to be used as
safe fallbacks to increase AEC transparency.
The changes have been shown to be bitexact for a test dataset.
Bug: webrtc:11487,chromium:1067597
Change-Id: I7f9f78db4964990bcdfa9adae6ef36a56bce7224
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172840
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30998}
This allows clients to exclude the transient suppression submodule from WebRTC builds, by defining WEBRTC_EXCLUDE_TRANSIENT_SUPPRESSOR.
The changes have been shown to be bitexact for a test dataset (when the flag is _not_ defined.)
No-Try: True
Bug: webrtc:11226, webrtc:11292
Change-Id: I6931c82a280a9b40a53ee1c2a9820ed9e674a9a5
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/171421
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Reviewed-by: Per Åhgren <peah@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30978}
This allows the user to run audioproc_f with various field trials set.
The approach is copied from test/test_main_lib.cc.
Tested:
1. Verified bitexactness vs ToT audioproc_f on a large dataset of aecdumps
2. Ran it with flags --aec=1 --force_fieldtrials="WebRTC-Aec3ClampInstQualityToZeroKillSwitch/Enabled/WebRTC-Aec3ClampInstQualityToOneKillSwitch/Enabled/" and verified in GDB that the AEC3 config was changed accordingly.
No-Try: True
Bug: webrtc:5298
Change-Id: I70eec7777f70893b36af33794a5842f67d56af31
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172623
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Reviewed-by: Artem Titov <titovartem@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30976}
Currently, audioproc_f crashes on a DCHECK as the data vector of Int16Frame is not resized.
Bug: webrtc:5298
Change-Id: I897cf0fce07e0ed2c0a365a965fa50fd3d8ddd18
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172624
Reviewed-by: Per Åhgren <peah@webrtc.org>
Commit-Queue: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30969}
This CL adds a number of kill-switches to the AEC3 code to be used as
safe fallbacks to increase AEC transparency.
The changes have been shown to be bitexact for a test dataset.
Bug: webrtc:11475,chromium:1066836
Change-Id: Ibebcbbfbbd958cb6fcc6993247e3030fa65b582c
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172600
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Commit-Queue: Per Åhgren <peah@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30964}
r30936 accidentally made it defualt off. This reverts to the old
behavior by default.
Bug: webrtc:8975, chromium:1066442
Change-Id: I415d2f74bb7321f17b4039ca43cbd53c3e3725f9
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/172445
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30954}
EnqueuePackets() would reset the last process time if the queue
and media budgets were empty. This was done without reducing the
padding debt.
The result of this was that, given an existing debt, and an interval
between audio packets that is less than the drain time for the padding
debt, padding would not be sent at all.
Now, before adding a new packet, we reduce the padding debt if the
packet queue is empty.
Bug: webrtc:10809
Change-Id: I116169522c215257febd32e17abab45f1a7d609f
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/171808
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#30949}
