This is a reland of 6b9c60b06d04bc519195fca1f621b10accfeb46b
Original change's description:
> Removes lock release in PacedSender callback.
>
> The PacedSender currently has logic to temporarily release its internal
> lock while sending or asking for padding.
> This creates some tricky situations in the pacing controller where we
> need to consider if some thread can enter while we the process thread is
> actually processing, just temporarily busy sending.
>
> Since the pacing call stack is no longer cyclic, we can actually remove
> this lock-release now.
>
> Bug: webrtc:10809
> Change-Id: Ic59c605252bed1f96a03406c908a30cd1012f995
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173592
> Reviewed-by: Sebastian Jansson <srte@webrtc.org>
> Commit-Queue: Erik Språng <sprang@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31206}
Bug: webrtc:10809
Change-Id: Id39fc49b0a038e7ae3a0d9818fb0806c33ae0ae0
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175656
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31332}
With an optional parameter this allows the task-queue based paced
sender to mimic the old behavior and coalesce sending of packets in
order to reduce thread wakeups and provide opportunity for batching.
This is done by simply overriding the minimum time the thread should
sleep. The pacing controller will already handle the "late wakup" case
and send any packets as if it had been woken at the optimal time.
Bug: webrtc:10809
Change-Id: Iceea00693a4e87d39b0e0ee8bdabca081dff2cba
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175648
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Markus Handell <handellm@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31328}
Some VideoStreamEncoderTests such as
AdaptsFramerateForLowQuality_MaintainResolutionMode had been observed
to be flaky. In a local run, this test failed 1/12000 times, but on
bots it may have failed more often.
All tests could sometimes fail due to expecting something to be the
case without waiting for that thing to happen, which was made evident
when adding SleepMs() at strategic points in the code and running the
tests locally.
With this CL, the tests should no longer be flaky after having added
waiting for adaptation to be applied or EXPECT_TRUE_WAIT for sink wants
to have the appropriate values.
Bug: webrtc:11586
Change-Id: I60e76c742d9ccc8305feb14a834a4f61a60a62a3
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175654
Commit-Queue: Henrik Boström <hbos@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31327}
Remove android_manifest_for_lint from BUILD.gn files
The Chromium Roll into WebRTC isn't flowing
The first CL that caused the problem is https://webrtc-review.googlesource.com/c/src/+/175140/
The error is: ERROR at //examples/BUILD.gn:104:33: Assignment had no effect. android_manifest_for_lint = "androidapp/AndroidManifest.xml"
android_manifest_for_lint has ben removed so update BUILD files that use that feature to reflect this.
BUG=None
Change-Id: If526d9a4dd80cddca7f2c9dd7f67ba9efe3f1a84
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175661
Commit-Queue: Courtney Edwards <courtneyfe@google.com>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31325}
This CL extends the WebRTC testing API to allow audioproc_f -based
testing using a pre-created AudioProcessing object. This is an
important feature to allow testing any AudioProcessing objects
that are injected into WebRTC.
Beyond adding this, the CL also changes the simulation code to
operate on a scoped_refptr<AudioProcessing> object instead of a
std::unique<AudioProcessing> object
Bug: webrtc:5298
Change-Id: I70179f19518fc583ad0101bd59c038478a3cc23d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175568
Commit-Queue: Per Åhgren <peah@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Reviewed-by: Sam Zackrisson <saza@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31319}
- Moves AnalyzerConfig and helper functions IsAudioSsrc, IsVideoSsrc, IsRtxSsrc, GetStreamNam and GetLayerName to analyzer_common.h
- Moves log_segments() code to rtc_event_log_parser.h
- Moves TriageAlert/Notification code to a new file with a couple of minor fixes to make it less spammy.
Bug: webrtc:11566
Change-Id: Ib33941d8185f7382fc72ed65768e46015e0320de
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/174824
Commit-Queue: Björn Terelius <terelius@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31318}
...for the NullDataDumper, WrongCaptureBlockSize and
DISABLED_WrongRenderBlockSize tests. This is to avoid creation
of additional threads on Mac, which can cause issues on asan bots.
Bug: webrtc:11577
Change-Id: I4e6a64d47ec3b0a0e0018b19a0486208ba7e6ae2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175600
Reviewed-by: Per Åhgren <peah@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Commit-Queue: Tommi <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31307}
This will improve support for tests that use Clock as well as offer
a way to remove use of Sleep() in the tests.
Bug: none
Change-Id: I25fd0c6fc1b52ec0c917e56fae6807b136213d8d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175566
Commit-Queue: Tommi <tommi@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31305}
This reverts commit 576ad5d510894040d7bbc041d5c86745c67f30f8.
Reason for revert: Causes compile error in Chrome.
Original change's description:
> Make TransformableVideoFrameInterface::GetMetadata pure virtual.
>
> GetMetadata() has been implemented downstream and can be made pure
> virtual.
>
> Bug: chromium:1069295
> Change-Id: I62a3be6106552d2d82d8c413c6f523d31626b0d8
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175001
> Commit-Queue: Marina Ciocea <marinaciocea@webrtc.org>
> Reviewed-by: Harald Alvestrand <hta@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31281}
TBR=hta@webrtc.org,marinaciocea@webrtc.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: chromium:1069295
Change-Id: I5915270d5b8dab9fc30a07f22fddedb29beca01a
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175620
Reviewed-by: Guido Urdaneta <guidou@webrtc.org>
Commit-Queue: Guido Urdaneta <guidou@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31304}
Logging of content name (mid) is valuable to debug issues
in scenarios with multiple m= line sections in SDP.
For example, video conferencing applications which
uses SFU and Unified Plan SDPs will likely to leverage
from more detailed logs when issues need to be debugged.
Bug: webrtc:10139
Change-Id: Id52ba3ad54af5caa0f8c03daaa51bdb0caf9fe67
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175115
Commit-Queue: Tommi <tommi@webrtc.org>
Reviewed-by: Tommi <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31302}
The Chromium Roll into WebRTC isn't flowing
The first CL that caused the problem is https://webrtc-review.googlesource.com/c/src/+/175140/
The error is: ERROR at //examples/BUILD.gn:104:33: Assignment had no effect. android_manifest_for_lint = "androidapp/AndroidManifest.xml"
android_manifest_for_lint has ben removed so update BUILD files that use that feature to reflect this.
BUG=None
Change-Id: Ic3eb16eab8e4a4ab87daac9998d7a07373fec493
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175569
Commit-Queue: Courtney Edwards <courtneyfe@google.com>
Reviewed-by: Artem Titov <titovartem@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31299}
ModuleRtpRtcpImpl::Process seems to be called as many
times as 200 times a second (kRtpRtcpMaxIdleTimeProcessMs == 5).
This CL changes it so that LastReceivedReportBlockMs() is called
once a second instead of potentially every time Process() runs.
This should result in grabbing locks fewer times, however there
are still other call sites for the same lock.
Bug: webrtc:11581
Change-Id: I4c2fd9aa43343fdac2763250ae7f4d2545e98ec2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175350
Commit-Queue: Tommi <tommi@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31298}
This reverts commit d0d55515c4b88a07446bc66a5c183f50ee896282.
Reason for revert: Relanding with workaround.
Original change's description:
> Revert "Use Windows 10 thread naming API"
>
> This reverts commit e35004dffb42dd96b8cf37b33c9a3af4a5fd376c.
>
> Reason for revert: Reverting while downstream issue is resolved.
>
> Original change's description:
> > Use Windows 10 thread naming API
> >
> > While profiling video chat in Chrome I noticed that some of the webrtc
> > threads were not named. This change adds conditional use of the thread
> > naming APIs. These thread names work even if you attach a debugger after
> > the thread is named, and they show up in ETW traces, for easier
> > profiling.
> >
> > The sctp_create_thread_adapter threads are still not named but since
> > those are in C files they would require a C++-with-extern-C interface
> > to fix, so I'm leaving them for now.
> >
> > Bug: webrtc:10745
> > Change-Id: I68f6aa780e2417ce706764d69e5b64cc48aba333
> > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175280
> > Commit-Queue: Tommi <tommi@webrtc.org>
> > Reviewed-by: Tommi <tommi@webrtc.org>
> > Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
> > Cr-Commit-Position: refs/heads/master@{#31285}
>
> TBR=kwiberg@webrtc.org,tommi@webrtc.org,brucedawson@chromium.org
>
> Change-Id: Icf877afbd82918ebe0c42a93b8a763cdab9a73ce
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: webrtc:10745
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175347
> Reviewed-by: Tommi <tommi@webrtc.org>
> Commit-Queue: Tommi <tommi@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31289}
TBR=kwiberg@webrtc.org,tommi@webrtc.org,brucedawson@chromium.org
Bug: webrtc:10745
Change-Id: I51ee413fd8a0ff62f6b8b2a11f546b2a70168842
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175349
Reviewed-by: Tommi <tommi@webrtc.org>
Commit-Queue: Tommi <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31292}
I named the network sequence checker 'worker' some time ago.
It represents the network TQ in RtpTransportControllerSendInterface
though, so should rather be called 'network'.
Bug: none
Change-Id: If82d7528b8cfcc180e1515b2fad63ed01610ec32
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175340
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Tommi <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31291}
Also re-enable the TestAnnotationsOnWrongQueueDebug test and rename
the test suite to SequenceCheckerDeathTest so that it gets executed
before other tests.
Bug: webrtc:11577
Change-Id: I3b8037644e4b9139755ccecb17e42b09327e4996
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175346
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Commit-Queue: Tommi <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31290}
This reverts commit e35004dffb42dd96b8cf37b33c9a3af4a5fd376c.
Reason for revert: Reverting while downstream issue is resolved.
Original change's description:
> Use Windows 10 thread naming API
>
> While profiling video chat in Chrome I noticed that some of the webrtc
> threads were not named. This change adds conditional use of the thread
> naming APIs. These thread names work even if you attach a debugger after
> the thread is named, and they show up in ETW traces, for easier
> profiling.
>
> The sctp_create_thread_adapter threads are still not named but since
> those are in C files they would require a C++-with-extern-C interface
> to fix, so I'm leaving them for now.
>
> Bug: webrtc:10745
> Change-Id: I68f6aa780e2417ce706764d69e5b64cc48aba333
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175280
> Commit-Queue: Tommi <tommi@webrtc.org>
> Reviewed-by: Tommi <tommi@webrtc.org>
> Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31285}
TBR=kwiberg@webrtc.org,tommi@webrtc.org,brucedawson@chromium.org
Change-Id: Icf877afbd82918ebe0c42a93b8a763cdab9a73ce
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:10745
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175347
Reviewed-by: Tommi <tommi@webrtc.org>
Commit-Queue: Tommi <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31289}
This adds priority to the API configuration of datachannels,
and passes the value in the OPEN message.
It does not yet influence SCTP prioritization of messages.
Bug: chromium:1083227
Change-Id: I46ddd1eefa0e3d07c959383788b9e80fcbfa38d6
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175107
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Taylor <deadbeef@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31287}
While profiling video chat in Chrome I noticed that some of the webrtc
threads were not named. This change adds conditional use of the thread
naming APIs. These thread names work even if you attach a debugger after
the thread is named, and they show up in ETW traces, for easier
profiling.
The sctp_create_thread_adapter threads are still not named but since
those are in C files they would require a C++-with-extern-C interface
to fix, so I'm leaving them for now.
Bug: webrtc:10745
Change-Id: I68f6aa780e2417ce706764d69e5b64cc48aba333
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175280
Commit-Queue: Tommi <tommi@webrtc.org>
Reviewed-by: Tommi <tommi@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31285}
This reverts commit af1b9ceb62dce3462083f9a44e26ee6d79639cef.
Reason for revert: Speculative reland after looking into downstream
failures. It's possible that carryover state from unrelated tests
running in parallel was causing failures.
Original change's description:
> Revert "Revert back to using the task_queue_ for guarding access."
>
> This reverts commit 475006d4a30f8bc47f82eb540a6a066da2829095.
>
> Reason for revert: Speculative revert. Breaks downstream project
>
> Original change's description:
> > Revert back to using the task_queue_ for guarding access.
> >
> > This removes the SequenceChecker that was temporarily used while
> > the rtc::Thread TQ implementation was being fixed.
> >
> > Bug: none
> > Change-Id: Iaa46e47371211ac0a97b2dcaf23cef12b43ee8ea
> > Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175081
> > Commit-Queue: Tommi <tommi@webrtc.org>
> > Reviewed-by: Sebastian Jansson <srte@webrtc.org>
> > Cr-Commit-Position: refs/heads/master@{#31256}
>
> TBR=tommi@webrtc.org,srte@webrtc.org
>
> Change-Id: I17a12bdca888a63f2fd161da30c0def5b9c3d04e
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: none
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175103
> Reviewed-by: Artem Titov <titovartem@webrtc.org>
> Commit-Queue: Artem Titov <titovartem@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31258}
TBR=tommi@webrtc.org,srte@webrtc.org,titovartem@webrtc.org
# Not skipping CQ checks because this is a reland.
Bug: none
Change-Id: I23992643126d7d6dae63da1bb14420b2b8794fd9
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/175135
Reviewed-by: Tommi <tommi@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Reviewed-by: Artem Titov <titovartem@webrtc.org>
Commit-Queue: Tommi <tommi@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31283}
