At speed 8, vp9 on ARM is currently ~2x times slower than vp8 on ARM (speed -12).
Update some parameters in videoprocessor_integrationtest.cc
to make tests pass on android (which uses the new speed setting).
TBR=stefan@webrtc.org
BUG=
Review URL: https://codereview.webrtc.org/1526973004 .
Cr-Commit-Position: refs/heads/master@{#11072}
We can now use std::move instead!
This CL leaves the Pass methods in place; a follow-up CL will add deprecation annotations to them.
Review URL: https://codereview.webrtc.org/1460043002
Cr-Commit-Position: refs/heads/master@{#11064}
The purpose is so that a decoder (Android) that only have a limited number of output buffers can make sure that decoding is done just before the frame is needed.
Removed unused iSupportsRenderTiming and the settings structs since it was not used.
Added VCMReceiver::FrameForDecoding unit test for the case when PreferDecodeLate is set.
Note that this does not change the current behaviour. We actually currently always decode frames late. This cl is to make sure the behaviour is kept for Android, if the default behaviour is changed.
Review URL: https://codereview.webrtc.org/1428293003
Cr-Commit-Position: refs/heads/master@{#10974}
Specify kf_min_dist to get correct key frame interval in svc mode.
Also set QP-max/min per temporal and spatial layer (was previously only allowed to be set per spatial layer).
BUG=chromium:500602
R=stefan@webrtc.org
Review URL: https://codereview.webrtc.org/1492633005 .
Cr-Commit-Position: refs/heads/master@{#10890}
Also doing some simplifications inside video_coding. No CHECKs added,
since they appear to have introduced breakages in downstream tests.
Overall reducing the number of potential ways a decoder could possibly
be set null. Removing deregistration of external decoders should also
give a quicker shutdown time since that may attempt to register
internal decoders.
BUG=chromium:563299
TBR=stefan@webrtc.org
Review URL: https://codereview.webrtc.org/1483423002 .
Cr-Commit-Position: refs/heads/master@{#10858}
Reason for revert:
Speculative revert since a downstream test started failing with this.
Original issue's description:
> Add _decoder CHECK to VCMGenericDecoder constructor.
>
> This should never be using a null decoder, but it looks like it's
> crashing out in the field. Adding a CHECK to see if it catches any
> interesting stack traces.
>
> Also making the _decoder pointer const to show that it should never be
> changing.
>
> BUG=chromium:563299
> R=stefan@webrtc.org
>
> Committed: a443ec1a75TBR=stefan@webrtc.org,pbos@webrtc.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:563299
Review URL: https://codereview.webrtc.org/1490703002
Cr-Commit-Position: refs/heads/master@{#10851}
The callback keeps a reference to an object until the callback goes out of scope.
Review URL: https://codereview.webrtc.org/1487493002
Cr-Commit-Position: refs/heads/master@{#10847}
This should never be using a null decoder, but it looks like it's
crashing out in the field. Adding a CHECK to see if it catches any
interesting stack traces.
Also making the _decoder pointer const to show that it should never be
changing.
BUG=chromium:563299
R=stefan@webrtc.org
Review URL: https://codereview.webrtc.org/1485713002 .
Cr-Commit-Position: refs/heads/master@{#10843}
Prevents double-initialization of decoders due to resolution changes
between initial database settings and first incoming frame.
BUG=webrtc:5251
R=mflodman@webrtc.org
Review URL: https://codereview.webrtc.org/1474193002 .
Cr-Commit-Position: refs/heads/master@{#10822}
Also adds a RTC_CHECK in VideoReceiveStream that verifies that decoders
aren't null, since this will attempt to deregister a codec which would
previously fail with an obscure stack trace not indicating what actually
was wrong.
BUG=webrtc:5249
R=stefan@webrtc.org
Review URL: https://codereview.webrtc.org/1479793002 .
Cr-Commit-Position: refs/heads/master@{#10821}
Remove the headers that were kept to provide non-breaking updates
of downstream code for https://codereview.webrtc.org/1418913006/
and https://codereview.webrtc.org/1417283007/.
BUG=webrtc:5095
TESTED=Passing compile-trybots with --clobber flag:
git cl try --clobber --bot=win_compile_rel --bot=linux_compile_rel --bot=android_compile_rel --bot=mac_compile_rel --bot=ios_rel --bot=linux_gn_rel --bot=win_x64_gn_rel --bot=mac_x64_gn_rel --bot=android_gn_rel -m tryserver.webrtc
NOTRY=True
Review URL: https://codereview.webrtc.org/1467173003
Cr-Commit-Position: refs/heads/master@{#10773}
This makes it clearer this code not meant to be used as an API.
I could not find any use of this in downstream code.
BUG=webrtc:5095
TESTED=git cl try -c --bot=android_compile_rel --bot=linux_compile_rel --bot=win_compile_rel --bot=mac_compile_rel --bot=ios_rel --bot=linux_gn_rel --bot=win_x64_gn_rel --bot=mac_x64_gn_rel --bot=android_gn_rel -m tryserver.webrtc
R=stefan@webrtc.orgTBR=magjed@webrtc.org
Review URL: https://codereview.webrtc.org/1440873005 .
Cr-Commit-Position: refs/heads/master@{#10699}
The main purpose was the interface-> include rename, but other files
were also moved, eliminating the "main" dir.
To avoid breaking downstream, the "interface" directories were copied
into a new "video_coding/include" dir. The old headers got pragma
warnings added about deprecation (a very short deprecation since I plan
to remove them as soon downstream is updated).
Other files also moved:
video_coding/main/source -> video_coding
video_coding/main/test -> video_coding/test
BUG=webrtc:5095
TESTED=Passing compile-trybots with --clobber flag:
git cl try --clobber --bot=win_compile_rel --bot=linux_compile_rel --bot=android_compile_rel --bot=mac_compile_rel --bot=ios_rel --bot=linux_gn_rel --bot=win_x64_gn_rel --bot=mac_x64_gn_rel --bot=android_gn_rel -m tryserver.webrtc
R=stefan@webrtc.org, tommi@webrtc.org
Review URL: https://codereview.webrtc.org/1417283007 .
Cr-Commit-Position: refs/heads/master@{#10694}
Removes VP8::Encode trace in favor of VCMGenericEncoder ones and adds
one to InitEncode. Also adds an instant event to ::Encoded since this
can be done on a different thread.
Also adds the corresponding traces to VCMGenericDecoder.
BUG=webrtc:5167
R=stefan@webrtc.org
Review URL: https://codereview.webrtc.org/1412573010
Cr-Commit-Position: refs/heads/master@{#10674}
To avoid breaking downstream, the "interface" directories were copied
into a new "common_video/include" dir. The old headers got pragma
warnings added about deprecation (a very short deprecation since I plan
to remove them as soon downstream is updated).
The header guards are also identical to avoid mixing them up in the transition.
BUG=webrtc:5095
TESTED=Passing compile-trybots with --clobber flag:
git cl try --clobber --bot=win_compile_rel --bot=linux_compile_rel --bot=android_compile_rel --bot=mac_compile_rel --bot=ios_rel --bot=linux_gn_rel --bot=win_x64_gn_rel --bot=mac_x64_gn_rel --bot=android_gn_rel -m tryserver.webrtc
Review URL: https://codereview.webrtc.org/1418913006
Cr-Commit-Position: refs/heads/master@{#10659}
Put VideoSender/VideoReceiver flat within the object, not as
scoped_ptrs, giving fewer allocations and looking a bit nicer.
BUG=
R=tommi@webrtc.org
Review URL: https://codereview.webrtc.org/1443613002
Cr-Commit-Position: refs/heads/master@{#10634}
Receiving RTCP often caused the worker thread to stall for >20 ms
(>100ms observed) due to contention on VideoSender's send_crit_ (used to
protect encoding).
This change removes an unnecessary acquire of send_crit_ and caches
encoder settings in ViEEncoder instead of acquiring them through
::SendCodec() in VCM (which is blocking).
BUG=webrtc:5106
R=stefan@webrtc.org
Review URL: https://codereview.webrtc.org/1433703002 .
Cr-Commit-Position: refs/heads/master@{#10582}
Reason for revert:
Failed test not related to this CL (test fails on
master at an earlier date), re-landing original CL..
(This time from my @webrtc account.)
Original issue's description:
> Revert of Work on flexible mode and screen sharing. (patchset #28 id:520001 of https://codereview.webrtc.org/1328113004/ )
>
> Reason for revert:
> Seems to break VideoSendStreamTest.ReconfigureBitratesSetsEncoderBitratesCorrectly on Linux Memcheck buildbot.
>
> Original issue's description:
> > Work on flexible mode and screen sharing.
> >
> > Implement VP8 style screensharing but with spatial layers.
> > Implement flexible mode.
> >
> > Files from other patches:
> > generic_encoder.cc
> > layer_filtering_transport.cc
> >
> > BUG=webrtc:4914
> >
> > Committed: https://crrev.com/77ccfb4d16c148e61a316746bb5d9705e8b39f4a
> > Cr-Commit-Position: refs/heads/master@{#10572}
>
> TBR=sprang@webrtc.org,stefan@webrtc.org,philipel@google.com,asapersson@webrtc.org,mflodman@webrtc.org,philipel@webrtc.org
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=webrtc:4914
>
> Committed: https://crrev.com/0be8f1d347bdb171462df89c2a4c69b3f3eb7519
> Cr-Commit-Position: refs/heads/master@{#10578}
TBR=sprang@webrtc.org,stefan@webrtc.org,philipel@google.com,asapersson@webrtc.org,mflodman@webrtc.org,terelius@webrtc.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=webrtc:4914
Review URL: https://codereview.webrtc.org/1431283002
Cr-Commit-Position: refs/heads/master@{#10581}
On Android, we would like to use MediaCodec output buffers to hold decoded frames until they can be rendered to a texture. There can only be one texture buffer used at the same time and therefore the calculated decode time in VCMTiming will be wrong since that calculation will also include the time where the decoder waited for the upper layers (that depend on network jitter and actual render time) to release the frame.
This new method will be used in
https://codereview.webrtc.org/1422963003/
BUG=webrtc:4993
R=stefan@webrtc.orgTBR=mflodman@webrtc.org
Review URL: https://codereview.webrtc.org/1414693006 .
Cr-Commit-Position: refs/heads/master@{#10576}
Reason for revert:
Breaks bot.
Original issue's description:
> Change type of pid_diff (int16_t -> uint8_t) according to updates in RTP payload profile. Max p_diff is 8 bits.
>
> Change type of number of reference pictures (size_t -> uint8_t). Max is 2 bits.
>
> Size of WebRtcRTPHeader: 4352 -> 1784 bytes.
>
> BUG=webrtc:5144, chromium:500602
>
> Committed: https://crrev.com/81c5c7f8157f767747bd97419eb0a589207354cf
> Cr-Commit-Position: refs/heads/master@{#10504}
TBR=stefan@webrtc.org,mflodman@webrtc.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=webrtc:5144, chromium:500602
Review URL: https://codereview.webrtc.org/1423493005
Cr-Commit-Position: refs/heads/master@{#10508}
Change type of number of reference pictures (size_t -> uint8_t). Max is 2 bits.
Size of WebRtcRTPHeader: 4352 -> 1784 bytes.
BUG=webrtc:5144, chromium:500602
Review URL: https://codereview.webrtc.org/1427253002
Cr-Commit-Position: refs/heads/master@{#10504}
This changes the following module directories:
* webrtc/modules/audio_conference_mixer/interface
* webrtc/modules/interface
* webrtc/modules/media_file/interface
* webrtc/modules/rtp_rtcp/interface
* webrtc/modules/utility/interface
To avoid breaking downstream, I followed this recipe:
1. Copy the interface dir to a new sibling directory: include
2. Update the header guards in the include directory to match the style guide.
3. Update the header guards in the interface directory to match the ones in include. This is required to avoid getting redefinitions in the not-yet-updated downstream code.
4. Add a pragma warning in the header files in the interface dir. Example:
#pragma message("WARNING: webrtc/modules/interface is DEPRECATED; "
"use webrtc/modules/include")
5. Search for all source references to webrtc/modules/interface and update them to webrtc/modules/include (*.c*,*.h,*.mm,*.S)
6. Update all GYP+GN files. This required manual inspection since many subdirectories of webrtc/modules referenced the interface dir using ../interface etc(*.gyp*,*.gn*)
BUG=5095
TESTED=Passing compile-trybots with --clobber flag:
git cl try --clobber --bot=win_compile_rel --bot=linux_compile_rel --bot=android_compile_rel --bot=mac_compile_rel --bot=ios_rel -m tryserver.webrtc
R=stefan@webrtc.org, tommi@webrtc.org
Review URL: https://codereview.webrtc.org/1417683006 .
Cr-Commit-Position: refs/heads/master@{#10500}
This is a re-land of https://codereview.webrtc.org/1353263005/
which was reverted because of perf-regressions. Changes since that CL:
* Change LayerFilteringTransport to send a padding packet instead of
dropping it for data that should be filtered out. This prevents
confusion due to changed sequence numbers.
* Changed timing of stats poller thread in VideoAnalyzer. Startup was
racy wrt initializion of send_stream_.
* Minor formatting issues.
PERF NOTE: This change will affect some performance numbers slightly.
In particular, {encode_frame_rate, encode_time_ms,
encode_usage_percent, media_bitrate_bps} will change due to timing
of the measurements.
BUG=
R=pbos@webrtc.orgTBR=mflodman@webrtc.org
Review URL: https://codereview.webrtc.org/1412233003
Cr-Commit-Position: refs/heads/master@{#10483}
When estimating that we can send more than the codec max bitrate (under
the assumption that codec max bitrate is good enough for the quality),
we should use additional bitrate so that we can maintain good quality.
Global bitrate caps should still be enforced through bitrate caps (b=AS)
and not codec max bitrates.
BUG=webrtc:5102
R=stefan@webrtc.org
Review URL: https://codereview.webrtc.org/1428473002
Cr-Commit-Position: refs/heads/master@{#10457}
