To make it natural to reuse them for vp9
Bug: webrtc:11999
Change-Id: If2ef7ca16b8be96e0e03bb19211d9f5eb74b2d3d
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/188620
Reviewed-by: Philip Eliasson <philipel@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32414}
to make it a bit simpler
Bug: None
Change-Id: Ie6288594c5a1b8535007623032b422eefc716ca6
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/188460
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Philip Eliasson <philipel@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32405}
This is a reland of 76d3e7a8d1
I have run the WPT tests and ensured they are now passing with this
change. I have changed the following,
- The old CL was assuming that ToI420 frames had type I420, but they
could be I420A which was causing a crash.
- I fixed a copy-paste error in the offset of the V stride.
Original change's description:
> NV12 support for VP8 simulcast
>
> Tested using video_loopback with generated NV12 frames.
>
> Bug: webrtc:11635, webrtc:11975
> Change-Id: I14b2d663c55a83d80e48e226fcf706cb18903193
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/186722
> Commit-Queue: Evan Shrubsole <eshr@google.com>
> Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#32325}
Bug: webrtc:11635
Bug: webrtc:11975
Change-Id: Ifa790af97cd7ab30c6cb4648ebd140abc1593b0b
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/187490
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Evan Shrubsole <eshr@google.com>
Cr-Commit-Position: refs/heads/master@{#32381}
(Reland with no changes after the fix to the downstream project)
This can be overriden for kNative frame types to perform scaling efficiently.
Default implementations for existing buffer types require actual
buffer implementation, thus this CL also merges "video_frame"
with "video_frame_I420" build targets.
Originally Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/186303
(Landing with TBR as it's unchaged reland of already approved CL)
TBR=nisse@webrtc.org,sakal@webrtc.org
Bug: webrtc:11976, chromium:1132299
Change-Id: Ia23f7d3e474bd9cdc177104cc5c6d772f04b210f
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/187345
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32362}
Adds a field to EncoderInfo called preferred_pixel_formats which a
software encoder populates with the pixel formats it supports. When a
kNative frame is received for encoding, the VideoStreamEncoder will
first try to get a frame that is accessible by the software encoder in
that pixel format from the kNative frame. If this fails it will fallback
to converting the frame using ToI420.
This minimizes the number of conversions made in the case that the
encoder supports the pixel format of the native buffer or where
conversion can be accelerated. For example, in Chromium, the capturer can
emit an NV12 frame, which can be consumed by libvpx which supports NV12.
Testing: Tested in Chrome with media::VideoFrame adapters.
Bug: webrtc:11977
Change-Id: I9becc4100136b0c0128f4fa06dedf9ee4dc62f37
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/187121
Reviewed-by: Niels Moller <nisse@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Markus Handell <handellm@webrtc.org>
Commit-Queue: Evan Shrubsole <eshr@google.com>
Cr-Commit-Position: refs/heads/master@{#32353}
This reverts commit f5e261aaf6.
Reason for revert: Breaks downstream projects.
Original change's description:
> Introduce RTC_NO_UNIQUE_ADDRESS.
>
> This macro introduces the possibility to suggest the compiler that a
> data member doesn't need an address different from other non static
> data members.
>
> The usage of a macro is to maintain portability since at the moment
> the attribute [[no_unique_address]] is only supported by clang
> with at least -std=c++11 but it should be supported by all the
> compilers starting from C++20.
>
> Bug: webrtc:11495
> Change-Id: I9f12b67b4422a2749649eaa6b004a67d5fd572d8
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173331
> Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
> Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#32246}
TBR=mbonadei@webrtc.org,kwiberg@webrtc.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: webrtc:11495
Change-Id: Ice318d1b11ca3dff09c190187a0b0a32ca945fe3
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/186944
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Per Kjellander <perkj@webrtc.org>
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32335}
The trial name WebRTC-VP9-PerLayerSpeed is used to
a) set encoding speed per spatial layer, based on resolution
b) allow explicitly overriding speed per layer, for testing
Additionally, this CL updates the vp9 wrapper in preparation for
injectable trials.
Bug: webrtc:11551, webrtc:11926
Change-Id: I2bb3a664feaef60483ffc241b71070284d3e0172
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/186400
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32294}
Otherwise if the pixel format is not I420, the image buffer will
need to be reallocated on each reconfiguration.
Bug: webrtc:11974
Change-Id: Ib13f1865d7dbba4635f57dc09c7bff846e127585
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/186340
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Evan Shrubsole <eshr@google.com>
Cr-Commit-Position: refs/heads/master@{#32281}
Before this change the allocated buffer and encoder complexity was set
based on the highest resolution configured regardless if that spatial
layer was active or not.
This should reduce memory pressure and improve visual quality when only
a low resolution is requested. In test, increasing the encoder
complexity has paradoxically also resulted in increased decoder speed.
Bug: webrtc:11551
Change-Id: I3ae47a5856de82ff7d40fddfcb160935b12b1d2b
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/186301
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32280}
This macro introduces the possibility to suggest the compiler that a
data member doesn't need an address different from other non static
data members.
The usage of a macro is to maintain portability since at the moment
the attribute [[no_unique_address]] is only supported by clang
with at least -std=c++11 but it should be supported by all the
compilers starting from C++20.
Bug: webrtc:11495
Change-Id: I9f12b67b4422a2749649eaa6b004a67d5fd572d8
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/173331
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32246}
The former was unused, the latter is replaced with the explicit C++11
deletions. The related RTC_DISALLOW_COPY_AND_ASSIGN is left for now,
it is used in a lot more places.
Bug: None
Change-Id: I49503e7f2b9ff43c6285f8695833479bbc18c380
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/185500
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Commit-Queue: Niels Moller <nisse@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32224}
NV12 frames can be encoded by libvpx now, and this change allows for
encoding of them with VP9.
VP9 encode/decode tests now run with NV12 as well as I420.
Manually tested using video loopback with VP9 and NV12 generated frames.
out/Default/video_loopback.app/Contents/MacOS/video_loopback --clip=GeneratorNV12 --codec="VP9"
Bug: webrtc:11635, webrtc:11974
Change-Id: Ifc5cbf77d2a27821cd5560c253d5d447c7a7cf53
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/185123
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Commit-Queue: Evan Shrubsole <eshr@google.com>
Cr-Commit-Position: refs/heads/master@{#32220}
Unfortunate typo and weak tests made it so if only a middle spatial layer
is active, vp9 encoder would be configured to send two top layers.
Bug: webrtc:11319
Change-Id: I460c245044f60ea7e0127c0e4134d0edab85f4f3
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/185043
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32164}
This is a reland of
https://webrtc-review.googlesource.com/c/src/+/174261
Patchset 1 contains the old cl (plus a merge conflict fix).
Later patchets are bufixes: A PeerConnection can be created without a
Call instance (in the case of DataChannel only), so we can't always
use that to fetch the current trials.
Old CL descritpion:
This replaces field_trial:: -based functions from system_wrappers.
Field trials are still used as fallback, but injectable trials are now
possible.
// Since re-land is otherwise unchanged, setting previous reviewers as TBR
TBR=kthelgason@webrtc.org,mbonadei@webrtc.org,stefan@webrtc.org,srte@webrtc.org
Bug: webrtc:11926
Change-Id: I57a9e8c3454f226f77fb93215bcac83da65034b0
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/185003
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Per Kjellander <perkj@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32163}
CpuSpeedExperiment: Add option to have a separate config for cores below a configurable threshold.
Bug: none
Change-Id: I51562979f3a89a949d014a1ee6fc0802f3c1dae5
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/184926
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Commit-Queue: Åsa Persson <asapersson@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32154}
This replaces field_trial:: -based functions from system_wrappers.
Field trials are still used as fallback, but injectable trials are now
possible.
Bug: webrtc:11926
Change-Id: I70f28c4fbabf6d9e55052342000e38612b46682c
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/174261
Reviewed-by: Per Kjellander <perkj@webrtc.org>
Reviewed-by: Kári Helgason <kthelgason@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32129}
It should already be enabled by default in libaom, but explicitly enable
it here in case that changes.
Bug: None
Change-Id: I93a1dfc92f9c02bc5ec823c326d8cf6ff163bceb
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/184262
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Emil Lundmark <lndmrk@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32114}
Removes the need for specifying a fixed number of parameters.
Bug: none
Change-Id: I1324861807cb4929963aedccb6c2755b9c6ea3fc
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/180421
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Commit-Queue: Åsa Persson <asapersson@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32055}
We can then finally delete the top-level common_types.h, and the
corresponding build target webrtc_common.
Bug: webrtc:7660
Change-Id: I1c1096541477586d90774c7a3405b9d36edec14a
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/182800
Commit-Queue: Niels Moller <nisse@webrtc.org>
Reviewed-by: Sebastian Jansson <srte@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#32044}
Some libraries hooking into WebRTC still manage to have the conference mode
flag enabled on non screenshare sources resulting in a bad rate allocation.
Bug: webrtc:11310
Change-Id: Id5205affb562511eda40c460e380c105d8589c51
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/182003
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Florent Castelli <orphis@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31965}
Deletes all webrtc usage of this member. Next step is to delete
any downstream references, and when that's done, the member can be
deleted.
Bug: None
Change-Id: I3f3a94a063dccf56468a1069653efd3809875b01
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/181201
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Niels Moller <nisse@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31911}
This way can double adapt right away instead of relying
on the qp scaler checking soon into the future.
Bug: webrtc:11830
Change-Id: I8e878168303cf6a4c3edcf3997dd8ac2413a4479
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/181060
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Åsa Persson <asapersson@webrtc.org>
Commit-Queue: Evan Shrubsole <eshr@google.com>
Cr-Commit-Position: refs/heads/master@{#31895}
This is a reland of 32ca95145c
Fix includes not enabling the screenshare conference behavior on non
screenshare sources even if the flag is enabled.
Original change's description:
> Only enable conference mode simulcast allocations with flag enabled
>
> Non-conference mode simulcast screenshares were mistakenly using the
> conference mode semantics in the simulcast rate allocator, which broke
> spec compliant usage in some situation.
>
> This behavior should only be used when explicitly using the SDP entry
> "a=x-google-flag:conference" in both offer and answer.
>
> Bug: webrtc:11310, chromium:1093819
> Change-Id: Ibcba75c88a8405d60467546b33977a782e04e469
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/179081
> Reviewed-by: Harald Alvestrand <hta@webrtc.org>
> Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Commit-Queue: Florent Castelli <orphis@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31828}
Bug: webrtc:11310
Bug: chromium:1093819
Change-Id: Ic933f93a5c4bad20583354fe821f8a1170e911cd
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/180802
Commit-Queue: Florent Castelli <orphis@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31847}
This reverts commit 32ca95145c.
Reason for revert: Internal test failure
Original change's description:
> Only enable conference mode simulcast allocations with flag enabled
>
> Non-conference mode simulcast screenshares were mistakenly using the
> conference mode semantics in the simulcast rate allocator, which broke
> spec compliant usage in some situation.
>
> This behavior should only be used when explicitly using the SDP entry
> "a=x-google-flag:conference" in both offer and answer.
>
> Bug: webrtc:11310, chromium:1093819
> Change-Id: Ibcba75c88a8405d60467546b33977a782e04e469
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/179081
> Reviewed-by: Harald Alvestrand <hta@webrtc.org>
> Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Commit-Queue: Florent Castelli <orphis@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31828}
TBR=ilnik@webrtc.org,hta@webrtc.org,orphis@webrtc.org
Change-Id: I5ccb6e87594f491ba09fe6b837ee24d63db878ca
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: webrtc:11310
Bug: chromium:1093819
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/180801
Reviewed-by: Florent Castelli <orphis@webrtc.org>
Commit-Queue: Florent Castelli <orphis@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31829}
Non-conference mode simulcast screenshares were mistakenly using the
conference mode semantics in the simulcast rate allocator, which broke
spec compliant usage in some situation.
This behavior should only be used when explicitly using the SDP entry
"a=x-google-flag:conference" in both offer and answer.
Bug: webrtc:11310, chromium:1093819
Change-Id: Ibcba75c88a8405d60467546b33977a782e04e469
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/179081
Reviewed-by: Harald Alvestrand <hta@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Florent Castelli <orphis@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31828}
i.e. when chain are used,
require each decode target to be protected by some chain.
where previously it was allowed to mark decode target as unprotected.
See https://github.com/AOMediaCodec/av1-rtp-spec/pull/125
Bug: webrtc:10342
Change-Id: Ia2800036e890db44bb1162abfa1a497ff68f3b24
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/178807
Reviewed-by: Philip Eliasson <philipel@webrtc.org>
Reviewed-by: Björn Terelius <terelius@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31772}
No need to keep error_resilience 1 for layers in AV1
Bug: None
Change-Id: I6570d653a34ed2187307154ccdfd9e941ed8f917
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/179742
Reviewed-by: Marco Paniconi <marpan@webrtc.org>
Commit-Queue: Jerome Jiang <jianj@google.com>
Cr-Commit-Position: refs/heads/master@{#31769}
While converting the aggregated (stap-a) packet transform packet
framing input into an annex-b framing copy, the two loops (both the
required size calculation and the stap-a-to-annex-b copy) may
over-read the input buffer.
In both buffers, `nalu_ptr` follows the input (stap-a) buffer, which
is located in `data`, and whose length is `data_size`. Buffer is read
until `nalu_ptr` reaches the end of the buffer. Issues is that the 5th
line in the loop:
```
uint16_t segment_length = nalu_ptr[0] << 8 | nalu_ptr[1];
```
This line accesses `nalu_ptr[1]`, which needs to be protected in
the loop condition. Let's assume `data_size = 4`, and that we restart
the loop with `nalu_ptr = data + 3`. The condition of the loop does
hold (`nalu_ptr = data + 3 < data + data_size`), but the 5th line
will access to `data[3+1] = data[4]`, which is an over-read.
Tested:
```
$ ninja -C out/Default
$ out/Default/modules_unittests --gtest_filter=PacketBuffer*:H264*:RtpPacketizerH264Test*:VideoRtpDepacketizerH264Test*:TestH264SpsPpsTracker* --logs
...
[ PASSED ] 97 tests.
```
Change-Id: I8b8aaf7d12b0bb154430b8922f099cd49e684762
Bug: webrtc:11698
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/177140
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Commit-Queue: Niklas Enbom <niklas.enbom@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31561}
1) Fix several typos and small mistakes which could lead to crashes
2) Adjust bitrates if leading layers are disabled
3) Wire up webrtc quality scaler
Bug: webrtc:11319
Change-Id: I16e52bdb1c315d64906288e4f2be55fe698d5ceb
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/177525
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Reviewed-by: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31546}
enable this opt can give 20% performance improvement for video
decoding with 720P video loopback and fake camera on chromebook sarien.
Bug: None
Test: ./modules_tests on chromebook sarien
Change-Id: I8c6487b291b5861e6ba6b6d55b24d7ddb51c341e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/177335
Reviewed-by: Sergey Silkin <ssilkin@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31543}
This is a reland of d592575698
Patchset 2 is a reland of
https://webrtc-review.googlesource.com/c/src/+/177012
Patchset 3 is a fix for a potential crash when InitDecode()is called from
VideoStreamDecoderImpl::GetDecoder(), where the decoder_settings
parameter is a but surprisingly set to nullptr.
Original change's description:
> VP9 decoder: Sets thread count based on resolution, reinit on change.
>
> Previously, number of decoder threads for VP9 were always set to 8 but
> with a cap at number of cores. This was done since we "can't know" the
> resolution that will be used.
>
> With this change, we now intialize the number of threads based on
> resolution given in InitDecode(). If a resolution change happens in
> flight, it requires a keyframe. We therefore parse the header from
> any key frame and if it has a new resolution, we re-initialize the
> decoder.
>
> The number of threads used is based on pixel count. We set one thread
> as target for 1280x720, and scale up lineraly from there. The 8-thread
> cap is gone, but still limit it core count.
>
> This means for instance: 1 <= 720p, 2 for 1080p, 4 for 1440p, 9 for 4K.
>
> Bug: webrtc:11551
> Change-Id: I14c169a6c651c50bd1b870c4b22bc4495c8448fd
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/174460
> Commit-Queue: Erik Språng <sprang@webrtc.org>
> Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31507}
Bug: webrtc:11551
Change-Id: I2b4b146d0b8319f07ce1660202d6aa4b374eb015
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/177246
Reviewed-by: Johannes Kron <kron@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31527}
Params and format is the same as for existing ARM experiment, but a new
group name is created for non-ARM experiment.
Bug: webrtc:11551
Change-Id: I3a6c0f07a8c1d714477ae4703c16e48df36ac10e
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/177102
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Rasmus Brandt <brandtr@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31524}
Show it can make vp9 tests cleaner too.
Bug: None
Change-Id: I8333a61dec1ef90ade9faffea94e1555ccbfcfaa
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/177013
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Reviewed-by: Philip Eliasson <philipel@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31523}
This reverts commit d592575698.
Reason for revert: May cause crashes.
Original change's description:
> VP9 decoder: Sets thread count based on resolution, reinit on change.
>
> Previously, number of decoder threads for VP9 were always set to 8 but
> with a cap at number of cores. This was done since we "can't know" the
> resolution that will be used.
>
> With this change, we now intialize the number of threads based on
> resolution given in InitDecode(). If a resolution change happens in
> flight, it requires a keyframe. We therefore parse the header from
> any key frame and if it has a new resolution, we re-initialize the
> decoder.
>
> The number of threads used is based on pixel count. We set one thread
> as target for 1280x720, and scale up lineraly from there. The 8-thread
> cap is gone, but still limit it core count.
>
> This means for instance: 1 <= 720p, 2 for 1080p, 4 for 1440p, 9 for 4K.
>
> Bug: webrtc:11551
> Change-Id: I14c169a6c651c50bd1b870c4b22bc4495c8448fd
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/174460
> Commit-Queue: Erik Språng <sprang@webrtc.org>
> Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31507}
TBR=ilnik@webrtc.org,sprang@webrtc.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: webrtc:11551
Change-Id: Id235c8ded83b3e1fc1d132c8f56c9f20001f6f22
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/177242
Reviewed-by: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31521}
This reverts commit 26e5046951.
Reason for revert: Part of change that may cause crashes.
Original change's description:
> Adjusts allowable thread count for vp9 decoders.
>
> Set 2 thread as target for 1280x720 pixel count, and then scale up
> linearly from there - but cap at physical core count.
> For common resolutions this results in:
> 1 for 360p
> 2 for 720p
> 4 for 1080p
> 8 for 1440p
> 18 for 4K
>
> Bug: webrtc:11551
> Change-Id: I666bd971eccddee096749f20d3b08eb40fe868ad
> Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/177012
> Reviewed-by: Johannes Kron <kron@webrtc.org>
> Cr-Commit-Position: refs/heads/master@{#31513}
TBR=sprang@webrtc.org,kron@webrtc.org
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: webrtc:11551
Change-Id: I4ea5166efeed3d55255a0243a69deb584a0e19e2
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/177240
Reviewed-by: Erik Språng <sprang@webrtc.org>
Commit-Queue: Erik Språng <sprang@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31517}
Set 2 thread as target for 1280x720 pixel count, and then scale up
linearly from there - but cap at physical core count.
For common resolutions this results in:
1 for 360p
2 for 720p
4 for 1080p
8 for 1440p
18 for 4K
Bug: webrtc:11551
Change-Id: I666bd971eccddee096749f20d3b08eb40fe868ad
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/177012
Reviewed-by: Johannes Kron <kron@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31513}
Previously, number of decoder threads for VP9 were always set to 8 but
with a cap at number of cores. This was done since we "can't know" the
resolution that will be used.
With this change, we now intialize the number of threads based on
resolution given in InitDecode(). If a resolution change happens in
flight, it requires a keyframe. We therefore parse the header from
any key frame and if it has a new resolution, we re-initialize the
decoder.
The number of threads used is based on pixel count. We set one thread
as target for 1280x720, and scale up lineraly from there. The 8-thread
cap is gone, but still limit it core count.
This means for instance: 1 <= 720p, 2 for 1080p, 4 for 1440p, 9 for 4K.
Bug: webrtc:11551
Change-Id: I14c169a6c651c50bd1b870c4b22bc4495c8448fd
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/174460
Commit-Queue: Erik Språng <sprang@webrtc.org>
Reviewed-by: Ilya Nikolaevskiy <ilnik@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31507}
This change introduces a new non-reentrant mutex to WebRTC. It
enables eventual migration to Abseil's mutex.
The mutex types supportable by webrtc::Mutex are
- absl::Mutex
- CriticalSection (Windows only)
- pthread_mutex (POSIX only)
In addition to introducing the mutexes, the CL also changes
PacketBuffer to use the new mutex instead of rtc::CriticalSection.
The method of yielding from critical_section.cc was given a
mini-cleanup and YieldCurrentThread() was added to
rtc_base/synchronization/yield.h/cc.
Additionally, google_benchmark benchmarks for the mutexes were added
(test courtesy of danilchap@), and some results from a pthread/Abseil
shootout were added showing Abseil has the advantage in higher
contention.
Bug: webrtc:11567, webrtc:11634
Change-Id: Iaec324ccb32ec3851bf6db3fd290f5ea5dee4c81
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/176230
Commit-Queue: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Mirko Bonadei <mbonadei@webrtc.org>
Reviewed-by: Karl Wiberg <kwiberg@webrtc.org>
Reviewed-by: Tommi <tommi@webrtc.org>
Reviewed-by: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31443}
This reverts commit 6958d2c6f0.
Disable the test on iOS.
Bug: None
Change-Id: Ie42fada10a92bd4a802c6c79caeb4965410ddf6a
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/176461
Reviewed-by: Stefan Holmer <stefan@webrtc.org>
Commit-Queue: Jerome Jiang <jianj@google.com>
Cr-Commit-Position: refs/heads/master@{#31437}
to unblock rolling new version where private function is no longer available
Bug: None
Change-Id: I9c35fede3f331f7688cc97acfbda1250b42348a3
Reviewed-on: https://webrtc-review.googlesource.com/c/src/+/176441
Reviewed-by: Philip Eliasson <philipel@webrtc.org>
Commit-Queue: Danil Chapovalov <danilchap@webrtc.org>
Cr-Commit-Position: refs/heads/master@{#31427}
