git-subtree-dir: cobalt/updater
git-subtree-split: 5d069f92aa014ded86ab785b26971ac2dbe74b8e

git-subtree-dir: third_party/brotli
git-subtree-split: 1218c4f5fdcc00016ccd2531932e9f3dec0151e6

Fix bug that the EG file name is not correct when copied to RDK.

b/293172465

…outube#1736)

This avoids possible NullPointerException and NoSuchElementException
from MediaCodecBridge.formatHasCropValues.

b/298692099

Always set cflags so that different codepaths can append to it.

b/303505529

b/302689405

b/299304547

b/277591845

Review URL: https://codereview.chromium.org/435603010

git-svn-id: http://src.chromium.org/svn/trunk/deps/third_party/flac@287124 4ff67af0-8c30-449e-8e8b-ad334ec8d88c
git-subtree-dir: third_party/flac
git-subtree-split: 0635a091379d9677f1ddde5f2eec85d0f096f219

Remove arg that is introduced in python 3.8 as it's breaking our pylint
for python 3.7, even though our builder is running python 3.11.

This arg should be unneeded for our automated builds and was simply a
convenience for manual testing.

b/303258519

Change-Id: I414eabdd4a673fefa581bd6531830006303c9de0

b/303749197

Change-Id: Ie39ad1ab0b8a61357fcff133786cc9469bbadcd4

b/265873566

Change-Id: I696c619ec5107c3d399146e23fa42131f7016337

git-subtree-dir: url
git-subtree-split: ec3092c8d21e20e39698f5272c8a7f71c3ee88a2

b/298660052

Change-Id: I3d8c86431736684562bb352f91e4a80298c97bf3

Docker build step has diff mechanism that compares PR SHA to the
previous SHA on the branch (parent SHA). Setting fetch-depth to 2
creates problems for PRs that need to be rebased, because diff mechanism
doesn't have enough git history to locate parent SHA.

b/304280296

- Clearly document that the implementation should return the time zone
name in the IANA format.

b/302569322

Change-Id: Ibd6e29f23cb6293f6339fe5a7d860eeb3a9a2b5d

The experiment has launched to 100%, so it should be safe to set
this as default enabled for new installations.

b/285656784

b/268244534

Change-Id: Id54608490332514dc256e738ece58aa87855b94c

---------

Co-authored-by: Andrew Savage <[email protected]>

The DCHECK(stream) in SbPlayerPipeline::OnDemuxerStreamRead() may
trigger when SbPlayerPipeline::Stop() has been called. It won't affect
production build as it's a DCHECK and the check below on
|player_bridge_| correctly guarded the code from a stopped pipeline.

Now the check is made explicit, and redundant DCHECK() and if statements
are added.

b/276483058

direct_dependent_configs is the deprecated (see tools/gn/target_generator.cc)
name and public_configs replaces it keeping the same behavior though.

gn-dev thread:
https://groups.google.com/a/chromium.org/d/topic/gn-dev/h9N85MIx_Bk/discussion

BUG=None
TEST=gn gen out-gn/Release --args='is_debug=false is_component_build=false symbol_level=1'
[email protected], [email protected]

Review URL: https://codereview.chromium.org/1259713008 .

(cherry picked from commit e1b6e7983a8c54954f13f40772a40f490f1eefca)

Complie FLAC on Starboard and Windows

b/27696921

b/204809498

b/302372195

Change-Id: I6ee7f6b5978b9e17d1e2c28c5a1a42f376204948

git-subtree-dir: third_party/zlib
git-subtree-split: f12b46f00f17488ad16cb242bf4f248bdde9618b

It sets the log level to `LOG_VERBOSE - 15` to effectively enables all
known verbose messages.
Supporting detailed verbose level is possible but not implemented so the
implementation remains straight forward.

b/276483058

Replaces the IAudioEndpointVolume API with ISimpleAudioVolume ahead of a
future API change that will remove IAudioEndpointVolume functionality.

b/303500203

Fix to reset RC for temporal layers: the
first_spatial_layer_to_encode is usually/default 0,
so the logic to reset for temporal layers was not
being executed. Use VPXMAX(1, ) to make sure  all
temporal layers will be reset (when max-q is used
for overshoot).

Change-Id: Iec669870c865420d01d52eab9425cd6c7714eddc
git-subtree-dir: third_party/libvpx
git-subtree-split: 53747dfe65eaf670a7192f55117f3bf1e0280743

b/285656784

Instead of storing a reference to a JS callback in the
metrics_log_uploader, we instead publish an event to EventDispatcher
when metrics are ready to be uploaded. H5vccMetrics listens for these
events and returns the result to the JS callback.

This eliminates race conditions where the JS environment (and therefore
H5vcc) was being destroyed and metrics were queued for upload,
referencing freed pointers.

b/298050585

This allows a web app to override the default memory_limit on a
SourceBufferStream which backs an instance of source_buffer.

A SourceBufferStream is not created until after the first Init Segment
is appended, so if you try to read the memory_limit before then, the web
app will get an exception. However you can still pre-emptively set the
memory_limit which will be applied to SourceBufferStream upon creation.

b/291291120

Change-Id: Id577df24fadad4bb6405a1ecf1439942cd988837

We ran into an off-by-1 error with Telemetry as we were using
UmaHistogramEnumeration.

The histograms MACROs checks that sample is less than max - 

https://source.corp.google.com/h/lbshell-internal/cobalt_src/+/COBALT:base/metrics/histogram_functions.h;l=53

Chrome actually uses UmaHistogramExactLinear so they can add a +1 to the
PIPELINE_STATUS_MAX
https://source.chromium.org/chromium/chromium/src/+/main:media/mojo/services/media_metrics_provider.cc;l=144

This changes puts us more in line with the Chrome implementation.

b/287670693

Add more cleanup steps to keep devices healthier over large numbers of
automated tests.

b/291640038

Change-Id: Iac9e62a4bc40faa8df8441b77a543c04e2f2196d

b/302569322

Change-Id: Icce69ed2db148020c44ab9d95af8d83283e3c138

b/182513726

Change-Id: Icc182e400611df957d83e2c12df5559bc945c738

And a simple test.

Bug: linux-syscall-support:31
Change-Id: Ifea152cdea89183f69d881807f3bd60da56a4513
Reviewed-on: https://chromium-review.googlesource.com/c/linux-syscall-support/+/2351965
Reviewed-by: Mike Frysinger <[email protected]>
git-subtree-dir: third_party/linux-syscall-support
git-subtree-split: 29f7c7e018f4ce706a709f0b0afbf8bacf869480

Changes SbPlayerWriteSampleTest.PartialAudioDiscardAll to write buffers
with an infinite duration discarded from the front or back.

b/302020010

Aligns the size of the AudioDevice output buffer to the fixed size of
the written audio data. This ensures that the audio data is not written
outside of the buffer's bounds.

b/276502329

b/298236240

b/238885425
b/266714023

…1792)

b/304882604

b/304882604

This avoids exceptions being thrown from MediaCodecBridge.textureWidth()
and MediaCodecBridge.textureHeight().

b/298692099

)

b/300175596

Per
https://developer.android.com/about/versions/14/changes/fgs-types-required#media
we should be setting that we are using the
FOREGROUND_SERVICE_MEDIA_PLAYBACK permission and also passing the
corresponding type constants when we start the foreground service.

b/305802411

This new test type was added as a proof of concept and has been running
for a few months. We've since decided to go in a different direction,
though - go/sustainable-eg-tests - and the evergreen-as-blackbox tests
can be turned down and removed.

b/278953372

Change-Id: I678c82617e2208ea878dafdf903bf86257da8db4

When running the install steps for executables that are built as a
result of the shared library (i.e. loaders), sometimes the name of the
original shared library that invoked it is needed in the install_target
context in order to create necessary dependencies on the shared_library
install_target counterparts.

b/246855300

…outube#1812)

b/300175596

b/305285508

Change-Id: Iee0da0c85bba750d8929eacf60e45c1797e354af

…outube#1811)

Reverts youtube#1792

This should be OK to re-enable now.

b/304882604

…ol()

This avoids a compile error when building with GCC in C++17 mode.

(cherry picked from commit b7558540a9d52738d827b3993b62e8d140d9acb4)

Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/3307102

b/288332442

b/226999079

Fix a bug where the launcher script copies the binaries from.
The correct path is install/lib.

b/293172465

The cl makes SbAudioSinkGetMinBufferSizeInFrames() return default values
if audio output type changes from remote to local or from local to
remote. It doesn't re-run MinRequiredFramesTester.

b/298847156

b/175822425

Bundle content steps need to copy files from both cobalt and starboard
toolchain versions of the `content/data` directory. However to preserve
the folder structure so that we get the correct placement of
content/data files in the final packaged NX application, we need to
ensure the copy-script uses the correct toolchain-specific base
directory.

b/246855300

Change-Id: Iaffced7832bfc157b7c3c0a6d600b58768977ca0

…e#915)

[b/286272327](https://partnerissuetracker.corp.google.com/issues/286272327)

When building tests' evergreen loaders, ninja expects content directory
under install/base_unittests_evergreen_loader when building
base_unittests_evergreen_loader_install. This CL corrects the content
path so that ninja can find the content to be copied to the loader apk.

b/268244534

Change-Id: If542521268ce259c166729b065f7e9dd4aeb3a72

Remove a duplicate shell command, and remove the unnecessary prompt
waiting.

b/293172465

b/288332442

Much like the utility button to capture and download a Trace, this CL
adds a button to the Cobalt tab of Devtools for capturing a NetLog.

View it here - https://screenshot.googleplex.com/ATA2L2GcEH6W272 

See here for more details on NetLog -
https://github.com/youtube/cobalt/blob/main/cobalt/doc/net_log.md

b/305257093

Reverts youtube#1770

b/175822425

b/280671902

Change-Id: Ic4f7f5f61c69c45be71529c567a6324122c6471b