GitLab

This user perceived latency histogram should reflect the time from a browser
cold start to the moment the first navigation is committed: the progress bar
appears and the omnibox draws artifacts of the URL.

BUG=436960

Review URL: https://codereview.chromium.org/818513002

Cr-Commit-Position: refs/heads/master@{#310761}

Specifically, for https://goto.google.com/gcmdriver-already-instantiated

Sample output:
[FATAL:gcm_profile_service.cc(155)] An instance of GCMProfileService already exists! Old profile: 0x9e9d75c0 Default 0 0 1 13065207799278578, new profile: 0x9e9d75c0 Default 0 0 1 13065207799278578

BUG=437827

Committed: https://crrev.com/934554c3f369c9821b0af3ae8a09ed55f29eafc9
Cr-Commit-Position: refs/heads/master@{#310593}

Review URL: https://codereview.chromium.org/846473003

Cr-Commit-Position: refs/heads/master@{#310760}

Add a qcms_profile_get_description() api: returns the description
of a color profile for debug use.

TBR=darin@chromium.org
BUG=401971

Review URL: https://codereview.chromium.org/796373005

Cr-Commit-Position: refs/heads/master@{#310759}

When Chrome for Android encounters a Flash based YouTube embed, it
renders a plugin replacement that redirects to the Android YoutTube
app. The HTML that describes the plugin replacement got broken in the
refactorings for crrev.com/297078

BUG=447290
TBR=arv@chromium.org

Review URL: https://codereview.chromium.org/846513002

Cr-Commit-Position: refs/heads/master@{#310758}

Previously the first argument of toggleMode function passed to the event handler
was not bounded. Thus event object is passed to the first argument of toggleMode
unintentionally.

BUG=447519
TEST=None

Review URL: https://codereview.chromium.org/839173003

Cr-Commit-Position: refs/heads/master@{#310757}

BUG=445396

Review URL: https://codereview.chromium.org/842083004

Cr-Commit-Position: refs/heads/master@{#310756}

Adds code which generates a SHA256 fingerprint of the external V8 snapshot
blob files during the build process, and embeds these values into the
binary.  During startup the SHA256 hash of the snapshot blobs are compared
against these fingerprints and Chrome will fail to load if they differ.

BUG=421063,439661

Review URL: https://codereview.chromium.org/832393003

Cr-Commit-Position: refs/heads/master@{#310755}

Some platforms have multiple video decoders and encoders.
For decode and encoder, return the first succeed initialized VDA and VEA
from all possible platforms. GetSupportedProfile return all possible
profiles from all encoders.

This cl is submitted again because crrev.com/826663002 was reverted in
crrev.com/832383004.

BUG=445016
TEST=Tested on squawks. For decoder, make sure V4L2 initialization
failed and VAAPI successed. For encoder, test on extension.

TBR=piman@chromium.org,posciak@chromium.org

Review URL: https://codereview.chromium.org/842123002

Cr-Commit-Position: refs/heads/master@{#310754}

This script was previously run three times, once for
each arch, with two of the three always being no-ops.

Instead just run the script once and have the script
figure out which image to download.

Second attempt at landing this.  First attempt caused
failures on windows builders:
 https://codereview.chromium.org/834133003

TEST=trybots + 'gclient runhooks' locally

Review URL: https://codereview.chromium.org/843643003

Cr-Commit-Position: refs/heads/master@{#310753}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/c29c39c..fc1d527

TBR=keishi@chromium.org

Review URL: https://codereview.chromium.org/841293002

Cr-Commit-Position: refs/heads/master@{#310752}

(even when it's not the current profile that's being deleted).

BUG=445043

Review URL: https://codereview.chromium.org/840733002

Cr-Commit-Position: refs/heads/master@{#310751}

It does work on win and linux at least :)

BUG=446859
TBR=henrika@chromium.org

Review URL: https://codereview.chromium.org/830423010

Cr-Commit-Position: refs/heads/master@{#310750}

ServiceWorkerDatabase::ReadRegistration and ReadUserData should return NOT_FOUND
when the database hasn't been initialized yet, but the current implementation
wrongly returns STATUS_OK without any valid data. This CL fixes it.

BUG=447469
TEST=content_unittests --gtest_filter=ServiceWorkerDatabaseTest.*

Review URL: https://codereview.chromium.org/846443005

Cr-Commit-Position: refs/heads/master@{#310749}

WebRTC 7972:8027
Changes: https://chromium.googlesource.com/external/webrtc/trunk/webrtc.git/+log/c2c971d..f19180d

Libjingle 7972:8026
Changes: https://chromium.googlesource.com/external/webrtc/trunk/talk.git/+log/1aaae71..22a0397

BUG=NONE
TBR=tommi

Review URL: https://codereview.chromium.org/842143002

Cr-Commit-Position: refs/heads/master@{#310748}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/f819761..c29c39c

TBR=keishi@chromium.org

Review URL: https://codereview.chromium.org/840263002

Cr-Commit-Position: refs/heads/master@{#310747}

This is in preparation for converting this target to GN. There shoule be no functional change.

TBR=dpranke

Review URL: https://codereview.chromium.org/840213004

Cr-Commit-Position: refs/heads/master@{#310746}

If reading is aborted before the file is opened, then obviously abort event is
not dispatched to the providing extension, as there is no operation to abort.

This patch fixes this race by waiting until the read operation is started, so
the abort event will be always dispatched.

TEST=browser_tests: *FileSystemProvider*ReadFile*
BUG=None

Review URL: https://codereview.chromium.org/836613008

Cr-Commit-Position: refs/heads/master@{#310745}

BUG=NONE
TEST=NONE

Review URL: https://codereview.chromium.org/847453002

Cr-Commit-Position: refs/heads/master@{#310744}

BUG=303778
TEST=check Files.app, VideoPlayer.app, and AudioPlayer.app if fonts look same.

Review URL: https://codereview.chromium.org/842813002

Cr-Commit-Position: refs/heads/master@{#310743}

Unittest which updates database from version 1 to current version.

BUG=423828
R=gcasto@chromium.org

Review URL: https://codereview.chromium.org/808173002

Cr-Commit-Position: refs/heads/master@{#310742}

Before, an uninitialized class was being used in queue.h, which is illegal, but
for some strange reason fails on MSAN build bots only.

TEST=Compiles on MSAN.
BUG=None

Review URL: https://codereview.chromium.org/813403005

Cr-Commit-Position: refs/heads/master@{#310741}

This patch exposes the limiting feature to providing extensions.

TEST=browser_tests: *FileSystemProvider*ReadFile*
BUG=445125
R=benwells@chromium.org, yawano@chromium.org

Review URL: https://codereview.chromium.org/824753003

Cr-Commit-Position: refs/heads/master@{#310740}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/8f3519a..e34d6bd

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=caryclark@google.com

Review URL: https://codereview.chromium.org/819173004

Cr-Commit-Position: refs/heads/master@{#310739}

BUG=none

Review URL: https://codereview.chromium.org/802613004

Cr-Commit-Position: refs/heads/master@{#310738}

This requires that previously-used textures be returned to the browser compositor before a commit, so they can be uploaded to. This saves memory by reducing the number of textures that need to be created.

Also avoid releasing old textures from renderers on SubmitFrame, as that can be unnecessarily expensive if both the old and new frames reference the same textures.

BUG=446588,446417,446414

Review URL: https://codereview.chromium.org/843633003

Cr-Commit-Position: refs/heads/master@{#310737}

This finally adds a end-to-end WebRTC audio quality test which includes
getUserMedia and the capture path (except, obviously, the os-specific
code which actually interacts with the capture device etc).

This patch also makes the audio input device generic. It will now
convert its input .wav file to suit the audio bus. I had to do this
because the PESQ tests require files to be mono, 16-bit and so on.

BUG=446859

Review URL: https://codereview.chromium.org/831693004

Cr-Commit-Position: refs/heads/master@{#310736}

This is workaround for crbug.com/442217.

BUG=442217
TEST=None

Review URL: https://codereview.chromium.org/844903003

Cr-Commit-Position: refs/heads/master@{#310735}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/8a288b5..f819761

TBR=jsbell@chromium.org,keishi@chromium.org

Review URL: https://codereview.chromium.org/845683002

Cr-Commit-Position: refs/heads/master@{#310734}

Cr-Commit-Position: refs/heads/master@{#310733}

The testserver attempts to inherit from StoppableHTTPServer, but then
overrides the stoppable functionality. Instead of overriding serve_forever, override handle_request, to maintain stoppability.

BUG=

Review URL: https://codereview.chromium.org/797253003

Cr-Commit-Position: refs/heads/master@{#310732}

interactive_ui_tests was recently ported to gn in r310072 (all good wrt
MacViews).  Now is a good opportunity to make things a bit neater, and
get things closer to flipping toolkit_views=1 on Mac

This collapses the `sources!` filter for Mac+toolkit_views into
chrome_interactive_ui_test_views_non_mac_sources which now follows the
pattern that is already used in chrome_browser_ui.gypi. That is,
chrome_browser_ui_views_non_mac_sources and friends.

This does move a few more views test files out of interactive_ui_tests
on MacViews. These were compiling and linking, but had failing tests.
E.g. they do stuff like static_cast<BrowserView*>(browser->window())
which won't work while the browser is still Cocoa-based.

BUG=402679

Review URL: https://codereview.chromium.org/793663006

Cr-Commit-Position: refs/heads/master@{#310731}

Regressed in r309924. NativeWidgetMac just needs to check for a null
bridge_ to handle calls to {Set,Get}NativeWindowProperty during
asynchronous close calls.

BUG=378134

Review URL: https://codereview.chromium.org/826083003

Cr-Commit-Position: refs/heads/master@{#310730}

Previously ThumbnailLoader refers both metadata.thumbnail.url and
meatadata.external.thumbnailUrl. Thus it loads external thumbnail even if the
file is cached locally.

The CL fixes the logic and stop to use drive thumbnail for image files that are
cached. Note that if the file is not an image, use drive thumbnail always
because Files.app does not have logic to generate thumbnail from non-image
files.

BUG=446827
TEST=Added

Review URL: https://codereview.chromium.org/831833004

Cr-Commit-Position: refs/heads/master@{#310729}

To avoid having a lot of -Wunused-private-field warnings when using "=delete" to implement DISALLOW_COPY_AND_ASSIGN.

BUG=447445

Review URL: https://codereview.chromium.org/790793008

Cr-Commit-Position: refs/heads/master@{#310728}

We can't enable on the ref build because SW doesn't work on M37.
And also we disable on XP because of test flakiness.

BUG=442752

Review URL: https://codereview.chromium.org/828063005

Cr-Commit-Position: refs/heads/master@{#310727}

This migration was added in Aug 2013, M30: https://chromiumcodereview.appspot.com/20087002 .
According to UMA stats, it is triggered now for very few users/devices and can be removed.

BUG=447211

Review URL: https://codereview.chromium.org/845533002

Cr-Commit-Position: refs/heads/master@{#310726}

This is in preparation for sharing these lists with the GN build.

There should be no functional change as a result of this.

TBR=dpranke

Review URL: https://codereview.chromium.org/842803002

Cr-Commit-Position: refs/heads/master@{#310725}

Don't show existing wallet cards in the UI when this flag is NOT passed.

BUG=none

Review URL: https://codereview.chromium.org/810763008

Cr-Commit-Position: refs/heads/master@{#310724}

BUG=none

Review URL: https://codereview.chromium.org/791053009

Cr-Commit-Position: refs/heads/master@{#310723}

BUG=384667

Review URL: https://codereview.chromium.org/819443003

Cr-Commit-Position: refs/heads/master@{#310722}