GitLab

This CL refactors audioparam-testing.js so we can hide global variables
and functions from the window scope.

BUG=722884
TEST=(All layout tests passes after this change.)

Review-Url: https://codereview.chromium.org/2886883002
Cr-Commit-Position: refs/heads/master@{#472202}

Make DnsSdRegistry a leaky singleton so that it can be used by both CastMediaSinkService and MDnsAPI.

BUG=687377

Review-Url: https://codereview.chromium.org/2874243003
Cr-Commit-Position: refs/heads/master@{#472201}

In ChromeOS the graphics cards are opened in the browser process and
IPC-ed to the GPU.

BUG=482610
TEST=Ran on link, and daisy and verified Chrome starts up

Review-Url: https://codereview.chromium.org/2870213003
Cr-Commit-Position: refs/heads/master@{#472200}

With our recent change to merge paint timings across all frames in
a page, some of the timing ordering assumptions we make, such as
parse and layout always starting before paint, can be violated.

More specifically, we can observe sequences like the following:
* parse starts in main frame
* child frame created
* child frame paints
* child frame reports paint event
* parent frame reports parse start event

Even though the events happened in the expected order, due to cross
frame buffering, we are notified of the child paint before the
main frame parse start.

This change addresses this issue, by buffering observer callbacks until
we've received events with the expected ordering.

zhenw's recent change https://codereview.chromium.org/2872793002 encountered
this issue and had to be rolled back as a result.

I confirmed that, with zhen's patch applied, the flaky test failed on 8 of
10 runs. With zhen's patch and this patch applied, that same test passed 10
of 10 times.

BUG=722860

Review-Url: https://codereview.chromium.org/2885053002
Cr-Commit-Position: refs/heads/master@{#472199}

BUG=591099
TBR=eae@chromium.org
NOTRY=true

Review-Url: https://codereview.chromium.org/2880993003
Cr-Commit-Position: refs/heads/master@{#472198}

https://chromium.googlesource.com/angle/angle.git/+log/82830ed..995c2ed

BUG=,chromium:721783, angleproject:1593

TBR=jmadill@chromium.org

TEST=bots

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2884343002
Cr-Commit-Position: refs/heads/master@{#472197}

The investigation revealed:
- BreakpointManager does not know how to set breakpoints in the UISourceCodes
  with exact URLs. For this reason, breakpoints are removed inside workers
  in case of multilpe workers launched with the same URL.
- node.js reports file URLs which clash with the file URLs from persistence.
  This provokes the same illicit behavior in BreakpointManager.
- the same would happen with file-url web sites with persistence folder

This behavior is a conceptual issue of BreakpointManager, which is relied
upon in different parts of BreakpointManager.

Given the poor state of the things in the BreakpointManager, this patch
suggests a trade-off solution for the issue:
- it is localized to persistence only
- it is localized to the case of URL collisition between FileSystem and
  network URLs

Unfortunately, the attempts to write a test with workers failed due to the
BreakpointManager's poor handling of breakpoints in workers.

BUG=722636
R=dgozman

Review-Url: https://codereview.chromium.org/2886643002
Cr-Commit-Position: refs/heads/master@{#472196}

BUG=722827
TBR=sdefresne@chromium.org for ios/

Review-Url: https://codereview.chromium.org/2886803002
Cr-Commit-Position: refs/heads/master@{#472195}

There are no more callers of this method.

BUG=667892

Review-Url: https://codereview.chromium.org/2882523002
Cr-Commit-Position: refs/heads/master@{#472194}

This disables automatic timeline recording on page reload.
To record a reload users now should either use reload button on Timelin
panel or use a dedicated shortcut (Ctrl+Shift+E).

BUG=708099

Review-Url: https://codereview.chromium.org/2883143003
Cr-Commit-Position: refs/heads/master@{#472193}

Before this CL, we asserted that ModuleMap::GetFetchedModuleScript always queried load completed module scripts, but it was not always the case.
This CL changes the member function to return nullptr if the entry doesn't exist or the entry is being fetched.

This CL fixes a crash that happens when we early-exit module script graph fetch by a sub-graph instantiation failed, but there are still partial subgraph in flight.

BUG=718442,594639

Review-Url: https://codereview.chromium.org/2860993002
Cr-Commit-Position: refs/heads/master@{#472192}

Add one UMA for CrossFadeAnimation smoothness, used in maximize windows,
so that we can evaluate how much improvement we can achieve by forcing
use of render surface.

BUG=708513
TEST=has histogram at local device.

Review-Url: https://codereview.chromium.org/2869263002
Cr-Commit-Position: refs/heads/master@{#472191}

This patch prunes a call site of the deprecated function to improve code
health. The change is safe because it is equivalent to hoisting the layout
update from the deprecated function.

BUG=698633
TEST=n/a; no behavioral changes

Review-Url: https://codereview.chromium.org/2883083003
Cr-Commit-Position: refs/heads/master@{#472190}

BUG=722943
TBR=gab
NOTRY=true

Review-Url: https://codereview.chromium.org/2887893002
Cr-Commit-Position: refs/heads/master@{#472189}

BUG=b:38123645
TEST=First run tutorial is started after voice interaction
  OOBE completed broadcast.

Review-Url: https://codereview.chromium.org/2881283002
Cr-Commit-Position: refs/heads/master@{#472188}

BUG=717085
TBR=pfeldman

Review-Url: https://codereview.chromium.org/2884133002
Cr-Commit-Position: refs/heads/master@{#472187}

Avoids allocating the CSSGlobalRuleSet on html import StyleEngines.

Review-Url: https://codereview.chromium.org/2883003002
Cr-Commit-Position: refs/heads/master@{#472186}

Review-Url: https://codereview.chromium.org/2861373004
Cr-Commit-Position: refs/heads/master@{#472185}

BUG=716658,716656
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2888493002
Cr-Commit-Position: refs/heads/master@{#472184}

The DisplayReconfigCallback used by GpuDataManagerImplPrivate was
checking if GPUInfo::gpu's vendor or device had changed. This will
never happen, because GPUInfo::gpu is the primary GPU, not the
active GPU.

Add a GPUInfo::active_gpu() function to return the active GPU, and
make DisplayReconfigCallback use this.

Add more instrumentation to ImageTransportSuraceOverlay's
SwapBuffersInternal function, to allow for a more transparent view
into this function.

BUG=712090
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2878163002
Cr-Commit-Position: refs/heads/master@{#472183}

This patch adds a stable id into PaintImage. This is stable in the sense
that if the blink::Image that generated the paint image hasn't changed,
then the id would be the same. It's used in a case where either we have
a partially loaded image that finishes loading or an animated image
ticks to a new frame. In those cases, the SkImage unique id is different
since the contents of the image are different. However, the stable id
should remain the same to indicate that it's a part of logically the
same image.

The primary use for the stable id would be to determine whether the
compositor is allowed to checker the image. Specifically, if we have
ever displayed an image stable id X, then all other images with the
same stable id X are not allowed to be checker imaged.

R=khushalsagar@chromium.org
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2873003002
Cr-Original-Commit-Position: refs/heads/master@{#471351}
Committed: https://chromium.googlesource.com/chromium/src/+/ae192fde12e7fba4ac2d8f027037d10c3d0f22a4
Review-Url: https://codereview.chromium.org/2873003002
Cr-Commit-Position: refs/heads/master@{#472182}

Since this benchmark is not scheduled on perf waterfall, we move it to the contrib/
directory & set memory-infra as OWNERS.

BUG=720257

Review-Url: https://codereview.chromium.org/2874033005
Cr-Commit-Position: refs/heads/master@{#472181}

Previously, entering/exiting fullscreen whilst the media controls are
hidden due to inactivity would cause the media controls to briefly show
then almost immediately hide again, resulting in an ugly flicker.

They were shown by HTMLMediaElement::UpdateControlsVisibility (called
from HTMLMediaElement::Did{Enter,Exit}Fullscreen) and then hidden again
by MediaControlsImpl::OnTimeUpdate since ShouldHideMediaControls()
returns true.

(This was often masked by the fact that clicking the fullscreen button
with a mouse results in the mouse hovering over the media controls at
the start of the fullscreen transition, in which case it's ok to show
the media controls. It became particularly noticeable with the recent
video-rotate-to-fullscreen feature, which enters/exits fullscreen
without interacting with the media controls, but it can be reproduced
without that feature as well).

This patch renames MediaControls::Show to MaybeShow and makes it only
show the controls if they won't soon be hidden again.

BUG=717520
NOPRESUBMIT=true

Review-Url: https://codereview.chromium.org/2873493004
Cr-Commit-Position: refs/heads/master@{#472180}

This patch is to improve performance of network panel when a domain url
is very long we crop it in order to make scrolling not lag as much.

R=pfeldman
BUG=None

Review-Url: https://codereview.chromium.org/2876623003
Cr-Commit-Position: refs/heads/master@{#472179}

Revert of [Autofill] Enable Autofill UKM by default. (patchset #4 id:60001 of https://codereview.chromium.org/2874413004/ )

Reason for revert:
Failing consistently on
chromium.memory/Linux ASan LSan Tests (1)

See crbug.com/722947

Original issue's description:
> [Autofill] Enable Autofill UKM by default.
>
> BUG=718014
>
> Review-Url: https://codereview.chromium.org/2874413004
> Cr-Commit-Position: refs/heads/master@{#472094}
> Committed: https://chromium.googlesource.com/chromium/src/+/06fc093968064421daf828caf503ef5fafb07cba

TBR=csashi@google.com,mathp@chromium.org,sebsg@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=718014

Review-Url: https://codereview.chromium.org/2888633002
Cr-Commit-Position: refs/heads/master@{#472178}

This CL replaces the use of the --cr-control-spacing with two variables
--cr--control-label-spacing and --cr-controlled-by-spacing. Those two
distances are now required (by UI) to be different values (20px and 24px
respectively). Removing the old variable and using more explicitly named
variables should help avoid confusion about where each should be used.

BUG=721230
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2886603002
Cr-Commit-Position: refs/heads/master@{#472177}

SequencedWorkerPool is being deprecated in favor of TaskScheduler.

BUG=667892
R=qinmin@chromium.org

Review-Url: https://codereview.chromium.org/2868393006
Cr-Commit-Position: refs/heads/master@{#472176}

SequencedWorkerPool is being deprecated in favor of TaskScheduler.

BUG=667892
R=waffles@chromium.org

Review-Url: https://codereview.chromium.org/2882633003
Cr-Commit-Position: refs/heads/master@{#472175}

This patch:
* Adds reschedule capability to BackgroundTask and BackgroundTaskScheduler
* Updates all implementations of BackgroundTask
* Adds capability to remove all stored tasks from shared preferences
* Adds a call to reschedule from BTSGcmTaskService.onInitializeTasks
* Adds appropriate tests

BUG=710630
R=nyquist@chromium.org

Review-Url: https://codereview.chromium.org/2814653003
Cr-Commit-Position: refs/heads/master@{#472174}

* Add a ScriptForbiddenScope to FrameLoader::Init(), since script shouldn't ever run in Init(). Simplify some callers of Init() that assume Init() might detach the frame.
* Use WebLocalFrameImpl::InitializeCoreFrame in CreateProvisional(), resolving a FIXME there.

BUG=

Review-Url: https://codereview.chromium.org/2869203002
Cr-Commit-Position: refs/heads/master@{#472173}

Stack traces are presented as DEBUG:I, but the previous logcat filter
does not include such logcat.

In this cl, we are including DEBUG:I into the logcat.

BUG=721811

Review-Url: https://codereview.chromium.org/2882793002
Cr-Commit-Position: refs/heads/master@{#472172}

The change in https://codereview.chromium.org/2743253006 stops the Mac
memory pressure monitor from checking the current pressure level
directly on each call to GetCurrentPressureLevel(). However, if the OS
sends a pressure change notification, callers of
GetCurrentPressureLevel() may receive the old pressure level. This cl
fixes the race condition by forcing an internal pressure level update
upon receiving a pressure level change notification from the OS.

R=brettw@chromium.org,mark@chromium.org
BUG=720924

Review-Url: https://codereview.chromium.org/2877573002
Cr-Commit-Position: refs/heads/master@{#472171}

This CL renames the 'name' field to 'mojom_name' on various classes, and introduces Stylizer to map 'mojom_name' to 'name' which now means the stylized name.

The default Stylizer just sets 'name' to the same value as 'mojom_name'. A follow-up CL will add a JavaScriptStylizer.

BUG=718614

Review-Url: https://codereview.chromium.org/2888503002
Cr-Commit-Position: refs/heads/master@{#472170}

This fixes an ODR violation (multiple definitions of a class named
download::MockObserver), and should fix a test failure on the CFI
Linux bot.

https://build.chromium.org/p/chromium.fyi/builders/CFI%20Linux/builds/8068

R=xingliu@chromium.org

Review-Url: https://codereview.chromium.org/2888433004
Cr-Commit-Position: refs/heads/master@{#472169}

BUG=706118

Review-Url: https://codereview.chromium.org/2877073002
Cr-Commit-Position: refs/heads/master@{#472168}

This CL is a follow up to https://crrev.com/2877273002, where tests
could not be renamed due to some platforms' file systems being case
insensitive.

Here we Rename a few tests:

- getphotocapabilities.html => getPhotoCapabilities.html
- setoptions.html => setOptions.html
- takephoto-with-photosettings.html => takePhoto-with-PhotoSettings.html
- takephoto.html => takePhoto.html

**No new code.**

TBR=emircan@chromium.org

Bug: 718765
Change-Id: I4dc03a95a86a43a4422f3d2311ed163322d656ba
Reviewed-on: https://chromium-review.googlesource.com/506506Reviewed-by: Miguel Casas <mcasas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#472167}

BUG=701939

Review-Url: https://codereview.chromium.org/2872933003
Cr-Commit-Position: refs/heads/master@{#472166}

It had been flaky after https://codereview.chromium.org/2872423002/
which defined PaintInvalidationReason::kImage for invalidations
caused by image change. The test sometimes produced "full" and
sometimes produced "image" as the invalidation reason of the image.
The flakiness might be because of arbitrary timing of image animation.

- Add image onload handler to start the test;
- Advance image animation twice to avoid flakiness.
- Convert it into a ref test.
- Increase the size of the image to avoid the layout from being
  affected by different font sizes on different platforms.

BUG=722131

Review-Url: https://codereview.chromium.org/2882873002
Cr-Commit-Position: refs/heads/master@{#472165}

Mocks are here https://gallery.googleplex.com/projects/MCHbtQVoQ2HCZfSHzqDNTRuU/files/MCFvOc4T9u7Awmru3H3e7QDjwMSJbxAVNpw. (page 11/12)

Moves setup pin error/warning message inside pin keyboard element.Have not received positioning/sizing details. But this CL covers the functionality.

Screenshots:
https://screenshot.googleplex.com/qsWXoWteyDr

TEST=browser_tests --gtest_filter="CrSettingsPeoplePageSetupPinDialogTest.*
BUG=703998
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2848483002
Cr-Commit-Position: refs/heads/master@{#472164}

Removes problematic usage of iron-a11y-keys and replaces it with a
simpler on-keypress handler.

Also fixes a bug.

BUG=721256
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation
TEST=MANUAL

Review-Url: https://codereview.chromium.org/2882143003
Cr-Commit-Position: refs/heads/master@{#472163}