GitLab

  "compositor: Tick the UI animations from cc, instead of from timer callbacks."
  "asan: Fix asan leak from r274404."

These CLs seem to break ash_unittests on win8 aura bot:
http://build.chromium.org/p/chromium.win/buildstatus?builder=Win8%20Aura&number=19390

BUG=371071
TBR=sadrul@chromium.org
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/311783002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274423 0039d316-1c4b-4281-b951-d872f2087c98

launcher inits the view manager, running the "window manager" app @ root.
window manager embeds another app.

** not tested by main chrome integration testing **

R=sky@chromium.org
TEST=mojo_shell --origin=http://127.0.0.1:4444 --disable-cache mojo:mojo_demo_launcher
http://crbug.com/365012

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=274322

Review URL: https://codereview.chromium.org/303163005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274422 0039d316-1c4b-4281-b951-d872f2087c98

Revert of Reduce plugin_metrics_provider_ usage in MetricsService (https://codereview.chromium.org/308433004/)

Reason for revert:
This CL seems to cause bot failure on CI:
http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%281%29/builds/2745

Original issue's description:
> Reduce plugin_metrics_provider_ usage in MetricsService
> 
> This CL eliminates MetricsService's call to
> PluginMetricsProvider::RecordPluginChanges (replaced by a new API on
> MetricsProvider). It additionally adds a
> MetricsServicesManager::OnPluginLoadingError() API and has the plugin observer
> call that API rather than calling MetricsService directly. This change will
> enable easily moving MetricsService::LogPluginLoadingError() to
> ChromeMetricsServiceClient once the latter is the class that keeps a weak
> pointer to the plugin metrics provider.
> 
> BUG=375776
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=274403

TBR=asvitkine@chromium.org,isherman@chromium.org,jochen@chromium.org,blundell@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=375776

Review URL: https://codereview.chromium.org/314583002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274418 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274411 0039d316-1c4b-4281-b951-d872f2087c98

Make sure the initiated animations are terminated correctly before teardown.

BUG=none
TBR=sky@chromium.org

Review URL: https://codereview.chromium.org/307323002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274409 0039d316-1c4b-4281-b951-d872f2087c98

Two cases don't have breaks and fall through incorrectly. The first
may be responsible for issue http://crbug.com/370919.

BUG=370919

Review URL: https://codereview.chromium.org/307293002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274408 0039d316-1c4b-4281-b951-d872f2087c98

All licensing issues were fixed in pdfium/checklicenses last week.

BUG=374943
TBR=jam@chromium.org

Review URL: https://codereview.chromium.org/307933013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274407 0039d316-1c4b-4281-b951-d872f2087c98

The commented bug seems to be resolved, and I don't see any reason to suspect
that this would still timeout.

BUG=

Review URL: https://codereview.chromium.org/304353018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274406 0039d316-1c4b-4281-b951-d872f2087c98

BUG=365352
TEST=none
NOCHECK=true

Review URL: https://codereview.chromium.org/308073012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274405 0039d316-1c4b-4281-b951-d872f2087c98

Update the animations in the UI in response to the animation step in the compositor,
instead of from a timer callback. This should make it more difficult for rogue UI
animations to negatively impact the system too much.

BUG=371071
R=ajuma@chromium.org, piman@chromium.org, sky@chromium.org

Review URL: https://codereview.chromium.org/291843012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274404 0039d316-1c4b-4281-b951-d872f2087c98

This CL eliminates MetricsService's call to
PluginMetricsProvider::RecordPluginChanges (replaced by a new API on
MetricsProvider). It additionally adds a
MetricsServicesManager::OnPluginLoadingError() API and has the plugin observer
call that API rather than calling MetricsService directly. This change will
enable easily moving MetricsService::LogPluginLoadingError() to
ChromeMetricsServiceClient once the latter is the class that keeps a weak
pointer to the plugin metrics provider.

BUG=375776

Review URL: https://codereview.chromium.org/308433004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274403 0039d316-1c4b-4281-b951-d872f2087c98

BUG=361778
TEST=NONE
TBR=gab, bauerb

Review URL: https://codereview.chromium.org/309873002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274398 0039d316-1c4b-4281-b951-d872f2087c98

BUG=336096
TEST=Play audio.  Run trace.  Observe entries.

Review URL: https://codereview.chromium.org/309483015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274397 0039d316-1c4b-4281-b951-d872f2087c98

In a few places, the command buffer used glStencilMask directly
instead of calling through to the helper function which updated
the mask cache values. This could cause some strange behaviour
in ANGLE, such as the front and back stencil masks getting split.
The split mask values could cause unexpected GL errors during draw.

BUG=378754

Review URL: https://codereview.chromium.org/308703010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274395 0039d316-1c4b-4281-b951-d872f2087c98

Version bump in Privet info

R=vitalybuka@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/311583003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274392 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://codereview.chromium.org/302273003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274391 0039d316-1c4b-4281-b951-d872f2087c98

As telemetry result is mainly meant for monitoring regression, exposing
different types of input latency, e.g. scroll/mouse/touch latency as we
currently do, does not add more benefit to this purpose but instead
adds more confusion for people when they look at the different input
latency metrics.

Lets only report one input latency metric mean_input_event_latency,
which is derived from all input events which have event creation timestamp
and swap buffer timestamp, regardless of its event type.

BUG=378027
TEST=run telemetry smoothness tests and see only one mean_input_event_latency is
reported. unittests pass.

Review URL: https://codereview.chromium.org/300273002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274390 0039d316-1c4b-4281-b951-d872f2087c98

So the terminal will auto-linkify and it's much easier to manually point a
browser at a test server instance.

BUG=none

Review URL: https://codereview.chromium.org/312593002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274386 0039d316-1c4b-4281-b951-d872f2087c98

With this change we take into account current status of update from update service.

+jhawkins for owners

BUG=272540
R=sorin,cpu,jhawkins

Review URL: https://codereview.chromium.org/209313002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274385 0039d316-1c4b-4281-b951-d872f2087c98

syzygy/chrome_child.dll doesn't get instrumented when doing a SyzyASan official build of Chrome (because we don't want to ship an instrumented renderer to the users).

However we still produce an instrumented version of chrome_child.dll, and we put it in the syzygy/instrumented folder. This CL should add this dll to the chrome-win32.zip archive, this way we'll be able to use it in our fuzzing infrastructure.

R=robertshield@chromium.org

BUG=

Review URL: https://codereview.chromium.org/298253010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274384 0039d316-1c4b-4281-b951-d872f2087c98

In tests user wallpaper info can point to empty paths.
Use default wallpaper in this case.


BUG=371295
TEST=none

Review URL: https://codereview.chromium.org/309643008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274383 0039d316-1c4b-4281-b951-d872f2087c98

Avoid a crash in tests caused by
WeakPtr<ExtensionMessageFilter>s being used on the IO thread
while the filter is destroyed on the UI thread.

BUG=376634

Review URL: https://codereview.chromium.org/309123002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274382 0039d316-1c4b-4281-b951-d872f2087c98

R=rsleevi@chromium.org

Review URL: https://codereview.chromium.org/296513013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274378 0039d316-1c4b-4281-b951-d872f2087c98

BUG=322664

Review URL: https://codereview.chromium.org/304513007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274377 0039d316-1c4b-4281-b951-d872f2087c98

Currently, the lint warns against using the hardcoded "sdcard" strings. We
should use Environment.getExternalStorageDirectory().getPath() instead of it,
but the path could be inconsistent depending upon the state of the real or
emulated device upon which we are testing. So this CL adds @SuppressLint to
suppress the warnings.

BUG=327768

Review URL: https://codereview.chromium.org/304633003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274376 0039d316-1c4b-4281-b951-d872f2087c98

BUG=379875

Review URL: https://codereview.chromium.org/311613002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274375 0039d316-1c4b-4281-b951-d872f2087c98

Revert "Fix textfield blur, to prevent resetting the text input client when navigating between text fields."

The fix to use OnTextInputTypeChanged in place of OnFocus and OnBlur to resolve focus loss on text field navigation has introduced a number of regressions. As it turns out, the fix is no longer even necessary to address the original bug that it was intended to address (365741).  This patch is a revert of r270818.

BUG=375125,377359,378914

Review URL: https://codereview.chromium.org/302363002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274374 0039d316-1c4b-4281-b951-d872f2087c98

(Still to do: switches::kDisableCache.)

This is to make the shell more testable.

R=sky@chromium.org

Review URL: https://codereview.chromium.org/305013015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274373 0039d316-1c4b-4281-b951-d872f2087c98

The logic in BookmarkBarView::HitTestRect() does not depend on
|rect| and only has the purpose of preventing or allowing
event-targeting of itself or its descendants. For this reason,
the logic belongs in an override of
CanProcessEventsWithinSubtree() instead of HitTestRect().

BUG=374340
TEST=none

Review URL: https://codereview.chromium.org/308673011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274372 0039d316-1c4b-4281-b951-d872f2087c98

Add animation for the hiding and showing of the resize button. Add an animation
for sliding the minimize button into the position of the resize button. Delay
changing the size of FrameCaptionButtonContainerView to account for conflicting
layout changes caused by the transition to TouchView.

TEST=FrameCaptionButtonContainerViewTest
TEST=CustomFrameViewAshTest
TES=BrowserNonClientFrameViewAshTest
BUG=363717

Review URL: https://codereview.chromium.org/271913002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274371 0039d316-1c4b-4281-b951-d872f2087c98

BUG=273710

Review URL: https://codereview.chromium.org/302043005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274370 0039d316-1c4b-4281-b951-d872f2087c98

BUG=369289

Review URL: https://codereview.chromium.org/291093012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274369 0039d316-1c4b-4281-b951-d872f2087c98

and make the bg color of all bubbles match the dialog bg color

BUG=376832

Review URL: https://codereview.chromium.org/309713002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274368 0039d316-1c4b-4281-b951-d872f2087c98

Remove undefined behaviors by using offsets instead of pointers
and by checking that offset never exceeds the size of the data.

Update Native Messaging test to use Native Messaging format
instead of Pickle format which accidentally coincided before.

BUG=none

Review URL: https://codereview.chromium.org/290173008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274367 0039d316-1c4b-4281-b951-d872f2087c98

Some overrides of View::HitTestRect() just return 
false in order to force event-targeting to fail for
that view type and its descendants. Because this 
logic is unrelated to hit-testing against a rectangular
region, remove these overrides and instead override 
View::CanProcessEventsWithinSubtree() to just return 
false.

The relevant classes are:

ProfileImageView
ShelfButton::BarView
ShelfButton::IconView
PasswordRow
BoundedLabel
Label (and Link as a result)

BUG=374303
TEST=none

Review URL: https://codereview.chromium.org/287193002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274366 0039d316-1c4b-4281-b951-d872f2087c98

This creates a 'Help' submenu with 'Chrome Help Center' and
'Report an Issue...'. This menu will be consistent across platforms.

BUG=371094

Review URL: https://codereview.chromium.org/274973002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274365 0039d316-1c4b-4281-b951-d872f2087c98

Revert of Move some sync/notifier to components/invalidation (https://codereview.chromium.org/294123004/)

Reason for revert:
Passed the CQ but broke the buildbot.

Original issue's description:
> Move some sync/notifier to components/invalidation
> 
> Moves many of the files in sync/notifier to components/invalidation.
> 
> This change does not introduce any new dependencies.  The relevant
> dependency rules both before and after this change should be:
> - chrome/browser/invalidation and chrome in general depend on
>   components/invalidation.
> - components/invalidation depends on sync/notifier and sync in
>   general.
> - sync/notifier, components/invalidation, and various parts of
>   chrome all depend on sync/internal_api/public.
> 
> The eventual goal is to move all of sync/notifier into
> components/invalidation.  The invalidation-related parts of
> sync/internal_api/public should be moved to components/invalidation,
> too.  This will allow us to remove the deopendencies from
> components/invalidation to sync, and remove sync's dependencies on
> cacheinvalidation and libjingle.
> 
> This change is a regression in terms of shared library componentization.
> the files in the sync/notifier folder could be built as a shared
> library.  The files in compononents/invalidation do not support this
> yet.  The SYNC_EXPORT declarations in the moved files have been changed
> to INVALIDATION_EXPORT so as to not lose this information, but the
> macros are currently #defined to no-ops.
> 
> This change does not attempt to rename any classes or namespaces.
> Many of the files ported from sync/notifier still use the syncer
> namespace.  Some, like SyncSystemResources, still have names tied
> to their sync heritage.  This will be addressed in future CLs.
> 
> Some non-trivial or non-obvious changes include:
> - invalidator_state.h was moved to sync/internal_api/public/base so it
> could be shared by both sync/ and components/invalidation.  This should
> be fixed in a future CL.
> - FromNotifierReason was split out of invalidator_state.h and moved to
> the newly-created components/invalidator_reason_util.h
> 
> TBR=zea,rtenneti,mallinath,dcheng
> BUG=259559
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=274350

TBR=pavely@chromium.org,dcheng@chromium.org,zea@chromium.org,rtenneti@chromium.org,mallinath@chromium.org,maniscalco@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=259559

Review URL: https://codereview.chromium.org/308413002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274364 0039d316-1c4b-4281-b951-d872f2087c98

Adds about 40MB.

BUG=None
TEST=local
R=sbc@chromium.org, hinoka@chromium.org

Review URL: https://codereview.chromium.org/302083016

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274362 0039d316-1c4b-4281-b951-d872f2087c98

BUG=379943
R=thestig@chromium.org
TBR=thestig@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/306013011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274361 0039d316-1c4b-4281-b951-d872f2087c98

Background: We restart adbd between runs because we noticed that the
likelyhood of an adbd hang goes up with the length of time the process
has been running.

Killing was just a convenient way to do that, but it isn't the most robust.
The chromium.perf bots are very frequently going offline now and they
always are getting stuck at the point of the adbd restart. So I'm trying
this gentler method of restarting to see if it improves reliability.

BUG=

Review URL: https://codereview.chromium.org/304053002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274360 0039d316-1c4b-4281-b951-d872f2087c98