GitLab

Revert of Adding Site Isolation Win bot filtering for WindowOpenPanelTest.WindowOpenFocus. (patchset #1 id:1 of https://codereview.chromium.org/2269523003/ )

Reason for revert:
The filter file is apparently not needed anymore: despite the fact that the test got renamed in r413759, we are not seeing failures in the WindowOpenFocus test in the latest builds on Site Isolation Win bot.

Original issue's description:
> Adding Site Isolation Win bot filtering for WindowOpenPanelTest.WindowOpenFocus.
>
> BUG=639510
>
> Committed: https://crrev.com/a98073ac85514395939268f9d71a5d609867d450
> Cr-Commit-Position: refs/heads/master@{#413773}

TBR=creis@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=639510

Review-Url: https://codereview.chromium.org/2287743002
Cr-Commit-Position: refs/heads/master@{#414857}

BUG=636109

Review-Url: https://codereview.chromium.org/2281963002
Cr-Commit-Position: refs/heads/master@{#414856}

This CL is a follow-up to CL 2273083002 where the <h2> sub-labels were
updated to the current mocks. The displays sub-page (chromeos only) also
needed to be updated.

BUG=638461
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2283933002
Cr-Commit-Position: refs/heads/master@{#414855}

BUG=none

Review-Url: https://codereview.chromium.org/2283963002
Cr-Commit-Position: refs/heads/master@{#414854}

Revert of exo: Disable RGBX fullscreen overlays. (patchset #1 id:1 of https://codereview.chromium.org/2278803003/ )

Reason for revert:
crrev.com/2281033003 makes it so the compositor sends only one plane to Ozone.

In this way Ozone will accept an RGBX fullscreen plane.

Original issue's description:
> exo: Disable RGBX fullscreen overlays.
>
> Fullscreen overlays have been enabled on minnie.
> Unfortunately there are some issues in HardwareDisplayPlaneManager::
> AssignOverlayPlanes with RGBX buffers.
>
> This patch disables the overlay optimization for those buffers
> while we work on a fix.
>
> BUG=b/29430506
>
> Committed: https://crrev.com/18d1778af83ec9877cdf33401eddf4b082f766ef
> Cr-Commit-Position: refs/heads/master@{#414281}

TBR=reveman@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=b/29430506

Review-Url: https://codereview.chromium.org/2287773002
Cr-Commit-Position: refs/heads/master@{#414853}

Review-Url: https://codereview.chromium.org/2282963002
Cr-Commit-Position: refs/heads/master@{#414852}

Plugins (soon to be called Flash) needs to support 3 modes in the
default content setting.

This patch updates the MD SiteSettingsHandler to return the default
content setting as a string (instead of just a boolean).

BUG=622922
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2282603002
Cr-Commit-Position: refs/heads/master@{#414851}

BUG=none
TEST=none
TBR=pastarmovj

Review-Url: https://codereview.chromium.org/2282073002
Cr-Commit-Position: refs/heads/master@{#414850}

WindowAndroid object is cached in PermissionInfobars, and the object is
not updated since their creation. Yet after reparenting, WindowAndroid
has changed. This CL fixes this issue by adding a new public call,
onTabReparented() to Infobar.java

BUG=632477

Review-Url: https://codereview.chromium.org/2282953003
Cr-Commit-Position: refs/heads/master@{#414849}

When there's only one button, we show the gear icon with text. Otherwise, just show the icon directly.

BUG=641300
TEST=Verified on local build.

Review-Url: https://codereview.chromium.org/2287453002
Cr-Commit-Position: refs/heads/master@{#414848}

This adds the experiment "PrerenderSilence" to disable all performance-related
uses of prerendering. The group name encodes a timestamp. After reaching this
timesamp, prerendering should continue as normal.

The expiration timestamp should be in GMT, since all Field Trial logic both on
the server and on the client are also in GMT.

Note: This makes FieldTrial metrics incorrect for the days shortly after the
expiration timestamp because the experiment is sticky to the session.

BUG=640592

Review-Url: https://codereview.chromium.org/2275933002
Cr-Commit-Position: refs/heads/master@{#414847}

Using StringView for inbound parameters and StringBuffer for outbound helps to avoid copies in important places (mostly when passing large protocol messages around).

This opens possibility to:
- use WTF::String in core/inspector instead of String16;
- make String16 an alias to std::basic_string<uint16_t> to leverage all existing
optimizations.

One possible optimization is to migrate search utility (in V8StringUtil) to
StringView and avoid any copies of content there.

BUG=637032

Review-Url: https://codereview.chromium.org/2260233002
Cr-Commit-Position: refs/heads/master@{#414846}

Generalizing updateInstancesAnimatedAttribute{,NoInvalidate} to one
higher-level helper to get rid of the subtle differences brought on
by the differences in invalidation semantics.

BUG=640676

Review-Url: https://codereview.chromium.org/2280923002
Cr-Commit-Position: refs/heads/master@{#414845}

Previously it was possible for a merge to trigger a tab model
model selection change. The tab model that was selected before
the merge should still be selected after.

BUG=637064

Review-Url: https://codereview.chromium.org/2281293002
Cr-Commit-Position: refs/heads/master@{#414844}

Review-Url: https://codereview.chromium.org/2283513002
Cr-Commit-Position: refs/heads/master@{#414843}

This is a variation of r378802 which was reverted because of
a Dr Memory error (solved by using GenerateAlternatePEFileVersion
instead of GenerateSpecificPEFileVersion). r378802 also caused
problems with shortcuts crbug.com/592040 (solved by not letting
the deletion of old files cause a rollback).

With this CL, the installer deletes all files that belong to old
versions of Chrome that are not in use after an update.

Eventually, the DeleteOldVersions() will be invoked again to delete
old versions that were in use at time of the update.

BUG=451546
TEST=
-- Not-in-use --
1. Run mini_installer.exe --chrome --multi-install --verbose-logging --do-not-launch-chrome
- The install directory should contain chrome.exe and a matching version directory.
- There should be no other executable or version directory in the install directory.
2. Run next_version_mini_installer.exe --multi-install --verbose-logging --do-not-launch-chrome
- The install directory should contain chrome.exe and a matching version directory. The version should be higher than at the previous step.
- There should be no other executable or version directory in the install directory.

-- In-use --
1. Run mini_installer.exe --chrome --multi-install --verbose-logging --do-not-launch-chrome
- The install directory should contain chrome.exe and a matching version directory.
- There should be no other executable or version directory in the install directory.
2. Launch Chrome.
3. Run next_version_mini_installer.exe --multi-install --verbose-logging --do-not-launch-chrome.
- The install directory should contain chrome.exe and a matching version directory + new_chrome.exe and a matching version directory.
- There should be no other executable or version directory in the install directory.

Review-Url: https://codereview.chromium.org/2273113002
Cr-Commit-Position: refs/heads/master@{#414842}

Optimizes QueryCache to only return requests if responses aren't needed. This
prevents extra resources from being wasted.

BUG=634428

Review-Url: https://codereview.chromium.org/2286833004
Cr-Commit-Position: refs/heads/master@{#414841}

Looks like explicitly not setting the layout to something with
has zero bounds for the anchor view for the power save blocker has
some effect on the benchmark.

BUG=635515
TBR=jinsukkim@chromium.org

Review-Url: https://codereview.chromium.org/2282983004
Cr-Commit-Position: refs/heads/master@{#414840}

The changes in 8f2032ae remove the code
that used the variables that were used introduced in
f52b1e62.

This change removes those variables and the associated code in the
destructor.

BUG=640275
TEST=still builds.
CQ_INCLUDE_TRYBOTS=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/2264403003
Cr-Commit-Position: refs/heads/master@{#414839}

The //chrome directory is no longer used by the iOS build so these conditionals
are all dead build code.

Review-Url: https://codereview.chromium.org/2282573002
Cr-Commit-Position: refs/heads/master@{#414838}

This binary will be used by the non-elevated It2Me process to allow the remote
user to interact with higher UIPI level windows on the desktop.

BUG=617185

Review-Url: https://codereview.chromium.org/2185693003
Cr-Commit-Position: refs/heads/master@{#414837}

The current logic is not taking into account that streams can be
disabled and is heavily tied to the assumption of max 1 audio/video
stream. This refactoring strives to preserve the original logic as
much as possible, but also disregards disabled streams when searching
for the preferred seeking stream, and generalizes that logic to
potentially handle multiple audio/video streams.

BUG=641451

Review-Url: https://codereview.chromium.org/2281843002
Cr-Commit-Position: refs/heads/master@{#414836}

Design doc: https://docs.google.com/document/d/1M669yu7nsF9Wrkm7nQFi3Pp2r-QmCMqm4K7fPPo-doA/edit#heading=h.t80gaiwoyk1m

This will replace ObjectPaintInvalidator::invalidateUsingContainer()
for SPv2.

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2
BUG=510908

Review-Url: https://codereview.chromium.org/2277443003
Cr-Commit-Position: refs/heads/master@{#414835}

Previously posting a WTF::Closure (which contains a
base::Closure) resulted in it getting wrapped in a WebTaskRunner::Task
which was then wrapped in a base::Closure!

We fix that here by posting the wrapped base::Closure directly.

We also and refactor the various blink scheduler mocks to use
either a real scheduler or a centralized FakeWebTaskRunner.

At Sami's request taskRunner has been renamed to
toSingleThreadTaskRunner.

BUG=638542

Review-Url: https://codereview.chromium.org/2266443002
Cr-Commit-Position: refs/heads/master@{#414834}

BUG=602538
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2256773005
Cr-Commit-Position: refs/heads/master@{#414833}

SyncExtensionFunctions, AsyncExtensionFunctions,
ChromeSyncExtensionFunctions, and ChromeAsyncExtensionFunctions are
deprecated.

Remove SyncExtensionFunctions from:
- tab_capture
- virtual_keyboard_private

BUG=634140

Review-Url: https://codereview.chromium.org/2272563011
Cr-Commit-Position: refs/heads/master@{#414832}

TEST=manual
BUG=641555

Review-Url: https://codereview.chromium.org/2276233002
Cr-Commit-Position: refs/heads/master@{#414831}

OverlayStrategyFullscreen used to promote a quad to an overlay
finding a suitable quad, removing it from the quad_list and
pushing the overlay candidate to the list of overlay candidates.

This behavior was relying on ozone optimization that would
replace the primary plane if there were a fullscreen overlay
on top of it.

This CLs changes OverlayStrategyFullscreen behavior making it more
aggressive.
Once a quad that can be promoted is found, quad_list is erased since
all the quads are either not visible, or covered by the fullscreen
overlay, or the fullscreen overlay itself.

It also clear the overlay_list, that contains the primary buffer,
since the promoted overlay will be replacing it, in this way
we stop relying on the ozone optimization.

BUG=b/29430506
TEST=cc_unittests --gtest_filter="*FullscreenOverlayTest*", also tried it on minnie.
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2281033003
Cr-Commit-Position: refs/heads/master@{#414830}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/470a9e84e35b..1af2c16ca8ae

$ git log 470a9e84e..1af2c16ca --date=short --no-merges --format='%ad %ae %s'
2016-08-26 chiniforooshan Update stale links in the Telemetry doc
2016-08-26 eakuefner [Catapult] Add catapult_base/catapult_base/node_modules to .gitignore
2016-08-26 benjhayden Modernize DiagnosticMap to subclass ES6 Map.

TBR=catapult-sheriff@chromium.org

Review-Url: https://codereview.chromium.org/2282063003
Cr-Commit-Position: refs/heads/master@{#414829}

This is a prepartory CL to bring compositor_frame.typemap to Blink. There is no
change in behavior except the include paths for compositor_frame.h is trimmed.

Currently, compositor_frame.h indirectly include base/trace_event/trace_event.h
via latency_info.h and resource_format.h (Note that resource_format.h indirectly
include trace_event.h via gpu_memory_buffer.h). Once the compositor_frame.h is
brought to Blink mojom file (which I'm going to do in future CLs), the
trace_event.h will have conflicting macro definition with
third_party/WebKit/Source/platform/TraceEvent.h.

By analyzing latency_info.h and resource_format.h, I find that the trace_event.h
was previously over-included. Therefore, I shifted the problematic included
file from these two header files to their respective cc files. All the rest of
changes in this CL are just results due to this shift to make Chrome compiled.

BUG=629566
TBR=dtrainor@chromium.org
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2273193002
Cr-Commit-Position: refs/heads/master@{#414828}

These methods have been marked with DISABLE_CFI_PERF
attribute in https://codereview.chromium.org/2258003002/
and don't need to be double-listed.

Also, fix a couple of patterns with double star.

BUG=641343,641536

Review-Url: https://codereview.chromium.org/2282753003
Cr-Commit-Position: refs/heads/master@{#414827}

EphemeralRange is supposed to operate on the same version of the tree
it was created. Unfortunately, the updateStyleAndLayout call in
plainText was causing a DOM mutation in some cases, thus
invalidating the range. Let's not do that.

BUG=639728
R=esprehn

Review-Url: https://codereview.chromium.org/2272793002
Cr-Commit-Position: refs/heads/master@{#414826}

The Mac IxD doesn't use them. Instead, buttons have a separate pressed state which is defined in the Harmony spec.

BUG=640032

Review-Url: https://codereview.chromium.org/2270933002
Cr-Commit-Position: refs/heads/master@{#414825}

Sheriffs and general bug fixers often find this logspam confusing,
thinking it relevant to the problem at hand. However it's rare that the
actual perpetrator sees it when adding the problem. So keep the blocking
behaviour, but remove the log.

R=thakis@chromium.org
BUG=258201

Review-Url: https://codereview.chromium.org/2286943002
Cr-Commit-Position: refs/heads/master@{#414824}

There is one blocking bug that prevents this from being added, but this addresses the other issues.

Aura and Ash hit a "gn check" worst-case which requires lots of //nogncheck annotations.

This also separates out browser.h includes into !OS_ANDROID blocks (previously these includes were unconditional but "Browser" doesn't exist on Android).

R=sky
TBR=dbeam for webui/*

Review-Url: https://codereview.chromium.org/2272173002
Cr-Commit-Position: refs/heads/master@{#414823}

This makes sure we have a placeholder bitmap when we resume
and don't have a frame from the renderer yet.

BUG=636630

Review-Url: https://codereview.chromium.org/2186453004
Cr-Commit-Position: refs/heads/master@{#414822}

InputConnection#getSelectedText() should return null when the selection
is empty. ThreadedInputConnection was not doing this correctly, and confused
some OEM keyboards.

BUG=638201

Review-Url: https://codereview.chromium.org/2279293002
Cr-Commit-Position: refs/heads/master@{#414821}

Reland of move content::NotificationPermissionDispatcher. (patchset #1 id:1 of https://codereview.chromium.org/2258673002/ )

Reason for revert:
Fixed the leak by adding a connection error handler that resets m_permissionService, breaking the circular reference between NotificationManager and pending Mojo callbacks.

Original issue's description:
> Revert of Remove content::NotificationPermissionDispatcher. (patchset #5 id:80001 of https://codereview.chromium.org/2244913002/ )
>
> Reason for revert:
> Causing failure on Builder WebKit Linux Leak
>
> https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux%20Leak/builds/22036
>
> request-permission-detached-context.html
>
> 15:25:46.371 7635 worker/0 http/tests/notifications/request-permission-detached-context.html leaked
> 15:25:46.371 7635   Xlib:  extension "RANDR" missing on display ":9".
> 15:25:46.371 7635   Xlib:  extension "RANDR" missing on display ":9".
> 15:25:46.374 17565 [4/95] http/tests/notifications/request-permission-detached-context.html failed unexpectedly (leak detected: ({"numberOfLiveActiveDOMObjects":[2,6],"numberOfLiveDocuments":[1,3],"numberOfLiveNodes":[4,36],"numberOfLiveResources":[0,3]}))
> 15:25:46.372 7635 worker/0 http/tests/notifications/request-permission-detached-context.html failed:
> 15:25:46.372 7635 worker/0  leak detected: ({"numberOfLiveActiveDOMObjects":[2,6],"numberOfLiveDocuments":[1,3],"numberOfLiveNodes":[4,36],"numberOfLiveResources":[0,3]})
>
> Original issue's description:
> > Remove content::NotificationPermissionDispatcher.
> >
> > blink::Notification can make calls to the Mojo PermissionService itself.
> >
> > BUG=561879
> >
> > Committed: https://crrev.com/be876eec511985fc0f476628b7b43dc951bd04c5
> > Cr-Commit-Position: refs/heads/master@{#412636}
>
> TBR=esprehn@chromium.org,haraken@chromium.org,johnme@chromium.org,mvanouwerkerk@chromium.org,peter@chromium.org,reillyg@chromium.org
> # Skipping CQ checks because original CL landed less than 1 days ago.
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=561879
>
> Committed: https://crrev.com/5241216894d534ab30c763375a78ad68e0d155db
> Cr-Commit-Position: refs/heads/master@{#412678}

# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=561879

Review-Url: https://codereview.chromium.org/2258353002
Cr-Commit-Position: refs/heads/master@{#414820}

Revert "Revert of Updates the IME list when the IME has refreshed. (patchset #7 id:120001 of https://codereview.chromium.org/2271483003/ )"

This reverts commit 70178552.

Reland the cl https://codereview.chromium.org/2271483003/ and fix the crash in test. It's caused by double deleting the tray bubble view when shutting down test/os without closing the IME menu bubble, which should not happen in real production.

Fix the underlying crash problem and add new test:
ImeMenuTrayTest.QuitChromeWithMenuOpen
to cover it.

BUG=640432
TEST=Verified on local build.

Review-Url: https://codereview.chromium.org/2277753006
Cr-Commit-Position: refs/heads/master@{#414819}

Revert of Let PowerSaveBlocker manager anchor view in Java side (patchset #3 id:40001 of https://codereview.chromium.org/2259493005/ )

Reason for revert:
not needed. doesn't fix the regression.

Original issue's description:
> Let PowerSaveBlocker manager anchor view in Java side
>
> This reduces the number of native-to-Java calls by one
> with a view to mitigating the benchmark test regression
> introduced with https://crrev.com/2103243002.
> PowerSaveBlocker was the only class that created an anchor
> view during the test. Though the root cause not being
> entirely clear yet, having the view created on Java side
> can possibly help decrease the browser thread cpu consumption.
>
> BUG=635515
> TBR=tedchoc@chromium.org
>
> Committed: https://crrev.com/bc403d8872340a62c310c9665e3b3972344c8c5b
> Cr-Commit-Position: refs/heads/master@{#413900}

TBR=twellington@chromium.org,boliu@chromium.org,tedchoc@chromium.org,jinsukkim@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=635515
NOTRY=True

Review-Url: https://codereview.chromium.org/2279313005
Cr-Commit-Position: refs/heads/master@{#414818}