GitLab

The "= " was getting removed from start of calculator answers
but now it will be left in when new answer layout is enabled.

Bug: 838733
Change-Id: Ice7bf2087cc51cd052744164018ff74f4aeb68a7
Reviewed-on: https://chromium-review.googlesource.com/1141126
Commit-Queue: Orin Jaworski <orinj@chromium.org>
Reviewed-by: Justin Donnelly <jdonnelly@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576109}

The limit of this implementation is unintentional launch, e.g user click
some notification and go to external activity and then go back, it will
also be counted. Solution is create a histogram of launch type, if majority
if launches come from hitting Chrome icon, then the noise have less impact
to the accuracy of daily launch count histogram.

Bug: 806240
Change-Id: Iec367758267bba806b4ed2c34c9e030e0ec65909
Reviewed-on: https://chromium-review.googlesource.com/1103838Reviewed-by: Ted Choc <tedchoc@chromium.org>
Reviewed-by: Gayane Petrosyan <gayane@chromium.org>
Commit-Queue: Charlie Ma <charliema@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576108}

Bug: 837334
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:linux_vr;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel
Change-Id: Ie5724d2c97e663afb5b2dc0ee9ad994a81db2052
Reviewed-on: https://chromium-review.googlesource.com/1138645Reviewed-by: Christopher Grant <cjgrant@chromium.org>
Commit-Queue: Aldo Culquicondor <acondor@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576107}

Removing the isNewTabPageUsedAsHomepage from HomepageClient interface,
since it can be combined with isHomepageEnabled and was only used in a
check on native side along with isHomepageEnabled. So those two methods
are combined on Java side to make interface simpler.

Bug: 862786
Change-Id: I32f112590afe206da8d3d2264e36202d8cf2b9ed
Reviewed-on: https://chromium-review.googlesource.com/1140691Reviewed-by: Theresa <twellington@chromium.org>
Reviewed-by: Friedrich Horschig <fhorschig@chromium.org>
Commit-Queue: Yafei Duan <romax@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576106}

Only enable the toolbar button for devices with a minimum screen width
of 320dp or greater.

BUG=865035

Change-Id: I6738739990cf0d1081a703afaaa703d644925779
Reviewed-on: https://chromium-review.googlesource.com/1141992
Commit-Queue: Theresa <twellington@chromium.org>
Reviewed-by: Matthew Jones <mdjones@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576105}

Without this dependency making an SDK update containing e.g. a FIDL
generator change would not cause FIDL stubs to be regenerated.

Bug: 865059
Change-Id: If760cccfd0a2357abdd3f4f32281fb2dd79cbd43
Reviewed-on: https://chromium-review.googlesource.com/1142127
Commit-Queue: Wez <wez@chromium.org>
Reviewed-by: Sergey Ulanov <sergeyu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576104}

This reverts commit ac436aaa.

Reason for revert: 
Sorry for the revert, but this might be breaking some layout tests on 
https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/Linux%20Tests/71090 (see also a few subsequent builds)

Failing tests include:
 external/wpt/trusted-types/HTMLImageElement-src.tentative.html
 external/wpt/trusted-types/HTMLMediaElement-src.tentative.html
 external/wpt/trusted-types/HTMLSourceElement-src.tentative.html
 external/wpt/trusted-types/block-string-assignment-to-HTMLImageElement-src.tentative.html
 external/wpt/trusted-types/block-string-assignment-to-HTMLMediaElement-src.tentative.html
 external/wpt/trusted-types/block-string-assignment-to-HTMLSourceElement-src.tentative.html

Sample failure output (from external/wpt/trusted-types/HTMLImageElement-src.tentative.html, others are similar):

This is a testharness.js-based test.
FAIL src = TrustedURL.create(). URLS is not defined
FAIL src = TrustedURL.unsafelyCreate(). URLS is not defined
Harness: the test ran to completion.



Original change's description:
> Adjusted references in tests to match new helper.sub.js file
> 
> helper.js changed to helper.sub.js.
> URLS.safe modified to work with wpt.
> 
> Bug: 739170
> Change-Id: I89e0efdb07c94e78be826c8807d0577ede8a97ec
> Reviewed-on: https://chromium-review.googlesource.com/1141725
> Reviewed-by: Mike West <mkwst@chromium.org>
> Commit-Queue: Maja Kabus <kabusm@google.com>
> Cr-Commit-Position: refs/heads/master@{#576037}

TBR=vogelheim@chromium.org,mkwst@chromium.org,kabusm@google.com

Change-Id: Ieaa3bba36a970c7aa7547939c8f97a57a1b32544
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 739170
Reviewed-on: https://chromium-review.googlesource.com/1142165Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Commit-Queue: Alex Moshchuk <alexmos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576103}

Bug: 864611
Change-Id: I15fe4879139e6b4500cc9dc67633cc2ddc3eed0f
Reviewed-on: https://chromium-review.googlesource.com/1141663
Commit-Queue: Ned Nguyen <nednguyen@google.com>
Reviewed-by: Nodir Turakulov <nodir@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576102}

FeatureUtilities#isBottomToolbarEnabled is agnostic to whether Chrome
is in an activity that supports Duet. This patch moves the logic for
whether to remove the menu button from the top toolbar to
ToolbarPhone, where Duet is always supported.

Bug: 865061
Change-Id: Icfceddf92d04722bb9e8d20266e671642c50033d
Reviewed-on: https://chromium-review.googlesource.com/1142137Reviewed-by: Ted Choc <tedchoc@chromium.org>
Commit-Queue: Matthew Jones <mdjones@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576101}

Change-Id: I4facf3700f274141640707b18c97a8db9fd7ef0c
Bug: 865007
Reviewed-on: https://chromium-review.googlesource.com/1141280Reviewed-by: Julian Pastarmov <pastarmovj@chromium.org>
Commit-Queue: Georges Khalil <georgesak@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576100}

Replace bellow codes with std::move in chrome/browser/net
      const base::Closure callback = callback_;
      callback_.Reset();
      callback.Run();

BUG=807724

Change-Id: I008cf9af9785272ddceb18fff4feba13e62fdac8
Reviewed-on: https://chromium-review.googlesource.com/1134715
Commit-Queue: Matt Menke <mmenke@chromium.org>
Reviewed-by: Matt Menke <mmenke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576099}

To keep track of WebUSB state, use the same backend as the WebUSB tab-strip
icon.

Bug: 863922
Change-Id: I89e346c3a310cfb76618a7492e2d89821c758106
Reviewed-on: https://chromium-review.googlesource.com/1140281Reviewed-by: Sébastien Marchand <sebmarchand@chromium.org>
Reviewed-by: Chris Hamilton <chrisha@chromium.org>
Commit-Queue: Nicolas Ouellet-Payeur <nicolaso@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576098}

This was failing on Nautilus. I am not sure why it worked fine with
other devices, but I tested this manually.

Test: manual
Bug: b/111137179
Change-Id: I8d457f7180b5b486d86226960913b20836849f70
Reviewed-on: https://chromium-review.googlesource.com/1139227Reviewed-by: Mitsuru Oshima <oshima@chromium.org>
Commit-Queue: Sammie Quon <sammiequon@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576097}

Bug: 836246
Change-Id: I991c886d8eeec65277c38d24f6a48b7e2ba35c98
Reviewed-on: https://chromium-review.googlesource.com/1135876
Commit-Queue: Patricia Hongo <phongo@google.com>
Reviewed-by: Yury Khmel <khmel@chromium.org>
Reviewed-by: Luis Hector Chavez <lhchavez@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576096}

This change moves the frame check inside the reader check to address
cases when GamepadSharedMemoryReader::Start() is called again after
Stop(), but the frame is null. Since the frame parameter is only used
when creating a new instance of GamepadSharedMemoryReader, the check was
moved inside the corresponding condition.

Bug: 864557
Change-Id: I7b32e82f5c79af2007b211cf80cff43ff33be2c9
Reviewed-on: https://chromium-review.googlesource.com/1141156Reviewed-by: Matt Reynolds <mattreynolds@chromium.org>
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Commit-Queue: Oksana Zhuravlova <oksamyt@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576095}

With this CL, the password accessory controller caches credentials by
origin to deliver the correct suggestions in a Tab, esp. when the focus
changes to another frame with a different origin.

Bug: 854152
Change-Id: I43cd3aa6a86e310414dec1ae43169338d5b1a245
Reviewed-on: https://chromium-review.googlesource.com/1126765
Commit-Queue: Friedrich Horschig <fhorschig@chromium.org>
Reviewed-by: Vasilii Sukhanov <vasilii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576094}

Large refactor of the XR test frameworks and utility classes. Main
changes:

1. Restructure framework hierarchy so all functionality in
   XyzTestFramework makes sense for feature Xyz.
2. Merge all utility methods that make use of WebContents into the
   frameworks.
3. Split VR, XR, and AR build dependencies.
4. Add non-static versions of static framework methods that use
   WebContents since they were almost exclusively used with the result
   of getFirstTabWebContents.

Bug: 863256
Change-Id: Ib984e1c3cb0b51dcd2424734cbcfafa6e51ac3d1
Reviewed-on: https://chromium-review.googlesource.com/1141150Reviewed-by: Michael Thiessen <mthiesse@chromium.org>
Commit-Queue: Brian Sheedy <bsheedy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576093}

This should fix the x86 downstream bots that are currently blocking
Canary release.

Bug: 864924
Change-Id: I0929373233e58897f91536a5e60813bf3db5e2f7
Reviewed-on: https://chromium-review.googlesource.com/1142126Reviewed-by: Michael Thiessen <mthiesse@chromium.org>
Reviewed-by: Ted Choc <tedchoc@chromium.org>
Commit-Queue: Tibor Goldschwendt <tiborg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576092}

This makes more sense to describe the actual state of the SyncService (rather
than the reason for it). Presence or absence of consent isn't part of the
SyncService state.
It'll also make it easier to decouple Sync-the-machinery from
Sync-the-feature, since the concept of consent doesn't apply for the
machinery.

In practice, WAITING_FOR_CONSENT and PENDING_DESIRED_CONFIGURATION are almost
equivalent, but they are defined in different terms.

Bug: 839834, 856179
Change-Id: I5fc20a1359bd302cae233a5bce9b25bda2681e2b
Reviewed-on: https://chromium-review.googlesource.com/1141804
Commit-Queue: Marc Treib <treib@chromium.org>
Reviewed-by: Mikel Astiz <mastiz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576091}

This CL adds another queue type (resource loading) that
FrameTaskQueueController will be responsible for controlling.
FrameTaskQueueController provides methods for creating these queues and
accessing the related voters.

Bug: 859963
Change-Id: I61b10986a68f87b1c4428b1b3926ecd821ee36fa
Reviewed-on: https://chromium-review.googlesource.com/1137099
Commit-Queue: Scott Haseley <shaseley@google.com>
Reviewed-by: Alexander Timin <altimin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576090}

Generation of IETF QUIC ACK frames and calculation of the length of same
differed, due to missing "-1" in the gap-block and ack-block sizes.
This was a problem when the decrement caused the length to cross a VarInt62
encoding size boundary, causing the calculated-length to be different from
the actual, serialized, length.

In this case, the gap block size was 64 in the length calculation method, 63
in the actual serialization. 64 is encoded in two bytes, 63 in one, leading
to the length discrepancy.  Unless the length boundary was crossed, this did
not cause problems in that the serialized value (63) was the correct one.

This was detected in the Chromium clusterfuzz tests and reported as Chromium
issue 859949:
https://bugs.chromium.org/p/chromium/issues/detail?id=859949

Merge internal change: 204947824

R=rch@chromium.org

Change-Id: I6578162b274f8474ca0ef7979a1308afd51cd295
Reviewed-on: https://chromium-review.googlesource.com/1140838Reviewed-by: Ryan Hamilton <rch@chromium.org>
Commit-Queue: Frank Kastenholz <fkastenholz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576089}

Changes in sandbox behavior can have couple-pixel adjustments to controls
drawn with Cocoa, like scrollbars. This moves some affected layout test
baselines from mac10.12 to mac10.10 as a result. The baseline search path
for macOS 10.10 is 10.10->10.11->10.12->mac->generic. Enabling the V2
sandbox means that mac10.11 and 10.12 now behave like "mac", and now
only 10.10 has a differing result for these tests.

Bug: 749839, 689306
Change-Id: I9133a74c39fe995282daf950efe6d8b0fbcb2b4f
Reviewed-on: https://chromium-review.googlesource.com/1132398Reviewed-by: Avi Drissman <avi@chromium.org>
Commit-Queue: Robert Sesek <rsesek@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576088}

Hash: d1cae24fda4182406e02d3f4df6309d48878fc50

Bug: 772558
Change-Id: Ia7db6d3088e573562aa72172f52fda7be58cc17d
Reviewed-on: https://chromium-review.googlesource.com/1141325Reviewed-by: Pavel Feldman <pfeldman@chromium.org>
Commit-Queue: Paul Irish <paulirish@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576087}

Correct reload to correctly reload header and footer, and then
reload section instead of entire collection view.

Bug: 854173
Cq-Include-Trybots: luci.chromium.try:ios-simulator-full-configs;master.tryserver.chromium.mac:ios-simulator-cronet
Change-Id: I36931ebd69525b85936713de0b48bf647a647c45
Reviewed-on: https://chromium-review.googlesource.com/1141279Reviewed-by: Gauthier Ambard <gambard@chromium.org>
Commit-Queue: Justin Cohen <justincohen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576086}

This reverts commit c355e2e1.

Reason for revert: The tests are still flaky
http://build.chromium.org/p/chromium.memory/builders/Linux%20ChromiumOS%20MSan%20Tests/builds/7968


Original change's description:
> Fix an element value check in PasswordManager interactive tests.
>
> Bug: 849582
> Change-Id: I6e75d9a62b4b58569c2c29b913c0a6a0cab0bf83
> Reviewed-on: https://chromium-review.googlesource.com/1139057
> Reviewed-by: Vasilii Sukhanov <vasilii@chromium.org>
> Commit-Queue: Alexey Khoroshilov <sense@yandex-team.ru>
> Cr-Commit-Position: refs/heads/master@{#575584}

TBR=vasilii@chromium.org,dvadym@chromium.org,sense@yandex-team.ru

# Not skipping CQ checks because original CL landed > 1 day ago.

Bug: 849582
Change-Id: I4d363b0dfefa99256d50cae5fd51102204f7610a
Reviewed-on: https://chromium-review.googlesource.com/1142084
Commit-Queue: Christian Dullweber <dullweber@chromium.org>
Reviewed-by: Christian Dullweber <dullweber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576085}

BUG: 843236
Change-Id: I210265af65409181b095bbd6a00da530759fead4
Reviewed-on: https://chromium-review.googlesource.com/1140922Reviewed-by: Rohit Rao <rohitrao@chromium.org>
Reviewed-by: Olivier Robin <olivierrobin@chromium.org>
Reviewed-by: John Budorick <jbudorick@chromium.org>
Commit-Queue: Rohit Rao <rohitrao@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576084}

This also moves the official build logic to the base class.

Change-Id: I8369bf832e5f9223a610cb04e18965ed9ca599a1
Bug: 812641
Reviewed-on: https://chromium-review.googlesource.com/1140372
Commit-Queue: Georges Khalil <georgesak@chromium.org>
Reviewed-by: Julian Pastarmov <pastarmovj@chromium.org>
Reviewed-by: Roger Tawa <rogerta@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576083}

The new FormData->PasswordForm parser did not check fields for being
CVC fields (except for autocomplete attributes). Ideally, it would not
need to, because it could rely on server hints, but the hints are not
ready yet, so this CL introduces a short-term ability for the new
parser to understand when a field name sounds like that of a CVC
field.

While in the particular case of the linked bug (on
http://www.shoedazzle.com/), this will break the previously correct
behaviour of the parser, the goal is to minimize the difference in
behaviour against the old parser, so as not to introduce an unknown
amount of new breakages.

Bug: 860700
Change-Id: I4595dbf01142e9f26f7264a1d62cde4cf5e4c0c0
Reviewed-on: https://chromium-review.googlesource.com/1141958Reviewed-by: Vadym Doroshenko <dvadym@chromium.org>
Commit-Queue: Vaclav Brozek <vabr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576082}

Bug:863767
Cq-Include-Trybots: master.tryserver.chromium.perf:obbs_fyi
Change-Id: Id2786b7907c9fd88141391a81c544747843bc017

NOTRY=true  # test by PRESUBMIT

Change-Id: Id2786b7907c9fd88141391a81c544747843bc017
Reviewed-on: https://chromium-review.googlesource.com/1141928Reviewed-by: Juan Antonio Navarro Pérez <perezju@chromium.org>
Commit-Queue: Ned Nguyen <nednguyen@google.com>
Cr-Commit-Position: refs/heads/master@{#576081}

For the following 2 tests, their test blocks render
correctly, but fail because its LayoutNG uses different
font for spaces in the ref block. LayoutNG segments whole
paragraph, and that fonts for the space characters may be
affected by surrounding characters.

  css2.1/t1202-counter-04-b.html
  css2.1/t1202-counters-04-b.html

TBR=eae@chromium.org
NOTRY=true

Bug: 591099
Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng
Change-Id: I8f2bdc838bf00355cbed18786080f727c41eea66
Reviewed-on: https://chromium-review.googlesource.com/1142046
Commit-Queue: Koji Ishii <kojii@chromium.org>
Reviewed-by: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576080}

This patch adjusts the logic for UI thread priority:

1) The priority is only increased to DISPLAY unless it was lower to
   begin with. Recent versions of Android (O+) automatically set the top
   activity's UI thread's niceness to TOP_APP_PRIORITY_BOOST (-10), so
   by setting the priority to DISPLAY (-4) we were in fact lowering the
   priority.

2) The priority is increased earlier in the start-up flow to minimize
   the chances of descheduling.

We also add a field trial for the first change to evaluate its impact.

BUG=841804
TEST=Manual inspection

Change-Id: I3ec70042be86f2b88bb1ae3a2dd8cba7d87eb807
Reviewed-on: https://chromium-review.googlesource.com/1136648
Commit-Queue: Sami Kyöstilä <skyostil@chromium.org>
Reviewed-by: Camille Lamy <clamy@chromium.org>
Reviewed-by: Gabriel Charette <gab@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576079}

* This also fixes the IsSelecting so that it returns false
during shutdown.

BUG=834400
TEST=Covered by unit test

Change-Id: Iac0cb8a3bdf03609727c31c51000c52ececfc66a
Reviewed-on: https://chromium-review.googlesource.com/1140749
Commit-Queue: Mitsuru Oshima <oshima@chromium.org>
Reviewed-by: Sammie Quon <sammiequon@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576078}

Change-Id: I43482a641c1b59ab2b88f3bce271ba09aa75b8c4
Reviewed-on: https://chromium-review.googlesource.com/1141738Reviewed-by: Evan Stade <estade@chromium.org>
Commit-Queue: Dominic Battré <battre@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576077}

The pref kUrlKeyedAnonymizedDataCollectionEnabled can now be
controlled by enterprise policy.

Bug: 841832
Change-Id: I24dd2558e7c4762a314f0c390b0cb2e331cf03b7
Reviewed-on: https://chromium-review.googlesource.com/1140716
Commit-Queue: Thomas Tangl <tangltom@chromium.org>
Reviewed-by: Julian Pastarmov <pastarmovj@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576076}

This fixes a fuzzer OOM. 256 characters ought to be enough for anyone.

Bug: 863775
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel
Change-Id: I0103a2ab95aa6d87375dbc055d6a0f517c726799
Reviewed-on: https://chromium-review.googlesource.com/1140774Reviewed-by: Antoine Labour <piman@chromium.org>
Commit-Queue: Jonathan Backer <backer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576075}

Following bot results are included.
8062 8064 8068 8070

TBR=eae@chromium.org, mstensho@chromium.org
NOTRY=true

Bug: 591099
Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng
Change-Id: I3220b03ef8d221246112a1c395ea179156bd27bb
Reviewed-on: https://chromium-review.googlesource.com/1141558
Commit-Queue: Koji Ishii <kojii@chromium.org>
Reviewed-by: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576074}

Bug: 831648
Change-Id: I37ce716d7529a9330ff8fa566c1b5ab686bc78d4
Reviewed-on: https://chromium-review.googlesource.com/1128382Reviewed-by: Robert Sesek <rsesek@chromium.org>
Reviewed-by: Alexei Svitkine <asvitkine@chromium.org>
Reviewed-by: Filip Gorski <fgorski@chromium.org>
Reviewed-by: Patrick Noland <pnoland@chromium.org>
Commit-Queue: Sky Malice <skym@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576073}

This reverts commit 8b20bfaf.

Reason for revert: This should only apply on Chrome OS, see follow-up CL:1141742 .

Original change's description:
> Introduce command-line switch to force-enable developer tools
> 
> Since CL:1125855, the new default is that policy-installed extensions
> can not be inspected by developer tools.
> However, some autotests rely on inspecting policy-installed extensions
> and it is infeasible to change the policy the device sees there.
> Introduce the "--force-devtools-available" command-line switch which
> enforces developer tools availability.
> 
> Bug: 838146
> Test: browser_tests --gtest_filter=*DevTools*
> Change-Id: I666d9785195eff53f1a1346d95177985d1573d72
> Reviewed-on: https://chromium-review.googlesource.com/1140308
> Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
> Commit-Queue: Pavol Marko <pmarko@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#575994}

TBR=dgozman@chromium.org,atwilson@chromium.org,pmarko@chromium.org,jochen@chromium.org

Change-Id: I8ca2b4920ebec5dbc811bdf05fbbf39afc94dc08
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 838146
Reviewed-on: https://chromium-review.googlesource.com/1142104Reviewed-by: Pavol Marko <pmarko@chromium.org>
Commit-Queue: Pavol Marko <pmarko@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576072}

Currently there's an inversion where active tabs in minimized windows don't get
loaded as they are created. This is an optimization in the occlusion tracking code
that doesn't calls ReloadIfNecessary unless the window hosting the active tab
actually becomes visible. Due to this inversion the TabLoader can start loading
background tabs in the minimized window before the active tab in the minimized
window is ever loaded.

Giving TabLoader responsibility to load all tabs fixes this. In the case of
active and visible tabs whose loads are initiated by the browser this is fine
because calling ReloadIfNecessary twice is effectively a nop, and the TabLoader
is already smart enough to track tabs that have started loading for external
reasons.

BUG=864725

Change-Id: I4b86a71470b2cb1a1a9106c4d4d92ee6b18c7284
Reviewed-on: https://chromium-review.googlesource.com/1141071Reviewed-by: Sébastien Marchand <sebmarchand@chromium.org>
Commit-Queue: Chris Hamilton <chrisha@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576071}

Cold mode spellchecker sometimes creates shadow-crossing checking
ranges (due to EndOfSentence() crossing shadow boundary).
SpellCheckRequest converts the checking range into a |Range|, which
may have a different or even null root editable element, and crashes
when it's null.

This patch stops the crash in such case.

Note: So far we haven't found stable and meaningful test cases with
shadow-crossing checking ranges, so we only stop the crash right
now without further fixes.

Bug: 863784
Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng
Change-Id: I27e61f8bb2921559799a986b6d99c8c3d2fb114f
Reviewed-on: https://chromium-review.googlesource.com/1141426Reviewed-by: Yoshifumi Inoue <yosin@chromium.org>
Commit-Queue: Xiaocheng Hu <xiaochengh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#576070}