GitLab

Adds material-refresh to top-chrome-md under chrome://flags for
upcoming Material style changes.

Bug: chromium:822040

Change-Id: Ibc5ce5ead8462a56f04b80b637406d5726665c09
Reviewed-on: https://chromium-review.googlesource.com/961275
Commit-Queue: Peter Boström <pbos@chromium.org>
Reviewed-by: Peter Kasting <pkasting@chromium.org>
Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org>
Reviewed-by: Bret Sepulveda <bsep@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543815}

Mappy used deprecated APIs.
Updated sendRequest and onRequest to sendMessage and onMessage.

Bug: 808592
Change-Id: I52a24f12d16cfde1bab8f9729bfe996ba97a356d
Reviewed-on: https://chromium-review.googlesource.com/907214Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
Commit-Queue: Crystal Lambert <crystallambert@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543814}

See intent: https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/pfDg-VAeKWs

This makes new formatting contexts behave exactly like floats would
when they are positioned. I.e. they don't look at the shape-outside
of the float, and instead as positioned for their margin box.

I've added a .tentative test as per the intent.

This will allow us to have a (relatively) simple implementation of
css-shapes in LayoutNG as we only have to consider the shape area for
line-boxes. \o/

Bug: 820520
Change-Id: Ib07600a291eb4838675a9a1068c5fc16cac0633f
Reviewed-on: https://chromium-review.googlesource.com/739961
Commit-Queue: Ian Kilpatrick <ikilpatrick@chromium.org>
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543813}

Bug: 777879, 821878
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_mojo
Change-Id: I1bbc4db1fc665ab1844780c4d689d7a8ab0d044c
Reviewed-on: https://chromium-review.googlesource.com/967111
Commit-Queue: Marijn Kruisselbrink <mek@chromium.org>
Commit-Queue: Victor Costan <pwnall@chromium.org>
Reviewed-by: Victor Costan <pwnall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543812}

This currently seems to work, but isn't a great idea, and the code isn't
expecting another callback after that point, anyways.

Bug: 821067
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_mojo
Change-Id: I57a7cd5b71d1859c72a1dcbf76fba3f68943e45f
Reviewed-on: https://chromium-review.googlesource.com/966986
Commit-Queue: Matt Menke <mmenke@chromium.org>
Reviewed-by: Helen Li <xunjieli@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543811}

This CL moves a few mojo types out of url_loader.mojom so they can be shared
(e.g. with the new tls socket APIs).

Bug: 721401
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_mojo
Change-Id: Ie613dc08b32eb0541ef43137dee1ac4f515cc078
Reviewed-on: https://chromium-review.googlesource.com/962991Reviewed-by: Matt Menke <mmenke@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Commit-Queue: Helen Li <xunjieli@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543810}

Using wpt-import in Chromium 177e20da.
With Chromium commits locally applied on WPT:
d42a5bc8 "bluetooth: FakeBluetoothChooser stub impl."


Build: https://ci.chromium.org/buildbot/chromium.infra.cron/wpt-importer/13880

Note to sheriffs: This CL imports external tests and adds
expectations for those tests; if this CL is large and causes
a few new failures, please fix the failures by adding new
lines to TestExpectations rather than reverting. See:
https://chromium.googlesource.com/chromium/src/+/master/docs/testing/web_platform_tests.md

Directory owners for changes in this CL:
chongz@chromium.org, garykac@chromium.org:
  external/wpt/staticrange
dcheng@chromium.org, scottmg@chromium.org:
  external/wpt/domxpath
kojii@chromium.org:
  external/wpt/css/css-scoping
ksakamoto@chromium.org:
  external/wpt/resource-timing
meade@chromium.org:
  external/wpt/web-animations
rego@igalia.com:
  external/wpt/css/css-grid

TBR=lukebjerring

No-Export: true
Change-Id: I26ed6baee62d9b9797384c328f5a676c0912c19d
Reviewed-on: https://chromium-review.googlesource.com/966889
Commit-Queue: Blink WPT Bot <blink-w3c-test-autoroller@chromium.org>
Reviewed-by: Blink WPT Bot <blink-w3c-test-autoroller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543809}

Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel
Change-Id: I5994c5a3fc087c80441009ac1316752f1cfafff6
Reviewed-on: https://chromium-review.googlesource.com/964719Reviewed-by: weiliangc <weiliangc@chromium.org>
Commit-Queue: David Bokan <bokan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543808}

This makes it simpler to create a DataPipeGetter out of a Blob, and also
ensures that a blob that is used as the body of a request can still be
read even after the process that created the DataPipeGetter (i.e. the
renderer that initiated the fetch) dies.

Bug: 821878, 761117
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_mojo
Change-Id: I519a6061900f4cb29b786b8445dc14f034959e27
Reviewed-on: https://chromium-review.googlesource.com/963088
Commit-Queue: Marijn Kruisselbrink <mek@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543807}

This CL also makes the WebRTC UMA utility available in Blink for
features that have been migrated from content/renderer.

UMA histograms for MediaDevices.enumerateDevices() got removed while
migrating it from content to blink in https://crrev.com/c/808128.
This CL restores the same.


Bug: 821331
Change-Id: Iaba8ede9cfe36dfde0f6d18203f09be42bd15061
Reviewed-on: https://chromium-review.googlesource.com/961778Reviewed-by: Guido Urdaneta <guidou@chromium.org>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Alexei Svitkine <asvitkine@chromium.org>
Reviewed-by: Emircan Uysaler <emircan@chromium.org>
Commit-Queue: Chandan Padhi <c.padhi@samsung.com>
Cr-Commit-Position: refs/heads/master@{#543806}

The test testLitePageBTFWithoutFallback is failing due passing in the "alt1" experiment which is the basic team food experiment and will return a stale preview.

The test is expecting a lite page, the experiment needs to be changed to alt6 which will force lite page.

Bug: 822863
Change-Id: I657ab3d08c6d40c224ca190434eb37bdf1a102d0
Reviewed-on: https://chromium-review.googlesource.com/967147Reviewed-by: Ryan Sturm <ryansturm@chromium.org>
Commit-Queue: Richard Bustamante <bustamante@google.com>
Cr-Commit-Position: refs/heads/master@{#543805}

Bug: 799921
Change-Id: Ie1a0773bc7ae23c141b170611ecc1d44acfcf45a
Reviewed-on: https://chromium-review.googlesource.com/966861Reviewed-by: Benjamin Pastene <bpastene@chromium.org>
Commit-Queue: John Budorick <jbudorick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543804}

Selection bounds are sent from Blink to CC as part of a commit cycle so
that we can draw touch handles for it. Currently, the selection bounds
are relative to the main graphics layer of a CompositedLayerMapping. In
the case of a scroller, this will be its clip rect - rather than its
scrolling contents layer. Unfortunately, this means that scrolling on
the compositor isn't applied as part of the ToScreen transformation on
the selection bounds so scrolling wont update the selection bounds
location until another Blink commit. This went unnoticed until now
because the root layer was not considered a scroller. The page would
paint into a document-sized layer and the compositor would provide extra
scrolling layers to handle frame scrolling. In this configuration, the
ToScreen transformation from the root layer does include the scroll
offset.

Now that root layer scrolling has been turned on, frame scrolling works
much the same as other scrollers. Thus, this shortcoming is seen on
frame scrolling also.

The solution in this CL is to move the selection bounds rect to be
relative to the scrolling contents layer - if one exists. The ToScreen
transformation done in CC will correctly compensate for any scroll
offset applied in the compositor and touch selection handles stick to
the selection as its scrolled. For the CC side, see
ComputeViewportSelectionBound in layer_tree_impl.cc

Reland Note: This patch makes the RenderedPositionTests run with mock
scrollbars.

TBR=chrishtr@chromium.org

Bug: 812048
Change-Id: Ib9eacd14b75b71b4e0d4f0a9b57416800f9dfb91
Reviewed-on: https://chromium-review.googlesource.com/967001Reviewed-by: David Bokan <bokan@chromium.org>
Commit-Queue: David Bokan <bokan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543803}

blink_perf.events.hit-test-lots-of-layers is a microbenchmark of
hit PaintLayer::HitTestLayer which uses GeometryMapper. One of the
hottest functions is GeometryMapper::LocalToClipRectInternal which
creates a lot of FloatClipRects that check FloatRoundedRect::IsZero
in their constructor. By inlining the IsZero functions, we get a
3%-4% improvement on the entire benchmark.

Bug: 818772
Change-Id: I0fb4c02152b799399aa19d521ba91ed46e326c3c
Reviewed-on: https://chromium-review.googlesource.com/966797Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Commit-Queue: Philip Rogers <pdr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543802}

UDPSocket supports RANDOM_BIND flag which ensures that local port number
is allocated randomly. To implement this feature it was previously taking
RandIntCallback as one of constructor parameters. That parameter was
useful only in tests - in other cases base::RandInt() was always
sufficient. Than additional callback adds extra complexity where is can
be avoided. Also some platforms (e.g. Fuchsia) allocate port numbers
randomly by default. That callback made it impossible to rely on the OS
for port random allocation.

In this CL:
 1. Removed RandIntCallback from UDPSocket constructor. UDPSocket
    implementations are supposed to generate random port numbers
    internally.
 2. Updated unittests that were relying on this custom RandIntCallback.
    New RandomBind unittest should be more reliable (see linked bug).

TBR=miu@chromium.org, bbudge@chromium.org, mikhail.pozdnyakov@intel.com

Bug: 821516
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_mojo
Change-Id: If813d9db2ce5a5b0c6cdcfd8670361ccf6836fd8
Reviewed-on: https://chromium-review.googlesource.com/964895
Commit-Queue: Sergey Ulanov <sergeyu@chromium.org>
Reviewed-by: mark a. foltz <mfoltz@chromium.org>
Reviewed-by: Matt Menke <mmenke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543801}

When run using a 64-bit build of python.exe, the tool needs to
explicitly look in the 32-bit hive for Chrome's uninstall key.

BUG=822448

Change-Id: Icc0274f73411a717878651ec50ad65e7292f3102
Reviewed-on: https://chromium-review.googlesource.com/966444
Commit-Queue: Stephen Martinis <martiniss@chromium.org>
Reviewed-by: Stephen Martinis <martiniss@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543800}

Rename U2fDevice, U2fHidDevice, U2fBleDevice to FidoDevice,
FidoHidDevice, FidoBleDevice and change prefix of U2fHidMessageCallback
to HidMessageCallback.

Bug: 814059
Change-Id: Ia77ade2f571325d177872178eaa2f2995bcf3809
Reviewed-on: https://chromium-review.googlesource.com/927809
Commit-Queue: Jun Choi <hongjunchoi@chromium.org>
Reviewed-by: Jan Wilken Dörrie <jdoerrie@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543799}

BUG=822752
TBR=guidou@chromium.org

Change-Id: I5a2c60b22df4953c8aa207abd90650b86628edf4
Reviewed-on: https://chromium-review.googlesource.com/967019Reviewed-by: Karan Bhatia <karandeepb@chromium.org>
Commit-Queue: Karan Bhatia <karandeepb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543798}

BUG=822635

Change-Id: Ic3966f502e8d8ee5937a96b1a006f62c73d70230
Reviewed-on: https://chromium-review.googlesource.com/967062Reviewed-by: Reilly Grant <reillyg@chromium.org>
Commit-Queue: Charlie Reis <creis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543797}

This CL makes the V4 Store perftest a bit faster by:
1. Using a smaller # of hashes in debug builds
2. Using one big string instead of a vector of full hashes, to avoid
   allocating and de-allocating many small strings. To get this working,
   needed to slightly modify V4Store to allow querying by StringPiece.

(2) improved local performance on my release build by 25-30%

Bug: 822624
Change-Id: I646af2b0ccfb6447ccbb94ce68b661bd235587d1
Reviewed-on: https://chromium-review.googlesource.com/965931
Commit-Queue: Charlie Harrison <csharrison@chromium.org>
Reviewed-by: Varun Khaneja <vakh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543796}

This was an oversight in a previous CL.

Bug: 818709
Change-Id: I2697d136be7af56e00f342f77451b765a37c8eaa
Reviewed-on: https://chromium-review.googlesource.com/966796Reviewed-by: Steve Kobes <skobes@chromium.org>
Commit-Queue: Chris Harrelson <chrishtr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543795}

EncodedFormData already (optionally) contains a blink::BlobDataHandle,
so rather than looking up the same blob again by uuid just use that
handle if we have it. This should ensure no races are possible between a
blob being dereferenced (in blink) and the blob being converted into a
DataPipeGetter (as looking up the blob by UUID is an async operation).

Bug: 740744, 821878
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_mojo
Change-Id: Ic8464b778ec91d10fdbd7688e2b680b9464fe1dc
Reviewed-on: https://chromium-review.googlesource.com/964716
Commit-Queue: Marijn Kruisselbrink <mek@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543794}

TBR=kbr@chromium.org
BUG=822733

Cq-Include-Trybots: luci.chromium.try:linux_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
Change-Id: I8d8b0b6c411b8974b130e40776591897cd41760e
Reviewed-on: https://chromium-review.googlesource.com/966884Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Commit-Queue: Kenneth Russell <kbr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543793}

Add RenderWidgetHost(RWH) to the RenderWidgetHostViewBase(RWHVB)
constructor, so that each child class does not need to store its
own RWH.

Bug:716911

Change-Id: I10728cef663807e687a986f0f131f9decb067b67
Reviewed-on: https://chromium-review.googlesource.com/959456
Commit-Queue: Yi Xu <yiyix@chromium.org>
Reviewed-by: Avi Drissman <avi@chromium.org>
Reviewed-by: Fady Samuel <fsamuel@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543792}

Similar to https://chromium-review.googlesource.com/961861, we should
also apply paint offset with a transform operation during painting.

The added tests don't work with --site-per-process yet because of
crbug.com/822372.

Bug: 821486
Change-Id: I83f2fd7cb1d71e2639b9c37a4e139cf274be422d
Reviewed-on: https://chromium-review.googlesource.com/964728Reviewed-by: Philip Rogers <pdr@chromium.org>
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543791}

This CL moves magic numbers to constants files:

- Various dimensions in GridCell
- Margins and height of TabGridTopToolbar
- Margins and height of TabGridBottomToolbar

This CL also changes the toolbar classes to use
|-willMoveToSuperview:| to setup views instead of
|-init|.

Bug: 818198, 822524

Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Change-Id: Ib4dcfd344c29c3e0ccee103b6002dd6e1d14fb30
Reviewed-on: https://chromium-review.googlesource.com/965576
Commit-Queue: edchin <edchin@chromium.org>
Reviewed-by: edchin <edchin@chromium.org>
Reviewed-by: Mark Cogan <marq@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543790}

For VR UKM Metrics recording.

Bug: 808584
Cq-Include-Trybots: luci.chromium.try:linux_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_vr;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
Change-Id: I98418d91b0df01c265683362349cff6365d375c8
Reviewed-on: https://chromium-review.googlesource.com/959440
Commit-Queue: Anna Offenwanger <offenwanger@chromium.org>
Reviewed-by: Robert Kaplow <rkaplow@chromium.org>
Reviewed-by: Christopher Grant <cjgrant@chromium.org>
Reviewed-by: Tibor Goldschwendt <tiborg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543789}

TBR=chongz@chromium.org, kinuko@chromium.org
BUG=822585

Change-Id: I926212903064753268cc217ab8ef3be4e9dfaf4c
Reviewed-on: https://chromium-review.googlesource.com/967009
Commit-Queue: Karan Bhatia <karandeepb@chromium.org>
Reviewed-by: Karan Bhatia <karandeepb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543788}

Before MR2, a ThreadLocal was used to store the currentActivityThread.
I don't think there are any existing calls to getProcessName() off of
the main thread, so this isn't fixing an existing crash.

Change-Id: If4cf17dbfe4f1cead2cc5a4ec043405ec665e922
Reviewed-on: https://chromium-review.googlesource.com/962866
Commit-Queue: agrieve <agrieve@chromium.org>
Reviewed-by: Richard Coles <torne@chromium.org>
Reviewed-by: Yaron Friedman <yfriedman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543787}

blink-dev i2s thread: https://groups.google.com/a/chromium.org/d/topic/blink-dev/8RBFue7RMXQ/discussion

Change-Id: I5595b3aed72ea0cece88948f69f480f5808fce6b
Bug: 735518,782381,816661
TBR: raymes@chromium.org
Reviewed-on: https://chromium-review.googlesource.com/957265Reviewed-by: Tarun Bansal <tbansal@chromium.org>
Reviewed-by: David Dorwin <ddorwin@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Commit-Queue: Tarun Bansal <tbansal@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543786}

It seems to "just work".

Tested on:
 - Windows 10
 - Windows 10 with --disable-gpu
 - Windows 7
 - Windows 7 Classic mode (Aero disabled)

Bug: 713359
Change-Id: I04466aab7e978ad51e5e579488e29789cda9ab72
Reviewed-on: https://chromium-review.googlesource.com/961552
Commit-Queue: Tommy Li <tommycli@chromium.org>
Reviewed-by: Mohsen Izadi <mohsen@chromium.org>
Reviewed-by: Trent Apted <tapted@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543785}

The comment incorrectly asserted that the method was only called by
PingLoader. In fact the method is called for every request by devtools.
So instead just ignore data pipes in the size calculation for now, just
like blobs without blob data handle are ignored.

If we do want actual size information for these elements at some point
we'd probably just want to add an explicit size field along with the
DataPipeGetter, just like network::DataElement already does for almost
every other type of element.

Bug: 821878
Change-Id: I6a64aaeffbcf0e22bf90bcf44c38367bf682cf3e
Reviewed-on: https://chromium-review.googlesource.com/964612Reviewed-by: Matt Falkenhagen <falken@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Commit-Queue: Marijn Kruisselbrink <mek@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543784}

The root PRESUBMIT.py contains
_CheckNoProductionCodeUsingTestOnlyFunctions which checks against
including a call to a 'for testing only' method in production code.
That check has not been tested. This CL adds two tests for it.

Additionally, the CL also fixes two issues in the presubmit tests:
 * FilterSourceFile in PRESUBMIT_test_mocks.py did not match the
   production version from tools/depot_tools/presubmit_support.py: the
   production requires the filename to be in the whitelist, while the
   mock version was OK with just not being in the blacklist. The CL
   modifies the mock version to match the production.
 * The test for _CheckAndroidTestJUnitInheritance did not adhere to
   the whitelisted filename pattern (*Test.java). This CL changes the
   names of the fake files to match the pattern.

Bug: 821981
Change-Id: I65edf07ddb2ae26ad7d08ceb7cf4d51b482b5e56
Reviewed-on: https://chromium-review.googlesource.com/966605Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Commit-Queue: Vaclav Brozek <vabr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543783}

https://skia.googlesource.com/skia.git/+log/8b858526faf4..16c1196ea6ba

$ git log 8b858526f..16c1196ea --date=short --no-merges --format='%ad %ae %s'
2018-03-16 ethannicholas fix SkSL SPIR-V geometry shaders' entry point declaration
2018-03-16 rmistry Remove Android Framework bot from CQ while server is being upgraded
2018-03-16 ethannicholas we now complain if an SkSL program tries to override the default output
2018-03-16 egdaniel Fix vulkan layout barriers when we are doing in line uploads.

Created with:
  roll-dep src/third_party/skia


The AutoRoll server is located here: https://autoroll.skia.org

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.


CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;luci.chromium.try:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
TBR=stephana@chromium.org

Change-Id: I24f30c0216c31dd6705f3287bac5baeb82ffe6a2
Reviewed-on: https://chromium-review.googlesource.com/966941
Commit-Queue: skia-chromium-autoroll <skia-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Reviewed-by: skia-chromium-autoroll <skia-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#543782}

LayoutBox::ScrolledContentOffset can be much simpler for LayoutViews, and
also for LayoutBoxes that are not LayoutFlexibleBox. To tease this apart,
this patch makes ScrolledContentOffset virtual to provide the more expensive
implementation only when needed.

Also, check for fixed-position in IsFixedPositionObjectInPagedMedia
before finding the View(), which may be expensive because it involves ping-
ponging through Document.

Bug:818709

Change-Id: I8470b7c1487b3fbf4f227aad21e8f9da81c1e9bb
Reviewed-on: https://chromium-review.googlesource.com/964933Reviewed-by: vmpstr <vmpstr@chromium.org>
Commit-Queue: Chris Harrelson <chrishtr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543781}

Disables two more tests found to be flaky.

TBR=zhaobin

Bug: 822231
Change-Id: I0c6a36a7d448e942ffd191ea12471f33c2729c81
Reviewed-on: https://chromium-review.googlesource.com/966983Reviewed-by: mark a. foltz <mfoltz@chromium.org>
Commit-Queue: mark a. foltz <mfoltz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543780}

This CL adds a variation parameter which encodes a threshold for
the Site Engagement score [1] required to do tab-unders. By default,
we allow tab-unders to origins that have any non-zero engagement.

This moves the Tab.TabUnder.EngagementScore metric as the last check
of IsSuspiciousClientRedirect, to simulate the scores blocked tab-unders
would have if we removed that check.

[1]: https://www.chromium.org/developers/design-documents/site-engagement

Bug: 661629
Change-Id: I48ba83ab8cc58641cc21dcad062c313f1d1290ef
Reviewed-on: https://chromium-review.googlesource.com/965403Reviewed-by: Avi Drissman <avi@chromium.org>
Commit-Queue: Charlie Harrison <csharrison@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543779}

1. Clients used to submit local_surface_id in hit-test data; however, this can be
racy because we have primary_surface_id and fallback_surface_id and we can use
any of the latest one in that range. Change it to not using local_surface_id
submitted from clients, but instead asking SurfaceAggregator at HitTestAggregator
::Aggregator() time for the latest local_surface_id to use - this is done by
adding LatestLocalSurfaceIdLookupDelegate in surfaces to talk to viz::Display,
where we have access to the surfaces SurfaceAggregator used at aggregation time.
Add FrameSinkIdMap in SurfaceAggregator to keep track of previously contained
frame sinks.

2. Keep track of the list of child regions we've aggregated so far in
HitTestAggregator to prevent cycles.

3. When preparing hit-test data, skip the quad if its FrameSinkId has changed
between fallback and primary because we don't know which FrameSinkId would be used
to draw this quad.

4. Add viz::TestLatestLocalSurfaceIdLookupDelegate to be used in tests.

Bug: 819057, 819435
Test: viz_unittests, content_unittests

Cq-Include-Trybots: master.tryserver.chromium.android:android_optional_gpu_tests_rel
Change-Id: I35ec40ae00af1ba4e3937002c1c0d261dae8b136
Reviewed-on: https://chromium-review.googlesource.com/957881
Commit-Queue: Ria Jiang <riajiang@chromium.org>
Reviewed-by: Fady Samuel <fsamuel@chromium.org>
Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543778}

https://chromium.googlesource.com/catapult.git/+log/6733f341991f..16322a374abb

$ git log 6733f3419..16322a374 --date=short --no-merges --format='%ad %ae %s'
2018-03-16 perezju [Telemetry] Remove Cherry_Mobile_Android_One condition

Created with:
  roll-dep src/third_party/catapult


The AutoRoll server is located here: https://catapult-roll.skia.org

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.


TBR=sullivan@chromium.org

Change-Id: I0e17eb56979559a363dcb76f40a2e1e755d20768
Reviewed-on: https://chromium-review.googlesource.com/966584Reviewed-by: catapult-chromium-autoroll <catapult-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Commit-Queue: catapult-chromium-autoroll <catapult-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#543777}

Bug: 822723
Change-Id: Ifdfe9ac1a72c39fd9a65a82da25f27b1ec246b9c
Reviewed-on: https://chromium-review.googlesource.com/966865Reviewed-by: Marijn Kruisselbrink <mek@chromium.org>
Commit-Queue: Daniel Murphy <dmurph@chromium.org>
Cr-Commit-Position: refs/heads/master@{#543776}