GitLab

This reverts commit bb8a36eb.

Reason for revert: Created too much load on too-small Xenial
pool, resulting in large pending test shard times, resulting
in long builds, resulting in pending times, resulting in even
longer builds due to cold caches, resulting in even longer
pending times.

Original change's description:
> [testing] switch to use xenial for swarming test
> 
> This CL switches to use xenial for test as many as possible.
> 
> Sheriff:
> If this broke some ci/fyi builder, feel free to revert this.
> 
> TBR=jbudorick
> 
> Bug: 954450
> Change-Id: I98abfb50bb10ca3eeec0e23d197fe7672c8acc95
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1689513
> Reviewed-by: Takuto Ikuta <tikuta@chromium.org>
> Commit-Queue: Takuto Ikuta <tikuta@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#675527}

TBR=yyanagisawa@chromium.org,tikuta@chromium.org,jbudorick@chromium.org

Change-Id: I1244abdfa1cf627c7f9effa3048ccf8ed53091b8
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 954450, 982335
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692035Reviewed-by: John Budorick <jbudorick@chromium.org>
Reviewed-by: Aaron Gable <agable@chromium.org>
Commit-Queue: John Budorick <jbudorick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675722}

Rename cryptauth_constants.cc/h to cryptauth_enrollment_constants.cc/h
to distinguish from CryptAuth constants related to DeviceSync.

Bug: 899080
Change-Id: Iba7801c2cb7ea46e3a23f86e66a4a8a1d5cef75d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1693162
Commit-Queue: Josh Nohle <nohle@chromium.org>
Commit-Queue: Kyle Horimoto <khorimoto@chromium.org>
Auto-Submit: Josh Nohle <nohle@chromium.org>
Reviewed-by: Kyle Horimoto <khorimoto@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675721}

https://chromium.googlesource.com/angle/angle.git/+log/9be098b52fbb..cb8b4a5716ba


git log 9be098b52fbb..cb8b4a5716ba --date=short --no-merges --format='%ad %ae %s'
2019-07-09 clemendeng@google.com Rename "opengl32" to "libGL"
2019-07-09 geofflang@chromium.org Move member initializers from Caps.cpp into the struct definitions.
2019-07-09 angle-autoroll@skia-public.iam.gserviceaccount.com Roll ./third_party/glslang/src 4b4b41a63499..8aa9a7bb8f74 (1 commits)
2019-07-09 angle-autoroll@skia-public.iam.gserviceaccount.com Roll ./third_party/spirv-tools/src 9702d47c6fe4..0c4feb643b89 (7 commits)


Created with:
  gclient setdep -r src/third_party/angle@cb8b4a5716ba

The AutoRoll server is located here: https://autoroll.skia.org/r/angle-chromium-autoroll

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=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
TBR=cwallez@chromium.org

Change-Id: I1efe58ce6a6bc2dc49baaf1e75f6776924aab678
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1693085Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#675720}

Move optimization guide related experiments, features, and switches to //components/optimization_guide

Bug: 969558
Change-Id: Ia2403feb3f0afa826a58db02a6b4479f3a9e3d5b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1691013
Commit-Queue: Sophie Chang <sophiechang@chromium.org>
Reviewed-by: Doug Arnett <dougarnett@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Joshua Pawlicki <waffles@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675719}

Eager sweeping was added to dispose of CachesStorage.
CachesStoreage is a member, so eager sweeping didn't do anything.

Bug: 981043
Change-Id: I418dabf6596188660d10b91fefddf80c9dd8dae4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692984Reviewed-by: Kentaro Hara <haraken@chromium.org>
Commit-Queue: Omer Katz <omerkatz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675718}

R=jwd@chromium.org

Change-Id: Ifd419096bc7f527d0698a3f5b869e4c714a2e9da
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692577Reviewed-by: Jesse Doherty <jwd@chromium.org>
Commit-Queue: Alexander Timin <altimin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675717}

Pre-finalizer performs:
1) Inform context_proxy_ that parent is destroyed
2) Cancel requests to classic_script_loader_
3) Reset uniqueptr to factory_client_

Bug: 981043
Change-Id: I425513bcc91d7d8676d575d302be85c9ca114cf4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1690584Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Commit-Queue: Omer Katz <omerkatz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675716}

https://imgur.com/a/qnVs4lG

Bug: 686905, 934202
Change-Id: If51daf61ec4ddaa780aff7a1f11af6746330d198
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1690564Reviewed-by: Esmael El-Moslimany <aee@chromium.org>
Commit-Queue: John Lee <johntlee@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675715}

Bug: 883985
Change-Id: If43f93fe27ce84fddc098aca8870b3cc729d7541
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692168
Commit-Queue: Denis Kuznetsov <antrim@chromium.org>
Reviewed-by: Roman Sorokin [CET] <rsorokin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675714}

https://chromium.googlesource.com/external/github.com/google/shaderc/+log/e9bb8f287417..1f2dec1e1234


git log e9bb8f287417..1f2dec1e1234 --date=short --no-merges --format='%ad %ae %s'
2019-07-09 dneto@google.com Remove app_dummy call in Android test (#756)


Created with:
  gclient setdep -r src/third_party/shaderc/src@1f2dec1e1234

The AutoRoll server is located here: https://autoroll.skia.org/r/shaderc-chromium-autoroll

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=radial-bots+chrome-roll@google.com

Change-Id: I9e6a9a3efdde264299f00fc268e3ae753a6664df
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1693081Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#675713}

The comment above HTMLFrameOwnerElement::GetLayoutEmbeddedContent()
says:
  // Most subclasses use LayoutEmbeddedContent (either LayoutEmbeddedObject or
  // LayoutIFrame) except for HTMLObjectElement and HTMLEmbedElement which may
  // return any LayoutObject when using fallback content.
  LayoutEmbeddedContent* GetLayoutEmbeddedContent() const;

No new test because I couldn't reproduce the bug locally, but I believe this
will fix the clusterfuzz bug.

Bug: 981459
Change-Id: I3ecf8022111dc25a2e862c0311ffa56467d18c2e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1689336Reviewed-by: Philip Rogers <pdr@chromium.org>
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675712}

Bug: 930109
Change-Id: I82f98f11719978ba7944351383c456bc5e60f9b5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692750Reviewed-by: Achuith Bhandarkar <achuith@chromium.org>
Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Commit-Queue: A Olsen <olsen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675711}

Replaced DomTimer destructor with a prefinalizer that calls Stop().
Added flag is_stopped_ to make sure second call to Stop() (from
TimerBase dtor) is noop.

Bug: 981043
Change-Id: Iab1473fe2d96fd9d6bfaf0ff2d7e46099796670f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1688928
Commit-Queue: Omer Katz <omerkatz@chromium.org>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675710}

Bug: 981828
Change-Id: I39d4af3b988165bd5980f884647b7c59f3aff1e3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1691329Reviewed-by: Hector Carmona <hcarmona@chromium.org>
Commit-Queue: Esmael El-Moslimany <aee@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675709}

Spec requires dropping y value for 2d translate if it's zero (*). The
current implementation contains bug that:

1. It rounds the value before checking, which results in incorrect
   dropping of '0.1px'

2. It calls DoubleValue() on calc() without type checking, while
   DoubleValue() on calc() is meaningless when the calc() doesn't
   resolve to a simple numeric value (e.g., 1px - 1%). This results in
   incorrect dropping of some calc().

Both are fixed by this patch.

This is also a preparation for adding DCHECK in DoubleValue() to ensure
that it's called only when calc() can be resolved into a simple numeric
value.

Bug: 979895
Change-Id: Ie8b846729b91e55006485a25fab0b65533983eac
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692202Reviewed-by: Anders Hartvoll Ruud <andruud@chromium.org>
Commit-Queue: Xiaocheng Hu <xiaochengh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675708}

This test fails consistently on win-asan.

Tbr: aboxhall@chromium.org
Bug: 835455
Change-Id: Ia4b348a8c2a6693f7c0fd18906445f19f9422e3d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1693026Reviewed-by: Kevin McNee <mcnee@chromium.org>
Commit-Queue: Kevin McNee <mcnee@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675707}

Bug: 959357
Change-Id: I14b723d2fa1f250b5dd97bf377391324d0752d48
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1686701Reviewed-by: Fernando Serboncini <fserb@chromium.org>
Reviewed-by: Xida Chen <xidachen@chromium.org>
Reviewed-by: Robert Kaplow <rkaplow@chromium.org>
Commit-Queue: Fernando Serboncini <fserb@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Auto-Submit: Fernando Serboncini <fserb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675706}

Makes the tests more stand-alone & easier to run when dropped
into non-chromium codebase.

Allows //base to maintain the assumption that all tests that
use its class runner also use its instrumentation.

Bug: 971715
Change-Id: I2620936ad6b745563cd6114362bcb131b951bcd4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1690094
Commit-Queue: Andrew Grieve <agrieve@chromium.org>
Reviewed-by: Paul Jensen <pauljensen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675705}

Discovered this use case @
https://chromium-review.googlesource.com/c/chromium/src/+/1686776/5
and extracted fix from it.

An example of how this can happen is
FtlMessageReceptionChannelTest.TimeoutIncreasesToMaximum
which uses FastForwardBy from a GMOCK WillRepeatedly() hook.

Also fix an issue in calls to AllowRunTasks(). Because of
nesting and the way RunUntilIdle() works (with separate
allow/disallow phases), it was possible to enter a nested call
from either an allow/disallow phase and break the state on exit.
This CL makes RunUntilIdle() always restore the previous state.

This will also be required in
https://chromium-review.googlesource.com/c/chromium/src/+/1686776
which will need to AllowRunTasks() from FastForwardBy().

R=fdoray@chromium.org

Bug: 946657
Change-Id: I634a6ab89add3424383b4f105069b080d6fa25b9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1687796
Commit-Queue: Gabriel Charette <gab@chromium.org>
Reviewed-by: Alex Clarke <alexclarke@chromium.org>
Reviewed-by: François Doray <fdoray@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675704}

Bug: 930109
Change-Id: I64848ce01b47350b3d04967be873f93ce5c4977a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692754Reviewed-by: Roman Sorokin [CET] <rsorokin@chromium.org>
Commit-Queue: Roman Sorokin [CET] <rsorokin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675703}

This is needed so that checks that occur as part of navigation
don't get confused by the replication state (notably the origin)
of the portal guest contents being missing from the remote frame
in the host contents' renderer.

Since this state is pushed down into the blink::RemoteFrame,
we need InitializeCoreFrame to be called before we can set the
replication state. As a result WebRemoteFrame::CreateForAdoptedPortal
is created, which performs this initialization with the provided
HTMLPortalElement.

Included web test crashes before this CL.

Bug: 980937
Change-Id: I843192e5f5aa8e4e022a88d318fa4a5e1d46a3d9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1677163
Commit-Queue: Jeremy Roman <jbroman@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Nasko Oskov <nasko@chromium.org>
Reviewed-by: Lucas Gadani <lfg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675702}

Partial revert of crrev.com/c/1689513.

chrome-swarming doesn't have xenial bots in the main chrome test pool:
https://ci.chromium.org/p/chrome/builders/ci/linux-chromeos-google-rel/4291

# Notry'ing to fix bots.
NOTRY=true

Bug: 954450
Change-Id: Ifbfb4c4dc099184f24714ac07a4c501830b75e5e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692823
Commit-Queue: Ben Pastene <bpastene@chromium.org>
Reviewed-by: John Budorick <jbudorick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675701}

Some libc++ tests fail when the mac_upload_clang bot is run on LUCI
instead of Buildbot, and it's not clear why.

Since we don't really care about testing libc++ here, don't run those
tests to unblock the LUCI migration.

Bug: 731375
Change-Id: I273e0345098795cf27cb527873b04ddf8945bb83
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1680546
Commit-Queue: Nico Weber <thakis@chromium.org>
Reviewed-by: Nico Weber <thakis@chromium.org>
Reviewed-by: Reid Kleckner <rnk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675700}

Bug: 980024
Change-Id: I49fdaac8ea558a6068a3de6ba7d74e922ee9de81
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1691255Reviewed-by: Jeremy Roman <jbroman@chromium.org>
Commit-Queue: Mounir Lamouri <mlamouri@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675699}

Undefined behavior occurs (observed as a crash) when the Feed Internals
page is loaded and the last_fetch_trigger_type_ var is uninitialized,
so wrapping it here in a nullable pointer to avoid this.

Bug: 981513
Change-Id: I8637412423ab14d1a1329a414997d1f38887daaa
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1691046Reviewed-by: Sky Malice <skym@chromium.org>
Commit-Queue: Natalie Chouinard <chouinard@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675698}

Bug: 39240
Change-Id: I9cdf9a9a9de76fcdbc77bd36c24196f48b9035dd
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692934
Auto-Submit: Raphael Kubo da Costa <raphael.kubo.da.costa@intel.com>
Commit-Queue: Andrew Whalley <awhalley@chromium.org>
Reviewed-by: Andrew Whalley <awhalley@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675697}

The hardcoded "Upload from device" and "No background" tiles
were slightly misaligned. Partly due to a single character space
being inserted between inline-block elements. Fix the issues by
moving some properties between elements and setting the font size
of the parent to 0px. This fixes some problems with keyboard
navigation as well as the slight visual misalignment.

Bug: 937570
Change-Id: Iab866d50ae288d9bf4d5dc12280053253e6f3767
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1691260Reviewed-by: Kristi Park <kristipark@chromium.org>
Commit-Queue: Kyle Milka <kmilka@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675696}

Timings that should remain real-time even under MOCK_TIME are now
explicitly using subtle::TimeTicksNowIgnoringOverride().

ThreadPoolClock defaults to DefaultTickClock in this CL but will be
overridden under ScopedTaskEnvironment's upcoming multi-threaded
MOCK_TIME mode @
https://chromium-review.googlesource.com/c/chromium/src/+/1686776

Ultimately, when MOCK_TIME mode always mocks TimeTicks::Now(), we will
be able to remove this and use TimeTicks::Now() again directly for
these calls.

Bonus #1: Gets rid of a call to DeprecatedGetMockTickClock().

Bonus #2: Added thread annotations to DelayedTaskManager::queue_lock_.

R=fdoray@chromium.org

Bug: 946657, 789079, 905412
Change-Id: Ief86338388b670ad9ddafb926e6817df6d96a95a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1687028
Commit-Queue: Gabriel Charette <gab@chromium.org>
Reviewed-by: François Doray <fdoray@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675695}

On trying to access the locked screen mode in split screen (tablet mode), the browser crashes due to NULL pointer reference.

Bug: b/119853019
Change-Id: Idaf715a9d48f3979243b83bd9fa9e59706f09122
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1685101Reviewed-by: Xiaoqian Dai <xdai@chromium.org>
Commit-Queue: Aya Elsayed <ayaelattar@google.com>
Cr-Commit-Position: refs/heads/master@{#675694}

Bug: 982289
Change-Id: I475fa95c7ed0f31be93a53f99f35b1e56bb710c7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692967
Commit-Queue: Michael Ludwig <michaelludwig@google.com>
Commit-Queue: Florin Malita <fmalita@chromium.org>
Auto-Submit: Michael Ludwig <michaelludwig@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675693}

This CL removes the divider between the footer commands and adds one
before the footer commands.

Screenshots and spec can be found on the bug.

Bug: 980455
Change-Id: Ia40e3a3bd83a5e135087eff2a11573dfc944ee1c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1691031Reviewed-by: Friedrich [CET] <fhorschig@chromium.org>
Reviewed-by: Becky Zhou <huayinz@chromium.org>
Commit-Queue: Ioana Pandele <ioanap@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675692}

Currently, the is_icon_url field in ParsedFaviconPath is only set to
true for urls pointing to .ico files. We change the underlying regex
used so as to also accept .png files.

Bug: 982217
Change-Id: Iab345655eadebe5f279468b0d333b297064338bf
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692167
Commit-Queue: Victor Vianna <victorvianna@google.com>
Reviewed-by: Esmael El-Moslimany <aee@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675691}

The calc unit types no longer make too much sense after we have split
CSSMathFunctionValue from CSSPrimitiveValue, as:
- They complicate UnitType by introducing some entries that are not
  really unit types
- If a client really relies on these calc types, they can refer to
  CSSMathFunctionValue::Category() instead

Hence, this patch removes them for code health.

Bug: 979895
Change-Id: I9c8720bd68d64f756a6c4a251f59666350649b29
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1691554Reviewed-by: Anders Hartvoll Ruud <andruud@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Commit-Queue: Xiaocheng Hu <xiaochengh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675690}

This patch changes the presubmit script so that when any module.json, .grd,
and .grdp files change, the script that autofixes any localizable string
inconsistency will run. This is because these files also contain localizable
strings, in addition to .js files.

Bug: 941561
Change-Id: I3aaba696122757ccb5e4e83f49c50503e459000f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1648912Reviewed-by: Joel Einbinder <einbinder@chromium.org>
Commit-Queue: Mandy Chen <mandy.chen@microsoft.com>
Cr-Commit-Position: refs/heads/master@{#675689}

NGFragmentItem is a lighter weight version of
NGPhysicalFragment we plan to use for inline boxes and text
runs.

As seen in InlineBox v.s. LayoutObject, inline boxes and text
runs do not need all the information normal block boxes need,
and they are easier to handle when presented as a flat list.

Design doc:
https://docs.google.com/document/d/10vJ6wdyEdeGkmcotKBZ9h3YtDzw5FIpDksa8rCHVFuM/edit#heading=h.iakhx55u4iw4

Bug: 982194
Change-Id: I01c2c7b12ec9c2ef9c866c08f3132eb57d15b7d8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692109
Commit-Queue: Emil A Eklund <eae@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675688}

Get rid of raw |thread_pool_| member in favor of keeping more const
traits around. This makes an upcoming CL cleaner, extracted from it
http://chromium-review.googlesource.com/c/chromium/src/+/1686776.

R=fdoray@chromium.org

Bug: 946657
Change-Id: I26d8675e3e643dd938cff7105232f90fba9c52de
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1687025
Commit-Queue: Gabriel Charette <gab@chromium.org>
Reviewed-by: François Doray <fdoray@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675687}

This patch changes |NGOutOfFlowLayoutPart::LayoutCandidate()|
not to use a cached result if it has an inline container.

|NGLayoutResult::OutOfFlowPositionedOffset()| has an offset
relative to the inline formatting context block. When an out-
of-flow object has inline contianer, this offset may change
without setting |NeedsLayout()| to the out-of-flow object.

Bug: 981686
Change-Id: Ib8376f791af0b95f6c2b6e5c2361b6a887c74e69
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1692107
Commit-Queue: Emil A Eklund <eae@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675686}

Bug: 980220
Change-Id: Ib624572d19f5f9ea28b5e2db67374bdfb2fa1e6c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1683518
Commit-Queue: Victor Vianna <victorvianna@google.com>
Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Mikel Astiz <mastiz@chromium.org>
Reviewed-by: Jan Krcal <jkrcal@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675685}

Bug: 979927
Change-Id: Ie35300e7fce923490c4885582d2b1ab4b5b53ea8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1688617Reviewed-by: Tien Mai <tienmai@chromium.org>
Commit-Queue: Rakesh Soma <rakeshsoma@google.com>
Cr-Commit-Position: refs/heads/master@{#675684}

This reverts commit a0d2f696.

Fix crbug.com/980086 by calling MarkAncestorChainForFlagsUpdate() from
DirtyStackingContextZOrderLists() to ensure z-order lists are updated in
UpdateDescendantDependentFlags() when scrollbar existence changes.

Original change's description:
> Revert "Implement correct z-order of overlay scrollbars"
>
> This reverts commit 128c7578.
>
> Reason for revert: crbug.com/980086
>
> Original change's description:
> > Implement correct z-order of overlay scrollbars
> >
> > Previously we implemented z-ordering of composited overlay scrollbars
> > in GraphicsLayerTreeBuilder which reparents overlay scrollbar
> > GraphicsLayers of scrollers with higher z-order scrollable contents.
> > The overlay scrollbar GraphicsLayers are inserted into the
> > GraphicsLayer of the stacking context above all scrolling contents
> > GraphicsLayers, to ensure the overlay scrollbars to be above any
> > scrolling contents.
> >
> > However, the logic doesn't work for all overlay scrollbars in
> > CompositeAfterPaint and non-composited scrollbars in
> > pre-CompositedAfterPaint because these scrollbars are not controlled
> > by GraphicsLayerTreeBuilder.
> >
> > This CL let PaintLayerStackingNode also manage z-order of overlay
> > scrollbars. If any scroller with overlay scrollbars has any stacked
> > contents that have higher z-order than the scroller, the overlay
> > scrollbars will be inserted after the highest z-order scrolling
> > contents in the z-order list.
> >
> > Also add a paint phase for overlay scrollbars. This is not strictly
> > necessary for now because all scrollers with overlay scrollbars are
> > self-painting. However, this allows us in the future to let scrollers
> > not self-painting to achieve perfect painting order.
> >
> > Bug: 918155
> > Change-Id: Iccae1cbcde8de46158267e45bab2a13a027db3ec
> > Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1461536
> > Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
> > Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#672748}
>
> TBR=wangxianzhu@chromium.org,chrishtr@chromium.org
>
> Bug: 918155, 980086
> Change-Id: Icccb9ec7af7ef424306afade0c80015e4c5c5c6f
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1684022
> Reviewed-by: Xianzhu Wang <wangxianzhu@chromium.org>
> Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#674526}

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

Bug: 918155, 980086
Change-Id: I5f001477512de5ebbaa7a4fe62c1ef1bb76e442a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1691331Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#675683}