GitLab

The tests were failing at a lower rate on other OS's, but it seems
they were indeed failing on other OS's. Generalize the [ Pass Failure ]
TestExpectations from [ Mac ] to all platforms.

Bug: 1104024
Change-Id: I78461aa43041776799930416f261f7271ca52ad5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2311312Reviewed-by: Victor Costan <pwnall@chromium.org>
Commit-Queue: Victor Costan <pwnall@chromium.org>
Commit-Queue: Darwin Huang <huangdarwin@chromium.org>
Auto-Submit: Darwin Huang <huangdarwin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790628}

Bug: 1104064
Change-Id: Idae3d296992c08ed9c2af032e345e57795b8fe34
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2309753
Commit-Queue: Xianzhu Wang <wangxianzhu@chromium.org>
Reviewed-by: Philip Rogers <pdr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790627}

To move the implementation into blink we first move the emulation messages.

BUG=1093963

Change-Id: Ib87e8da814cd1756a2a4136515df1688763d221d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2303367
Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790626}

This CL attaches impressions to context menu navigations that occur
from anchor tags with impression attributes.

This also declares the frame the context menu occurred in as the
initiator routing frame for the navigation, which was previously null.

Bug: 1049674
Change-Id: I7a2031b41865d44bc926c87af1fba86aa72048f8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2240198
Commit-Queue: John Delaney <johnidel@chromium.org>
Reviewed-by: Ken Buchanan <kenrb@chromium.org>
Reviewed-by: Nate Chapin <japhet@chromium.org>
Reviewed-by: Charlie Harrison <csharrison@chromium.org>
Reviewed-by: Avi Drissman <avi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790625}

Roll ChromeOS Silvermont AFDO profile from 86-4181.3-1594638142-benchmark-86.0.4205.0-r1 to 86-4183.25-1595242326-benchmark-86.0.4207.0-r1

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/cros-afdo-silvermont-chromium
Please CC c-compiler-chrome@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug

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

Tbr: c-compiler-chrome@google.com
Change-Id: If7719142d813b4919b31610d597d522f72839af4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2310959Reviewed-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@{#790624}

This also removes the public_android_sdk check from the system WebView
and trichrome versions, since those are now defined even when
!public_android_sdk.

Also a small cleanup to the --switch-webview-to arg, so we don't have to
re-specify the package name, and can just pass the APK directly.

Change-Id: Ib408c5db892cf51763684aca2f92d70f99366ec3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2310876Reviewed-by: Scott Violet <sky@chromium.org>
Commit-Queue: Clark DuVall <cduvall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790623}

The active process limit is only checked on process creation so can
safely be dropped to zero. This prevents reuse of the process handle
to launch another process in the job after the contained process has
terminated.

Tested by existing tests.

Bug: 1050359
Change-Id: I803ca74e6a654b46484a945925085e4444a1df07
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2308174Reviewed-by: Will Harris <wfh@chromium.org>
Commit-Queue: Alex Gough <ajgo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790622}

TBR=tmartino@chromium.org

Bug: 1084931
Change-Id: Ic89e67ad2f87e972aab271e3fb208bdd808920d2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2310808Reviewed-by: Bruce Dawson <brucedawson@chromium.org>
Commit-Queue: Katie Dektar <katie@chromium.org>
Commit-Queue: Bruce Dawson <brucedawson@chromium.org>
Auto-Submit: Katie Dektar <katie@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790621}

Ampersands are used to indicate mnemonics in menu items. We must escape
real ampersands before using them in menu item labels.

Before:
https://bugs.chromium.org/p/chromium/issues/attachment?aid=457370&signed_aid=NdbKLbBf4ZUuOW9VvGbKOQ==&inline=1
https://bugs.chromium.org/p/chromium/issues/attachment?aid=457372&signed_aid=kcRyF6PZEi7SyoW3L6hG8A==&inline=1

After:
https://bugs.chromium.org/p/chromium/issues/attachment?aid=457369&signed_aid=AIbAsCPsbM2MYSHtCskv7g==&inline=1
https://bugs.chromium.org/p/chromium/issues/attachment?aid=457371&signed_aid=NJP9QClLhZVOh5d-lP4piQ==&inline=1

Bug: 1106668
Change-Id: Ia3fe2950218ba3e227c588c74e4031a5edb210e4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2309830
Auto-Submit: Alan Cutter <alancutter@chromium.org>
Reviewed-by: Michael Wasserman <msw@chromium.org>
Commit-Queue: Michael Wasserman <msw@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790620}

WinCFG can be Enabled or Disabled depending on whether
the CFG dispatch guards are compiled into the binary or
not.

BUG=584575

Change-Id: Ib4b76208ae839ddb8c5be264f741dee4eb389de9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2308030
Commit-Queue: Will Harris <wfh@chromium.org>
Reviewed-by: Alexei Svitkine <asvitkine@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790619}

GLRenderer/Webview use (Begin/End)BatchReadAccessSharedImage commands
to mark begin/end frame in SharedImageBatchAccessManager. Currently it's
tied to ScopedBatchedReturnResources which issues
EndBatchReadAccessSharedImage after DrawAndSwap was done and so the
command won't be flushed till next frame which leaves SharedImage in
state where EndRead was called, but no fence was produced.

To fix it and avoid extra flush move it to GLRenderer BeginDrawingFrame
and FinishDrawingFrame, so it will be flushed with SwapBuffers call.

Change-Id: I4452213c3081a700a991349d40aa65ea00a49dbe
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2302414Reviewed-by: Jonathan Backer <backer@chromium.org>
Commit-Queue: Vasiliy Telezhnikov <vasilyt@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790618}

This CL adds a new mb config that will be used by both the official and
try codesearch builders for lacros. It also updates the try builder console.

Bug: 1105242
Change-Id: I5e970b7213730c3bdf54ea786df252ed40200182
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2311035
Commit-Queue: Erik Chen <erikchen@chromium.org>
Commit-Queue: Yuke Liao <liaoyuke@chromium.org>
Auto-Submit: Erik Chen <erikchen@chromium.org>
Reviewed-by: Yuke Liao <liaoyuke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790617}

gn check failed because util/status_builder.h doesn't exist.

Change-Id: Ie0f2e2a93c61769854879dffb0f7327a2f837b7f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2311031Reviewed-by: Dirk Pranke <dpranke@google.com>
Commit-Queue: Edward Lesmes <ehmaldonado@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790616}

We should use gpuBenchmarking.smoothScrollByXY in
fast/events/hit-test-counts.html instead of eventSender to simulate
mouse wheel scroll. After scroll unification, we can scroll on
compositor, that eventSender only scrolls on main thread.

Bug: 1047176
Change-Id: Ic2bee891e89a25dd6622843beea1f9dc1c2f5821
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2270337Reviewed-by: David Bokan <bokan@chromium.org>
Commit-Queue: Lan Wei <lanwei@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790615}

This is a reland of a1374a65

TBR=steimel@chromium.org

Original change's description:
> [Kaleidoscope] Public side of the color code
>
> Change-Id: I98b421dfac9adda34448c645c0432d7a9ad2455c
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2303854
> Reviewed-by: Tommy Steimel <steimel@chromium.org>
> Commit-Queue: Becca Hughes <beccahughes@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#790186}

Change-Id: I174e12ca3b61a1aecc5ed2838cde6d7b47d0319f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2310897Reviewed-by: Becca Hughes <beccahughes@chromium.org>
Commit-Queue: Becca Hughes <beccahughes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790614}

Bug: 1059389
Change-Id: I09e70640a74fcfd3c10f278e8a777fe73ee87305
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2304540
Commit-Queue: Ken Rockot <rockot@google.com>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Dale Curtis <dalecurtis@chromium.org>
Reviewed-by: Robert Sesek <rsesek@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790613}

They got added in https://crrev.com/787824

Bug: 1009837
Change-Id: Icd572904d3f9acde45a7220503f7b04a2a6b1019
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2310970Reviewed-by: Scott Violet <sky@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790612}

This fixes a regression I introduced after [1] which removed colormap
creation code.  It turns out the colormap is still necessary for
transparent visuals, so this CL adds it back.

[1] https://chromium.googlesource.com/chromium/src/+/ae0a09cefaca3445bf43b0abefe3d068794388c0

R=sky

Change-Id: I3d29ced609ab3f915f9b7cdbdce6f4e8b2a5dce1
Bug: 1106334,1107385
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2311070
Commit-Queue: Scott Violet <sky@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Auto-Submit: Thomas Anderson <thomasanderson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790611}

Modern kdialog versions support passing in filters in the format
SelectFileDialog has them, avoiding the need for a lossy conversion
via mime types. Detect the kdialog version and make use of this.

A downside of this is that is now much more obvious that always passing
1 as "selected filter index" to select file callers is wrong. But
unfortunately kdialog does not provide a way to figure out which filter
was selected when multiple filters could match the returned file. At
least this isn't more broken than it was before.

Bug: 1108060
Change-Id: Ie39baf0dfb0664e366d7ae1574303a66ab8e2de9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2311056
Commit-Queue: Marijn Kruisselbrink <mek@chromium.org>
Reviewed-by: Thomas Anderson <thomasanderson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790610}

https://skia.googlesource.com/skia.git/+log/5933d7d54fb8..324578b5537d

2020-07-21 tdenniston@google.com Add toy stroker sample
2020-07-21 adlai@google.com Revert "Migrate SkImage::makeWithFilter to GrRecordingContext"
2020-07-21 johnstiles@google.com Remove support for multiple coverage processors from GrPaint API.
2020-07-21 michaelludwig@google.com Reland "Support moving blocks from one allocator to another"
2020-07-21 fmalita@chromium.org [skottie] Cleanup post https://skia-review.googlesource.com/c/skia/+/303997
2020-07-21 skia-autoroll@skia-public.iam.gserviceaccount.com Roll skcms from f73242142ce5 to fc0d614b6313 (1 revision)
2020-07-21 robertphillips@google.com Rename GrRenderTargetContext::getContext to recordingContext
2020-07-21 reed@google.com make more constructors constexpr
2020-07-21 bsalomon@google.com Rename GrTexture/Proxy/Priv mip map members/functions
2020-07-21 adlai@google.com Migrate SkImage::makeWithFilter to GrRecordingContext

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/skia-autoroll
Please CC robertphillips@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug

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

Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux-blink-rel;luci.chromium.try:linux-chromeos-compile-dbg;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel
Bug: None
Tbr: robertphillips@google.com
Change-Id: Ifa7fba68836bb83fe3ac4f5211ae475ac76bc7d8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2311132Reviewed-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@{#790609}

Bug: 1095223
Change-Id: Ic9d3ceb5eba994b9c0328aa7a8661fb8492ae5d3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2303909
Commit-Queue: Yann Dago <ydago@chromium.org>
Reviewed-by: Sergey Ulanov <sergeyu@chromium.org>
Reviewed-by: Owen Min <zmin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790608}

Change:
* Replace the four requested information in PRParams with a
PaymentOptions.
* Create a utility class PaymentOptionsUtils to simplify that code where
4 individual requested information are used.
* PaymentAppFactoryParams extends PRParams. So PRImpl does not need to
implement both params.

Bug: 1102522

Change-Id: I59656137f0ec80f9eb29edf12b29d1c5cfc5c52f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2310896Reviewed-by: Danyao Wang <danyao@chromium.org>
Commit-Queue: Liquan (Max) Gu <maxlg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790607}

Tests on CI/CQ bots should be hermetic, so download the and isolate
ash-chrome at build time instead of test time.

Bug: 1104318
Change-Id: Ia967cbc6fef205d30ec093ea4bfaadcc0eb46d62
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2307646
Commit-Queue: Yuke Liao <liaoyuke@chromium.org>
Reviewed-by: Dirk Pranke <dpranke@google.com>
Reviewed-by: Erik Chen <erikchen@chromium.org>
Reviewed-by: Sven Zheng <svenzheng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790606}

Bug: 1097623
Change-Id: I3b09b083884924c21f87e02d24900b8e17d12615
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2308433Reviewed-by: Avi Drissman <avi@chromium.org>
Commit-Queue: Melissa Zhang <melzhang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790605}

This unblocks http://review.skia.org/304776. After that lands, we can
come back and remove the explicit cast.

The inclusion of GrDirectContext.h is "jumping ahead" to where Chromium
CL 2297920 is taking us, and it is in the process of landing. We could
have imported GrContext.h but then we'd have a needless conflict.

Bug: skia:104662
Change-Id: Ibc6b5a5061577bca59a3b57ce16a1af0298d0f5b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2310974
Commit-Queue: Adlai Holler <adlai@google.com>
Auto-Submit: Adlai Holler <adlai@google.com>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790604}

Since WebLayer was sending old_state for the constraint, a switch from
SHOWN to HIDDEN, as when entering fullscreen, could hit a DCHECK in
browser_controls_offset_manager.cc

I think some confusion arises from the re-use of a single enum for both
the _constraint_ and the _current state_. Code used the terms old_state
and new_state, when really the RenderFrameHost method expected a new
constraint and new desired state. The BOTH value in the context of a
constraint means unconstrained, and in the context of a state means
"don't change unless you have to to conform to the constraint".

Test: requestFullscreen on an element in weblayer shell doesn't cause
      the renderer to crash in debug

Bug: 1107647
Change-Id: I4f72e1144a9fc9a0cf0934b9640c660867c0d971
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2310652
Commit-Queue: Evan Stade <estade@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790603}

This reverts commit 33f234e7.

Reason for revert: Windows 7 test bots are failing

This is because WinHttpCreateProxyResolver requires Windows 8+.
https://docs.microsoft.com/en-us/windows/win32/api/winhttp/nf-winhttp-winhttpcreateproxyresolver

Original change's description:
> Implement the WindowsSystemProxyResolutionService/Request
> 
> This change creates a functional proxy resolver for Windows that relies
> on WinHttp APIs. This is not in its final state, though. This CL
> represents the simplest, least-work way to implement this proxy
> resolution service. Work still to come includes:
> - Complete NetLogs
> - Better error reporting from WinHttp APIs
> - Proxy Delegate
> - Proxy retry info
> - Assurance that the proxy configs we're receiving are from the system
> 
> The WindowsSystemProxyResolutionService is the object that external
> callers use to resolve a proxy. These callers can keep track of the
> status of this proxy resolution via a ProxyResolutionRequest,
> implemented here as a WindowsSystemProxyResolutionRequest. The request
> object is mainly intended to kick off a specific proxy resolution and
> report a result via a caller-provided callback. Both of these objects
> deal with a WindowsSystemProxyResolver.
> 
> The WindowsSystemProxyResolver is a new class that does all the required
> interfacing with WinHttp APIs. It is a reference counted object created
> by the WindowsSystemProxyResolutionService. Once created, it is used for
> the lifetime of the WindowsSystemProxyResolutionService. Throughout that
> time, the Resolver indirectly keeps a WinHttp session handle open. Each
> WindowsSystemProxyResolutionRequest is given a reference to the new
> resolver. When the Request object attempts to resolve a proxy for a
> given URL, it'll call into the Resolver object. Under the hood, the
> Resolver object creates a handle for the proxy resolution and runs the
> async call to WinHttp. At any time, the Request object can choose to
> ignore a pending result from the Resolver (ex: shutdown). The Resolver
> object must be reference counted because async WinHttp calls call back
> directly into the Resolver object, so it needs to stay alive as long as
> we're waiting for an async response from WinHttp.
> 
> The only other new object is the WinHttpAPIWrapper, which is just a thin
> wrapper over WinHttp APIs which we use to enable easier testing and
> to simplify some interactions with WinHttp. A WinHttpAPIWrapper is owned
> by a WindowsSystemProxyResolver. The Resolver object is the only object
> that should interact with the Wrapper object in any meaningful way.
> Internally, the WinHttpAPIWrapper keeps track of the opened WinHttp
> session handle that we're using for the lifetime of the Resolver object.
> 
> This change includes tests for the WindowsSystemProxyResolver layer (by
> mocking out the WinHttpAPIWrapper) and the
> WindowsSystemProxyResolutionService (by mocking out the
> WindowsSystemProxyResolver).
> 
> Bug: 1032820
> Change-Id: Ic1c60033ff148e6e8f3708e37a2af366613fefac
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2159128
> Commit-Queue: Nicolas Arciniega <niarci@microsoft.com>
> Reviewed-by: Ramin Halavati <rhalavati@chromium.org>
> Reviewed-by: Eric Roman <eroman@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#790493}

TBR=eroman@chromium.org,rhalavati@chromium.org,niarci@microsoft.com

Change-Id: I54d4659d5165261af912c427fdeb8f372a7bb2bf
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 1032820
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2311063Reviewed-by: Bruce Dawson <brucedawson@chromium.org>
Commit-Queue: Bruce Dawson <brucedawson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790602}

r790297 crrev.com/c/2309494 added comments saying the DCHECK
can hit, but forgot to remove it.

This patch removes it.

Bug: 1107204
Change-Id: I716e8faca243a7a9b8457c4acc2bb43cd8d33dfe
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2309877
Commit-Queue: Kent Tamura <tkent@chromium.org>
Auto-Submit: Koji Ishii <kojii@chromium.org>
Reviewed-by: Kent Tamura <tkent@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790601}

TBR=bling-team@google.com

Change-Id: Ic2810ce2ccd866697f36b7bfd44a8ed27dcc76b8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2311111Reviewed-by: Internal Frameworks Autoroller <bling-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com>
Commit-Queue: Internal Frameworks Autoroller <bling-autoroll-builder@chops-service-accounts.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#790600}

As part of this, introduced "nvidia_driver" schema for workarounds.

Bug: 1041166
Change-Id: I6debe871f65fd261aa6d03b46906286995cdfa28
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2308853
Auto-Submit: Frank Liberato <liberato@chromium.org>
Commit-Queue: Zhenyao Mo <zmo@chromium.org>
Reviewed-by: Zhenyao Mo <zmo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790599}

Threaded scrolls unnecessarily cause commits on every change. This
patch implements an optimization from pre-BlinkGenPropertyTrees in
layer list mode to avoid these commits.

When an impl-side scroll occurs, cc can immediately apply the scroll
before notifying blink so that the blink update only causes a commit if
needed (e.g., interest rect changed). This has been implemented in
LayerTreeHost::UpdateScrollOffsetFromImpl. A second change was needed in
PropertyTreeManager::DirectlyUpdateScrollOffsetTransform to avoid
commits if cc already matches blink.

This is a reland of https://crrev.com/745110.

Bug: 1048384f
Change-Id: Id50a0d0a944f7d1f763906304bbcfccdf81cb301
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2297709
Commit-Queue: Khushal <khushalsagar@chromium.org>
Reviewed-by: Khushal <khushalsagar@chromium.org>
Auto-Submit: Philip Rogers <pdr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790598}

In the past we had to pass in the alpha type of a SharedImage when
copying it. Recent CL's have added this information to the SharedImage
itself so now we can avoid redundancy and set the alpha type once
when creating the SharedImage. This CL removes
|unpack_premultiply_alpha| from RasterDecoder::CopySubTextureINTERNAL
in favor of using the alpha state stored with the image.

Bug: 1034086
Change-Id: I592bb7f6629212905614fcadf4f8717993923490
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2310971Reviewed-by: Khushal <khushalsagar@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Nathan Zabriskie <nazabris@microsoft.com>
Cr-Commit-Position: refs/heads/master@{#790597}

Since SDR white level was baked into the display color space, HDR video
YUV quads with PQ color space had white level adjusted to skip white
level scaling.  However, that workaround didn't apply to RGBA HDR video
texture quads with scRGB linear color space.

This CL applies the white level adjustment to that case.  Texture quads
containing video are marked so that the white level adjustment applies
only to video and not webgl or canvas.  The default white level for
scRGB linear is set to 80 nits so that it matches the SRGB white level.

Bug: 1084689
Change-Id: I59c50bb2fe0f702c5ea48ed5d2b71b69572f0f8d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2296006
Commit-Queue: Sunny Sachanandani <sunnyps@chromium.org>
Reviewed-by: Robert Liao <robliao@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: ccameron <ccameron@chromium.org>
Reviewed-by: Dale Curtis <dalecurtis@chromium.org>
Reviewed-by: kylechar <kylechar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790596}

Add methods for to get and set dictionary, time and url profile wide
ExtensionPrefs.

Design Doc: http://go/extprefrefactor

Bug: 1069560
Change-Id: Ia922a3be11fb77715e185f2a286eb3d4f59622cb
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2308857
Commit-Queue: Archana Simha <archanasimha@chromium.org>
Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790595}

Rather than talking to quarantine directly, this changes
NativeFileSystemFileWriter to go through the possibly out-of-process
quarantine service similar to how downloads interacts with this.

This also makes it much easier to write tests to make sure quarantine
is invoked correctly.

Bug: 990997
Change-Id: I0ee4ad15792acf4f40e7bc251ba6d64437912609
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2308920Reviewed-by: Xing Liu <xingliu@chromium.org>
Reviewed-by: Will Harris <wfh@chromium.org>
Reviewed-by: Nasko Oskov <nasko@chromium.org>
Commit-Queue: Marijn Kruisselbrink <mek@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790594}

This reverts commit c2b6797f.

Reason for revert: Broke compile on Window:
https://ci.chromium.org/p/chromium/builders/ci/Win%20Builder/96923?

Original change's description:
> Add test that verifies CFG loader config is generated correctly.
> 
> This CL adds a test that creates a callback from operating system
> code to a module that has CFG enabled where the callback is not
> listed as a valid indirect target.
> 
> Note: this test does not require dispatch guards to be present
> in chromium code, just for them to be present in OS code and
> for the CFG load config to be generated correctly.
> 
> BUG=584575
> 
> Change-Id: Id6f5f4061f7237c23f08c8fe1b7471ea200f0628
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2308854
> Commit-Queue: Will Harris <wfh@chromium.org>
> Reviewed-by: Alex Gough <ajgo@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#790587}

TBR=wfh@chromium.org,ajgo@chromium.org

Change-Id: I8d5070d373d381146ee1e906e57bdc0323a7e0ab
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 584575
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2311146Reviewed-by: Katie Dektar <katie@chromium.org>
Commit-Queue: Katie Dektar <katie@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790593}

This fixes lint issues in Dev UI code. The only logic change is we're
now calling super.onNewIntent().

Bug: 1107950
Test: autoninja -C out/Default system_webview_apk
Test: manually verify Dev UI navigation looks right
Change-Id: Ia70e8c843bbb2ceb4e0052f530496cc26dd38325
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2310871Reviewed-by: Andrew Grieve <agrieve@chromium.org>
Reviewed-by: Hazem Ashmawy <hazems@chromium.org>
Commit-Queue: Nate Fischer <ntfschr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790592}

This change:
- adds test coverage for Learn Mode with regard to keyboard, gesture, and braille
- adds live descriptions of each gesture (e.g. swipe right with one finger) in addition to describing the command
- adds the touchExplore gesture as something described (with special handling due to how frequently it fires)
- is more careful about queueing up too much speech; added robust tests for this throughout

R=akihiroota@chromium.org

Fixed: 1106462
AX-Relnotes: only announce a modifier once, when holding down modifiers like Search in ChromeVox Learn Mode. Also, adds descriptions for gestures like swipe right with one finger, in addition to the command triggered while in Learn Mode.
Change-Id: Ib600330c63cf3c412aa62c34050a4281e07cb816
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2309981
Commit-Queue: David Tseng <dtseng@chromium.org>
Reviewed-by: Akihiro Ota <akihiroota@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790591}

Bug: 1007686
Change-Id: Ibcb774c35eaaab97a832728304781718ee699fce
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2300994Reviewed-by: Naoki Fukino <fukino@chromium.org>
Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Commit-Queue: Josip Sokcevic <sokcevic@google.com>
Cr-Commit-Position: refs/heads/master@{#790590}

Screen shots in crbug.com/1103558.

WebApp themes now match how custom themes are treated by checking
ThemeProvider::HasCustomColor rather than
ThemeService::UsingDefaultTheme which doesn't indicate when a PWA
is in use as has set a custom theme.

Default theme text color changed to use default MD blue.

Bug: 1103558
Change-Id: I7d527f89f577374bed028644b4e79f40ec0c77de
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2306490
Commit-Queue: Joel Hockey <joelhockey@chromium.org>
Auto-Submit: Joel Hockey <joelhockey@chromium.org>
Reviewed-by: Peter Kasting <pkasting@chromium.org>
Cr-Commit-Position: refs/heads/master@{#790589}