GitLab

Currently starting Libassistant happens in 2 steps:
   1. (On background thread) ServiceController::Start() creates the
      |AssistantManager|.
   2. (On main thread) ServiceControllerProxy::FinishAssistantStart()
      sets some extra fields on |AssistantManager| and calls
      AssistantManager::Start().

This causes a race condition if Libassistant is immediately stopped,
as that happens on the background thread and can collide with step 2
above.

To prevent this from happening, this CL collapses both starting steps in
a single step that runs on the background thread. We do this by
registering a callback which is called by the background thread in step
1 above.

This callback is temporary but necessary as some parts of the
initialization code still use objects that have not been migrated to the
Libassistant mojom service yet.

Bug: 1157177, b/171748795, 1128032
Test: chromeos_unittests "Assistant*:ServiceController*"
Change-Id: Ifab9705839ad5d3e5c9f2b7f7fba4c336b9f7c9d
Cq-Include-Trybots: luci.chrome.try:linux-chromeos-chrome
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2594090
Commit-Queue: Jeroen Dhollander <jeroendh@chromium.org>
Reviewed-by: Xiaohui Chen <xiaohuic@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838277}

This is a follow-up to https://crrev.com/c/2568665.

Bug: 526773
Change-Id: I48beb93d9ca3366f519251c909da951215c6e0cf
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2587237Reviewed-by: Natalie Chouinard <chouinard@chromium.org>
Commit-Queue: Bence Béky <bnc@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838276}

What:
Route impressions for the Conversion Measurement API to the browser
process for anchor tags with rel="noopener".

Why:
We currently drop impressions for anchor clicks which have noopener
set because the impressions do not get forwarded to the browser process.

Anchor click navigations with the rel="noopener" attribute use a
different navigation path than those without the attribute, in which
the window creation mojo also navigates the frame to the target url.
This navigation is not annotated with an impression.

How:
Add an Impression member to CreateNewWindowParams which will get
forwarded as a LoadURLParam for the subsequent navigation of the
window.

This change also fixes a bug in which initiator_frame_tokens are not
set for noopener navigations.

As of https://chromium-review.googlesource.com/c/chromium/src/+/2327963,
the lack of an initiator frame tokens are used to to indicate whether
an opener was suppressed for a navigation. Because attaching the token
breaks this logic, we route a new |was_opener_suppressed| param through
the NavigationController:LoadUrlWithParams() flow to recover the
behavior.

Bug: 1155577
Change-Id: I84739a9f4ec0df9a867f91397f80b46eea1bcee6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2575223
Commit-Queue: John Delaney <johnidel@chromium.org>
Reviewed-by: Charlie Harrison <csharrison@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Charlie Reis <creis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838275}

What:
Delete data from conversions database on exit for origins which are
configured to be session only.

Why:
This better aligns conversion storage with other site data
implementations.

How:
When the ConversionManager for a given storage partition is tore down,
we queue a clear data task which deletes all conversion data whose
origins match a session only origin.

Note that as is, using ClearData() with the new filter will be
rather un-performant as it will trigger a host settings lookup on
every invocation of the filter. This is being addressed on
https://crrev.com/c/2568367

Bug: 1154453
Change-Id: If6f250ddd3329dfb595db3b83e2601b22a49c2c1
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2567626
Commit-Queue: John Delaney <johnidel@chromium.org>
Reviewed-by: Avi Drissman <avi@chromium.org>
Reviewed-by: Charlie Harrison <csharrison@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838274}

This is about release blocker: https://crbug.com/1153966

We think the bug is about iterating over an observer that has already
been deleted. After:
https://chromium-review.googlesource.com/c/chromium/src/+/2594772
all the observers are using ScopedObserver / base::ScopedObservation.
This ensures they aren't deleted before unregistering themself.

There are 5 RenderWidgetHostObserver:
1) FileSelectHelper
2) SessionRestoreStatsCollector
3) RenderWidgetHostVisibilityTracker
4) HungPagesTableModel
5) PageHandler.

Bug happens when deleting an unresponsive tab. Maybe HungPagesTableModel
is the culprit here?

This patch makes every observer to stop observing as soon as the
observee sends RenderWidgetHostDestroyed(). This also clears any
references toward the RenderWidgetHost.

I don't really have high hopes this will fix the bug. However this
doesn't hurt to be more cautious.

Bug: 1153966
Change-Id: I210e414e281665f56b148ac4248770448c55a0be
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2595443
Commit-Queue: danakj <danakj@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838273}

AddEventListenerOptionsResolved holds a Member<AbortSignal>, which was
causing a circular reference keeping the AbortSignal alive forever.

Bug: 1148326
Fixed: 1155771
Change-Id: I673bc78c3ab2b10f40624958d9ea147be01c002f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2585674Reviewed-by: Mason Freed <masonfreed@chromium.org>
Commit-Queue: Joey Arhar <jarhar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838272}

Use testdriver Action API to simulate mouse actions in
css/selectors/focus-visible-006.html and
css/selectors/focus-visible-008.html.

Bug: 1145677
Change-Id: Id25c6120f1dbd489d4c803ee979e67ad5ef49f8c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2551818Reviewed-by: Mustaq Ahmed <mustaq@chromium.org>
Commit-Queue: Lan Wei <lanwei@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838271}

Change-Id: I0a3c0c69994a8bad47b1c8f9c5d7c5d660aa6d05
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2597777
Commit-Queue: Ryan Hansberry <hansberry@chromium.org>
Auto-Submit: Ryan Hansberry <hansberry@chromium.org>
Reviewed-by: James Vecore <vecore@google.com>
Cr-Commit-Position: refs/heads/master@{#838270}

When building against libstdc++, the macro DCHECK_NE does not seem to
handle implicit conversion between std::unique_ptr and nullptr. This
patch fixes one such build error in compositor_frame_reporter.cc.

Bug: 957519
Change-Id: If1801dba777a90089a4c0585cec7e62de8006fff
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2595279Reviewed-by: danakj <danakj@chromium.org>
Commit-Queue: Hans-Filip Elo <helo@vewd.com>
Cr-Commit-Position: refs/heads/master@{#838269}

Add a basic browser test for WebUI JS Error Reporting. This just tests
the basic functionality for a start.

Bug: chromium:1121816
Change-Id: I8f8ed722da59d362f1ca1ad588698deac733fc4a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2580231
Commit-Queue: Ian Barkley-Yeung <iby@chromium.org>
Reviewed-by: Kyle Horimoto <khorimoto@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838268}

This CL adds SplitOnceCallback, a utility function that allows to get
two OnceCallbacks out of one. Invoking any of the two split callbacks
will run the originally passed callback. Invoking the remaining callback
will result in a no-op.

Bug: 1156809
Change-Id: Iebf4542732fb6230af92fa27c4d7c5705933ad6b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2586688
Commit-Queue: Thomas Guilbert <tguilbert@chromium.org>
Reviewed-by: danakj <danakj@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838267}

When VideoFrameCallbackRequesterImpl schedules its callbacks to be run
in the next rendering steps, and the HTMLMediaElement changes its
WebMediaPlayer, we have to make sure to cancel the scheduled callback
execution.

This CL adds a test to make sure we do, as a follow up to the CL in the
linked bug.

Bug: 1158160
Change-Id: I723361671d923a44d84457812fad8da2cefffac0
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2595984
Commit-Queue: Thomas Guilbert <tguilbert@chromium.org>
Commit-Queue: Dale Curtis <dalecurtis@chromium.org>
Auto-Submit: Thomas Guilbert <tguilbert@chromium.org>
Reviewed-by: Dale Curtis <dalecurtis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838266}

https://chromium.googlesource.com/catapult.git/+log/5039608d3366..8d43e589947e

2020-12-17 johnchen@chromium.org [devil] Add logging around APK install

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/catapult-autoroll
Please CC wenbinzhang@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:chromeos-kevin-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
Bug: chromium:1156424
Tbr: wenbinzhang@google.com
Change-Id: I043a91eadee351ce4abb1bb93d358d4ca6ea5b80
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2597624Reviewed-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@{#838265}

This change prepares the non-Windows part of the codebase for the
base::string16 switch to std::u16string. This is done by adding explicit
casts where current code relies on base::char16 being int16_t.

Furthermore, WTF::String, WTF::StringView and WTF::AtomicString are
modified to only expose a constructor from const char16_t* when char16_t
and icu's UChar are different types.

Lastly, this change removes a couple banned usages of NoCopy creation of
(Core)Foundation type. Apart from that this change does not change any
functionality.

TBR=jbroman,mmenke,jamiewalch

Bug: 911896
Change-Id: Ie208e7a5e7f00097d0a07eb4fd9197284cd8c482
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2596261Reviewed-by: Jamie Walch <jamiewalch@chromium.org>
Reviewed-by: Matt Menke <mmenke@chromium.org>
Reviewed-by: Jeremy Roman <jbroman@chromium.org>
Reviewed-by: Matt Mueller <mattm@chromium.org>
Reviewed-by: Avi Drissman <avi@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Commit-Queue: Jan Wilken Dörrie <jdoerrie@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838264}

https://skia.googlesource.com/skia.git/+log/860d01ca8bae..fa9a08369ed4

2020-12-17 johnstiles@google.com Remove unnecessary Blocks from the inliner.
2020-12-17 csmartdalton@google.com Enable GrTessellationPathRenderer by default
2020-12-17 johnstiles@google.com Replace inliner do-while loops with for loops.
2020-12-17 djsollen@google.com Revert "[svg] Absolute positioning support for text"
2020-12-17 brianosman@google.com New SkSL code generator that emits directly to skvm
2020-12-17 johnstiles@google.com Add support for outerProduct in SPIR-V.
2020-12-17 johnstiles@google.com Fix 4x4 outerProduct, and add unit tests.
2020-12-17 jvanverth@google.com Fix shadow directional light bounds.
2020-12-17 reed@google.com Stash m44
2020-12-17 fmalita@chromium.org [svg] Absolute positioning support for text
2020-12-17 egdaniel@google.com Handle vulkan copies correctly when using resolve attachments.
2020-12-17 vichang@google.com Android: Build skia on top of libicu.so

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 tdenniston@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
Cq-Do-Not-Cancel-Tryjobs: true
Bug: None
Tbr: tdenniston@google.com
Test: Test: m droid
Change-Id: Ife8173f676e0df9ff05e4d414bf0b4d35a526bb3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2597698Reviewed-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@{#838263}

Per discussion on chromium-dev
(https://groups.google.com/u/1/a/chromium.org/g/chromium-dev/c/F2xhH0h36rU),
set up stack tracing and enable error reporting when
ReadyToCommitNavigation is called instead of when RenderFrameCreated is
called.

Bug: chromium:1121816
Change-Id: Ie0aaf56ced830180284f153bc43322701c95e504
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2585659
Commit-Queue: Ian Barkley-Yeung <iby@chromium.org>
Reviewed-by: dpapad <dpapad@chromium.org>
Reviewed-by: danakj <danakj@chromium.org>
Auto-Submit: Ian Barkley-Yeung <iby@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838262}

This CL ensures paper-tooltips is not a hop on keyboard arrows key
navigations.

Bug: 1158411
Change-Id: I6067312772859ef62bbb98d4d368edd93e3e9a57
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2592789
Commit-Queue: Monica Basta <msalama@chromium.org>
Reviewed-by: Tibor Goldschwendt <tiborg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838261}

At smaller window sizes, parts of the settings section become hidden
from view. Allow users to scroll through the settings section when this
occurs.

Screenshot: http://shortn/_kQ7nVMKigV

Bug: 1059779
Change-Id: I1e118760c8dfa6a191163416028681b3644d6869
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2596170Reviewed-by: Zentaro Kavanagh <zentaro@chromium.org>
Commit-Queue: Jesse Schettler <jschettler@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838260}

In CL:2577757, we enabled fragmentation for absolutely positioned nodes
with `height: auto` and positioned with the bottom property. That
uncovered a bug with nested abspos in a fragmentation context - the
innermost abspos was never laid out.

This CL fixes it. We now query the OOF descendants generated during the
layout pass of the outermost abspos right after the layout is completed
to lay them out - if there are any.

For the moment, the offset of the innermost abspos is still wrong, but
the crash is fixed.

Bug: 1157086
Change-Id: I1031f4837f493032feb5de047e2366afc6d69599
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2587553Reviewed-by: Alison Maher <almaher@microsoft.com>
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Commit-Queue: Benjamin Beaudry <benjamin.beaudry@microsoft.com>
Cr-Commit-Position: refs/heads/master@{#838259}

When doing sentence navigation, we will first search for next sentence
start within the current node group. If that fails, we will look for the
rest of the content within this paragraph. If we still cannot find a
sentence start, we will look for the next paragraph.

Once we find a sentence start, we will get the nodes from the sentence
start until the end of that paragraph. Then, we will send the new
content to TTS.

AX-Relnotes: N/A.
Bug: 1143823
Change-Id: I051c572d4a9f866aeb37fba32137ed81b9cac744
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2587799
Commit-Queue: Lei Shi <leileilei@google.com>
Reviewed-by: David Tseng <dtseng@chromium.org>
Reviewed-by: Akihiro Ota <akihiroota@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838258}

If an entire iframe is display locked by an element in the parent
document, it throttles rendering in the iframe's content frame and
also forces all IntersectionObserver's to report "not intersecting".
This is already the case for same-process iframes; this CL extends it
to OOPIF's by informing the OOPIF when it is display locked by the
parent.

Change-Id: I4cb89c515c74b6a39a0d5cf8a4fd721eb9b95b98
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2587657
Commit-Queue: Stefan Zager <szager@chromium.org>
Reviewed-by: Arthur Sonzogni <arthursonzogni@chromium.org>
Reviewed-by: Ken Buchanan <kenrb@chromium.org>
Reviewed-by: vmpstr <vmpstr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838257}

https://dawn.googlesource.com/dawn.git/+log/155241b665f8..ee977a0df8e3

2020-12-17 enga@chromium.org Update MultisampledSamplingTests to use WGSL
2020-12-17 enga@chromium.org Update DepthStencilCopyTests to use WGSL - FragDepth
2020-12-17 enga@chromium.org Autogenerate all of the wire callback mocks
2020-12-17 cwallez@chromium.org Introduce SubresourceStorage (3/N): Inline data

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/dawn-chromium-autoroll
Please CC rharrison@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:dawn-linux-x64-deps-rel;luci.chromium.try:dawn-mac-x64-deps-rel;luci.chromium.try:dawn-win10-x64-deps-rel;luci.chromium.try:dawn-win10-x86-deps-rel
Bug: None
Tbr: rharrison@google.com
Change-Id: Ie4775c298e9a1806b2b3c47d74c3060a80edcd13
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2597111Reviewed-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@{#838256}

When reading user-selected content (e.g., from mouse selection), STS pauses at the end of the selected content. This CL enables the user to use resume to continue reading from the pause location till the end of the paragraph.

When resuming from the content processed by the navigation feature (e.g., the user hits pause), STS will continue reading from the start of the current sentence (implemented in 2572905).

In both cases, if there is no content left in the paragraph, STS will continue reading content from the next paragraph.

AX-Relnotes: N/A
Bug: 1143817
Change-Id: Idb48219fd805c8b5167f1b30850fea904a06259e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2582524
Commit-Queue: Lei Shi <leileilei@google.com>
Reviewed-by: David Tseng <dtseng@chromium.org>
Reviewed-by: Dominic Mazzoni <dmazzoni@chromium.org>
Reviewed-by: Josiah Krutz <josiahk@google.com>
Cr-Commit-Position: refs/heads/master@{#838255}

Screenshot of change has been uploaded to the bug.

Bug: 1156245
Change-Id: I80ed29bb9486c75b1ce6bd4217401cb4f25758d8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2594448Reviewed-by: Yulun Wu <yulunwu@chromium.org>
Reviewed-by: Toni Baržić <tbarzic@chromium.org>
Reviewed-by: Alex Newcomer <newcomer@chromium.org>
Commit-Queue: Yulun Wu <yulunwu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838254}

This reverts commit 6a6c700a.

Reason for revert: https://logs.chromium.org/logs/chrome/buildbucket/cr-buildbucket.appspot.com/8860631130980203728/+/steps/compile__with_patch_/0/stdout

Original change's description:
> GMSCore API availability function for specific versions
>
> Change-Id: Ie431852d225659e8b24bfec9c09d7c043924660f
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2595929
> Commit-Queue: Sam Maier <smaier@chromium.org>
> Reviewed-by: David Van Cleve <davidvc@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#838236}

TBR=smaier@chromium.org,chromium-scoped@luci-project-accounts.iam.gserviceaccount.com,davidvc@chromium.org

Change-Id: Iac1a894b3d3adaa58c0fcfefd407cd8db82b3e56
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2597817Reviewed-by: Sam Maier <smaier@chromium.org>
Commit-Queue: Sam Maier <smaier@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838253}

Fixes a recent regression where mb.py started incorrectly classifying
tests due to a change in how GN args were parsed from the build
directory. This re-adds quotes around string args, which brings the
previous behavior back.

Bug: 1159618
Change-Id: Ide52674e258d2a2698d0011928bd0d452a92cb8b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2596640
Commit-Queue: Brian Sheedy <bsheedy@chromium.org>
Commit-Queue: Ben Pastene <bpastene@chromium.org>
Auto-Submit: Brian Sheedy <bsheedy@chromium.org>
Reviewed-by: Ben Pastene <bpastene@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838252}

Bug: 1158635, 1112471
Change-Id: I51e6bfff20cabc4d79ae49af66b8e1cfd095c593
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2597797Reviewed-by: Erik Staab <estaab@chromium.org>
Commit-Queue: Dirk Pranke <dpranke@google.com>
Cr-Commit-Position: refs/heads/master@{#838251}

https://android.googlesource.com/platform/external/perfetto.git/+log/cc9b7694ce3d..0d23e521cc40

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/perfetto-trace-processor-linux-chromium
Please CC perfetto-bugs@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: perfetto-bugs@google.com
Change-Id: I28a0e39a19247117b0601d1a84acc448b53e8650
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2596872Reviewed-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@{#838250}

The overall project aims to expose CNAME aliases to the
SubresourceFilter to aid in ad tagging and blocking.

A previous CL added a base::Optional parameter to
net::ConnectJob::SetSocket and a getter and setter to net::Socket.

This change continues the plumbing by adding an accessor to
net::HttpBasicState as well as one to net::HttpStream.

All of HttpStream's derived classes receive trivial implementations
except for HttpBasicStream and WebSocketBasicHandshakeStream. SPDY and
QUIC support will come in later CLs.

Relevant tests:
net:net_unittests
out/Default/net_unittests --gtest_filter=*Dns*Alias*

Bug: 1151047
Change-Id: I8b38591e4b186ded8cb41f89571385eee9fe0aa6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2563450
Commit-Queue: Cammie Smith Barnes <cammie@chromium.org>
Reviewed-by: Matt Menke <mmenke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838249}

Screenshot: https://screenshot.googleplex.com/3iiZMSnumBBEyB6.png

Bug: 1093185
Change-Id: If5c3b664f3affc958cad4a0042a0d6552ed56180
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2597766Reviewed-by: Azeem Arshad <azeemarshad@chromium.org>
Commit-Queue: Nnamdi Theodore Johnson-kanu <tjohnsonkanu@google.com>
Cr-Commit-Position: refs/heads/master@{#838248}

This avoids destroying and recreating a border every animation tick as
well.  Hopefully this will reduce CPU usage a tiny bit.

Bug: 1114976
Change-Id: I3b2b29a2c8bd04fd7d38aaf69bb9988070826a11
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2596438
Commit-Queue: Peter Kasting <pkasting@chromium.org>
Commit-Queue: Scott Violet <sky@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Auto-Submit: Peter Kasting <pkasting@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838247}

This is a reland of 42509254

Original change's description:
> Fix ScrollNestedLocalNonFastScrollableDiv under ScrollUnification
>
> With Scroll Unification, in LayerTreeImpl::DidUpdateScrollOffset there
> are cases where a ScrollUpdate may be mutating a ScrollNode that:
> - has a MainThreadScrollingReason - the main thread needs to repaint it
> each time we scroll.
> - it isn't composited
>
> In this case, when ScrollNode is mutated we don't want to make a
> corresponding change in the associated TransformNode. Instead, we want
> to SetNeedsCommit so that the changed offset in the ScrollNode is
> committed to the main thread, then blink regenerates its side of the
> property trees and cc::Layers.
>
> In this test, we hit the first case. In order for the test to pass we
> need to wait for the main thread to sync.
>
> Bug: 1086625
> Change-Id: Icdc3eb5cc786ebdecd516f3cb0fd58d47077d6ec
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2585765
> Reviewed-by: Ken Buchanan <kenrb@chromium.org>
> Reviewed-by: David Bokan <bokan@chromium.org>
> Commit-Queue: Liviu Tinta <liviutinta@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#836593}

Bug: 1086625
Change-Id: I053f4eabd5bc62e0959274236919cd53293315ce
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2591088Reviewed-by: Ken Buchanan <kenrb@chromium.org>
Reviewed-by: David Bokan <bokan@chromium.org>
Commit-Queue: Liviu Tinta <liviutinta@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838246}

Bug: 1152279
Change-Id: Ib3ba70a5b7ebb25bed2faada69c48f300b0ff18a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2595462Reviewed-by: Filip Gorski <fgorski@chromium.org>
Commit-Queue: Yi Gu <yigu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838245}

TBR=kinuko@chromium.org

Bug: 1156939
Change-Id: I23443fc4461dc61223fef249bac109a8b7e8183e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2586233Reviewed-by: Aaron Leventhal <aleventhal@chromium.org>
Reviewed-by: Dominic Mazzoni <dmazzoni@chromium.org>
Commit-Queue: Aaron Leventhal <aleventhal@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838244}

This cl add strings which will be used by grid tabs when
IncognitoModeAvailability policy is set to disabled/force.
Mock: https://docs.google.com/presentation/d/19BYpNAMopjUnxVdI3Gc4WcpJ6IVAZ_kw_4CCNs3tP3A/edit#slide=id.g5412b41971_0_147

Submit this cl first because it contains string changes.

The follow up cls are:
1. Create empty tab grid with enterprise message for normal tab grid,
incognito tab grid and the recent tabs grid.
2. Prevent the access to certain tab grid and using empty tab grid when
incognito mode is disabled/forced.
3. Add click action on the "learn more" in-text link.
4. Disable new tab button in certain mode.

Bug: 1148888

Change-Id: I8ba847935ed0228b45c0125f7f2d590e276daa3d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2595591
Commit-Queue: Tina Wang <tinazwang@chromium.org>
Reviewed-by: edchin <edchin@chromium.org>
Reviewed-by: Gauthier Ambard <gambard@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838243}

This is part of the effort to remove usages of cr.js and eliminate
dependencies to JS Compiler's --chrome_pass flag.

Bug: 1028829
Change-Id: Ic963a5dbfe0c18f0428e4432b2e6db3fb930a081
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1938074
Commit-Queue: John Lee <johntlee@chromium.org>
Auto-Submit: dpapad <dpapad@chromium.org>
Reviewed-by: John Lee <johntlee@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838242}

This is a purely refactoring change to set up ChromeVox to use all
possible intents. IntentHandler, the newly introduced class, will
cleanly process all intents, along with whatever computed metadata is
needed (e.g. EditableLine).

R=akihiroota@chromium.org, nektar@chromium.org

AX-Relnotes: n/a
Test: existing browser_tests --gtest_filter=ChromeVoxEditing*.*
Change-Id: I74d9454794fc6a40075bbf56fb0037b607c2b974
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2594240
Commit-Queue: David Tseng <dtseng@chromium.org>
Reviewed-by: Akihiro Ota <akihiroota@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838241}

* This patch creates empty targets in the payments component which will
  contain the minimal dependencies needed by downstream GPay code.
* This allows downstream to depend on these targets before things are
  moved into it.

Bug: 1144488
Change-Id: Ib0d6a1e0ea9aae2988322e6632e890d5185e4b4a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2597559
Commit-Queue: Nick Burris <nburris@chromium.org>
Reviewed-by: Rouslan Solomakhin <rouslan@chromium.org>
Reviewed-by: Liquan (Max) Gu <maxlg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838240}

This CL is a prefactoring to other in-progress work on
TemplateURLService.

We discovered that TemplateURLService has both CanReplace() and
safe_for_autoreplace() with slightly different logic.

This CL goes all-in with safe_for_autoreplace():
 - It marks Play API engines as NOT safe for autoreplace. It adds an
   UMA to be recorded so we know when the migration is complete.
 - It adds an additional constraint on auto-generated engines that they
   also leave prepopulated engines alone. (Debatable, may change).
 - It also explicitly forbids deleting the DSE during de-duplication,
   but it DOES allow deleting other pre-populated engines in favor of
   policy, extension, or Play API engines, which fixes 953711.

The full design is in:
go/chrome-search-engines-2020-improvements

See Heading: 6. Ranking, and simplifying CanReplace() vs.
safe_for_autoreplace() - (pkasting@'s idea)

Bug: 1022775, 953711
Change-Id: I217bbfc9010fae1c3546e1ffd061ff415e9674dc
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2589638Reviewed-by: Pavel Yatsuk <pavely@chromium.org>
Reviewed-by: Peter Kasting <pkasting@chromium.org>
Commit-Queue: Tommy Li <tommycli@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838239}

A Trust Tokens issuance request is a request created by
  fetch(..., {trustToken: {type: 'token-request'}})

With the new "platform-provided trust tokens" expansion of the Trust
Tokens feature, the request's destination origin can provide
configuration indicating that the browser should divert the origin's
Trust Tokens issuance requests to system-local providers, on platforms
that support this behavior (currently Android). [1]

In this case, there's no need to send an HTTP request directly to the
destination URL provided in the fetch() call, so we return early via
URLLoader::OnComplete(URLLoaderCompletionStatus). We provide a
non-net::OK error code to this URLLoader::OnComplete call since we won't
receive an HTTP response from the request's destination.

Since this case is semantically a success, we want the caller to be able
to distinguish it from errors encountered during the Trust Tokens
operation. This CL updates trust_token_to_mojom.cc to throw a
NoModificationAllowedError, instead of OperationError, when operations
are successfully diverted locally.

It might make more sense from an API design perspective to resolve the
promise, instead of rejecting it, in this case. This would require some
refactoring in the loader (or the willingness to resolve a promise from
FetchManager::Loader::DidFail!). For now, returning a distinct exception
at least allows callers to distinguish this success case from the
feature's failure cases.

I've updated the user guide to mention this behavior. [2]

[1]: https://bit.ly/platform-provided-trust-tokens
[2]: https://bit.ly/trust-token-prototype-user-guide

R=yhirano
CC=zhengwe@google.com

Test: The child CL adds browser tests covering this path
Bug: 1154847
Change-Id: Ie56ddd0aaf544289184f7d0b47f525fcb654877b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2585792
Commit-Queue: David Van Cleve <davidvc@chromium.org>
Reviewed-by: Yutaka Hirano <yhirano@chromium.org>
Cr-Commit-Position: refs/heads/master@{#838238}