GitLab

When launching (online enrolled) demo sessions with no connectivity, ARC
provisioning for the demo mode public account will timeout (it will
fail to get the account auth token), and restart the public session.

This CL adds a fallback for demo session when the auth token for the
demo account cannot be fetched - it continues ARC provisioning with
offline demo mode provisioning.

BUG=875504
TEST=Start demo session with no network - ARC successfully provisions,
  and the session is not forcibly restarted

Change-Id: I76fe36cf939a5ac8d34b89f821fc142d2618d7d4
Reviewed-on: https://chromium-review.googlesource.com/1180415
Commit-Queue: Toni Baržić <tbarzic@chromium.org>
Reviewed-by: Yusuke Sato <yusukes@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585308}

The compositor updates the viewports and buffers accordingly per
RenderLoop request. VrShellGl still inherits from RenderLoop, but it no
longer accesses the Ui, but through a new interface suited for it.

VrShellGl also implements the new SchedulerDelegate interface, which is
in charge of making draw calls to the RenderLoop from different signals.

Bug: 875291
Cq-Include-Trybots: luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:linux_vr;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel
Change-Id: I68c724145cdf8c4d63e90856f59bd07994cd9f1b
Reviewed-on: https://chromium-review.googlesource.com/1181949
Commit-Queue: Aldo Culquicondor <acondor@chromium.org>
Reviewed-by: Michael Thiessen <mthiesse@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585307}

XR devices and sessions cannot be obtained without a physical device
connected to the system or a system to mock out the backend APIs. As
setting up such as system would take a lot of work, it is unlikely to be
worthwhile while the spec is still undergoing so much change. Therefore,
the incorrect calls to get device and session have been removed, and can
be written correctly at a later date.

Bug: 870602
Change-Id: I0c9f512605d17dc3df890d699da016a8bc1e1c0c
Reviewed-on: https://chromium-review.googlesource.com/1185285
Commit-Queue: Anna Offenwanger <offenwanger@chromium.org>
Reviewed-by: Brandon Jones <bajones@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585306}

As part of https://chromium-review.googlesource.com/c/chromium/src/+/1183892
I wired up what was OutputProtection. That means these tests now pass.

BUG=678687
TEST=test only change

Change-Id: Ie68291b08973953e5d3d238ce87bb6d552a1b50e
Reviewed-on: https://chromium-review.googlesource.com/1185908Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Commit-Queue: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585305}

When we activate the virtual keyboard from the shelf in laptop mode, we
have to do two things:

1) Enable the accessibility keyboard.
2) Change the keyset to be the requested keyset.

When we close the keyboard, we have to undo steps 1 and 2. Currently,
it's indeterminate as to what order these two steps happen because
step 2 is a mojo call.

This became a problem after we launched the new keyboard UI, because
the new keyboard UI breaks when step 1 is undone before step 2 (i.e.
we disable the keyboard, but then the keyset changes). We change the
code so that we always disable the keyboard before resetting the keyset.

TBR=shuchen@chromium.org

Bug: 875456
Change-Id: Iaaa583851c7b63023a508f967f13f3d9aba42415
Reviewed-on: https://chromium-review.googlesource.com/1184588Reviewed-by: Darren Shen <shend@chromium.org>
Commit-Queue: Darren Shen <shend@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585304}

Change-Id: I28f66dbd11c0a9cd6ab47019503fc4dbb17847a5
Reviewed-on: https://chromium-review.googlesource.com/1185638Reviewed-by: Zhongyi Shi <zhongyi@chromium.org>
Commit-Queue: Ryan Hamilton <rch@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585303}

BUG=876369

Change-Id: I3daff1c0fbec9568edeb987b2f67a8ab168922a3
Reviewed-on: https://chromium-review.googlesource.com/1185912Reviewed-by: Matthew Jones <mdjones@chromium.org>
Commit-Queue: Ted Choc <tedchoc@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585302}

Add GPU process pid to the list of processes where tracing is enabled.

Change-Id: Ic401ee4ea7d218cad36f687be226d092cd4e54a0
Reviewed-on: https://chromium-review.googlesource.com/1182304
Commit-Queue: Alexei Filippov <alph@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585301}

1. Launch the Highlights app with AppLoadService (which is a wrapper
   of UnpackedInstaller).

2. Observing session state should be a cleaner solution, because we
   don't have to launch it from two places (user_session_manager for
   regular login, and chrome_session_manager for crash/restart).
   Launching it from two places are okay because we already have
   demo-mode related code in both places, but it may not be necessary
   for the Highlights app.

Bug: 819404
Change-Id: I3879df729f02c2477ca25511947b23ebe26e5ed1
Reviewed-on: https://chromium-review.googlesource.com/1176183
Commit-Queue: Wenzhao (Colin) Zang <wzang@chromium.org>
Reviewed-by: Michael Giuffrida <michaelpg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585300}

Rather than only displaying hosts, display the actual URLPatterns
granted to an extension. Given this is a power-user surface, we can be
a little less concerned with the technicality shown. Additionally,
since we don't use a full string (e.g., "Read and change your data on
all google.com sites"), we need to have the differentiation of which
hosts the extension is granted, such as whether it includes subdomains.

Also allow the user to enter patterns as an URLPattern (thus including
subdomains if they so desire). Enforce that paths are always '*', since
we don't use path-based permissions. To make it easier on users, also
support just entering the host name (e.g., 'google.com'), which will be
expanded to '*://google.com/*'.

This also properly fixes the issue of clicking 'remove' or 'edit' on
existing host options.

Add JS tests for pattern validation, and update unittests to support
pattern-based host manipulation.

Bug: 844128

Change-Id: I599151bb6b89bdb113af4c39dcf0e21d152aa0fe
Reviewed-on: https://chromium-review.googlesource.com/1166596
Commit-Queue: Devlin <rdevlin.cronin@chromium.org>
Reviewed-by: Karan Bhatia <karandeepb@chromium.org>
Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585299}

It really should be called PrintHostMsg_DidStartPreview. Getting the
page count is usually one of the first steps in starting a print
preview, but the IPC message does a bit more than that now.

Change-Id: I001a66c169fa5cf42b238605d6067bf12d58962e
Reviewed-on: https://chromium-review.googlesource.com/1184317Reviewed-by: Rebekah Potter <rbpotter@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585298}

https://chromium.googlesource.com/webm/libvpx.git/+log/6c62530c666f..dbcb89be244e

$ git log 6c62530c6..dbcb89be2 --date=short --no-merges --format='%ad %ae %s'
2018-08-22 jianj Revert "vp8: Fix memory address overflow in decoder."
2018-08-22 jingning Set refresh_frame_context flag off in show_existing_frame mode
2018-08-21 jingning Drop empty line in vp9_get_compressed_data()
2018-08-21 jingning Allow codec to skip temporal filter for intermediate ARFs
2018-08-21 jingning Control reference frame refresh flags for USE_BUF_FRAME
2018-08-21 jingning Safely swap the show frame buffer pointer in show_existing mode
2018-08-21 jingning Skip loop filter operation in show_existing_frame mode
2018-08-21 jingning Point show frame buffer towards existing frame buffer
2018-08-21 jingning Skip frame encoding when show_existing_frame is on
2018-08-21 jingning Add USE_BUF_FRAME enum to FRAME_UPDATE_TYPE
2018-08-20 jingning Unify set_arf_sign_bias function
2018-08-20 jingning Remove unneeded frame_till_gf_update_due assignment
2018-08-20 jingning Add multi_layer_arf flag
2018-08-20 jingning Add a comment in init_gop_frames()
2018-06-12 supradeep.tr Loopfilter MultiThread Optimization
2018-08-17 jingning Skip frame bit-stream writing for show-existing frame
2018-08-17 jingning Support code show_existing_frame in bit-stream header
2018-08-17 jingning Refactor init_gop_frame()
2018-08-17 jingning Clean up var define in apply_temporal_filter()
2018-08-17 jingning Add inline to mod_index()
(...)

Created with:
  roll-dep src/third_party/libvpx/source/libvpx
R=johannkoenig@google.com
BUG=875626,875680

Change-Id: I4d395733d13462e248119791f9483396c0614f5b
Reviewed-on: https://chromium-review.googlesource.com/1185797Reviewed-by: Johann Koenig <johannkoenig@google.com>
Commit-Queue: Jerome Jiang <jianj@google.com>
Cr-Commit-Position: refs/heads/master@{#585297}

This CL changes StickyPositionScrollingConstraints to a struct type.
It is a POD type anyway, and this allows us to remove tons of
boilerplate getter/setters.

Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_slimming_paint_v2;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: Ie3bee752aceab1334919613df84e25167fb9c759
Reviewed-on: https://chromium-review.googlesource.com/1184231Reviewed-by: Philip Rogers <pdr@chromium.org>
Reviewed-by: Stephen McGruer <smcgruer@chromium.org>
Commit-Queue: Tien-Ren Chen <trchen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585296}

When dragging a tab out of a browser window (source window), the source
window scales down. The background behind the source window should be
the same blurred and darkened wallpaper as in overview.

See recorded video:
https://drive.google.com/open?id=1I_tXdQzUsiCwAqFikmZi9EywnUG1-wKl

Bug: 874266
Change-Id: Ie25dd3594fbcef37cda5c427e4daebf0ecffb1fd
Reviewed-on: https://chromium-review.googlesource.com/1185521Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Commit-Queue: Xiaoqian Dai <xdai@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585295}

- Remove underline
- Use default cursor since the button is not clickable when disabled.

Bug: 875800
Change-Id: Id01f15cbde8afeb7a0e5341e4e04fd91c3b00ea0
Reviewed-on: https://chromium-review.googlesource.com/1182451Reviewed-by: Scott Chen <scottchen@chromium.org>
Commit-Queue: Rebekah Potter <rbpotter@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585294}

This reverts commit a9ed46b7.

Reason for revert: ChromeBrowserMainBrowserTest.VariationsServiceStartsRequestOnNetworkChange started flaking after this CL.

BUG: 876861

Original change's description:
> Migrate ResourceRequestAllowedNotifier to NetworkConnectionTracker
> 
> A getter is used for NetworkConnectionTracker because some services that
> use ResourceRequestAllowedNotifier are initialized early in browser
> startup (e.g. VariationsService), and only perform the initialization
> of ResourceRequestAllowedNotifier later on the UI thread. The getter
> allows us to run get the connection tracker at that point so we don't get
> DCHECKs about being on the UI thread when running
> content::GetNetworkConnectionTracker().
> 
> This also moves the NetworkConnectionTracker in ios/ from BrowserState to
> ApplicationContext, which is available everywhere. It also matches non-IOS
> usage more closely, since we have it as a global there.
> 
> Bug: 868021
> Cq-Include-Trybots: luci.chromium.try:ios-simulator-full-configs;master.tryserver.chromium.mac:ios-simulator-cronet
> Change-Id: I130c6b47feb90f0f7f0776ccc65666414a1ae802
> Reviewed-on: https://chromium-review.googlesource.com/1180360
> Reviewed-by: Eugene But <eugenebut@chromium.org>
> Reviewed-by: Robert Sesek <rsesek@chromium.org>
> Reviewed-by: John Abd-El-Malek <jam@chromium.org>
> Reviewed-by: Robbie McElrath <rmcelrath@chromium.org>
> Commit-Queue: Clark DuVall <cduvall@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#584849}

TBR=jam@chromium.org,eugenebut@chromium.org,rsesek@chromium.org,rmcelrath@chromium.org,cduvall@chromium.org

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

Bug: 868021
Change-Id: I62cf40b6b4661d1d93b385dd6502402f547aacc5
Cq-Include-Trybots: luci.chromium.try:ios-simulator-full-configs;master.tryserver.chromium.mac:ios-simulator-cronet
Reviewed-on: https://chromium-review.googlesource.com/1185685Reviewed-by: Sky Malice <skym@chromium.org>
Commit-Queue: Sky Malice <skym@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585293}

Bug: 864826
Change-Id: Ia594cc934f7c66121ff84660bd1039c5e05213d1
Reviewed-on: https://chromium-review.googlesource.com/1176463
Commit-Queue: Zach Helfinstein <zhelfins@chromium.org>
Reviewed-by: David Tseng <dtseng@chromium.org>
Reviewed-by: Katie Dektar <katie@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585292}

This roll has only one change as recorded in
     https://chromium.googlesource.com/chromium/deps/icu/+/a191af9d .

TBR=gsathya@chromium.org

Bug: chromium: 870338
Test: See the bug
Change-Id: Iceaa255ede13ef3f9e94fdcf6ab74268c1940427
Reviewed-on: https://chromium-review.googlesource.com/1184508Reviewed-by: Jungshik Shin <jshin@chromium.org>
Commit-Queue: Jungshik Shin <jshin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585291}

IMAGE_THUNK_DATA.ForwarderString is 64-bit on 64-bit platforms.

../../base/win/pe_image.cc(422,36):  error: implicit conversion loses integer precision: 'ULONGLONG' (aka 'unsigned long long') to 'DWORD' (aka 'unsigned long') [-Werror,-Wshorten-64-to-32]
          RVAToAddr(name_table->u1.ForwarderString));
          ~~~~~~~~~ ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~
../../base/win/pe_image.cc(504,47):  error: implicit conversion loses integer precision: 'ULONGLONG' (aka 'unsigned long long') to 'DWORD' (aka 'unsigned long') [-Werror,-Wshorten-64-to-32]
                     RVAToAddr(name_table->u1.ForwarderString));
                     ~~~~~~~~~ ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~

BUG=588506

Change-Id: I9b6b05611bc3301cf9459f2bf7847975cc2b3c93
Reviewed-on: https://chromium-review.googlesource.com/1183852Reviewed-by: Bruce Dawson <brucedawson@chromium.org>
Commit-Queue: Will Harris <wfh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585290}

There's enough capacity in these two fleets to run the same test
suites. Run these test suites:
  webgl2_conformance_d3d11_validating_tests
  webgl2_conformance_gl_passthrough_tests
on these bots:
  Win10 FYI Exp Release (Intel HD 630)
  Win10 FYI Release (Intel HD 630)

Simplify the test suite configurations.

Bug: 876777
Change-Id: I02e0dd1e10b6572a87e9450fd32a5958840d87ea
Reviewed-on: https://chromium-review.googlesource.com/1185375Reviewed-by: Geoff Lang <geofflang@chromium.org>
Commit-Queue: Kenneth Russell <kbr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585289}

Bug: 876811
Change-Id: Iaf8508f2cf9bd5b7243ba85572f783b86787dc98
Reviewed-on: https://chromium-review.googlesource.com/1185735
Commit-Queue: Brandon Wylie <wylieb@chromium.org>
Reviewed-by: Sky Malice <skym@chromium.org>
Reviewed-by: Filip Gorski <fgorski@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585288}

Since we use this class for both MSKP and PDF data processing, rename
its name to MetafileSkia. Similarly, rename the util files and test
files accordingly.

TBR=jam@chromium.org,dcheng@chromium.org

Change-Id: I295a106e1f48cb0110c5d8eb8539d1080e689c05
Reviewed-on: https://chromium-review.googlesource.com/1184157Reviewed-by: Wei Li <weili@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
Commit-Queue: Wei Li <weili@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585287}

The goal is to make policy-set trust anchors available as early
as possible, instead of waiting for the user's private slot to be
initialized.

The following changes were made to achieve this:
- UserNetworkConfigurationUpdater makes server and CA certificates
  available to observers as soon as ONC policy comes in.
- UserNetworkConfigurationUpdater does not pass server and CA
  certificates to CertificateImporter for permanent import anymore.
  Only Client certificates are permanently imported from user policy.
- PolicyCertService (which observes for user-policy-set cert changes)
  makes all user-policy-set server and authority certificates available
  to NSS as temp certificates through TempCertsCacheNSS.

Planned follow-up work:
The policy-set certs are not displayed in the Certificates settings
view. A follow-up CL (CL:916681) will add support for that.

      browser_tests --gtest_filter=*PolicyProvided*

Bug: 787602
Test: unit_tests --gtest_filter=*NetworkConfigurationUpdater* &&
Change-Id: I71aa8f1f91849b0b0ad24eb81ce3d1863fe7959c
Reviewed-on: https://chromium-review.googlesource.com/873810
Commit-Queue: Pavol Marko <pmarko@chromium.org>
Reviewed-by: Maksim Ivanov <emaxx@chromium.org>
Reviewed-by: Matt Mueller <mattm@chromium.org>
Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585286}

Bug: b/112564431
Test: Manual Test
Change-Id: I6750c3d4c8727948165bf821b417cb47d5db0427
Reviewed-on: https://chromium-review.googlesource.com/1185862Reviewed-by: Xiaohui Chen <xiaohuic@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Commit-Queue: Yue Li <updowndota@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585285}

Bug: 852577
Change-Id: Ibad70134b5f43877a39ef48e4a6fc99546d7ca3d
Reviewed-on: https://chromium-review.googlesource.com/1182288
Commit-Queue: Jian Li <jianli@chromium.org>
Reviewed-by: Peter Williamson <petewil@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585284}

Do not process pending request if initialize call is come with not inited
 dictionaries.

R=groby@chromium.org

Bug: 850424
Change-Id: I0dd7213f9865b2c5949fd6367c524edfd09febf1
Reviewed-on: https://chromium-review.googlesource.com/1085449
Commit-Queue: Rachel Blum <groby@chromium.org>
Reviewed-by: Rachel Blum <groby@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585283}

Launch was approved.

Bug: 752657
Change-Id: I01ac04040b8b928c27512d203667dda234588153
Reviewed-on: https://chromium-review.googlesource.com/1184659Reviewed-by: Scott Violet <sky@chromium.org>
Commit-Queue: Darren Shen <shend@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585282}

This reverts commit bc4a7725.

Reason for revert: Speculating that this may cause flaky test crashes: https://crbug.com/876903

Original change's description:
> CORB shouldn't make an exception for plugins, unless Flash actually runs
> 
> Overview
> ========
> 
> Flash has its own CORS-like mechanism (crossdomain.xml-based) and
> therefore CORB (Cross-Origin Read Blocking) cannot be enforced for
> requests initiated by Flash.
> 
> This CL avoids making an exception for plugins, unless the given
> renderer process is actually hosting a Flash plugin (and is therefore
> capable of proxying network requests on behalf of Flash).  This
> means that the exception won't take place unless the user has
> approved running Flash (via click-to-play / content settings /
> enterprise policy - see the bug for more details).
> 
> 
> Details
> =======
> 
> This CL introduces a global set that stores process IDs of renderers
> that host Flash.  This set lives either in the NetworkService process or
> (if NetworkService feature is disabled) in the IO thread of the browser
> process.  In both cases the global set is implemented and exposed by new
> static methods of network::CrossOriginReadBlocking class.
> 
> The CL populates the global set from
> PluginServiceImpl::FindOrStartPpapiPluginProcess after all the security
> checks have been done and the plugin process is ready to be used or
> launched.
> 
> The CL consults the global set before deciding to make a CORB exception
> for a plugin request.  This is done from network::URLLoader (used if
> NetworkService feature is enabled) and from
> CrossSiteDocumentResourceHandler (used otherwise).
> 
> The CL removes items from the global set when RenderProcessHostImpl is
> destroyed.
> 
> 
> Bug: 874515
> Cq-Include-Trybots: luci.chromium.try:linux_mojo
> Change-Id: I50484807c921a4daea08be8a00c67a3cf9c82cf0
> Reviewed-on: https://chromium-review.googlesource.com/1178885
> Commit-Queue: Łukasz Anforowicz <lukasza@chromium.org>
> Reviewed-by: Nasko Oskov <nasko@chromium.org>
> Reviewed-by: John Abd-El-Malek <jam@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#585124}

TBR=nasko@chromium.org,jam@chromium.org,lukasza@chromium.org

Change-Id: I683e189587a2edfe8fbf516b757a6590c6441c9c
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 874515
Cq-Include-Trybots: luci.chromium.try:linux_mojo
Reviewed-on: https://chromium-review.googlesource.com/1185941Reviewed-by: Łukasz Anforowicz <lukasza@chromium.org>
Commit-Queue: Łukasz Anforowicz <lukasza@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585281}

I chatted with Shiba and Sebastian about this after looking at
screenshots of Android Messages inline reply. It seems like
600 is the right match with the MD2 spec.

R=stevenjb@chromium.org

Bug: None
Change-Id: I2e66180eb752049abf19452fd02cbe103cb516a4
Reviewed-on: https://chromium-review.googlesource.com/1185843Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Commit-Queue: Jeremy Klein <jlklein@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585280}

Cleaned up android video capture factory methods to have consistent
usage of camera API version.

Change-Id: I835ea6debd5fe5fcc758207355c7936f0dccd84c
Reviewed-on: https://chromium-review.googlesource.com/1185534
Commit-Queue: Angad Sidhu <angadsidhu@google.com>
Reviewed-by: Christian Fremerey <chfremer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585279}

In single process mash, gaia dialog widget goes through the
following stack:
  WindowServiceDelegateImpl::SetModalType ->
  wm::GetDefaultParent ->
  GetSystemModalContainer

It does not have a |transient_parent| and hence assumed to be
a user window and put into SystemModalContainer instead of
LockSystemModalContainer.

Since the dialog widget is created with a proper container parent,
the CL fixes the issue by re-using existing container parent in
GetSystemModalContainer.

Bug: 876329
Change-Id: Iaf38273e852738b7d258a82f76dae9952051a08d
Reviewed-on: https://chromium-review.googlesource.com/1185406
Commit-Queue: Xiyuan Xia <xiyuan@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585278}

CL 1174982 only fixes layout issues when the screen has two cutouts that
cancels each other when calculating the center of the safe area. We
still have layout issues in single cutout scenario. This CL fixes these
bugs:

* Fix the calculation of the safe area center. It should be
  (width + inset.left - inset.right) / 2 instead of
  width / 2 + inset.left - inset.right. Same thing for y coordinate
  calculation.
* The letterbox padding doesn't need to account for the safe insets
  since GetViewportScreenCenter() and GetViewportBounds() already
  account for that. Undoing the letterbox changes.
* The SystemUIRect already takes cutouts into consideration. Since we
  no longer include safe insets in the letterbox padding, this CL makes
  DesktopCanvas ignore safe insets during calculation when the
  SystemUIRect is provided so that we don't double-apply the adjustment.

Scenarios manually tested:
{no cutout, 1 cutout, 2 symmetrical cutouts on top and bottom} x
{portrait mode, landscape mode} x
{touch input mode, trackpad input mode} x
{desktop < screen (letterboxes showing),
 desktop >= screen (no letterbox)} x
{keyboard showing, keyboard not showing}

Bug: 831670
Change-Id: I0ad192b2a199c8c1d4d7740b52d4b074c3626640
Reviewed-on: https://chromium-review.googlesource.com/1185631Reviewed-by: Joe Downing <joedow@chromium.org>
Commit-Queue: Yuwei Huang <yuweih@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585277}

Approved Intent to Ship:
https://groups.google.com/a/chromium.org/d/msg/blink-dev/HSfL38bAexE/n9QuIIILDwAJ

BUG=861735

Change-Id: I3784f1a6b9c4f117563d273d5710cbe2784ac5ac
Reviewed-on: https://chromium-review.googlesource.com/1185725Reviewed-by: Jeremy Roman <jbroman@chromium.org>
Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585276}

R=thestig
BUG=876586

Change-Id: I97c9609bd49f034027f3eb18ae7f57fef14c2267
Reviewed-on: https://chromium-review.googlesource.com/1096401
Commit-Queue: Thomas Anderson <thomasanderson@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585275}

- Simplify ctors.
- Rewrite GetCurrentPrintPreviewStatus() as ShouldCancelRequest().

Change-Id: Ie4cbb4dc7a56c1310ec4768687e800f7f40fe61a
Reviewed-on: https://chromium-review.googlesource.com/1184507Reviewed-by: Rebekah Potter <rbpotter@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585274}

Bug: b/112310573
Test: Manual Test
Change-Id: Ib6ca27acd261437043ebce55c57124458f21425e
Reviewed-on: https://chromium-review.googlesource.com/1185367
Commit-Queue: Yue Li <updowndota@chromium.org>
Reviewed-by: Xiaohui Chen <xiaohuic@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585273}

Relanding a CL that was reverted because of a variable that was
unused in non-DCHECK builds and caused a compile failure.
Original CL: Reviewed-on: https://chromium-review.googlesource.com/1171482

This CL adds a feature to use THREAD_MODE_BACKGROUND_BEGIN instead of
THREAD_PRIORITY_LOWEST for ThreadPriority::BACKGROUND threads. This
lowers the disk and network I/O priority of the thread in addition to
the CPU scheduling priority. MSDN recommends using this setting for
threads that perform background work.
https://docs.microsoft.com/en-us/windows/desktop/api/processthreadsapi/nf-processthreadsapi-setthreadpriority

TBR=gab@chromium.org

Bug: 872820
Change-Id: I212fb77bb035088595b4944ce145a74d333c38eb
Reviewed-on: https://chromium-review.googlesource.com/1185529
Commit-Queue: François Doray <fdoray@chromium.org>
Commit-Queue: Gabriel Charette <gab@chromium.org>
Reviewed-by: Gabriel Charette <gab@chromium.org>
Reviewed-by: François Doray <fdoray@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585272}

This will avoid full manifest rebuilds for Chromium developers when
the manifest version changes.

Bug: 876717
Change-Id: I6d552ad03936fb07cf559d0f072508e5ea2184a4
Reviewed-on: https://chromium-review.googlesource.com/1185009
Commit-Queue: Philip Jägenstedt <foolip@chromium.org>
Reviewed-by: Robert Ma <robertma@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585271}

Couldn't reproduce the flakiness in those tests, so re-enabling with
committed interstitials, will revert this CL if the flakiness returns.
Also: This tests seem to be flaky in general, even without committed
interstitials, e.g., crbug.com/874959

Bug: 863272
Change-Id: I64b5291513041e4dd9da9fa6125785fe4727a6c7
Reviewed-on: https://chromium-review.googlesource.com/1178928
Commit-Queue: Carlos IL <carlosil@chromium.org>
Reviewed-by: Mustafa Emre Acer <meacer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585270}

This converts ash to using ServerRemoteViewHost now that it hosts the
WindowService. This also makes WindowServiceOwner create the WindowService
immediately.

BUG=876513
TEST=covered by tests

Change-Id: I7aba636c9e87db4ab08e1ccab97acecacdfc59f9
Reviewed-on: https://chromium-review.googlesource.com/1185796Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Commit-Queue: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#585269}