GitLab

Make setupConnection oneway/async. Add a ICallbackInt interface to
return the PID of the child process. The reply is then posted back to
launcher thread. Connection setup is already mostly async so only
involves moving related code to the reply callback.

One caveat is that getCallingPid no longer works and just returns 0
for oneway calls. The pid check is not strictly necessary, so dropped
the from setupConnection instead.

BUG=689758, 690118

Review-Url: https://codereview.chromium.org/2810433002
Cr-Commit-Position: refs/heads/master@{#463332}

BUG=627747

Review-Url: https://codereview.chromium.org/2814513002
Cr-Commit-Position: refs/heads/master@{#463331}

Since its always the default. Whenever there is a non-default blend
mode, render surface takes care of blending

BUG=709269
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2801383002
Cr-Commit-Position: refs/heads/master@{#463330}

A second SnackbarManager is used to display snackbars on top
of the bottom sheet.

BUG=704128

Review-Url: https://codereview.chromium.org/2806593004
Cr-Commit-Position: refs/heads/master@{#463329}

BUG=

Review-Url: https://codereview.chromium.org/2788733002
Cr-Commit-Position: refs/heads/master@{#463328}

No intended behavior change.

BUG=none

Review-Url: https://codereview.chromium.org/2810723003
Cr-Commit-Position: refs/heads/master@{#463327}

- Remove dependency on TestProject
- Support xctestrun file
- Support retries of xctests on devices

BUG=708859

Review-Url: https://codereview.chromium.org/2805133005
Cr-Commit-Position: refs/heads/master@{#463326}

In a previous mask tiling patch, there was a typo that incorrectly check
mask type when doing the tiled/non-tiled conversion.

This does not break anything as currently the flag is disabled. This
patch fixes this typo.

BUG=567293
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2809653002
Cr-Commit-Position: refs/heads/master@{#463325}

. Use SANDBOXED as the default argument to GpuProcessHost::Get(), since
  that's the most common use.
. Remove unused ash::GPUSupport::GetGpuProcessHandles().
. Remove unused content::GpuDataManager::GetGpuProcessHandles().
. Replace GpuProcessHost::GetProcessHandles() with GetHasGpuProcess(),
  since the only remaining user (RenderMessageFilter) cares only about
  the existence of a gpu process, and not about what those processes are.

BUG=417532, 709332

Review-Url: https://codereview.chromium.org/2810593003
Cr-Commit-Position: refs/heads/master@{#463324}

This CL adds ExecutionContext::from(ScriptState*) to eventually replace ScriptState::getExecutionContext. This is to remove a dependency to ExecutionContext from ScriptState.

BUG=682322

Review-Url: https://codereview.chromium.org/2805333002
Cr-Commit-Position: refs/heads/master@{#463323}

Adding user actions and a histogram so that, when the settings reset
prompt is not accepted, we can distinguish between cases when user
clicks the cancel button versus dismisses the prompt in some other
way.

BUG=

Review-Url: https://codereview.chromium.org/2800773003
Cr-Commit-Position: refs/heads/master@{#463322}

For mushrome, in order to have screen management in one process and
Ozone in another process we need a mojom::NativeDisplayDelegate and
implementation. The ForwardingDisplayDelegate will be created in ash and
will connect to mus-ws. In mus-ws ScreenManagerForwarding will own the
real NativeDisplayDelegate. A mojom::NativeDisplayObserver works in the
other direction.

This only implements the Ozone DRM parts of NativeDisplayDelegate. Most
of the other methods exist for the X11 implementation which isn't used.

BUG=706589

Review-Url: https://codereview.chromium.org/2805633004
Cr-Commit-Position: refs/heads/master@{#463321}

When running GetAllNodes, USS types post to the model thread to access
their processor/bridge. This means after they've gathered the nodes,
they need to return to the UI thread to give the results back to the
GetAllNodesRequestHelper.

Unfortunately I removed a wrapping BindToCurrentThread call in
https://codereview.chromium.org/2672493002/diff/80001/components/sync/driver/model_type_controller.cc
that did the jump back to the UI thread. This meant that non-UI USS
types (namely AUTOFILL) were updating GetAllNodesRequestHelper from a
different thread, through very non-thread safe code. To fix simply
added the BindToCurrentThread back.

Also added a added a thread checker to GetAllNodesRequestHelper in
hopes of stopping similar bugs from occurring again.

BUG=686172

Review-Url: https://codereview.chromium.org/2808113003
Cr-Commit-Position: refs/heads/master@{#463320}

We've been on git for a long time now, this code has been unused for years.
No intended behavior change.

BUG=none

Review-Url: https://codereview.chromium.org/2814503002
Cr-Commit-Position: refs/heads/master@{#463319}

A ConstrainedWebDialogBase creates and own a WebContent. For various
reasons, it stores a pointer to itself inside the WebContent. The
problem was that the lifetime of the WebContent may exceed the one of
ConstrainedWebDialogBase when |release_contents_on_close_| is set to
false. The pointer becomes invalid.

BUG=704327

Review-Url: https://codereview.chromium.org/2798583002
Cr-Commit-Position: refs/heads/master@{#463318}

Note that this only changes the icon for the mini infobar, not for
the entire feature (content settings, desktop UI, etc).

BUG=689992

Review-Url: https://codereview.chromium.org/2791093002
Cr-Commit-Position: refs/heads/master@{#463317}

If sticky position is unconstrained we don't need to create constraints for it,
or add it to viewport constrained objects (causing scroll on main), or repaint
on scroll.

BUG=704817
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2769353002
Cr-Commit-Position: refs/heads/master@{#463316}

BUG=none

Review-Url: https://codereview.chromium.org/2809443002
Cr-Commit-Position: refs/heads/master@{#463315}

https://screenshot.googleplex.com/sHy71kU48TQ

BUG=691108
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2795603002
Cr-Commit-Position: refs/heads/master@{#463314}

BUG=500085

Review-Url: https://codereview.chromium.org/2805923002
Cr-Commit-Position: refs/heads/master@{#463313}

R=dbeam@chromium.org
BUG=689356
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2805213003
Cr-Commit-Position: refs/heads/master@{#463312}

BUG=706931, 695577

Review-Url: https://codereview.chromium.org/2811793002
Cr-Commit-Position: refs/heads/master@{#463311}

When building content_unittests for Cast on Android with mojo_media_host
as "browser", linker complains media::MojoRendererService::Create() is
undefined. Add //media/mojo/services as a dependency when
enable_mojo_media is true to resolve the error.

BUG=Internal b/37086171

Review-Url: https://codereview.chromium.org/2807743002
Cr-Commit-Position: refs/heads/master@{#463310}

The intent of the new algorithm is to allow resolution adjustment if the
resolution range allowed by constraints is a superset of the capabilities.
However, the actual capabilities are not currently known and, in practice,
resolution adjustment was allowed only when no constraints were specified.

This CL relaxes this rule to allow resolution adjustment by treating a max
constraint as if it were the actual capability. This should be updated so
that the actual native screen resolution is used as capability, but, in the
meantime, this fix is more compatible with existing applications that expect
the behavior of the old algorithm. Moreover, this patch does not affect spec
compliance since screen-resolution policies are not regulated by the spec as
long as the output resolution is within the constrained range.

BUG=709915

Review-Url: https://codereview.chromium.org/2800343005
Cr-Commit-Position: refs/heads/master@{#463309}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/af72fd8e..bd228187

Please follow these instructions for assigning/CC'ing issues:
https://github.com/v8/v8/wiki/Triaging%20issues

Please close rolling in case of a roll revert:
https://v8-roll.appspot.com/
This only works with a Google account.

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel

TBR=hablich@chromium.org,machenbach@chromium.org,kozyatinskiy@chromium.org,vogelheim@chromium.org

Review-Url: https://codereview.chromium.org/2806393002
Cr-Commit-Position: refs/heads/master@{#463308}

Both are needed for proper accelerate to software conversion.
getImageData wasn't calling needFinalizeFrame, so frames that only
called it, wouldn't be moved to software.
Also, when converting from accelerate, background-color/alpha wasn't
being properly propagated because we didn't updated compositing.

BUG=

Review-Url: https://codereview.chromium.org/2802353002
Cr-Commit-Position: refs/heads/master@{#463307}

The code was not handling a shipping option going away very well.

BUG=707240
TEST=browser_tests

Review-Url: https://codereview.chromium.org/2808883002
Cr-Commit-Position: refs/heads/master@{#463306}

BUG=652781

Review-Url: https://codereview.chromium.org/2802363002
Cr-Commit-Position: refs/heads/master@{#463305}

Ed kindly pointer out that If you quickly tap on the forward or back toolbar navigation
arrow, the App would crash. That was caused since the button doesn't get disabled  fast enough after there's no more pages left to navigate.

This CL fixes that crash.

BUG=683793

Review-Url: https://codereview.chromium.org/2809563002
Cr-Commit-Position: refs/heads/master@{#463304}

BUG=skia:6465

Review-Url: https://codereview.chromium.org/2798413002
Cr-Commit-Position: refs/heads/master@{#463303}

Stop using kSingleProcess, which does not mean what we think it means.
Pass the correct webview process type to breakpad, so that microdumps
contain the correct information.

BUG=708773

Review-Url: https://codereview.chromium.org/2806733002
Cr-Commit-Position: refs/heads/master@{#463302}

Remove references to //ash/common, rewrite the top-level README.md, and
remove unnecessary DEPS.

BUG=none
TEST=checkdeps passes

Review-Url: https://codereview.chromium.org/2811643003
Cr-Commit-Position: refs/heads/master@{#463301}

The context menus API is another that violates the schema is defines
when calling sendRequest() through its custom bindings. Allow these
as well, similar to how we allow them when an API updates its
arguments post validation.

The context menus API also has functions defined as properties on object
parameters, and serializes these into dictionaries. Match the the
serialization into a dictionary, but don't serialize properties of the
function.

Add an end-to-end test for the contextMenus API with native bindings.

BUG=653596

Review-Url: https://codereview.chromium.org/2805123002
Cr-Commit-Position: refs/heads/master@{#463300}

https://skia.googlesource.com/skia.git/+log/f9aa9e51691d..b8ab7f7634fc

$ git log f9aa9e516..b8ab7f763 --date=short --no-merges --format='%ad %ae %s'
2017-04-10 borenet Fix BundleRecipes depending on Release-Shared build
2017-04-10 herb Add multi-stop SkJumper stage.
2017-04-10 bsalomon Move ref counting out of GrProcessor and into subclasses.

Created with:
  roll-dep src/third_party/skia


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

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls


CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel
TBR=mtklein@chromium.org

Change-Id: I097954c955bc68e5c33c914740ee043638b16df5
Reviewed-on: https://chromium-review.googlesource.com/472927Reviewed-by: Skia Deps Roller <skia-deps-roller@chromium.org>
Commit-Queue: Skia Deps Roller <skia-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#463299}

Build: https://build.chromium.org/p/chromium.infra.cron/builders/rebaseline-o-matic/builds/658860

https://chromium.googlesource.com/chromium/src/+/112094796a99f

BUG=703971
TBR=tkent@chromium.org

Review-Url: https://codereview.chromium.org/2808903002 .
Cr-Commit-Position: refs/heads/master@{#463298}

The base::IsDirectoryEmpty() method returns true when the directory
doesn't exist or the directory is empty. We need to use
base::DirectoryExists() together with base::IsDirectoryEmpty() to
differentiate these two situations.

BUG=40407, 179576

Review-Url: https://codereview.chromium.org/2806873002
Cr-Commit-Position: refs/heads/master@{#463297}

Background: In order to tell which tests to download -expected.txt files for and which tests to update expectations for, wpt-update-expectations checks the type of the test. Previously TestParser was used to do this, but I want to get rid of TestParser.

This CL replaces TestParser by checking whether a test is a ref tests by calling port.reference_files, which in turn checks the MANIFEST.json for tests in external/wpt.

Review-Url: https://codereview.chromium.org/2805563002
Cr-Commit-Position: refs/heads/master@{#463296}

- Replace LockStateController::OnLockScreenHide with mojo call
  SessionController.RunUnlockAnimation so that it works on mash;
- Implement lock state change detection in SessionController
  and remove relevant part in LoginLockStateNotifier that needs
  to talk directly to Shell;

BUG=679450

Review-Url: https://codereview.chromium.org/2801333002
Cr-Commit-Position: refs/heads/master@{#463295}

The ContentScriptApiTest.ContentScriptOtherExtensions has the following
flow:

1. Register the inject.js content script to run at 'document_end'
   time.
2. Load iframe_content.{html,js} in an iframe.
3. On DOMContentLoaded, set a 10ms timer to report the iframe's contents
   to the test harness.
4. In 'inject.js', rewrite the page's content.
5. In the test harness, verify that the page reports the content mutated
   by inject.js.

The problem is that the timer registered in step 3 races with the
task[1] triggering the injection script since there's no explicit
synchronization between the two. Currently this works by accident, but
an upcoming change[2] in the Blink Scheduler will cause the test to
actually become flaky.

This change fixes the test by making the iframe report its contents
based on the load event rather than DOMContentLoaded. Since this is
guaranteed to happen after the injection has finished, there's no
possibility of out-of-order execution.

BUG=696001,701223

[1] blink::FrameLoader::finishedParsing
    => extensions::ScriptInjection::InjectJs
       => blink::SuspendableScriptExecutor::create
[2] https://bugs.chromium.org/p/chromium/issues/detail?id=701223#c4

Review-Url: https://codereview.chromium.org/2800103002
Cr-Commit-Position: refs/heads/master@{#463294}

Old name was presumably used to match was_within_same_page IPC, but
SameDocument name better reflects the the type of the navigation
(navigation did not change the document object).

This change will make naming more consistent with the rest of Chromium
code.

BUG=695189

Review-Url: https://codereview.chromium.org/2807723002
Cr-Commit-Position: refs/heads/master@{#463293}