GitLab

BlinkPlatformImpl currently creates an instance of WebThreadImpl when blink
creates a thread using 'createThread()'. When some code runs in the newly
created thread and tries to get hold of the WebThread that represents the
current thread using 'currentThread()', BlinkPlatformImpl creates a new
instance of a WebThreadImplForMessageLoop and returns that. For subsequent
calls to 'currentThread()', this same WebThreadImplForMessageLoop is
returned. So BlinkPlatformImpl ends up creating two WebThread instances
for each thread created in blink.

This patch changes this to only create a single instance of WebThread for
each thread created in blink. WebThreadImplForMessageLoop is no longer
necessary, so it is removed.

BUG=462067

Review URL: https://codereview.chromium.org/959803003

Cr-Commit-Position: refs/heads/master@{#318913}

BUG=none
R=thakis@chromium.org

Review URL: https://codereview.chromium.org/973073002

Cr-Commit-Position: refs/heads/master@{#318912}

The NSFont object could be nil, which would have a NULL font name, which when
converted to a std::string would cause a crash. The font name is already in the
FontDescriptor, so it's entirely unnecessary.

BUG=452651
R=avi@chromium.org

Review URL: https://codereview.chromium.org/969683003

Cr-Commit-Position: refs/heads/master@{#318911}

Before this change a malformed value of RemoteAccessHostUdpPortRange policy
(i.e. "123456-blah") was ignored and a default value was used.  Similarily for
the 3 third-party-auth-config policies (RemoteAccessHostTokenUrl,
...TokenValidationUrl and ...TokenValidationCertificateIssuer) we were falling
back to a secure default, but not reporting a policy error.  After this change
such malformed values will trigger an OnPolicyError callback.

Notes:
- Guaranteeing that PolicyWatcher always returns valid policy values, removes
  the need for a "rejecting" Me2MeHostAuthenticatorFactory.
- Moving PortRange and ThirdPartyAuthConfig to separate compilation units
  helps readability elsewhere + encourages better unit tests coverage.
- Initially I tried to wrap all policies in a new ChromotingPolicies class,
  but eventually went back to prevalidating and passing base::DictionaryValue.
  - Arguments for using ChromotingPolicies:
    - Helps avoid overtesting in policy_watcher_unittests.cc (i.e. helps focus
      the tests on a single policy value).
  - Arguments for using base::DictionaryValue:
    - Minimizes changes.
    - Keeps things simple (as opposed to having to introduce a custom
      equivalent of optional<T> [nothing similar present in Chromium AFAICT]).
  - Neutral:
    - Strong-typing of ChromotingPolicies didn't help readability as much as I
      expected and hoped for.

BUG=427513
TEST=remoting_unittests

Review URL: https://codereview.chromium.org/966433002

Cr-Commit-Position: refs/heads/master@{#318910}

MemoryPressureListener.

BUG=447208
R=ellyjones@chromium.org

Review URL: https://codereview.chromium.org/971413002

Cr-Commit-Position: refs/heads/master@{#318909}

This cl reverts a change made in

https://codereview.chromium.org/952823004

Since this code is controlling the display of the zoom icon (not
the bubble) we should not hide when not at default zoom level.

BUG=462482

Review URL: https://codereview.chromium.org/968213003

Cr-Commit-Position: refs/heads/master@{#318908}

EnhancedBookmarksModel(int) is never used and should be removed.

BUG=458632

Review URL: https://codereview.chromium.org/968383002

Cr-Commit-Position: refs/heads/master@{#318907}

Speculative fix for unit tests failure -- sometimes calling -removeVisits will
call back into history.js with -deleteComplete before |deleteCompleteCallback_|
has been set.

BUG=452893

Review URL: https://codereview.chromium.org/970213002

Cr-Commit-Position: refs/heads/master@{#318906}

BUG=462153

Review URL: https://codereview.chromium.org/969383004

Cr-Commit-Position: refs/heads/master@{#318905}

TEST=./tools/valgrind/chrome_tests.sh -t cmdline -- echo "hello"

Review URL: https://codereview.chromium.org/966363003

Cr-Commit-Position: refs/heads/master@{#318904}

Also adds a simple service to allow layout tests to test all this.

This is part of a series of changes:
[1/5] Blink side changes in https://codereview.chromium.org/924983002/ that expose a needed v8 context
[2/5] https://codereview.chromium.org/921013002/ allow MessagePort to send messages as values
[3/5] https://codereview.chromium.org/944443003/ which causes the send_messages_as_values flag to be propagated
[4/5] This CL
[5/5] https://codereview.chromium.org/940423004/ adds layout tests to blink to test these new features

BUG=426458

Review URL: https://codereview.chromium.org/938403005

Cr-Commit-Position: refs/heads/master@{#318903}

Also remove nduca@ as he's moved on to bigger and better things.

TBR=nduca@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/961003003

Cr-Commit-Position: refs/heads/master@{#318902}

Cherry pick the HarfBuzz commit 8ac345e5c0ed0aad6547592ea0839aabfb4ba980
as it fixes a set of serious shaping regressions in the core text shaper
on Mac OS. Cherry picking as opposed to rolling to allow for a clean DI.

BUG=459043
TBR=behdad@chromium.org

Review URL: https://codereview.chromium.org/971153002

Cr-Commit-Position: refs/heads/master@{#318901}

This is going to help the PermissionService to have its
RequestPermission returing the expected value.

Internally, PermissionContextBase is using ContentSetting but the public
API is using PermissionStatus.

BUG=432978

Review URL: https://codereview.chromium.org/955383003

Cr-Commit-Position: refs/heads/master@{#318900}

It fixes the following clang format message:

In file included from ../../ui/ozone/platform/dri/gbm_buffer.cc:14:
../../ui/ozone/platform/dri/gbm_device.h:18:3: error: [chromium-style] Classes that are ref-counted should have destructors that are declared protected or private.
  ~GbmDevice() override;
  ^
../../ui/ozone/platform/dri/gbm_device.h:14:19: note: [chromium-style] 'GbmDevice' inherits from 'ui::DrmDevice' here
class GbmDevice : public DrmDevice {
                  ^
../../ui/ozone/platform/dri/drm_device.h:41:32: note: [chromium-style] 'DrmDevice' inherits from 'base::RefCountedThreadSafe<DrmDevice>' here
class OZONE_EXPORT DrmDevice : public base::RefCountedThreadSafe<DrmDevice> {

BUG=none

Review URL: https://codereview.chromium.org/975773002

Cr-Commit-Position: refs/heads/master@{#318899}

BUG=361761
TEST=no visible change

Review URL: https://codereview.chromium.org/975533002

Cr-Commit-Position: refs/heads/master@{#318898}

BUG=345874

Review URL: https://codereview.chromium.org/974903002

Cr-Commit-Position: refs/heads/master@{#318897}

The need to make the distinction between the two types of infobars has
disappeared with the new design of the infobars.

BUG=None

Review URL: https://codereview.chromium.org/975663002

Cr-Commit-Position: refs/heads/master@{#318896}

WebRTC 8548:8573
Changes: https://chromium.googlesource.com/external/webrtc/trunk/webrtc.git/+log/1e5b7f7..c9dfd1c

Libjingle 8550:8571
Changes: https://chromium.googlesource.com/external/webrtc/trunk/talk.git/+log/bbfd399..3b22f07

TBR=

Review URL: https://codereview.chromium.org/969413002

Cr-Commit-Position: refs/heads/master@{#318895}

BUG=424075
TEST=manual

Review URL: https://codereview.chromium.org/957753002

Cr-Commit-Position: refs/heads/master@{#318894}

Also fix some mislabelled entries in the histogram enum.

Make the declaration site explicit to decrease the likelihood of such
omissions/mislabels happening again in the future.

BUG=None

Review URL: https://codereview.chromium.org/968253002

Cr-Commit-Position: refs/heads/master@{#318893}

We shouldn't hard code out/Release (since with GN the path could be different) or look for the checkout directory (since with swarming it doesn't exist).

This was needed with ui_tests or pyauto, but not with the current browser_tests binaries.

BUG=463180

Review URL: https://codereview.chromium.org/969313002

Cr-Commit-Position: refs/heads/master@{#318892}

Follows:
https://codereview.chromium.org/968943004/
https://chrome-internal-review.googlesource.com/#/c/203597/
BUG=428903
TBR=jbudorick

Review URL: https://codereview.chromium.org/964363003

Cr-Commit-Position: refs/heads/master@{#318891}

V8 caches all function templates internally for the lifetime of a web page in an
unbounded cache. Thus, creating a new function template for each call to a
method of injected object caused JS heap memory to grow linearly.

This is similar to a fix for Pepper Plugin objects (https://codereview.chromium.org/785213002/)

BUG=462664

Review URL: https://codereview.chromium.org/969353004

Cr-Commit-Position: refs/heads/master@{#318890}

Whether to actually cache the template depends on the use case, so only
adding a warning in a comment.

BUG=463487

Review URL: https://codereview.chromium.org/974863002

Cr-Commit-Position: refs/heads/master@{#318889}

It enables renderers without DumpRenderTree to trigger Blink GCs.

BUG=438074, 420515

Review URL: https://codereview.chromium.org/964283002

Cr-Commit-Position: refs/heads/master@{#318888}

See https://codereview.chromium.org/968663004/
BUG=377250

Review URL: https://codereview.chromium.org/965903003

Cr-Commit-Position: refs/heads/master@{#318887}

This API was missed in the CL to add missing APIs to histograms.
https://codereview.chromium.org/938283004/

This change also tweaks a comment that caused pretty_print.py to
complain.

BUG=

Review URL: https://codereview.chromium.org/970163002

Cr-Commit-Position: refs/heads/master@{#318886}

http://build.chromium.org/p/chromium.lkgr/builders/UBSan%20vptr%20Release/builds/1296

BUG=463464
TBR=kbr@chromium.org

Review URL: https://codereview.chromium.org/974843002

Cr-Commit-Position: refs/heads/master@{#318885}

facilitates WebUI creation.

All parts of WUG are implemented:
* Base classes for view (View) and view-model (ViewModel)
* WebUIView -- specialization of View for use in WebUI.
* View's CC, HTML and JS code generator
* Templates for GYP and GN which help to generate code and
  produce components from generated code.

Tests to follow.

TEST=none
TBR=avi@chromium.org,mark@chromium.org
BUG=459230

Review URL: https://codereview.chromium.org/928163002

Cr-Commit-Position: refs/heads/master@{#318884}

Highlights:
 - mojo::ChannelInit has been replaced with
   content::ChannelInit.
 - ScopedIPCSupport has been added so Mojo consumers can
   ensure the EDK is initialized.
 - single process mode now uses some evil tricks to get
   child threads to create their mojo client channels on
   the browser I/O thread.
 - several Android bits adapted to new interfaces
 - a number of tests have been adapted to work properly
   in spite of unconventional process arrangements

BUG=None

Review URL: https://codereview.chromium.org/954643002

Cr-Commit-Position: refs/heads/master@{#318883}

BUG=463459

Review URL: https://codereview.chromium.org/973813002

Cr-Commit-Position: refs/heads/master@{#318882}

TBR=machenbach

BUG=463436

Review URL: https://codereview.chromium.org/973063002

Cr-Commit-Position: refs/heads/master@{#318881}

Revert of Update V8 to version 4.3.15. (patchset #1 id:1 of https://codereview.chromium.org/969053002/)

Reason for revert:
Speculative revert for issue 463436.

Original issue's description:
> Update V8 to version 4.3.15.
>
> Summary of changes available at:
> https://chromium.googlesource.com/v8/v8/+log/c791d841..155278d6
>
> Please follow these instructions for assigning/CC'ing issues:
> https://code.google.com/p/v8-wiki/wiki/TriagingIssues
>
> TBR=machenbach@chromium.org
>
> Committed: https://crrev.com/32e6c72188d6b0b948d668afdc1cb209d44c2e70
> Cr-Commit-Position: refs/heads/master@{#318755}

TBR=v8-autoroll@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/975713002

Cr-Commit-Position: refs/heads/master@{#318880}

Revert of Update V8 to version 4.3.16. (patchset #1 id:1 of https://codereview.chromium.org/974813002/)

Reason for revert:
Speculative revert for issue 463436.

Original issue's description:
> Update V8 to version 4.3.16.
>
> Summary of changes available at:
> https://chromium.googlesource.com/v8/v8/+log/155278d6..ea02c333
>
> Please follow these instructions for assigning/CC'ing issues:
> https://code.google.com/p/v8-wiki/wiki/TriagingIssues
>
> TBR=machenbach@chromium.org
> NOTRY=true
>
> Committed: https://crrev.com/5d61006f40e3f14bbf8f2c60f698b051f8f84fd9
> Cr-Commit-Position: refs/heads/master@{#318874}

TBR=v8-autoroll@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/978433002

Cr-Commit-Position: refs/heads/master@{#318879}

BUG=461604,435358
TEST=manual

Review URL: https://codereview.chromium.org/969403002

Cr-Commit-Position: refs/heads/master@{#318878}

BUG=None

Review URL: https://codereview.chromium.org/969353005

Cr-Commit-Position: refs/heads/master@{#318877}

We adjusted the leaderd dbus API to pass back an ObjectPath instead of
a string. Adjust the chromium bindings to match that.

BUG=453872

Review URL: https://codereview.chromium.org/972983003

Cr-Commit-Position: refs/heads/master@{#318876}

Component build was broken since 4a84927b, where
chromeos::device_event_log was moved  to components.

TBR=reillyg@chromium.org,derat@chromium.org
BUG=none

Review URL: https://codereview.chromium.org/971363002

Cr-Commit-Position: refs/heads/master@{#318875}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/155278d6..ea02c333

Please follow these instructions for assigning/CC'ing issues:
https://code.google.com/p/v8-wiki/wiki/TriagingIssues

TBR=machenbach@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/974813002

Cr-Commit-Position: refs/heads/master@{#318874}