GitLab

https://crrev.com/2558743002 fixed a bug that loading metric doesn't
work for webview traces. This patch re-enables page_cycler_v2 and
loading.mobile benchmarks on webview bots.

BUG=651188
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.perf:linux_perf_cq;master.tryserver.chromium.perf:mac_retina_perf_cq;master.tryserver.chromium.perf:winx64_10_perf_cq

Review-Url: https://codereview.chromium.org/2564773002
Cr-Commit-Position: refs/heads/master@{#437822}

Now that ContextLifecycleObserver clears its observing context after calling
contextDestroyed. So there is no need to explicitly call clearContext().

BUG=610176

Review-Url: https://codereview.chromium.org/2552703006
Cr-Commit-Position: refs/heads/master@{#437821}

This is a rename requested by https://codereview.chromium.org/2554813002/.

Previously ActiveDOMObject was used to implement hasPendingActivity() of DOM wrappers,
where the name "ActiveDOMObject" made sense. Then we moved hasPendingActivity() to
ActiveScriptWrappable. Now ActiveDOMObject is a class that just provides resume/suspend.
It would make more sense to rename it to SuspendableObject.

BUG=610176
TBR=jam@chromium.org

Review-Url: https://codereview.chromium.org/2567913002
Cr-Commit-Position: refs/heads/master@{#437820}

BUG=671459
TBR=avi@chromium.org

Review-Url: https://codereview.chromium.org/2565273002
Cr-Commit-Position: refs/heads/master@{#437819}

BUG=NONE

Committed: https://crrev.com/98f0f335c82dbbd932d3d5bad9fbb22585dafb65
Review-Url: https://codereview.chromium.org/2556803002
Cr-Original-Commit-Position: refs/heads/master@{#437292}
Cr-Commit-Position: refs/heads/master@{#437818}

Cr-Commit-Position: refs/heads/master@{#437817}

Remove ArcBridgeService::Get().
Now all usage was gone. It's time to delete the function.

BUG=657687
BUG=b/31079732
TEST=Ran bots.

Review-Url: https://codereview.chromium.org/2558313002
Cr-Commit-Position: refs/heads/master@{#437816}

https://chromium.googlesource.com/angle/angle.git/+log/e735511..d023939

BUG=chromium:671021
TBR=jmadill@chromium.org
TEST=bots
NOTRY=true
NOPRESUBMIT=true

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

Review-Url: https://codereview.chromium.org/2567923002
Cr-Commit-Position: refs/heads/master@{#437815}

This CL has no behavior changes.

BUG=668300

Review-Url: https://codereview.chromium.org/2555743005
Cr-Commit-Position: refs/heads/master@{#437814}

Sort fields in ComputedStyleBase in decreasing order of size. This is
a micro optimization that improves field alignment and minimizes wasted
memory space.

From Intel on 'Coding for Performance':
https://software.intel.com/en-us/articles/coding-for-performance-data-alignment-and-structures
"One can try to minimize this memory wastage by ordering the structure
elements such that the widest (largest) element comes first, followed by
the second widest, and so on."

Another good explanation article:
https://msdn.microsoft.com/en-us/library/ewwyfdbe.aspx

BUG=628043

Review-Url: https://codereview.chromium.org/2558733002
Cr-Commit-Position: refs/heads/master@{#437813}

Revert of gpu: Move ShaderDiskCache into //gpu/ipc/host component. (patchset #5 id:80001 of https://codereview.chromium.org/2555693005/ )

Reason for revert:
Makes StoragePartitionShaderClearTest.ClearShaderCache fail on Linux Tests bot: https://build.chromium.org/p/chromium.linux/builders/Linux%20Tests%20%28dbg%29%281%29

Original issue's description:
> gpu: Move ShaderDiskCache into //gpu/ipc/host component.
>
> BUG=643746, 604369
> CQ_INCLUDE_TRYBOTS=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
>
> Review-Url: https://codereview.chromium.org/2555693005

TBR=piman@chromium.org,rch@chromium.org,rsleevi@chromium.org,ben@chromium.org,sadrul@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=643746, 604369

Review-Url: https://codereview.chromium.org/2565243002
Cr-Commit-Position: refs/heads/master@{#437812}

This CL gets rid of using ArcBridgeService::Get() in
arc_robot_auto_code_fetcher_browsertest, by getting rid of
depending of ArcAuthService.

Test for AttemptUserExit is moved to arc_session_manager_unittest.
For that purpose, fixed username_hash for ARC Kiosk fake user.

BUG=657687
BUG=b/31079732
TEST=Ran bots.

Review-Url: https://codereview.chromium.org/2553193002
Cr-Commit-Position: refs/heads/master@{#437811}

As a follow-up for [1], this CL fixes the flakiness of
timeline-receive-response-event.html by dumping only the first
ResourceReceivedData event for each request.

1: https://crrev.com/48f3e047d17e5f78543bb0bec1c421ba3a69cb79

BUG=666217

Review-Url: https://codereview.chromium.org/2563613004
Cr-Commit-Position: refs/heads/master@{#437810}

TBR=mkwst@chromium.org
NOTRY=true
BUG=610835

Review-Url: https://codereview.chromium.org/2567933002
Cr-Commit-Position: refs/heads/master@{#437809}

BUG=672516

Review-Url: https://codereview.chromium.org/2562063002
Cr-Commit-Position: refs/heads/master@{#437808}

Created by adding the following to testing/buildbot/manage.py in
def process_file(..):

   if 'views_unittests' in seen and 'native_theme_unittests' not in seen:
     data = [d for d in gtest_tests if d['test'] == 'views_unittests'][0].copy()
     data['test'] = 'native_theme_unittests'
     gtest_tests.append(data)

Then running manage.py --write. It told me to put an entry in
gn_isolate_map.pyl as well.

BUG=398316

Review-Url: https://codereview.chromium.org/2560633004
Cr-Commit-Position: refs/heads/master@{#437807}

Review-Url: https://codereview.chromium.org/2560423002
Cr-Commit-Position: refs/heads/master@{#437806}

It's only one test now (zero on Android, so it breaks on Android FYI)
and will be merged into service_manager_unittests.

BUG=None
R=dpranke@chromium.org

Review-Url: https://codereview.chromium.org/2563253003
Cr-Commit-Position: refs/heads/master@{#437805}

We moved to checking need-for-repaint and invalidating based
on anchor node in http://crrev.com/2401363003 but we
incorrectly left null-checking based on non-anchor node.

BUG=663319

Review-Url: https://codereview.chromium.org/2565103002
Cr-Commit-Position: refs/heads/master@{#437804}

Because refX and refY are in the coordinate space of each marker instance,
they are included in markerTransformation() instead of being part of the
marker viewport. As a result, only a viewport size is maintained in
LayoutSVGResourceMarker. This patch makes this distinction a little
clearer by refactoring m_viewport to m_viewportSize, as well as adding
some comments.

No new tests; no change in behavior.

Review-Url: https://codereview.chromium.org/2567853002
Cr-Commit-Position: refs/heads/master@{#437803}

https://chromium.googlesource.com/chromium/src/+/c896e79e5ba34

BUG=548616
TBR=robhogan@gmail.com

Review-Url: https://codereview.chromium.org/2563323002 .
Cr-Commit-Position: refs/heads/master@{#437802}

ReadingListStoreDelegate has virtual methods, so it should have a virtual
destructor.

BUG=673169

Review-Url: https://codereview.chromium.org/2565213002
Cr-Commit-Position: refs/heads/master@{#437801}

ReadingListStore inherits from multiple classes, including from
ReadingListModelStorage that does not define a virtual destructor.

Since ReadingListModelImpl has a scoped_ptr to ReadingListModelStorage,
it calls the base destructor with a pointer that does not correspond
to memory returned by "new" (it points to somewhere inside the memory
block).

The issue can be fixed by defining a virtual destructor on the base
class ReadingListModelStorage (as all class with virtual method should
do). Also mark the class a non-copyable since copying would cause slicing.

BUG=673169

Review-Url: https://codereview.chromium.org/2563303002
Cr-Commit-Position: refs/heads/master@{#437800}

Revert of Enable RL by default (patchset #6 id:100001 of https://codereview.chromium.org/2562643003/ )

Reason for revert:
Speculative revert as I think this CL broke the downstream autoroller.

It is a bit hard to debug as the autoroller was unable to land some CL as the downstream tree was closed, but the last successful roll would have happened at 39e086ae [1], and the first failing roll is at a229e828 [2].

This CL is the only one having an impact on the iOS build, so I think it is causing the failures:

$ git log --oneline 39e086ae...a229e828
a229e828 Strength-reduce the "scale-factor changed" condition in LayoutSVGRoot
d1693364 Update V8 to version 5.7.202.
7cfbed3d Enable RL by default
314c528d Revert of Improve EAT_STREAM_PARAMETERS for Windows x86 (patchset #10 id:240001 of https://codereview.chromium.org/2559323007/ )

[1]: https://uberchromegw.corp.google.com/i/internal.bling.main/builders/trunk-autoroller/builds/6861
[2]: https://uberchromegw.corp.google.com/i/internal.bling.main/builders/trunk-autoroller/builds/6862

Original issue's description:
> Enable RL by default
>
> BUG=577659
>
> Review-Url: https://codereview.chromium.org/2562643003

TBR=noyau@chromium.org,mardini@chromium.org,olivierrobin@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=577659
NOTRY=true

Review-Url: https://codereview.chromium.org/2564293002
Cr-Commit-Position: refs/heads/master@{#437799}

Computing the offset for a list marker after the rest of the objects on the line
it is on have been laid out means it will avoid floats it ought not to.

Instead, compute the offset when laying out the marker and cache it for use later.

BUG=548616
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2553793003
Cr-Commit-Position: refs/heads/master@{#437798}

BUG=671181

Review-Url: https://codereview.chromium.org/2559223002
Cr-Commit-Position: refs/heads/master@{#437797}

BUG=672516

Review-Url: https://codereview.chromium.org/2567483003
Cr-Commit-Position: refs/heads/master@{#437796}

Without this check we'd crash when creating the watcher and
asking for the server file descriptor from a null server instance.

TBR=sky@chromium.org
BUG=
TEST=chrome --enable-wayland-server (without XDG_RUNTIME_DIR set)

Review-Url: https://codereview.chromium.org/2569633002
Cr-Commit-Position: refs/heads/master@{#437795}

FaceDetector.detect() in ToT returns a sequence<DetectedObject>,
but DetectedObject doesn't exist, and it should return DetectedFace
instead according to the Spec. Also, DetectedFace is not used at the
moment (FaceDetector  goes around this issue by returning DOMRects
directly).

This CL sorts that out by returning DetectedFaces directly.

BUG=673075

Review-Url: https://codereview.chromium.org/2564123003
Cr-Commit-Position: refs/heads/master@{#437794}

Always creating a GL context is currently preventing this client
from being used when GL is not available.

BUG=
TEST=wayland_motion_events (without GL)

Review-Url: https://codereview.chromium.org/2566863002
Cr-Commit-Position: refs/heads/master@{#437793}

Globals need to be destroyed before the wl_display.

BUG=
TEST=wayland_motion_events --use-drm
TBR=dcastagna@chromium.org

Review-Url: https://codereview.chromium.org/2568703002
Cr-Commit-Position: refs/heads/master@{#437792}

Use ios_packed_resources_target as a dependency of //ios/chrome/app
target to select the correct packed resource target.

BUG=653086

Review-Url: https://codereview.chromium.org/2558913005
Cr-Commit-Position: refs/heads/master@{#437791}

In https://codereview.chromium.org/2479113002/ it was discovered that
current codebase has two duplicate TestingPrefServiceSyncable classes.
One in component/sync_preferences and another in
components/pref_registry. The decision from discussion is to delete the
one in components/pref_registry.

BUG=672900

R=gab@chromium.org, vasilii@chromium.org, hashimoto@chromium.org, pkasting@chromium.org, rdevlin.cronin@chromium.org, ben@chromium.org, blundell@chromium.org

TBR=rdevlin.cronin@chromium.org, ben@chromium.org, blundell@chromium.org

Review-Url: https://codereview.chromium.org/2562733003
Cr-Commit-Position: refs/heads/master@{#437790}

This rolls //third_party/typ to v0.9.9 to pick up dglazkov's patch
to allow "foreign arguments" to typ. A foreign argument is one
that the typ arg parser ignores.

This roll includes:

  d77f1a5 bump version to 0.9.9
  69c6119 Merge pull request #7 from dglazkov/foreign-argument
  0a2f193 fix indentation
  c8845b2 Introduce the concept of a foreign argument.

R=dglazkov@chromium.org

Review-Url: https://codereview.chromium.org/2568683002
Cr-Commit-Position: refs/heads/master@{#437789}

http://chromium.googlesource.com/chromium/deps/icu.git/+log/dda089a..73e2473

Two changes are in the range:

1. Timezone update to 2016i
2. Bign Endian support for MIPS

BUG=473288
TBR=mark@chromium.org
TEST=Try the following in d8 or Chrome's JS console:
> (new Date("12/04/2016 12:00Z")).toLocaleString("en", {timeZone:
"Europe/Saratov"})
"12/4/2016, 4:00:00 PM"
> (new Date("12/03/2016 12:00Z")).toLocaleString("en", {timeZone:
"Europe/Saratov"})
"12/3/2016, 3:00:00 PM"

Review-Url: https://codereview.chromium.org/2557743010
Cr-Commit-Position: refs/heads/master@{#437788}

BUG=641020

Review-Url: https://codereview.chromium.org/2557403005
Cr-Commit-Position: refs/heads/master@{#437787}

Cr-Commit-Position: refs/heads/master@{#437786}

This CL changes how the scope is computed if the scope is unspecified in the
Android Manifest.
We used to compute the default scope based on the "URL the WebAPK navigates to
when launched." Deep links can open a WebAPK at an arbirtrary URL.
This CL changes the computation to be based on the Web Manifest start URL.

This change enables cleaning up
ManifestUpgradeDetectorFetcher#onDataAvailable(). We can now create the fetched
WebApkInfo using the current WebAPK's WebApkInfo#uri().

BUG=639536

Review-Url: https://codereview.chromium.org/2548333006
Cr-Commit-Position: refs/heads/master@{#437785}

This is always defined on non-mobile platforms and it is not possible to
override in the build. Since //chrome is no longer build on iOS, most uses
can be easily keyed off of the Android defines. For files not compiled on
Android, the conditionals were removed altogether.

This change additionally coalesces some Android blocks in
chrome/browser/BUILD.gn and chrome/browser/ui/BUILD.gn

BUG=671706

Review-Url: https://codereview.chromium.org/2561013002
Cr-Commit-Position: refs/heads/master@{#437784}

This improves the synchronization by deferring the client side
wait until commit time.

BUG=
TEST=wayland_motion_events

Review-Url: https://codereview.chromium.org/2563263002
Cr-Commit-Position: refs/heads/master@{#437783}