GitLab

BUG=715962

TBR=rnephew@chromium.org

Review-Url: https://codereview.chromium.org/2850483002
Cr-Commit-Position: refs/heads/master@{#467730}

The blocking pool is being deprecated in favor of TaskScheduler.

BUG=667892
R=bartfab@chromium.org

Review-Url: https://codereview.chromium.org/2837083002
Cr-Commit-Position: refs/heads/master@{#467729}

The blocking pool is being deprecated in favor of TaskScheduler.

BUG=667892
R=bartfab@chromium.org

Review-Url: https://codereview.chromium.org/2834413003
Cr-Commit-Position: refs/heads/master@{#467728}

After processing overlay candidates we used to remove the overlay rects
from the damage rect.
The assumption was that all the overlays would be opaque.

Since we started allowing alpha blended overaly, we should avoid
subtracting from the damage when blending might be allowed.
Otherwise any element animating below a transparent overlay might
not be updated.

We haven't noticed any problem so far since partial update was disabled
on Mali. crrev.com/2829543003 enables empty swap, surfacing the problem
when the damage region is a subset of a transparent overlay.

BUG=705290
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2842383003
Cr-Commit-Position: refs/heads/master@{#467727}

The blocking pool is being deprecated in favor of TaskScheduler.

BUG=667892
R=mmenke@chromium.org

Review-Url: https://codereview.chromium.org/2843563002
Cr-Commit-Position: refs/heads/master@{#467726}

This flag should be on by default in all non-test cases, so this patch
removes the flag altogether.

BUG=696864
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2781183002
Cr-Commit-Position: refs/heads/master@{#467725}

BUG=none

Review-Url: https://codereview.chromium.org/2852443002
Cr-Commit-Position: refs/heads/master@{#467724}

Revert of [blink] Unique pointers in Platform.h (patchset #6 id:100001 of https://codereview.chromium.org/2846843002/ )

Reason for revert:
Broke the tree

Original issue's description:
> [blink] Unique pointers in Platform.h
>
> Make Platform::*Create methods return a unique_ptr instead of a raw pointer
> to explictly state ownership logic.
> CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_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
>
> Review-Url: https://codereview.chromium.org/2846843002
> Cr-Commit-Position: refs/heads/master@{#467710}
> Committed: https://chromium.googlesource.com/chromium/src/+/d548cc04b6568f8edc08853940134707fcd6c455

TBR=haraken@chromium.org,clamy@chromium.org,tdresser@chromium.org,thakis@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review-Url: https://codereview.chromium.org/2850593002
Cr-Commit-Position: refs/heads/master@{#467723}

This patch replaces some instances of unordered_map with flat_map. This
is done for cases where we can reason out that the number of elements
contained is small.

R=danakj@chromium.org
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2847553002
Cr-Commit-Position: refs/heads/master@{#467722}

libbase is needed when using compiled_action targets (these are
helper programs compiled for use later in the build), but it doesn't
currently build for the host inside the CrOS sysroot because the
toolchain doesn't include the libatomic library. libbase does, however,
seem to compile without the -latomic flag.

TEST=Revert 5e671fff, compile elm
BUG=710006, 710841

Review-Url: https://codereview.chromium.org/2823303003
Cr-Commit-Position: refs/heads/master@{#467721}

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
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2841833002
Cr-Commit-Position: refs/heads/master@{#467720}

This CL moves ScriptWrappable.h/cpp and all its dependencies to platform/bindings. The
original header for each file is not deleted, but is modified to point to the new header
in platform/bindings (This will be cleaned up in follow-up CLs).

BUG=682322

Review-Url: https://codereview.chromium.org/2843603002
Cr-Commit-Position: refs/heads/master@{#467719}

If a Frame has non-overlay scrollbars, the visualViewport's dimensions
should exclude them. In order to know whether the Frame has scrollbars,
we need to perform any pending layouts.

This was only broken for iframes as the main frame viewport size
depends on page scale. Page scale depends on layout to determine
the minimum/initial page scale so we already performed layout when the
main frame's viewport was queried.

BUG=714829

Review-Url: https://codereview.chromium.org/2842343002
Cr-Commit-Position: refs/heads/master@{#467718}

The object is quite large, this class doesn't need to know about
it. This is intended to ease refactoring.

BUG=715525

Review-Url: https://codereview.chromium.org/2845103002
Cr-Commit-Position: refs/heads/master@{#467717}

Revert of 📺 Move fullscreen web content to a new Activity. (patchset #11 id:200001 of https://codereview.chromium.org/2807663002/ )

Reason for revert:

Findit(https://goo.gl/kROfz5) identified CL at revision 467705 as the
culprit for failures in the build cycles as shown on:
https://findit-for-me.appspot.com/waterfall/culprit?key=ag9zfmZpbmRpdC1mb3ItbWVyRAsSDVdmU3VzcGVjdGVkQ0wiMWNocm9taXVtL2I3MWU5OGNkZjE0ZjE4Y2I5NjdhNzM4NTc4MjZmNmU4YzU2OGNlYTAM

Original issue's description:
> Move fullscreen web content to a new Activity.
>
> Introduce the FullscreenMediaActivity and move Tabs to it when entering
> fullscreen and away from it when leaving fullscreen.
>
> BUG=709042
>
> Review-Url: https://codereview.chromium.org/2807663002
> Cr-Commit-Position: refs/heads/master@{#467705}
> Committed: https://chromium.googlesource.com/chromium/src/+/b71e98cdf14f18cb967a73857826f6e8c568cea0

TBR=mlamouri@chromium.org,bauerb@chromium.org,twellington@chromium.org,yusufo@chromium.org,peconn@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=709042

Review-Url: https://codereview.chromium.org/2849533004
Cr-Commit-Position: refs/heads/master@{#467716}

Revert of Rename kDisableGestureRequirementForMediaPlayback and make it a test-only flag. (patchset #2 id:20001 of https://codereview.chromium.org/2843913003/ )

Reason for revert:
broke clang-clankium-tot-builder crbug.com/716075

(FullscreenWebContentsActivityTest not updated)

Original issue's description:
> Rename kDisableGestureRequirementForMediaPlayback and make it a test-only flag.
>
> The flag should only be set by tests and will allow all autoplay related
> restrictions such as autoplay policy and background playback to be
> disabled.
>
> BUG=714676
> CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_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
>
> Review-Url: https://codereview.chromium.org/2843913003
> Cr-Commit-Position: refs/heads/master@{#467692}
> Committed: https://chromium.googlesource.com/chromium/src/+/d1f8f6ca6c8b7cdaf02150099ce6db0cfc7223f6

TBR=dalecurtis@chromium.org,jam@chromium.org,mlamouri@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=714676

Review-Url: https://codereview.chromium.org/2851453003
Cr-Commit-Position: refs/heads/master@{#467715}

- Browser goes into VR Browsing Mode when the user inserts the phone into the headeset.
- Browser will then automatically go into presentation mode if page requests present in onvrdisplayactivate handler.
- In a follow-up CL, browser should tentatively go into presentation mode if page is listening for onvrdisplayactivate (and exit presentation if page does not request present).

BUG=703315

Review-Url: https://codereview.chromium.org/2850483003
Cr-Commit-Position: refs/heads/master@{#467714}

Since we currently don't support vulkan on chrome, we just add our dummy
Vulkan headers to chromes skia includes so we can the vulkan headers for
type declarations.

BUG=

Review-Url: https://codereview.chromium.org/2843823003
Cr-Commit-Position: refs/heads/master@{#467713}

* Use kFoo for enums.
* Don't need to use mojom to test serialization of structs.

BUG=NONE

Review-Url: https://codereview.chromium.org/2843413002
Cr-Commit-Position: refs/heads/master@{#467712}

It's easy to get this wrong and use Set instead, which can invoke author script
in unexpected places. This is a convenient helper that does the right thing,
similar to blink::V8ObjectBuilder.

A unit test and two "real-world" uses are included.

BUG=

Review-Url: https://codereview.chromium.org/2845463002
Cr-Commit-Position: refs/heads/master@{#467711}

Make Platform::*Create methods return a unique_ptr instead of a raw pointer
to explictly state ownership logic.
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_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

Review-Url: https://codereview.chromium.org/2846843002
Cr-Commit-Position: refs/heads/master@{#467710}

At the moment SensorProxy is the only source of error notification for the Sensor class instances, however soon other error notification sources will turn up (e.g. permission subsystem), so we must guarantee that stopping of a Sensor due to an error is reflected on platform side.

BUG=606766

Review-Url: https://codereview.chromium.org/2839963004
Cr-Commit-Position: refs/heads/master@{#467709}

Previously, whenever the shell_apk in shell_apk_version.gni there was a period
of time during which
- Chrome tries to update the WebAPK because the "shell_apk" version with which
  Chrome is compiled with was updated
- The WebAPK server is unable to fulfill the update request because the new
  ShellAPK has not yet been uploaded to the WebAPK server

This CL splits shell_apk_version in shell_apk_version.gni into two variables:
|template_shell_apk_version| and |expected_shell_apk_version|

The new flow for updating android/webapk/shell_apk will be:
1) Write CL which:
   - Makes changes to code in android/webapk/shell_apk
   - Increments |template_shell_apk_version|
   This will:
   - Update the template ShellAPK which is built on the official build bot
   - Trigger an alert that the ShellAPK on the WebAPK Minting Server needs to
     be updated
   - Does not affect the "WebAPK update logic".
2) Upload new ShellAPK template to the WebAPK Minting Server
   At this point:
   - new WebAPK installs will have the updated shell.
   - Chrome will still not request updates to the new ShellAPK version
3) Write CL which:
   - Increments |expected_shell_apk_version|
   This will:
   - cause Chrome to request updates to the new ShellAPK version

BUG=708053

Review-Url: https://codereview.chromium.org/2827233006
Cr-Commit-Position: refs/heads/master@{#467708}

Currently textQuads and ComputeTextRects algorithm is
mostly same. The difference of behaviour of these two functions
can be handled efficiently by using templates.

This CL is the Second patch set to achieve the above behaviour.
This CL introduces ComputeTextBounds as template function, which
will  be used by computeTextRects only in this patch.

BUG=691198

Review-Url: https://codereview.chromium.org/2839633002
Cr-Commit-Position: refs/heads/master@{#467707}

This "hidden" base class was forcing two virtual calls to reach a Frame from a WebFrame. It is now one virtual call away.

BUG=713798

Review-Url: https://codereview.chromium.org/2837593002
Cr-Commit-Position: refs/heads/master@{#467706}

Introduce the FullscreenMediaActivity and move Tabs to it when entering
fullscreen and away from it when leaving fullscreen.

BUG=709042

Review-Url: https://codereview.chromium.org/2807663002
Cr-Commit-Position: refs/heads/master@{#467705}

It is no longer necessary to list all the apk targets as separate
modules now that we have the _all pseudo module. Having the separate
modules resulted in bugs where Android Studio could not distinguish
between prod code and test code since some modules viewed dirs as prod
and some as test. The --split-projects flag can be used to see the
dependency graph, and when only a single module will be generated, the
_all pseudo module will not replace it.

For android studio's _all pseudo module, move all known test dirs to
androidTest instead of main so that they are properly displayed when
filtered.

BUG=620034

Review-Url: https://codereview.chromium.org/2837863002
Cr-Commit-Position: refs/heads/master@{#467704}

Automatically generated ARCMigrate commit
Notable issues:None
BUG=624363
TEST=None

Review-Url: https://codereview.chromium.org/2835413003
Cr-Commit-Position: refs/heads/master@{#467703}

Just a cleanup patch, no change in functionality.

BUG=

Review-Url: https://codereview.chromium.org/2843993002
Cr-Commit-Position: refs/heads/master@{#467702}

The method to check whether the payment instrument for the specified key is
stored.

Related Spec Link:
  https://w3c.github.io/webpayments-payment-handler/#paymentinstruments-interface

BUG=661608
TEST=payment_manager_unittest.cc, payment-instruments.html

Review-Url: https://codereview.chromium.org/2844673002
Cr-Commit-Position: refs/heads/master@{#467701}

The main thread doesn't pump +system+ messages.

R=gab@chromium.org
TBR=gab@chromium.org
BUG=

Review-Url: https://codereview.chromium.org/2847773002
Cr-Commit-Position: refs/heads/master@{#467700}

Replace implementation of allowpaymentrequest and allowfullscreen by feature policy.

Enable iframe feature policy by attributes:
  allow="feature1 feature2", allowfullscreen, and allowpaymentrequest.

See design doc: https://docs.google.com/a/chromium.org/document/d/1sskoBi7Ba7hLuuiJQ6VMQ1KYIGBP8Lr87hLbkF76c2I/edit?usp=sharing

Part 1: Introduce iframe allowAttr in HTMLIFrameElement::parseAttribute
and store featureNames in HTMLIFrameElement
(CL: https://codereview.chromium.org/2680083002/)

Part 2: Propagate featureNames from HTMLIFrameElement to frame owner and
remote frame owner.
(CL: https://codereview.chromium.org/2697713003)

Part 3: Set iframe feature policy in FrameLoader::didBeginDocument
3.a Replace string by enum in WebParsedFeaturePolicyDeclaration#feature
(CL: https://codereview.chromium.org/2727803004/)
3.b Rename WebParsedFeaturePolicyHeader to WebParsedFeaturePolicy
(CL: https://codereview.chromium.org/2738953002/)

Part 4: Replace implementation of allowpaymentrequest and allowfullscreen by feature policy.

BUG=682256

Review-Url: https://codereview.chromium.org/2767983003
Cr-Commit-Position: refs/heads/master@{#467699}

Also make sure that the user cannot click pay until we
get the callback from the merchant.

BUG=715170,715597

Review-Url: https://codereview.chromium.org/2836353002
Cr-Commit-Position: refs/heads/master@{#467698}

After https://codereview.chromium.org/2789363004 landed, Ozone Drm
supports the KMS "rotation" property that can be used to set a transform
to HW overlays.

This CL relaxes the transform restrictions of the fullscreen and
single on top overlay strategies.

Additionally, it makes sure that any configuration containing a transform
will be reject if the legacy page flip is being used.

After this CL WebGL and Pepper content can be promoted to HW overlay when
backed by a Gpu Memory Buffer.

BUG=b/36752417
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2794343003
Cr-Commit-Position: refs/heads/master@{#467697}

The fix for handling case thread_time equals None was landed.
BUG=715822
TBR=wangxianzhu@chromium.org

Review-Url: https://codereview.chromium.org/2849553002
Cr-Commit-Position: refs/heads/master@{#467696}

Reland of Break page load metrics test dependency on IPC. (patchset #1 id:1 of https://codereview.chromium.org/2847803002/ )

Reason for revert:
The flake that caused the previous revert by zmin existed before the original patch landed, so the original patch is not the cause of the flake.

This refactor should help us to un-flake the test, so I'm re-landing it.

Original issue's description:
> Revert of Break page load metrics test dependency on IPC. (patchset #2 id:20001 of https://codereview.chromium.org/2847513002/ )
>
> Reason for revert:
> PageLoadMetricsBrowserTest.DocumentWriteReloada is flaky
>
> https://luci-milo.appspot.com/buildbot/chromium.memory/Linux%20MSan%20Tests/387
>
> Original issue's description:
> > Break page load metrics test dependency on IPC.
> >
> > This change breaks direct dependencies on IPC for page load metrics
> > tests, in order to ease the transition to mojo.
> >
> > Previously, unit tests exercised the IPC dispatch path. This
> > achieved slightly increased coverage but coupled unit tests to
> > IPC. We now dispatch simulated timing events directly to callbacks,
> > bypassing the IPC dispatch.
> >
> > Additionally, we break the browsertest dependency on IPC and add
> > a MetricsWebContentsObserver::TestingObserver, which can observe
> > state changes at the observer level, instead of watching for IPC
> > messages. This both simplifies the logic and more directly verifies
> > expected behavior at the appropriate level.
> >
> > BUG=715744
> > TBR=csharrison
> >
> > Review-Url: https://codereview.chromium.org/2847513002
> > Cr-Commit-Position: refs/heads/master@{#467687}
> > Committed: https://chromium.googlesource.com/chromium/src/+/9567f9fa3fcfcaa50fc7b6a6adf66106a2eb6be5
>
> TBR=lpy@chromium.org,bmcquade@chromium.org
> # Skipping CQ checks because original CL landed less than 1 days ago.
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=715744
>
> Review-Url: https://codereview.chromium.org/2847803002
> Cr-Commit-Position: refs/heads/master@{#467691}
> Committed: https://chromium.googlesource.com/chromium/src/+/2cb37fe0beb13e6baa8852662362e9204d1ebfa3

TBR=lpy@chromium.org,zmin@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=715744

Review-Url: https://codereview.chromium.org/2849563002
Cr-Commit-Position: refs/heads/master@{#467695}

Many layout changes e.g. resizing doesn't affect computed collapsed
borders so don't need to invalidate collapsed borders.

We need to invalidate collapsed borders when
1. Table sections are set need recalc (when table structure changes);
2. A cell is appended into a row (which could belong to 1 but we have an
   optimization not to recalc table sections if the added cell is the
   last cell of the table);
3. Border style changes;
4. border-collapse CSS property changes;

This CL will reduce frame time of PerformanceTests/Mutation/large-
table-row-height-change-with-collapsed-border.html (in
https://codereview.chromium.org/2842313002/) by about 35% by avoiding
unnecessary collapsed border recalculations.

BUG=626748

Review-Url: https://codereview.chromium.org/2840723005
Cr-Commit-Position: refs/heads/master@{#467694}

The blocking pool is being deprecated in favor of TaskScheduler.

BUG=667892
R=jialiul@chromium.org

Review-Url: https://codereview.chromium.org/2833373004
Cr-Commit-Position: refs/heads/master@{#467693}

The flag should only be set by tests and will allow all autoplay related
restrictions such as autoplay policy and background playback to be
disabled.

BUG=714676
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_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

Review-Url: https://codereview.chromium.org/2843913003
Cr-Commit-Position: refs/heads/master@{#467692}

Revert of Break page load metrics test dependency on IPC. (patchset #2 id:20001 of https://codereview.chromium.org/2847513002/ )

Reason for revert:
PageLoadMetricsBrowserTest.DocumentWriteReloada is flaky

https://luci-milo.appspot.com/buildbot/chromium.memory/Linux%20MSan%20Tests/387

Original issue's description:
> Break page load metrics test dependency on IPC.
>
> This change breaks direct dependencies on IPC for page load metrics
> tests, in order to ease the transition to mojo.
>
> Previously, unit tests exercised the IPC dispatch path. This
> achieved slightly increased coverage but coupled unit tests to
> IPC. We now dispatch simulated timing events directly to callbacks,
> bypassing the IPC dispatch.
>
> Additionally, we break the browsertest dependency on IPC and add
> a MetricsWebContentsObserver::TestingObserver, which can observe
> state changes at the observer level, instead of watching for IPC
> messages. This both simplifies the logic and more directly verifies
> expected behavior at the appropriate level.
>
> BUG=715744
> TBR=csharrison
>
> Review-Url: https://codereview.chromium.org/2847513002
> Cr-Commit-Position: refs/heads/master@{#467687}
> Committed: https://chromium.googlesource.com/chromium/src/+/9567f9fa3fcfcaa50fc7b6a6adf66106a2eb6be5

TBR=lpy@chromium.org,bmcquade@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=715744

Review-Url: https://codereview.chromium.org/2847803002
Cr-Commit-Position: refs/heads/master@{#467691}