GitLab

This allows the Browser Task Scheduler to check variations before
metrics initializes and runs items on the blocking pool.

BUG=636518

Review-Url: https://codereview.chromium.org/2342373002
Cr-Commit-Position: refs/heads/master@{#419549}

After forking OutputSurface, there are a lot of things on both
OutputSurface and CompositorFrameSink that only belong on one or the
other. This patch removes such things from each of them.

Next up: Make OutputSurface::BindToClient not failable. Start
renaming SwapBuffers and friends for CompositorFrameSink.

TBR=dtrainor
R=boliu, dtrainor@chromium.org, enne
BUG=606056
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2349743004
Cr-Commit-Position: refs/heads/master@{#419548}

Code was added already:
https://cs.chromium.org/chromium/src/third_party/crashpad/crashpad/util/misc/metrics.cc?q=Crashpad.CrashReportSize&sq=package:chromium&l=24.

R=asvitkine@chromium.org
BUG=crashpad:100

Review-Url: https://codereview.chromium.org/2349993003
Cr-Commit-Position: refs/heads/master@{#419547}

CertErrors parameter.

BUG=634443

Review-Url: https://codereview.chromium.org/2349093002
Cr-Commit-Position: refs/heads/master@{#419546}

some containers, such as rows in Gmail's thread view, get focused. Ensure we speak the accessible name, but don't try to speak descendants.

We only speak descendants for region and group roles. Any other role in ptractice leads to a lot of queued up output; for example, press escape while in Sheets. This results in a focus event on an application node containing almost all the content of the page.

Finally, fix the leave rule so that it also ignores formatting a leave rule for a role that was part of the speak rule. This prevents things like "... row, exited row" when moving from row to row in Gmail's thread view.

TEST=manual; chromevox_tests --gtest_filter=Output*.*
BUG=
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2331163002
Cr-Commit-Position: refs/heads/master@{#419545}

BUG=644397

Review-Url: https://codereview.chromium.org/2332423002
Cr-Commit-Position: refs/heads/master@{#419544}

Revert of Set crash key if WM_QUIT message is received on GPU message loop. (patchset #3 id:40001 of https://codereview.chromium.org/2342713002/ )

Reason for revert:
We've got enough data now.

Original issue's description:
> Set crash key if WM_QUIT message is received on GPU message loop.
>
> Threads seem to be exiting unexpectedly, so to help debugging set a
> crash key if an unexpected WM_QUIT message is received.
>
> BUG=647068
> TBR=scottmg@chromium.org
>
> Committed: https://crrev.com/18db6b0cc6f81fba74278cfab2856ccb11c0e0c7
> Cr-Commit-Position: refs/heads/master@{#419057}

TBR=dcheng@chromium.org,scottmg@chromium.org,rsesek@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=647068

Review-Url: https://codereview.chromium.org/2346353002
Cr-Commit-Position: refs/heads/master@{#419543}

Revert of Prevent redundant DoWorks due to canceled delayed tasks  (patchset #6 id:100001 of https://codereview.chromium.org/2320403003/ )

Reason for revert:
Suspect that this CL caused breakage in Gmail and Hangouts per http://crbug.com/647484 .

Original issue's description:
> Prevent redundant DoWorks due to canceled delayed tasks
>
> To achieve this we make a few changes:
>
> 1. We only register the next wakeup with the TimeDomain, rather than all
> of them.
> 2. MoveReadyDelayedTasksToDelayedWorkQueue now registers the next
> wakeup (if any).  Since it removes all canceled delayed tasks from the
> front of the priority queue this has the effect of not scheduling
> wakeups for cancelled tasks.
> 3. Tweaking the TaskQueueManager level delayed DoWork de-duplication
> logic to only post a delayed DoWork if the task is meant to run before
> any previously registered delayed DoWorks.
>
> BUG=638542, 605718
>
> Committed: https://crrev.com/929cbb9f92b5570867c3842c80778243db81a013
> Cr-Commit-Position: refs/heads/master@{#418556}

TBR=skyostil@chromium.org,alexclarke@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=638542, 605718

Review-Url: https://codereview.chromium.org/2353473003
Cr-Commit-Position: refs/heads/master@{#419542}

We're looking at deprecating kStrokeAndFill_Style in Skia, so this CL's
main focus is removing the use of the flag.

While in this area, I've also updated the impl to draw the shadow using
Canvas::DrawRoundRect instead of going through a path.

R=reed@google.com,bruthig@chromium.org

Review-Url: https://codereview.chromium.org/2349273002
Cr-Commit-Position: refs/heads/master@{#419541}

BUG=internal 29459582
BUG=internal 31162371
TEST=cast_alsa_cma_backend_unittests --gtest_filter=StreamMixerAlsaTest.*
     exercise stream volume API; listen for smooth volume transitions

Review-Url: https://codereview.chromium.org/2341783004
Cr-Commit-Position: refs/heads/master@{#419540}

This patch moves some of the predecode image tracking to ImageManager.
The other part depends on crbug.com/647402. Once the TileTaskRunner
can keep a ref on scheduled tiles, then we can remove
TileManager::locked_image_tasks_ as well.

R=enne
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2345833002
Cr-Commit-Position: refs/heads/master@{#419539}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/ff40d4569c8b..c36cf56a4fcc

$ git log ff40d4569..c36cf56a4 --date=short --no-merges --format='%ad %ae %s'
2016-09-19 dskiba Support incremental Android installs in symbolize_trace script.

CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel
TBR=catapult-sheriff@chromium.org

Review-Url: https://codereview.chromium.org/2350203002
Cr-Commit-Position: refs/heads/master@{#419538}

https://chromium.googlesource.com/skia.git/+log/e202bd8b71f6..613664b1bc64

$ git log e202bd8b7..613664b1b --date=short --no-merges --format='%ad %ae %s'
2016-09-19 jvanverth Remove unnecessary assert in GrBatchAtlas.
2016-09-19 csmartdalton skpbench

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
TBR=borenet@google.com

Review-Url: https://codereview.chromium.org/2356433002
Cr-Commit-Position: refs/heads/master@{#419537}

Cr-Commit-Position: refs/heads/master@{#419536}

Review-Url: https://codereview.chromium.org/2347223003
Cr-Commit-Position: refs/heads/master@{#419535}

TEST=ash_unittests
BUG=645277

Review-Url: https://codereview.chromium.org/2324803003
Cr-Commit-Position: refs/heads/master@{#419534}

Add default ShouldDisplayOverFullscreen method to NotificationDelegate
that always returns false.
Add implementation for NotificationsApiDelegate that returns true for
ShouldDisplayOverFullscreen when the app that created the notificaiton
is displaying fullscreen content.

Note that this change does not enable this functionality.

BUG=481318

Review-Url: https://codereview.chromium.org/2335293002
Cr-Commit-Position: refs/heads/master@{#419533}

BUG=641626

Review-Url: https://codereview.chromium.org/2348693003
Cr-Commit-Position: refs/heads/master@{#419532}

To allow access to Physical Web data, the URL info and Physical Web
Service data needs to be paired together. The pairing gives all relevant
information for a scanned URL.

BUG=636490

Review-Url: https://codereview.chromium.org/2330253002
Cr-Commit-Position: refs/heads/master@{#419531}

Also adds `meaning=` tags so that the strings are translated separately for this exact context.

BUG=645290

Review-Url: https://codereview.chromium.org/2342683002
Cr-Commit-Position: refs/heads/master@{#419530}

This allows both updaters to use specific and correct prodids when
reporting the "version" in the update check and ping  requests.

BUG=641126

Review-Url: https://codereview.chromium.org/2340113002
Cr-Commit-Position: refs/heads/master@{#419529}

Fix failures on content_browser tests on Windows 7 from previous CL (patchset #4 id:200001 of https://codereview.chromium.org/2310683003/ )

BUG=509735

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:win7_blink_dbg,win7_blink_compile_dbg,win7_blink_compile_rel,win7_blink_rel

TBR=phoglund@chromium.org

Review-Url: https://codereview.chromium.org/2334223012
Cr-Commit-Position: refs/heads/master@{#419528}

BUG=NONE

Review-Url: https://codereview.chromium.org/2354433003
Cr-Commit-Position: refs/heads/master@{#419527}

https://pdfium.googlesource.com/pdfium.git/+log/a5b37f6b916f..b12fbddece43

$ git log a5b37f6b9..b12fbddec --date=short --no-merges --format='%ad %ae %s'
2016-09-19 dsinclair Update DEPS to always pull clang.
2016-09-19 tsepez Clean up CPDF_Stream.

BUG=647380

TBR=dsinclair@chromium.org

Review-Url: https://codereview.chromium.org/2350173003
Cr-Commit-Position: refs/heads/master@{#419526}

Add plumbing from LayerTree to LayerTreeImpl. Grab the color space
from the LayerTreeImpl in raster task creation in the TileManager by
using the TileManagerClient interface.

Note that while this is plumbing a valid gfx::ColorSpace, this still
has no impact on behavior, because the gfx::ColorSpace to
SkColorSpace conversion isn't in place yet. Once that is in place,
tests to query the tile resources' color spaces can be added.

BUG=44872
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2336853002
Cr-Commit-Position: refs/heads/master@{#419525}

These are unused in the chromium and chromiumos source trees and no one on
the mailing list is actually using them. Remove them significantly simplifies
D-Bus client initialization -- only the D-Bus thread creation needs to stay
centralized.

* Remove DBusClientTypes and the string representations of them
* Delete the parser for the switch
* Initialize the client bundles as either all real or all fake
* Remove a bunch of Init/Create methods from DBusThreadManager since there
are fewer conditional cases

BUG=647367,644414
TEST=chromeos_unittests, ash_unittests, unit_tests
TBR=tbarzic@chromium.org for easy_unlock

Review-Url: https://codereview.chromium.org/2350543002
Cr-Commit-Position: refs/heads/master@{#419524}

Internal Cast code requires that clients in remote processes access the
mojo:media service. Allow targets in the //chromecast/ subdirectory to
depend on client code. A pattern-match is used to obscure the structure
of internal directories.

BUG=

Review-Url: https://codereview.chromium.org/2353573002
Cr-Commit-Position: refs/heads/master@{#419523}

BUG=

Review-Url: https://codereview.chromium.org/2351773002
Cr-Commit-Position: refs/heads/master@{#419522}

BUG=

Review-Url: https://codereview.chromium.org/2352783002
Cr-Commit-Position: refs/heads/master@{#419521}

BUG=646951

Review-Url: https://codereview.chromium.org/2279123002
Cr-Commit-Position: refs/heads/master@{#419520}

This is an automated CL created by the recipe roller. This CL rolls recipe
changes from upstream projects (e.g. depot_tools) into downstream projects
(e.g. tools/build).

More info is at https://goo.gl/zkKdpD. Use https://goo.gl/noib3a to file a bug
(or complain)

build:
  https://crrev.com/c731a26b214ed6e743460469385b53bb3cc34228 Roll recipe dependencies (trivial). (recipe-roller@chromium.org)
depot_tools:
  https://crrev.com/9eadede7df2fb30f25edf64e378b129c88efdfb6 Add ability to set topic during "git cl upload" (rmistry@google.com)

TBR=martiniss@chromium.org,phajdan.jr@chromium.org
BUG=chromium:628237

Recipe-Tryjob-Bypass-Reason: Autoroller
Bugdroid-Send-Email: False
Review-Url: https://codereview.chromium.org/2346343002
Cr-Commit-Position: refs/heads/master@{#419519}

Currently the MD OOBE is enabled in a few different code paths, which unfortunately breaks when using, for example, zero-touch; or any path where enterprise enrollment is a possibility (as the login controller will jump straight into the network screen, bypassing the logic to determine whether to use the MD OOBE or not).

Instead, disable it for the few cases we know we have not tested (CfM and any device needing the HID screen).

Moreover, let's make sure that the MD/non-MD logic works when someone uses magic keys like CTRL+ALT+H to restart setup as a CfM.

This CL does not change the way local state stores whether to use MD or not, which is still a default of not showing it if state is not carefully set.

BUG=647365
TEST=manual (many combinations with and without existing local state)

Review-Url: https://codereview.chromium.org/2347703002
Cr-Commit-Position: refs/heads/master@{#419518}

Review-Url: https://codereview.chromium.org/2350223002
Cr-Commit-Position: refs/heads/master@{#419517}

Right now people are assuming it means something it doesn't.

BUG=

Review-Url: https://codereview.chromium.org/2340063002
Cr-Commit-Position: refs/heads/master@{#419516}

While investigating a crash, we blocked an optimization so we would have
more data in the minidumps.

The crash has been found so we can revert the changes made for
investigation.

BUG=643845
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel

Review-Url: https://codereview.chromium.org/2352533002
Cr-Commit-Position: refs/heads/master@{#419515}

Redirects through DataReductionProxy were counting as a DRP
navigations/pageloads even when the new URL that was redirected to was
non-DRP. This affects DRP PageLoad.* UMA and the DRP pingback.

This introduces a reset_to_default method on DRPData as well as changing
using original_url from URLRequest to just URL (as DRP cares more about
the actual URL of the site rather than the original URL before
redirects).

BUG=565421

Review-Url: https://codereview.chromium.org/2326443004
Cr-Commit-Position: refs/heads/master@{#419514}

Chrome on Chrome OS is incorrectly building against the Flash player
bundled in third_party/. However, on Chrome OS, flash player is loaded
from the root partition and passed to Chrome via command line. So Chrome
should not be referencing the versions in flapper_version.h. In
addition, this means the bundled flash player loading code is unused.

BUG=644795,644400

Review-Url: https://codereview.chromium.org/2340643002
Cr-Commit-Position: refs/heads/master@{#419513}

BUG=555865

Review-Url: https://codereview.chromium.org/2348953002
Cr-Commit-Position: refs/heads/master@{#419512}

This allows for the Browser Task Scheduler to setup and redirect after
SetupFieldTrials to get variations data but before SetupMetrics, which
posts tasks to the blocking pool.

The corresponding iOS change will be in a later CL

BUG=636518

Review-Url: https://codereview.chromium.org/2342993002
Cr-Commit-Position: refs/heads/master@{#419511}

https://chromium.googlesource.com/skia.git/+log/927ac9c53208..e202bd8b71f6

$ git log 927ac9c53..e202bd8b7 --date=short --no-merges --format='%ad %ae %s'
2016-09-19 halcanary SkPDF: SkBitSet gets reset() to make drop() better.
2016-09-19 egdaniel Move vulkan spir-v compile call to GrVkUtil

CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_precise_blink_rel
TBR=borenet@google.com

Review-Url: https://codereview.chromium.org/2349253003
Cr-Commit-Position: refs/heads/master@{#419510}