GitLab

The new backend supporting dynamic instantiation on Windows was rolled
out to the stable channel via a field trial flag.
This new backend solves some long-living problems, e.g., grabbing
devices forever, battery concerns, and stability.

Since the feature was enabled by default, and have worked well,
I want to remove the old backend now.

Another change that renames new backend will follow.

BUG=672793

Review-Url: https://codereview.chromium.org/2841493003
Cr-Commit-Position: refs/heads/master@{#466904}

BUG=711703

Review-Url: https://codereview.chromium.org/2828973002
Cr-Commit-Position: refs/heads/master@{#466903}

This CL uses MockScriptElementBase instead of nullptr to create
ClassicPendingScript, in order to remove
ClassicPendingScript::CreateForTesting() and make |PendingScript::element_|
always non-null when used.

BUG=686281, 694702

Review-Url: https://codereview.chromium.org/2827163003
Cr-Commit-Position: refs/heads/master@{#466902}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/5592ae343004..b367d46487c3

$ git log 5592ae343..b367d4648 --date=short --no-merges --format='%ad %ae %s'
2017-04-24 kainino Mac: download numpy and cv2 from Cloud Storage
2017-04-24 kainino Mac/Linux: disable formerly-skipped tests that fail when numpy/cv2 exist
2017-04-24 kainino Delete stale dependency zips

Created with:
  roll-dep src/third_party/catapult
BUG=711307,711307,711307


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.chromium.android:android_optional_gpu_tests_rel
TBR=sullivan@chromium.org

Change-Id: Ic2e84060eba2d7507e32549da2b8565d4d7ad8a3
Reviewed-on: https://chromium-review.googlesource.com/486063
Reviewed-by: <catapult-deps-roller@chromium.org>
Commit-Queue: <catapult-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#466901}

Process the output of 'ninja' live through the pipe rather than waiting
for it to finish.

Before:
20.74user 16.36system 0:24.57elapsed 151%CPU

After:
16.63user 9.24system 0:13.94elapsed 185%CPU

BUG=661774

Review-Url: https://codereview.chromium.org/2842513003
Cr-Commit-Position: refs/heads/master@{#466900}

This is preparation for introducing ModulePendingScript.

This CL shouldn't change the behavior.

This has been reverted due to crashing (Issue 711703) because
ResourceOwner's prefinalizer is called before PendingScript's prefinalizer,
causing CheckState() assertion failure.
This reland fixes this issue by registering PendingScript::Dispose() also
as the prefinalizer of ClassicPendingScript, which is called before
ResourceOwner's prefinalizer.

A unit test for the crash will be added by
https://codereview.chromium.org/2828973002/.

BUG=594639, 686281, 711703

Review-Url: https://codereview.chromium.org/2653923008
Cr-Original-Commit-Position: refs/heads/master@{#464494}
Committed: https://chromium.googlesource.com/chromium/src/+/d512803299c01324b92ffd8c962da97e3bc8ccff
Review-Url: https://codereview.chromium.org/2653923008
Cr-Commit-Position: refs/heads/master@{#466899}

This is required (and documented by a comment in the header file), so the DCHECKs are motivated.

BUG=711400
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/2828233002
Cr-Commit-Position: refs/heads/master@{#466898}

BUG=714831

Review-Url: https://codereview.chromium.org/2838753003
Cr-Commit-Position: refs/heads/master@{#466897}

BUG=690937
TEST=none

Review-Url: https://codereview.chromium.org/2815913005
Cr-Commit-Position: refs/heads/master@{#466896}

In a previous CL (https://codereview.chromium.org/2760753002) we made an
assumption that only local printers would appear in the local printer
list.  This turns out to be incorrect.  The actual condition is that only
cloud printers appear in the cloud list and sometimes appear in the local
list (when they're device printers).  This CL accounts for this behavior
and adds a test so we don't break it in the future.

BUG=713831
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2838603004
Cr-Commit-Position: refs/heads/master@{#466895}

This patch adds flaky results from 5-6 iterations of running tests.

I will run a few more iterations and see how it stabilizes.

BUG=591099

Review-Url: https://codereview.chromium.org/2836883002
Cr-Commit-Position: refs/heads/master@{#466894}

I was working on maybe adding something to this file and noticed that this flag
is always initialized to false and the setter is never used. So I think it can
be removed.

BUG=

Review-Url: https://codereview.chromium.org/2843433004
Cr-Commit-Position: refs/heads/master@{#466893}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/c76bb271..568edc40

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/2838803002
Cr-Commit-Position: refs/heads/master@{#466892}

Move blobalizer logic out of CachingWordShaper and ShapeResultBuffer and
into ShapeResultBloberizer, thereby containing all blobalizer logic in a
single file and removing dependencies on it from the results and shaper.

This will also help in reusing much of the text blob logic for LayoutNG.

R=drott@chromium.org
TEST=Source/platform/fonts/shaping/ShapeResultBloberizerTest.cpp

Review-Url: https://codereview.chromium.org/2835103002
Cr-Commit-Position: refs/heads/master@{#466891}

BUG=712987

Review-Url: https://codereview.chromium.org/2825203004
Cr-Commit-Position: refs/heads/master@{#466890}

$ git clone https://github.com/01org/libva.git
$ cd tmp/libva
$ git checkout libva-1.7.1
$ rm -rf ~/chromium/src/third_party/libva/va
$ find va -name \*.h -exec cp --parents \{\} ~/chromium/src/third_party/libva \;
$ sh build/gen_version.sh . va/va_version.h.in > ~/chromium/src/third_party/libva/va/va_version.h

Also Remove:
- remove va/egl/va_egl.h due to the lack of license header
- remove struct _VAEncMacroblockParameterBufferH264 from va/va_enc_h264.h,
  which contained an empty union, causing compile errors

BUG=None
TEST=build and test VDA and VEA on an Intel device.
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/2834313002
Cr-Commit-Position: refs/heads/master@{#466889}

Currently all the TRACE_EVENT* do not manage well in
embedded_worker_instance.cc, this CL integrates them to track
duration for every step throughout one complete starting worker process.

Also uses TRACE_EVENT_NESTABLE_ASYNC_* to replace deprecated
TRACE_EVENT_ASYNC*.

BUG=none

Review-Url: https://codereview.chromium.org/2821253002
Cr-Commit-Position: refs/heads/master@{#466888}

This CL introduces ModuleScript::CreateInstantiationError member func,
which serves as a getter for instantiation_error_.

This is split from hiroshige@'s CL: https://codereview.chromium.org/2819733002/

AUTHOR=hiroshige@chromium.org
BUG=594639

Review-Url: https://codereview.chromium.org/2838043002
Cr-Commit-Position: refs/heads/master@{#466887}

BUG=706926

Review-Url: https://codereview.chromium.org/2837943003
Cr-Commit-Position: refs/heads/master@{#466886}

Review-Url: https://codereview.chromium.org/2840633002
Cr-Commit-Position: refs/heads/master@{#466885}

The provider host may have been deleted when the request is resumed.

BUG=714065

Review-Url: https://codereview.chromium.org/2836593002
Cr-Commit-Position: refs/heads/master@{#466884}

This CL extracts PrepareDropDataForChildProcess out of
RenderWidgetHostImpl::GrantFileAccessFromDropData into a standalone
function.  This allows newly added unit tests to provide
|security_policy|, |child_id| and |file_system_context| without having a
dependency on RenderProcessHostImpl and/or RenderWidgetHostImpl.

This CL adds unit tests that PrepareDropDataForChildProcess
works as expected for DropData::filenames and for
DropData::file_system_files.  The latter one is a regression test
that would have prevented http://crbug.com/705295 from happening.

BUG=705295

Review-Url: https://codereview.chromium.org/2830743004
Cr-Commit-Position: refs/heads/master@{#466883}

It was introduced in 2014 to measure how many dead resources are reused. It is
not as useful as it used to be because now dead resources are basically evicted
from MemoryCache (with some exceptions such as preloaded resources).

BUG=None
R=clamy@chromium.org, japhet@chromium.org

Review-Url: https://codereview.chromium.org/2825893003
Cr-Commit-Position: refs/heads/master@{#466882}

Enable dynamic instantiation mode by default on Linux, ChromeOS,
and Windows. Other platforms still need to rely on the existing
experiment-controlled flag and will be launched separately.

BUG=672793,714511

Review-Url: https://codereview.chromium.org/2835083002
Cr-Commit-Position: refs/heads/master@{#466881}

This CL changes DetermineRevalidationPolicy a bit to simplify it. With this
CL, |allow_stale_resources_| will be checked after
ResourceLoaderOptions::CanReuseRequest is checked. CanReuseRequest checks
the synchronous policy and the cors policy, and neither of them is related
to staleness. Moreover, the synchronous policy was once checked BEFORE
allow_stale_resources_ (see [1]). Also, any user of
ResourceCacheValidationSuppressor, the class used to enable
|allow_stale_resource_|, doesn't seem to be interested in CORS. Hence I expect
no behavior change from this CL.

1: https://crrev.com/733a476495c43c1f472e5770c9d8aeeb94c5caca

BUG=652228
R=hiroshige@chromium.org, japhet@chromium.org

Review-Url: https://codereview.chromium.org/2828243002
Cr-Commit-Position: refs/heads/master@{#466880}

In the migration UI, we ask users that "Don't turn off or close your Chromebook".
In addition, we can resume migration even when user shut down / suspend the device
during migration.

That said, it should be even safer to prevent casual shutdown and power save mode by:
- Hide "Shut down" button during migration.
- Block power save during migration.

Note that if the user close the lid, the system will be suspended.
We have heard about heat issues when the lid is closed, and the system will be
heated by heavy CPU usage of migration.
It should be safer not to keep the device awake when the lid is closed.

BUG=711454
TEST=manually tested that the "Shut down" button was hidden and the device did
not go to sleep mode during migration.
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2842433002
Cr-Commit-Position: refs/heads/master@{#466879}

(Re-land after fix to avoid deleting nonexistent textures.)

Blitting from a multisampled RGB8 renderbuffer to an alpha:false WebGL
back buffer is supposed to work, and from an RGBA8 renderbuffer is
supposed to not work. The emulation of IOSurface-backed RGB textures
using RGBA textures didn't handle this case.

Add BindTexImage2DWithInternalformatCHROMIUM to the command buffer to
allow internal format to be overridden; currently this is only
supported on macOS. Use this to bind a second texture to the IOSurface
and use it only for BlitFramebuffer calls from WebGL, fixing up the
alpha channel afterward.

Covered by new conformance test in
https://github.com/KhronosGroup/WebGL/pull/2377 .

BUG=699566
TBR=zmo@chromium.org,ccameron@chromium.org
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/2838743002
Cr-Commit-Position: refs/heads/master@{#466878}

A new test got added to render_widget_host_connector_browsertest.cc
which ensures RWHVA gets connected to objects inheriting
RenderWidgetHostConnector regardless of the order in which they
are created. Previously this had a bug resulting in RWHVA not
connected ImeAdapter if RWHVA was created first, hence keeping
soft keyboard from showing up.

Also removed |WebContentsObserver::RenderViewReady()| override. It was
overridden to initialize rwvha in the connector after creation. But it
only covers the following cases:

- connector created -> rwhva created -> RenderViewReady invoked
- rwhva created -> connector created -> RenderViewReady invoked

Can't cover the following case:
 - rwhva created -> RenderViewReady invoked -> connector created

Now Initialize() covers them all, the method became redundant.

BUG=713924

Review-Url: https://codereview.chromium.org/2840443002
Cr-Commit-Position: refs/heads/master@{#466877}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/892a0b6a45cc..5592ae343004

$ git log 892a0b6a4..5592ae343 --date=short --no-merges --format='%ad %ae %s'
2017-04-24 achuith Disable testTraceCaptureUponFailure on chromeos.

Created with:
  roll-dep src/third_party/catapult
BUG=647340


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.chromium.android:android_optional_gpu_tests_rel
TBR=sullivan@chromium.org

Change-Id: I1b4136ceecbac675b4997f050d54c511c0f086df
Reviewed-on: https://chromium-review.googlesource.com/486165
Reviewed-by: <catapult-deps-roller@chromium.org>
Commit-Queue: <catapult-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#466876}

This CL adds a new interface that will be used to validate whether new
events should be stored, and during DB load on startup, it is also used
to check if old event should still be kept around.

The production implementation of this will be
FeatureConfigStorageValidator, which uses the FeatureConfig to check
what should be stored, but for now, the NeverStorageValidator will be
used for all things to ensure that nothing is ever stored for now.

This CL sets up the ownership of the StorageValidator, but does not
update the ModelImpl to use the new framework, and leaves that as
TODOs.

BUG=706309

Change-Id: Ic581edc73f8f059c18fb1dcb5a549c4e62f490f8
Reviewed-on: https://chromium-review.googlesource.com/484682Reviewed-by: David Trainor <dtrainor@chromium.org>
Commit-Queue: Tommy Nyquist <nyquist@chromium.org>
Cr-Commit-Position: refs/heads/master@{#466875}

Revert of chromeos: Add CPU temperature reader (patchset #11 id:220001 of https://codereview.chromium.org/2823583002/ )

Reason for revert:
This change causes a build error.

Original issue's description:
> chromeos: Add CPU temperature reader
>
> Reads from hwmon in sysfs.
>
> This does the same thing as ReadCPUTempInfo() in
> chrome/browser/chromeos/policy/device_status_collector.cc, but it is
> available to the rest of Chrome.
>
> BUG=chromium:709102
>
> Review-Url: https://codereview.chromium.org/2823583002
> Cr-Commit-Position: refs/heads/master@{#466858}
> Committed: https://chromium.googlesource.com/chromium/src/+/745fad9fe0f68a795a191a32cb038d864bfc2dbf

TBR=derat@chromium.org,stevenjb@chromium.org,sque@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:709102

Review-Url: https://codereview.chromium.org/2843433005
Cr-Commit-Position: refs/heads/master@{#466874}

Cr-Commit-Position: refs/heads/master@{#466873}

Reland of Add U2F request state machines (patchset #1 id:1 of https://codereview.chromium.org/2838573002/ )

Reason for revert:
Missing initialized variable can be initialized and this patchset can be merged again.

Original issue's description:
> Revert of Add U2F request state machines (patchset #5 id:180001 of https://codereview.chromium.org/2821263005/ )
>
> Reason for revert:
> Reverting because U2fRequestTest.TestBasicMachine fails consistently on MSan bots (Linux MSan, Linux ChromiumOS MSan).
>
> Original issue's description:
> > Add U2F request state machines
> >
> > U2fRequest base class is extended by each type of U2F
> > request. Common functions of enumeration and device iteration
> > are defined in the base class. U2fSign performs an entire
> > sign request on any attached U2F devices, and U2FRegister
> > performs the registration request on the devices.
> >
> > BUG=686310
> >
> > Review-Url: https://codereview.chromium.org/2821263005
> > Cr-Commit-Position: refs/heads/master@{#466529}
> > Committed: https://chromium.googlesource.com/chromium/src/+/892eb34117f811aa62c821c9129dc53110266e9a
>
> TBR=reillyg@chromium.org,kpaulhamus@chromium.org,juanlang@chromium.org,piperc@google.com,piperc@chromium.org
> # Not skipping CQ checks because original CL landed more than 1 days ago.
> BUG=686310
>
> Review-Url: https://codereview.chromium.org/2838573002
> Cr-Commit-Position: refs/heads/master@{#466666}
> Committed: https://chromium.googlesource.com/chromium/src/+/80f6eac95d5a0b54894b1526ba505308efd5ab23

TBR=reillyg@chromium.org,kpaulhamus@chromium.org,juanlang@chromium.org,piperc@google.com,pkalinnikov@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=686310

Review-Url: https://codereview.chromium.org/2835133003
Cr-Commit-Position: refs/heads/master@{#466872}

Revert of Add --headless flag to Windows (patchset #29 id:560001 of https://codereview.chromium.org/2762593002/ )

Reason for revert:
Testing revert on --headless flag, due to increase in child_dll size: https://bugs.chromium.org/p/chromium/issues/detail?id=714841

Original issue's description:
> Launch bug: https://bugs.chromium.org/p/chromium/issues/detail?id=705916
>
> Add --headless flag to Windows:
>
> Fix browser_tests not compiling properly on windows
> Fix headless/BUILD.gn by:
>    - Changing headless_lib to a headless component, following steps in https://chromium.googlesource.com/chromium/src/+/master/docs/component_build.md and fixing the build for linux, since it now compiles as a shared_library when component build is true.
>    - Adding neccesary HEADLESS_EXPORT (submitted in https://codereview.chromium.org/2775693003/)
>    - Disabling breakpad in windows (will fix in a separate CL)
>
>
> BUG=686608
>
> Review-Url: https://codereview.chromium.org/2762593002
> Cr-Original-Commit-Position: refs/heads/master@{#466176}
> Committed: https://chromium.googlesource.com/chromium/src/+/e3c745d4b37a659afaa0358c7de7b9bc881decb5
> Review-Url: https://codereview.chromium.org/2762593002
> Cr-Commit-Position: refs/heads/master@{#466259}
> Committed: https://chromium.googlesource.com/chromium/src/+/d238dae172d63175b562f99fc988e47e5213a244

TBR=alexclarke@chromium.org,sky@chromium.org,skyostil@chromium.org,wfh@chromium.org,jzfeng@chromium.org,nick@chromium.org,hans@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.
BUG=686608

Review-Url: https://codereview.chromium.org/2837093003
Cr-Commit-Position: refs/heads/master@{#466871}

We recently added a GetAll() method to retrieve all arguments from a
gin::Arguments object without conversion. Use it.

BUG=653596

Review-Url: https://codereview.chromium.org/2843673002
Cr-Commit-Position: refs/heads/master@{#466870}

R=stevenjb@chromium.org
BUG=714103
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2842543002
Cr-Commit-Position: refs/heads/master@{#466869}

The renderer OOMs on system_health.common benchmark on Android One
device because of tracing and other metrics being enabled. So, reduce
the number of article reads.

BUG=714650
R=nednguyen@google.com,perezju@chromium.org

Review-Url: https://codereview.chromium.org/2842593002
Cr-Commit-Position: refs/heads/master@{#466868}

There's no need to pass in the Isolate, since we already pass in the
context. Remove the Isolate argument from APIBinding::CreateInstance and
APIBindingsSystem::CreateAPIInstance.

BUG=653596

Review-Url: https://codereview.chromium.org/2842673002
Cr-Commit-Position: refs/heads/master@{#466867}

A browser window might be positioned over the edge between two screens. In that
case we try to show the drop down menu on the screen the mouse cursor is on, but
this does not work as expected on Windows. To achieve this, we obtain the bounds
of the screen the mouse cursor is on, and then adjust the position of the drop
down menu so that the menu stays within that bounds. But on Windows, we use the
bounding rectangle of all display monitors instead of a specific monitor, and
this leads to a different outcome than we expected. This CL fixes this problem.

BUG=713690

Review-Url: https://codereview.chromium.org/2831323002
Cr-Commit-Position: refs/heads/master@{#466866}

The blocking pool is being deprecated in favor of TaskScheduler.

BUG=667892
R=dewittj@chromium.org

Review-Url: https://codereview.chromium.org/2840653002
Cr-Commit-Position: refs/heads/master@{#466865}