GitLab

Using parameterized tests makes debugging test failures much easier as
gtest can actually print which test failed.

Review URL: https://codereview.chromium.org/1785563003

Cr-Commit-Position: refs/heads/master@{#380481}

The following tests are currently failing on Win10 with image diffs:
  fast/writing-mode/japanese-rl-selection.html
  fast/text/emphasis-vertical.html
  fast/writing-mode/japanese-lr-selection.html

BUG=593917
TBR=kojii@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/1786503002

Cr-Commit-Position: refs/heads/master@{#380480}

Component builds use CRT DLLs and the set of DLLs has changed
dramatically for VS 2015. This updates base/BUILD.gn to add the
necessary files to the .isolate file, including the Universal Windows
10 CRT files.

This is the .gn version of crrev.com/1676943002

BUG=440500

Review URL: https://codereview.chromium.org/1783973002

Cr-Commit-Position: refs/heads/master@{#380479}

More followup cleanup to the scroll position code. This is a minor
refactoring to push the scroll position double concent into
updateScrollbars since it does not need to be plumbed at every callsite.
scrollPositionDouble has some subtlety (see comment above it) which is
easy to mess up.

Review URL: https://codereview.chromium.org/1758413002

Cr-Commit-Position: refs/heads/master@{#380478}

Depends on https://codereview.chromium.org/1781003002

BUG=593420
R=iannucci@chromium.org

Review URL: https://codereview.chromium.org/1783903002 .

Cr-Commit-Position: refs/heads/master@{#380477}

This appears flaky just like AccessibilityModalDialogClosed which was
disabled earlier.

BUG=593846
TBR=dmazzoni@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/1783983002

Cr-Commit-Position: refs/heads/master@{#380476}

The |usb_detachable_whitelist| setting contains the list of USB devices
(identified by their USB VID:PID pairs) which can be detached from the kernel
drivers in order to use them in web application.
The list is used by the permission_broker daemon.

BUG=chrome-os-partner:50249
TEST=none

Review URL: https://codereview.chromium.org/1714473002

Cr-Commit-Position: refs/heads/master@{#380475}

Some tests need to be disabled because they create the same histogram with different parameters, a condition which does not always crash in the calling code but does with this CL.

This CL will be reverted once the necessary information has been collected, including re-enabling of any tests.

SHERIFFS: If a test breaks in the waterfall, you can either revert this CL or disable the test.  I'll re-enable it when reverting this after the necessary data has been collected.

BUG=588946

Committed: https://crrev.com/ce1315b94863f40d6323fa488490ed866aae4ed4
Cr-Commit-Position: refs/heads/master@{#378957}

Committed: https://crrev.com/63f1c5d846f6dac2cf148b81961c38b88ff1bd86
Cr-Commit-Position: refs/heads/master@{#379323}

Committed: https://crrev.com/89dca78efa00acb8fc8749adcc1a08a24eb105d5
Cr-Commit-Position: refs/heads/master@{#379489}

Committed: https://crrev.com/1375777f66fcf8edd3f393dadd9d9de6a7409030
Cr-Commit-Position: refs/heads/master@{#379690}

Committed: https://crrev.com/0abc5c2d2cb7ac70e26eab73d7e3f61f0472461e
Cr-Commit-Position: refs/heads/master@{#380070}

Review URL: https://codereview.chromium.org/1719363002

Cr-Commit-Position: refs/heads/master@{#380474}

BUG=

Review URL: https://codereview.chromium.org/1785713002

Cr-Commit-Position: refs/heads/master@{#380473}

BUG=570102

Review URL: https://codereview.chromium.org/1784643002

Cr-Commit-Position: refs/heads/master@{#380472}

For some reason flaky tests still run on the main waterfall.

BUG=592034

Review URL: https://codereview.chromium.org/1780943003

Cr-Commit-Position: refs/heads/master@{#380471}

This test was still flaky because sometimes the tests
were running before the image is loaded. Moving
the runtests to onload so that we are sure the
image is loaded.

BUG=469657

Review URL: https://codereview.chromium.org/1775613004

Cr-Commit-Position: refs/heads/master@{#380470}

This patch uses url_formatter::FormatUrlForSecurityDisplay to construct
|origin_string| at BluetoothChooserAndroid constructor.

BUG=593120

Review URL: https://codereview.chromium.org/1771353004

Cr-Commit-Position: refs/heads/master@{#380469}

This avoids raising the webgl infobar which forces a reload, when
the GPU process got likely OOM killed, which we unfortunately
can only infer from the fact that we were in the background.

Note that on Android, NORMAL_TERMINATION here implies that the
entire application was in the background.

So that we don't have webgl context restore fight with
the OOM killer, this also requires
https://codereview.chromium.org/1756633002/.

BUG=586905
NOTRY=True

Review URL: https://codereview.chromium.org/1749263003

Cr-Commit-Position: refs/heads/master@{#380468}

BUG=496258

Review URL: https://codereview.chromium.org/1769783002

Cr-Commit-Position: refs/heads/master@{#380467}

Resource already has copies.

BUG=

Review URL: https://codereview.chromium.org/1757633005

Cr-Commit-Position: refs/heads/master@{#380466}

BUG=593888
TBR=tkent@chromium.org,haraken@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/1782743004

Cr-Commit-Position: refs/heads/master@{#380465}

This also narrows some of the specified catapult directories to avoid
isolating directories containing dependencies downloaded from cloud storage.

BUG=589318

Review URL: https://codereview.chromium.org/1783603003

Cr-Commit-Position: refs/heads/master@{#380464}

We always build third_party/freetype2 as a dynamic library even for
static build to avoid symbol conflicts. We need to explicitly set
rpath for static build [1] to let content_shell find our own
libfreetype instead of the system one.

[1] We omit rpath for static gn builds. See
    https://codereview.chromium.org/1535863002.

Review URL: https://codereview.chromium.org/1781903002

Cr-Commit-Position: refs/heads/master@{#380463}

Moved back to libvpx in
https://codereview.chromium.org/1734613003/

WebRTC removed their reference in
https://codereview.chromium.org/1775633002

specificially https://codereview.webrtc.org/1766643002

R=brettw@chromium.org

Review URL: https://codereview.chromium.org/1776643003

Cr-Commit-Position: refs/heads/master@{#380462}

On ChromeOS (Pit), we observed remote video hang when running
apprtc loopback. The cause is that when the tab is inactive,
renderer does not pull frames from WebMediaPlayerMS, and it
relies on WebMediaPlayerMS to digest old frames.

In case of hardware decoding, we have only 4 decode buffers.
If we use up all 4 of them, no more decoding can happen, and
the video will hang. The old algorithm digesting old frames is
conservative, and then there is a chance that we use up all 4
buffers.

In this CL, we switch to an aggressive way. It is safe, as when
the tab is inactive, no one cares about what happens behind the
scenes.

BUG=588813

Review URL: https://codereview.chromium.org/1772353003

Cr-Commit-Position: refs/heads/master@{#380461}

When matching rules are fetched in DataUseTabModel, it involves one
thread hop from UI thread to ExternalDataUseObserver in the IO thread.
Remove this thread hop by DataUseTabModel directly having a pointer to
the ExternalDataUseObserverBridge.

BUG=586235

Review URL: https://codereview.chromium.org/1772273002

Cr-Commit-Position: refs/heads/master@{#380460}

Previously the |foo::Bar::Name_| field was "foo::Bar". With this change it
becomes "foo.Bar".

Using C++-style namespace in the name doesn't makes too much sense because the
name should remain the same across different languages.

The original CL was https://codereview.chromium.org/1778793004/

BUG=None
TBR=ben@chromium.org

Review URL: https://codereview.chromium.org/1782543004

Cr-Commit-Position: refs/heads/master@{#380459}

BUG=276753

Review URL: https://codereview.chromium.org/1753663002

Cr-Commit-Position: refs/heads/master@{#380458}

also: avoid drawing last group padding shadow when no groups are specified.

BUG=592363

Review URL: https://codereview.chromium.org/1782713002

Cr-Commit-Position: refs/heads/master@{#380457}

On linux we can get events outside the bounds of the window, which
causes problems for WindowManagerState. This fixes that.

BUG=none
TEST=none
R=ben@chromium.org

Review URL: https://codereview.chromium.org/1783553004

Cr-Commit-Position: refs/heads/master@{#380456}

It is flaky on those devices.  Restricting it while we investigate
further.

TBR=mlamouri@chromium.org
BUG=593840

Review URL: https://codereview.chromium.org/1785763002

Cr-Commit-Position: refs/heads/master@{#380455}

The "create_bundle" target recursively collect information from all the
"bundle_data" targets it depends and create a bundle from them. This is
configurable in many way to support all kinds of bundles on OS X or iOS.

Design: https://docs.google.com/document/d/1bKh57hg6TSBsEmeh0zWpOO5SVGA2INu-D3FGgsyICzk/view

BUG=297668

Review URL: https://codereview.chromium.org/1752033002

Cr-Commit-Position: refs/heads/master@{#380454}

BUG=593846
TBR=dmazzoni@chromium.org

Review URL: https://codereview.chromium.org/1782653004

Cr-Commit-Position: refs/heads/master@{#380453}

TBR=sky@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1782983002

Cr-Commit-Position: refs/heads/master@{#380452}

Review URL: https://codereview.chromium.org/1782583003

Cr-Commit-Position: refs/heads/master@{#380451}

Currently ArrayBufferContents' adjustAmountOfMemory is registered in initializeWithoutV8
but this is not nice because adjustAmountOfMemory is a method used *with* V8.
The registration should be moved to V8Initializer::initialize.

BUG=593269

Review URL: https://codereview.chromium.org/1776943002

Cr-Commit-Position: refs/heads/master@{#380450}

R=dpapad@chromium.org
BUG=none
TEST=more reviews!
NOTRY=true

Review URL: https://codereview.chromium.org/1776973004

Cr-Commit-Position: refs/heads/master@{#380449}

BUG=557222

Review URL: https://codereview.chromium.org/1758463002

Cr-Commit-Position: refs/heads/master@{#380448}

This shows RAIL strip on top of Timeline even in case there are no actual
records present (but still conditional on RAIL experiment).

Review URL: https://codereview.chromium.org/1780803002

Cr-Commit-Position: refs/heads/master@{#380447}

NOTRY=true
BUG=none

Review URL: https://codereview.chromium.org/1779203002

Cr-Commit-Position: refs/heads/master@{#380446}

SpdyFramer::WriteHeaderBlock() and
SpdyFramer::SerializeHeaderBlockWithoutCompression() are identical, except one
of them is static.  This CL slays this redundancy.

* Remove WriteHeaderBlock().
* Change SerializeHeaderBlockWithoutCompression() to public so that it can be
  called from net/quic/spdy_utils.cc.
* Instantiate a SpdyFramer in net/quic/spdy_utils.cc
  (WriteHeaderBlock used to be static) and call SerializeUncompressedHeaders on
  it.

This CL lands server change 116771332 by bnc.

BUG=521600, 488484

Review URL: https://codereview.chromium.org/1776373003

Cr-Commit-Position: refs/heads/master@{#380445}

This is a change to gather stability, performance, and size
data and will be rolled back by 2016-03-18.

This is the gyp portion of the change; the GN variables
controlling the same things currently reside in the pdfium
product itsef.

Original Review URL: https://codereview.chromium.org/1761673002

Cr-Commit-Position: refs/heads/master@{#379622}
(cherry picked from commit afa43e7f)

Review URL: https://codereview.chromium.org/1775173002

Cr-Commit-Position: refs/heads/master@{#380444}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/3c3b9fbb..70143b44

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.

TBR=hablich@chromium.org,machenbach@chromium.org,yangguo@chromium.org,vogelheim@chromium.org

Review URL: https://codereview.chromium.org/1778373003

Cr-Commit-Position: refs/heads/master@{#380443}

BUG=579697

Review URL: https://codereview.chromium.org/1785643004

Cr-Commit-Position: refs/heads/master@{#380442}