GitLab

Clusterfuzz identified a race between the media thread calling SetVolume
and the audio device thread calling ProvideInput. Add a lock to
synchronize access to |volume_| between threads.

Also adds some thread_checkers to just firm up the contract that the
majority of these methods are called only by the media thread.

BUG=588992

Committed: https://crrev.com/57c6958a99f2c62a4408d2a9d262a22129309106
Cr-Commit-Position: refs/heads/master@{#379349}

Review URL: https://codereview.chromium.org/1748183006

Cr-Commit-Position: refs/heads/master@{#381763}

Review URL: https://codereview.chromium.org/1811023002

Cr-Commit-Position: refs/heads/master@{#381762}

Although the root cause of crash is not clear now,
maybe we'd better add guard firstly to avoid crash.

BUG=594644

Review URL: https://codereview.chromium.org/1813673002

Cr-Commit-Position: refs/heads/master@{#381761}

This fixed the issue that enum generated code didn't work when variant is
specified.
This is a quick fix, we need more fundermental improvement for variant handling
code.

BUG=595791

Review URL: https://codereview.chromium.org/1811853004

Cr-Commit-Position: refs/heads/master@{#381760}

1.) Added |suppressDialogs| property to CRWWebController, which should
    be used instead of |setPageDialogOpenPolicy|.
2.) Suppress all dialogs in native code instead of JavaScript (except
    navigation dialogs which can not be suppressed w/o JS override).
3.) Suppress HTTP Auth dialog if |suppressDialogs| is YES.

BUG=595463
TEST=window.open, window.prompt, window.confirm, window.alert,
           geolocation and http auth dialogs should be blocked for preload
           pages. All these dialogs should work normally for regular pages.

Review URL: https://codereview.chromium.org/1806043004

Cr-Commit-Position: refs/heads/master@{#381759}

BUG=595490
TBR=zhaoqin@chromium.org

Review URL: https://codereview.chromium.org/1812963003

Cr-Commit-Position: refs/heads/master@{#381758}

This CL updates the section names (shown on the
side nav and above each section).  They use
sentence case.  The & symbol is replaced with
'and'. The Reset and Languages names have been
shortened.

BUG=593989

Review URL: https://codereview.chromium.org/1803363003

Cr-Commit-Position: refs/heads/master@{#381757}

When the whitelist is created we save the path to the locally stored
icon. During the suggestions creation save that information and when
trying to get the large icon for the NTP, first check if we have a
locally stored one first.

This does not yet handle the case when whitelist entry point becomes
part of most visited.

For more information on whitelists - go/chrome-whitelists

Screenshot - https://screenshot.googleplex.com/3eTZ3wghRvH.png

BUG=586097

Review URL: https://codereview.chromium.org/1772363002

Cr-Commit-Position: refs/heads/master@{#381756}

BUG=595772
NOTRY=true

Review URL: https://codereview.chromium.org/1814903002

Cr-Commit-Position: refs/heads/master@{#381755}

BUG=589630

Review URL: https://codereview.chromium.org/1798133002

Cr-Commit-Position: refs/heads/master@{#381754}

[ChromeDriver] Add nested frame xpath expression in order to switch to frame present inside nested frameset.

BUG=chromedriver:107

Review URL: https://codereview.chromium.org/1799183002

Cr-Commit-Position: refs/heads/master@{#381753}

This should allow device switching even when the new device
is specified as an advanced constraint or an old-style "optional".

BUG=589669

Review URL: https://codereview.chromium.org/1806303002

Cr-Commit-Position: refs/heads/master@{#381752}

Dependency resolution in gyp uses relative paths from the target, so usage
of the `for_blink` variable fails for any target that doesn't immediately
live in //mojo. This is different from GN, where the dependency is
declared absolutely: //mojo/public/cpp/bindings:wtf_support

BUG=583738

Review URL: https://codereview.chromium.org/1808253002

Cr-Commit-Position: refs/heads/master@{#381751}

Should be fixed with https://codereview.chromium.org/1800283002/

BUG=592594,592627,591224

Review URL: https://codereview.chromium.org/1814593002

Cr-Commit-Position: refs/heads/master@{#381750}

This change has been mirrored from google3 png fuzzer. Should prevent some timeouts.

R=aizatsky@chromium.org, inferno@chromium.org, kcc@chromium.org, krasin@chromium.org
BUG=584819
TBR=aizatsky@chromium.org

Review URL: https://codereview.chromium.org/1808123004

Cr-Commit-Position: refs/heads/master@{#381749}

This patch adds a unittest that fails without the patch, and passes
with it.

R=enne
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review URL: https://codereview.chromium.org/1814643002

Cr-Commit-Position: refs/heads/master@{#381748}

system_webview_apk target only exists when
use_webview_internal_framework is false. Removing this additional_apk
to fix the build.

BUG=595605
TBR=sgurun@chromium.org

Review URL: https://codereview.chromium.org/1811033002

Cr-Commit-Position: refs/heads/master@{#381747}

The //crypto library uses some constants that comes from the Security
framework (kSecClassGenericPassword, ...). Add this framework to the
list of libraries to link on iOS and OS X.

BUG=459705

Review URL: https://codereview.chromium.org/1797243002

Cr-Commit-Position: refs/heads/master@{#381746}

BUG=None
R=ttuttle@chromium.org

Review URL: https://codereview.chromium.org/1811983003

Cr-Commit-Position: refs/heads/master@{#381745}

It2me host can use spake2_hmac authentication method and it works
properly with the current client, so there is no reason to keep
spake2_plain for It2Me.

Review URL: https://codereview.chromium.org/1799293002

Cr-Commit-Position: refs/heads/master@{#381744}

BUG=579697

Review URL: https://codereview.chromium.org/1808993002

Cr-Commit-Position: refs/heads/master@{#381743}

According to the style guide a nested codeblock within a list, which is
the case we are changing here, should be indented, otherwise it will
break the numbering of the list, like it is now. Without this
indentation the numbering looks like: 1., 2., 1.. But that is just
wrong, and should be 1., 2., 3.. The indentation should restore the
right numbering here.

https://github.com/google/styleguide/blob/gh-pages/docguide/style.md#nest-codeblocks-within-lists

BUG=None
R=sdefresne@chromium.org

Review URL: https://codereview.chromium.org/1811613003

Cr-Commit-Position: refs/heads/master@{#381742}

Rationale: The benchmark is consistently failing on the "Android One
Perf (2)" bot (see the associated bug).

BUG=593973
TBR=skyostil
CQ_EXTRA_TRYBOTS=tryserver.chromium.perf:android_s5_perf_cq;tryserver.chromium.perf:winx64_10_perf_cq;tryserver.chromium.perf:mac_retina_perf_cq;tryserver.chromium.perf:linux_perf_cq

Review URL: https://codereview.chromium.org/1804413004

Cr-Commit-Position: refs/heads/master@{#381741}

https://codereview.chromium.org/1781433002 enabled auto-resize for
constrained web dialogs on Mac. It updated a number of callsites for
`new ConstrainedWindowMac`, but didn't update the callsite in
`password_prompt_view_bridge.mm`.

Tests continued to pass, as the chooser _was_ being popped up, it was
simply constrained to a 0x0 frame. According to
https://codereview.chromium.org/1781433002/#msg15, it's not clear that
we can add a test that would verify that this doesn't regress. That
seems unfortunate. :(

BUG=217034,
R=vabr@chromium.org, apacible@chromium.org, thakis@chromium.org
TBR=vasilii@chromium.org

Review URL: https://codereview.chromium.org/1809043002

Cr-Commit-Position: refs/heads/master@{#381740}

VideoResourceUpdater software planes use VideoFrame raw pointer and
video timestamp as the unique ID for each VideoFrame. However, if the
VideoFrame is destructed and the pointer is reused by a new VideoFrame,
and if the timestamp is not correctly set, the resource maybe
incorrectly reused. This CL marks the resource as destructed when the
original VideoFrame is destructed, and monitors if the destructed
resource is re-used. The CL only takes effect in Debug mode.

BUG=581480
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Review URL: https://codereview.chromium.org/1688033005

Cr-Commit-Position: refs/heads/master@{#381739}

While we have an inaccessible browser find, let ChromeVox take care of find using its incremental search.

TEST=ctrl+f brings up i-search.

Review URL: https://codereview.chromium.org/1807843005

Cr-Commit-Position: refs/heads/master@{#381738}

This adds a heuristic such that if a study only references a single
feature from any of its groups, its other groups will be associated
with that feature as well, for reporting only. This allows a config like
25% Enabled, 25% Disabled and 50% Default to still report the users
in the default group to UMA, so that they show up in the data.

More details in the code comments.

BUG=587135

Review URL: https://codereview.chromium.org/1809633003

Cr-Commit-Position: refs/heads/master@{#381737}

Revert of Shape unicode-range: font faces in only one iteration (patchset #7 id:180001 of https://codereview.chromium.org/1806653002/ )

Reason for revert:
Reverting on a suspicion that this is a root cause to the failures of webkit_unit_tests on Webkit Android (Nexus4) bots, e.g.
https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Android%20%28Nexus4%29/builds/44725

Original issue's description:
> Shape unicode-range: font faces in only one iteration
>
> Previously, we incorrectly split CSS composite faces into multiple faces
> for shaping, one for each comma-separated entry of unicode-range:. This
> breaks shaping of ligatures and other features when the characters that
> ought to be shaped combined were in different unicode-range entries. It
> is also inefficient for subsetted web fonts that use unicode-range:
> extensively, for example Google Fonts and Adobe TypeKit.
>
> The fix is to transfer the UnicodeRangeSet information from CSSFontFace
> to HarfBuzzFace and only restrict the glyph lookup function to the whole
> unicode-range information, instead of restricting it to a single entry
> and shaping multiple times with the same face. This should have a slight
> performance benefit as well.
>
> BUG=583450
> TEST=fast/css/font-face-unicode-range-ligatures.html
> R=eae, behdad
>
> Committed: https://crrev.com/9694005f93116f9c9cc73fa99132fa4475a0cdab
> Cr-Commit-Position: refs/heads/master@{#381683}

TBR=behdad@chromium.org,eae@chromium.org,drott@chromium.org
# Skipping CQ checks because original CL landed less than 1 days ago.
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=583450

Review URL: https://codereview.chromium.org/1806363002

Cr-Commit-Position: refs/heads/master@{#381736}

The inclusion of sys/syscall.h was introduced in crrev.com/4916 to
obtain the current thread ID via `syscall(__NR_gettid)`. The code was
later changed in crrev.com/118152 to use
`base::PlatformThread::CurrentId()` instead. The inclusion of
sys/syscall.h is no longer needed and thus removed by this CL.

BUG=None

Review URL: https://codereview.chromium.org/1812733002

Cr-Commit-Position: refs/heads/master@{#381735}

BUG=none
R=dmazzoni@chromium.org

Review URL: https://codereview.chromium.org/1812783002

Cr-Commit-Position: refs/heads/master@{#381734}

BUG=595304

Review URL: https://codereview.chromium.org/1810613002

Cr-Commit-Position: refs/heads/master@{#381733}

If a SharedBuffer was unlocked and the OS then purged the
backing DiscardableMemory, we will evict the Resource from
MemoryCache. However, before doing so, we will give inspector
a chance to copy the data. In the case where the buffer was
purged, InspectorPageAgent may improperly handle the buffer
given its state. In that case, clear the buffer so there is
no opportunity for the inspector to misuse it.

BUG=594117
TEST=ResourceTest.LockFailureNoCrash

Review URL: https://codereview.chromium.org/1802313003

Cr-Commit-Position: refs/heads/master@{#381732}

BUG=

Review URL: https://codereview.chromium.org/1807233003

Cr-Commit-Position: refs/heads/master@{#381731}

https://chromium.googlesource.com/external/github.com/catapult-project/catapult.git/+log/1d2c8cd5bebc..973a51b36f5e

$ git log 1d2c8cd5b..973a51b36 --date=short --no-merges --format='%ad %ae %s'

CQ_INCLUDE_TRYBOTS=
TBR=catapult-sheriff@chromium.org

Review URL: https://codereview.chromium.org/1809943003

Cr-Commit-Position: refs/heads/master@{#381730}

Previously, each time a SceneLayer was set in the compositor view the
root was checked against the cached root. If the check failed, the
layer would be removed from its parent and the current layer would be
stored.

In the case of scene overlays, the root element can change quite
frequently. In this case, the content of the tree does not necessarily
change, but the root does. This root change would cause the cached
layer to be removed from the new tree before it was rendered.

This change checks that the cached root has the new root as an
ancestor before attempting to detach it.

BUG=584340

Review URL: https://codereview.chromium.org/1801243003

Cr-Commit-Position: refs/heads/master@{#381729}

https://chromium.googlesource.com/skia.git/+log/0e3738db89e8..d427266a58af

$ git log 0e3738db8..d427266a5 --date=short --no-merges --format='%ad %ae %s'
2016-03-17 msarett Remove uses of SkImageDecoder from gms
2016-03-17 msarett Remove uses of SkImageDecoder from skdiff
2016-03-17 caryclark make sausages rounder and portable
2016-03-17 robertphillips Add SkSpecialImage::extractSubset & NewFromPixmap
2016-03-17 msarett Remove SkipZeroesBench

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=bungeman@google.com

Review URL: https://codereview.chromium.org/1811853003

Cr-Commit-Position: refs/heads/master@{#381728}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/fec11bb7..b364ca30

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/1810013002

Cr-Commit-Position: refs/heads/master@{#381727}

https://codereview.chromium.org/1777903003 introduced several calls to
TestRenderFrameHost::PrepareForCommit without calling
TestRenderFrameHost::SendRendererInitiatedNavigationRequest when simulating
renderer initiated navigations. This makes them fail with PlzNavigate enabled.
This CL fixes this fact.

BUG=439423
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_site_isolation

Review URL: https://codereview.chromium.org/1800123002

Cr-Commit-Position: refs/heads/master@{#381726}

The former is a possible candidate for deletion, the latter is a suitable replacement in the new navigation API.

BUG=576287

Review URL: https://codereview.chromium.org/1806593002

Cr-Commit-Position: refs/heads/master@{#381725}

Previous behavior: A list of optional constraints resulted
in a single advanced element, with one field for each value.

New behavior: A list of optional constraints results in a
list of advanced elements, with one field set for each value.

The difference is important when there are mulitple values
with the same name. For constraints with distinct names, there
should be no difference.

BUG=594929

Review URL: https://codereview.chromium.org/1808173002

Cr-Commit-Position: refs/heads/master@{#381724}