GitLab

Initialize in-process gpu before threads are created.
In-process gpu init appends command line switches to the
global current process command line, which is only safe to
do before any background threads start and read the command
line.

BUG=450396

Review URL: https://codereview.chromium.org/879703003

Cr-Commit-Position: refs/heads/master@{#313387}

BUG=
TBR=stgao@chromium.org

Review URL: https://codereview.chromium.org/883743003

Cr-Commit-Position: refs/heads/master@{#313386}

BUG=99379

Review URL: https://codereview.chromium.org/880093002

Cr-Commit-Position: refs/heads/master@{#313385}

* Remove old versions of createChildFrame.
* Add browser tests.

Related CLs:
Chromium part 1: https://codereview.chromium.org/837283003/
Blink part 1: https://codereview.chromium.org/793493003/
Blink part 2: https://codereview.chromium.org/838903002/

BUG=426512

Review URL: https://codereview.chromium.org/797813006

Cr-Commit-Position: refs/heads/master@{#313384}

Currently the SiteInstance and RenderProcessHost for a RenderFrameHost are retrieved from the associated RenderViewHost. The RFH code is sufficiently along the way to keep track of these on its own.

BUG=304341

Review URL: https://codereview.chromium.org/881683002

Cr-Commit-Position: refs/heads/master@{#313383}

BUG=

Review URL: https://codereview.chromium.org/843453002

Cr-Commit-Position: refs/heads/master@{#313382}

Some servers get confused when we send PASV twice. This CL delays sending
PASV/EPSV until RETR or LIST.

BUG=450687

Review URL: https://codereview.chromium.org/861193007

Cr-Commit-Position: refs/heads/master@{#313381}

BUG=429053
TEST=gpu_unittests
R=piman@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/878793002

Cr-Commit-Position: refs/heads/master@{#313380}

Before, it was not well defined what's the name of a root entry.

TEST=None
BUG=446437

Review URL: https://codereview.chromium.org/881903002

Cr-Commit-Position: refs/heads/master@{#313379}

On VS2015 without this causes:

d:\src\cr3\src\base\metrics\histogram.cc(372): error C2248: 'scoped_ptr<base::SampleVector,base::DefaultDeleter<T>>::scoped_ptr': cannot access private member declared in class 'scoped_ptr<base::SampleVector,base::DefaultDeleter<T>>'
        with
        [
            T=base::SampleVector
        ]
d:\src\cr3\src\base\memory\scoped_ptr.h(312): note: see declaration of 'scoped_ptr<base::SampleVector,base::DefaultDeleter<T>>::scoped_ptr'
        with
        [
            T=base::SampleVector
        ]

I'm not entirely sure why this doesn't happen on earlier compilers,
I guess something's move-ability changed.

R=isherman@chromium.org
BUG=440500

Review URL: https://codereview.chromium.org/882783002

Cr-Commit-Position: refs/heads/master@{#313378}

BUG=451983
TBR=thestig@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/881933002

Cr-Commit-Position: refs/heads/master@{#313377}

Version 0.8.4.1

[----------] Fixed some unittest failures.

Version 0.8.4.0

[5130039b66] Synchronize page protection modifications.
[e204e630ba] SyzyASan - Fix an issue when the first block of a corrupt range has
             page protections.
[aef343e391] Preserve alignment of code blocks generated by unsupported
             compilers.

See https://code.google.com/p/syzygy/source/browse/syzygy/build/RELEASE-NOTES.TXT for the full version of the release notes.
BUG=

Review URL: https://codereview.chromium.org/882803002

Cr-Commit-Position: refs/heads/master@{#313376}

The results are summarized here: goto/disallow-inlining-trial-results

This is effectively a revert of
https://src.chromium.org/viewvc/chrome?revision=273966&view=revision
and
http://src.chromium.org/viewvc/chrome?view=revision&revision=278811

BUG=

Review URL: https://codereview.chromium.org/879053002

Cr-Commit-Position: refs/heads/master@{#313375}

d:\src\cr3\src\base\json\json_parser.cc(431): error C2220: warning treated as error - no 'object' file generated
d:\src\cr3\src\base\json\json_parser.cc(431): warning C4456: declaration of 'next_char' hides previous local declaration
d:\src\cr3\src\base\json\json_parser.cc(427): note: see declaration of 'next_char'

I don't think there's any need for a new variable inside the loop.

R=thestig@chromium.org
BUG=440500

Review URL: https://codereview.chromium.org/881003003

Cr-Commit-Position: refs/heads/master@{#313374}

d:\src\cr3\src\base\win\registry.cc(665): error C2220: warning treated as error - no 'object' file generated
d:\src\cr3\src\base\win\registry.cc(665): warning C4456: declaration of 'result' hides previous local declaration
d:\src\cr3\src\base\win\registry.cc(659): note: see declaration of 'result'

I don't think there's any need for separate variables.

R=cpu@chromium.org
BUG=440500

Review URL: https://codereview.chromium.org/868253004

Cr-Commit-Position: refs/heads/master@{#313373}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/52b64b4..8ece6eb

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=fmalita@google.com

Review URL: https://codereview.chromium.org/881053002

Cr-Commit-Position: refs/heads/master@{#313372}

BUG=104806,106912,416957
TBR=thestig@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/878783003

Cr-Commit-Position: refs/heads/master@{#313371}

This feature is now known by the name 'bookmark apps'.

BUG=441127

Review URL: https://codereview.chromium.org/883603002

Cr-Commit-Position: refs/heads/master@{#313370}

BUG=449103
TBR=kpschoedel@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/885463002

Cr-Commit-Position: refs/heads/master@{#313369}

Disable DropFileSystemAccessIsSafe and CannotRegainPrivileges on ASAN.

BUG=451603

Review URL: https://codereview.chromium.org/881923002

Cr-Commit-Position: refs/heads/master@{#313368}

BUG=368775
TEST=Casting WebRTC works on daisy and peach_pit

Review URL: https://codereview.chromium.org/880643004

Cr-Commit-Position: refs/heads/master@{#313367}

On X11, ui::Event::GetCharacter() returns 0x0 when the correct return value
when control is pressed is unclear (e.g. Ctrl+[). This CL makes Ozone follow
the same behavior.

BUG=450849
TEST=Manual, see bug

Review URL: https://codereview.chromium.org/872213002

Cr-Commit-Position: refs/heads/master@{#313366}

Add me to OWNERS too to shut up the presubmits.

I tested this locally on OS X / x86-64.  Relying on the bots for the rest.

BUG=

Review URL: https://codereview.chromium.org/859393004

Cr-Commit-Position: refs/heads/master@{#313365}

d:\src\cr3\src\base\logging.cc(167): error C2220: warning treated as error - no 'object' file generated
d:\src\cr3\src\base\logging.cc(167): warning C4459: declaration of 'log_file' hides global declaration
d:\src\cr3\src\base\logging.cc(103): note: see declaration of 'logging::`anonymous-namespace'::log_file'

There was 3 different things called log_file in this file: a global FileHandle,
a local PathString, and a PathChar* member variable in LoggingSettings. :/

(Should probably rename all the globals to g_ prefixed in a followup.)

R=thakis@chromium.org
BUG=440500

Review URL: https://codereview.chromium.org/883853003

Cr-Commit-Position: refs/heads/master@{#313364}

BUG=None
TEST=None
R=binji@chromium.org

Review URL: https://codereview.chromium.org/884523002

Cr-Commit-Position: refs/heads/master@{#313363}

Summary of changes available at:
https://chromium.googlesource.com/external/google-toolbox-for-mac.git/+log/a095262..17eee69

Notable changes include:
17eee69  Make GTMSMJobRemove() have the same behavior on OSX 10.10+ as it has
         on previous OSX versions.

BUG=390276

Review URL: https://codereview.chromium.org/858093004

Cr-Commit-Position: refs/heads/master@{#313362}

d:\src\cr3\src\base\metrics\histogram.cc(498): warning C4457: declaration of 'count' hides function parameter
d:\src\cr3\src\base\metrics\histogram.cc(490): note: see declaration of 'count'

R=isherman@chromium.org
BUG=440500

Review URL: https://codereview.chromium.org/878223002

Cr-Commit-Position: refs/heads/master@{#313361}

With swarming, the run command gets baked into the .isolate file instead
of going through runtest.py.  Usually, an invocation of test_env.py is
baked into the isolate file. In asan mode, this used to call
asan_symbolize.py, which on Windows doesn't work -- prepend it with
sys.executable to make this work on all platforms.

BUG=451741

Review URL: https://codereview.chromium.org/877183002

Cr-Commit-Position: refs/heads/master@{#313360}

BUG=438287

Review URL: https://codereview.chromium.org/856393005

Cr-Commit-Position: refs/heads/master@{#313359}

In old implementaion, when SourceState::Append failed, it will report
DEMUXER_ERROR_COULD_NOT_OPEN error in ChunkDemuxer::AppendData case INITIALIZING.
But in fact, appended data may include initialization segment and media segment both.
The error can happen in any point.
In MSE spec
( http://w3c.github.io/media-source/#byte-stream-formats )
The user agent must run the end of stream algorithm with the error parameter
set to "decode" when in some conditions. They all happen in parsing the appended data.
So, here when SourceState::Append fails, it should report PIPELINE_ERROR_DECODE.

Another side, when the segment parser loop successfully parses a complete
initialization segment, it should call ReportMetadata immediately. Otherwise,
when meeting PIPELINE_ERROR_DECODE but WebMediaPlayer's ready_state_ is still
ReadyStateHaveNothing, it will be treated as NetworkStateFormatError not
NetworkStateDecodeError.

This Cl will resolve above two problems.

BUG=None
TEST=LayoutTest: https://codereview.chromium.org/742653002

Review URL: https://codereview.chromium.org/710693003

Cr-Commit-Position: refs/heads/master@{#313358}

SingleThreadProxy was previously not considering
DidFailToInitializeOutputSurface responses as an outstanding
RequestNewOutputSurface call. This would cause the embedder to start
servicing an output surface request, then composite, receive another,
and then have double output surface requests.

The fix is to consider this failure state as a request.

BUG=444277

Review URL: https://codereview.chromium.org/871743002

Cr-Commit-Position: refs/heads/master@{#313357}

http://test-results.appspot.com/dashboards/flakiness_dashboard.html#tests=RenderFrameHostImplBrowserTest.IsFocused_Change&testType=content_browsertests

BUG=452631
TBR=mlamouri@chromium.org

Review URL: https://codereview.chromium.org/865063003

Cr-Commit-Position: refs/heads/master@{#313356}

BUG=451624
R=raymes@chromium.org

Review URL: https://codereview.chromium.org/874703003

Cr-Commit-Position: refs/heads/master@{#313355}

Cherry-pick potential divide-by-zero fix
https://codereview.chromium.org/877783004

This also removes the last remnants of the AVX attempts:
https://codereview.chromium.org/873313005
It should have been a no-op since it only changes cpu detection. The functions
themselves have been disabled and are not compiled so this is the last remaining
related change.

CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_chromium_vista_rel_ng

R=tomfinegan@chromium.org
BUG=396572

Review URL: https://codereview.chromium.org/877123002

Cr-Commit-Position: refs/heads/master@{#313354}

The test fails or times out.

NOTRY=True
TBR=phoglund@chromium.org
BUG=452623

Review URL: https://codereview.chromium.org/882713003

Cr-Commit-Position: refs/heads/master@{#313353}

R=lcwu@chromium.org,servolk@chromium.org
BUG=None

Review URL: https://codereview.chromium.org/878843004

Cr-Commit-Position: refs/heads/master@{#313352}

This adds a Finch trial that can enable and disable bubbles as needed.
The flags will still override the Finch trial.

BUG=364686
TEST=
Visit https://adrifelt.github.io/bubble-demo.html and check to see if
you get a permission bubbles or an infobar, depending on the setting:
1. Infobar: set chrome://flags/#enable-permissions-bubbles to 'Disabled'
2. Bubble: set chrome://flags/#enable-permissions-bubbles to 'Enabled'
3. Infobar: set chrome://flags/#enable-permissions-bubbles to 'Default'
4. Infobar: --force-fieldtrials=PermissionBubbleRollout/Disabled/
5. Bubble: --force-fieldtrials=PermissionBubbleRollout/Enabled/

Review URL: https://codereview.chromium.org/791033004

Cr-Commit-Position: refs/heads/master@{#313351}

BUG=none

Review URL: https://codereview.chromium.org/873643002

Cr-Commit-Position: refs/heads/master@{#313350}

This patch adds conversion rule for aria-grabbed attribute and
exposes it only when it's defined.

The related patch is
https://codereview.chromium.org/736943002/

BUG=98817

Review URL: https://codereview.chromium.org/736023002

Cr-Commit-Position: refs/heads/master@{#313349}

R=mbarbella@chromium.org, tsepez@chromium.org
BUG=450268

Review URL: https://codereview.chromium.org/881883002

Cr-Commit-Position: refs/heads/master@{#313348}