GitLab

Breakage caused by this case can be found in http://build.chromium.org/p/chromium.perf/builders/Android%20Nexus4%20Perf/builds/1416/steps/page_cycler.bloat/logs/stdio

BUG=

Review URL: https://codereview.chromium.org/610483002

Cr-Commit-Position: refs/heads/master@{#296861}

Update FFmpeg with MIPS config files.

BUG= regression build issue
TEST= build Chromium for MIPS Linux

Review URL: https://codereview.chromium.org/590373004

Cr-Commit-Position: refs/heads/master@{#296860}

This regressed recently with the change to query the device scale factor to see if it needs to be scaled back to 1.0 for
display scales equal to or under 125%. user32/gdi32 API's crash or fail with the Win32K lockdown process mitigation.

Fix is to ensure that the device scale factor is queried from the one set via the command line. The other change is to move
the code in ContentMainRunner::Initialize which sets the global device scale factor for the process to before the content client
initialization. This is because the ShellContentRendererClient in its ctor calls the ShouldUseDirectWrite function which relies on
the GetDPIScale function to decide whether or not directwrite should be enabled. This causes a DCHECK in content_browsertests.

To ensure that the win32k lockdown does not regress I added a simple content browsertest RendererWin32KLockdownNavigationTest which runs
this on Windows 8+.

The accessibility fixes are to get the browser tests to run
on Windows 8. There were special cases to always enable
accessibility on Windows 8 for historical reasons. These are
no longer necessary as we don't rely on accessibility to display the virtual keyboard.

BUG=417147

Review URL: https://codereview.chromium.org/600263002

Cr-Commit-Position: refs/heads/master@{#296859}

Add histograms.xml info for the SRTPrompt, and don't create a browser if we are not going to prompt the user.

TBR=noelutz
For a simple histogram string rename in chrome\browser\safe_browsing\srt_global_error_win.cc

Review URL: https://codereview.chromium.org/607573002

Cr-Commit-Position: refs/heads/master@{#296858}

R=brettw@chromium.org

Review URL: https://codereview.chromium.org/604443002

Cr-Commit-Position: refs/heads/master@{#296857}

Without this, it's hard to know what will be spoken due to the new category
flush queueing mode.

BUG=none

Review URL: https://codereview.chromium.org/593283002

Cr-Commit-Position: refs/heads/master@{#296856}

In RenderPass use ListContainer for generating DrawQuads and acting as
QuadList.

BUG=344962

Review URL: https://codereview.chromium.org/448303002

Cr-Commit-Position: refs/heads/master@{#296855}

 I'll fix issues and add them back.

BUG=414340

Review URL: https://codereview.chromium.org/607603002

Cr-Commit-Position: refs/heads/master@{#296854}

BUG=417871

Review URL: https://codereview.chromium.org/608473004

Cr-Commit-Position: refs/heads/master@{#296853}

A CL was recently introduced that prevented unnecessary tab-strip relayouts.
(https://codereview.chromium.org/593223002/). That CL had a minor error where
it failed to relayout the tab strip when the left and right indentations
changed. This CL corrects that oversight.

This CL also refactors the browser_window_controller tab strip layout logic
into browser_window_layout. There is no intended behavioral change. The tab
strip layout logic now has test coverage.

BUG=415093

Review URL: https://codereview.chromium.org/603243002

Cr-Commit-Position: refs/heads/master@{#296852}

This change is needed to handle new location of
PNaCl libs/include.

I've also switched naclports from svn to git as
the svn repo is no longer recieving new commits.

This brings is 391 new revisions: 87f43d3->99f2417

TEST=build_sdk.py --build-ports
R=binji@chromium.org

Review URL: https://codereview.chromium.org/608673002

Cr-Commit-Position: refs/heads/master@{#296851}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/f1db071..5be896b

TBR=ager@chromium.org,zmo@chromium.org

Review URL: https://codereview.chromium.org/609573003

Cr-Commit-Position: refs/heads/master@{#296850}

The big changes here are:
1) Allow injecting of a target toolchain via args for os==chromeos.
2) For ARM builds, allow injecting -mtune and -mfloat-abi settings.
3) Allow injecting a pkg-config wrapper that correctly handles the
   CrOS build environment. This is how pkg-config is handled for
   all other packages in the build.
4) Create libmojo_sdk target, a static library of public Mojo code
   suitable for distribution.
5) Added testonly = true to a few targets under mojo/ that are not
   meant to be used in production.

BUG=388412
TEST=Create a target that builds //mojo/public, build with os==chromeos

Review URL: https://codereview.chromium.org/549453004

Cr-Commit-Position: refs/heads/master@{#296849}

R=jamesr@chromium.org

Review URL: https://codereview.chromium.org/603983003

Cr-Commit-Position: refs/heads/master@{#296848}

The purpose of this is to continue gathering all the related code together, and to make it easier to run pylint/tests for the main module which still contains most of the code.

After doing this, pylint and the tests caught a couple of things -- one unused variable name, and after the script changed directory, some of the paths had to be modified as well.

BUG=

Review URL: https://codereview.chromium.org/564663002

Cr-Commit-Position: refs/heads/master@{#296847}

Rather than talking to X11, in headless mode, the native viewport service just
drops requests on the floor. This configuration is useful for writing
integration tests for the Mojo system that don't need to interact with X11.

R=jamesr@chromium.org

Review URL: https://codereview.chromium.org/606883002

Cr-Commit-Position: refs/heads/master@{#296846}

This reverts commit 7fb8b995.

BUG=417939
TBR=rockot@chromium.org

Review URL: https://codereview.chromium.org/609633002

Cr-Commit-Position: refs/heads/master@{#296845}

BUG=417399

Review URL: https://codereview.chromium.org/601783004

Cr-Commit-Position: refs/heads/master@{#296844}

BUG=none

Review URL: https://codereview.chromium.org/608573003

Cr-Commit-Position: refs/heads/master@{#296843}

This permission is used by whitelisted apps and extensions to access all
FIDO Alliance Universal 2nd Factor devices without a hard-coded list of
all present and future vendor and product IDs.

BUG=374307

Review URL: https://codereview.chromium.org/601423002

Cr-Commit-Position: refs/heads/master@{#296842}

This brings in the following changes:

01fe820 Add tests for client version negotiation and session resumption.
30ddb43 Handle session resumption in SSLv23_client_method.
b0c8db7 runner: don't resume sessions if SessionTicketsDisabled is true.
7f520db Remove OPENSSL_NO_TLS1_2_CLIENT and OPENSSL_NO_DTLS1.
37d9246 Disallow all special operators once groups are used.
2a5ea98 Remove redundant check in cipher rule parsing.
bb0a17c Add a set of tests for cipher string parsing.
e113608 Switch the reason code check to a compile-time assert.
5127db3 Provide compatibility functions for PKCS#12 parsing.
588d252 Don't try to setup CPUID if NO_ASM.
bed8ce7 Add misc functions for easier porting.
d7c5368 Add missing errors codes for alerts.
fc7b086 Test that ALPN is preferred over NPN.
ae2888f Add tests for ALPN support.
fa055a2 Implement ALPN in runner.go.
812152a Don't deadlock if a resume test fails the first half.
e78bfde Improve test coverage for server_name extension.
594a58e Remove remnants of export cipher suite selection.
d633d63 Remove indirection in loading ciphers.
172fc2c Fix some OPENSSL_PUT_ERROR calls.

BUG=417134

Review URL: https://codereview.chromium.org/604103003

Cr-Commit-Position: refs/heads/master@{#296841}

HTML5FS (PPAPI) does not provide an INO so it is impossible to determine
a files uniqueness.  In addition, node information does not persist in
memory since nodes could be transparently destroyed from JavaScript, thus
creating leaks.  Because of this the current stack based INO allocation
causes sequentially accesed file handles to report the same INO.

HTML5FS (including media galleries) does not traverse links.  This implies
that all paths normalized to the mount are unique.  We use this property
to generate a 64 bit hash.

R=binji@chromium.org
BUG=414938

Review URL: https://codereview.chromium.org/605513002

Cr-Commit-Position: refs/heads/master@{#296840}

Revert of Changes NineImagePainter to snap to (enclosed) pixel boundaries (patchset #2 id:20001 of https://codereview.chromium.org/604463003/)

Reason for revert:
I'm reverting this as there are two issues:
1. Translate isn't being set correctly.
2. Even with 1 fixed there are problems with scrollbars. I'll investigate after the revert.

Original issue's description:
> Changes NineImagePainter to snap to (enclosed) pixel boundaries
>
> Otherwise the sides may be clipped out.
>
> BUG=417493
> TEST=see bug
> R=ananta@chromium.org
>
> Committed: https://crrev.com/d1ab0241ee29c70bd08531d1f038964664b124b6
> Cr-Commit-Position: refs/heads/master@{#296608}

TBR=ananta@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=417493

Review URL: https://codereview.chromium.org/605893002

Cr-Commit-Position: refs/heads/master@{#296839}

To support CDM_6, make the following changes:
- add SetServerCertificate
- add GetUsableKeyIds
- rename ReleaseSession to CloseSession
- add RemoveSession
- add SessionKeysChange event
- add SessionExpirationChange event

This gets the new functionality up to the blink boundary. Changes to
use these new interfaces in blink in a future CL.

For backwards compatibility with existing prefixed EME code, calls
to CancelKeyRequest() call RemoveSession() instead of CloseSession().

BUG=358271,417481
TEST=existing EME tests still pass + manual testing

Review URL: https://codereview.chromium.org/555223004

Cr-Commit-Position: refs/heads/master@{#296838}

R=darin@chromium.org

Review URL: https://codereview.chromium.org/596363003

Cr-Commit-Position: refs/heads/master@{#296837}

BUG=417888
R=keescook@google.com
TBR=rsesek

Review URL: https://codereview.chromium.org/604123002

Cr-Commit-Position: refs/heads/master@{#296836}

I broke the uiautomator tests with my change to instrumentation test
running in https://codereview.chromium.org/558883003/. The revised
implementation never called _RunTest, which the uiautomator test runner
overrides to run its tests. As a result, we were trying to run the
uiautomator tests with 'am instrument', which (obviously) doesn't work.

This patch revises the implementation of the instrumentation test runner
s.t. it does call _RunTest again.

BUG=417492

Review URL: https://codereview.chromium.org/607583002

Cr-Commit-Position: refs/heads/master@{#296835}

BUG=

Review URL: https://codereview.chromium.org/606653002

Cr-Commit-Position: refs/heads/master@{#296834}

Split off from cl 490443002 because for some reason this test seems to
fail with a NaCl error on one of the trybots. Let's try to land it
separately.

BUG=404470

Review URL: https://codereview.chromium.org/590913002

Cr-Commit-Position: refs/heads/master@{#296833}

- In Canonical Fullscreen, the toolbar/tabstrip now slide over the content when
the AppKit menu bar is revealed. This prevents a series of fast resizes of the
content view, which is janky.
- Adjust the z-ordering of the infobar so that it shows in front of the
bookmark bar and toolbar in Canonical Fullscreen.
- Removed method -isInFullscreenWithOmniboxSliding, as it has the same behavior
as -isInAnyFullscreenMode.

BUG=413210

Review URL: https://codereview.chromium.org/607603003

Cr-Commit-Position: refs/heads/master@{#296832}

Revert of Restored disable print preview policy. (patchset #8 id:140001 of https://codereview.chromium.org/597933002/)

Reason for revert:
Noticed a bug in render_view_context_menu.cc change

Original issue's description:
> Restored disable print preview policy.
> Not exact revert of https://codereview.chromium.org/488853002
>
> BUG=374321,416396
>
> Committed: https://crrev.com/15daa893f50daebf33ccf06ba67d74788700e495
> Cr-Commit-Position: refs/heads/master@{#296751}

TBR=thestig@chromium.org,joaodasilva@chromium.org,pastarmovj@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=374321,416396

Review URL: https://codereview.chromium.org/603253003

Cr-Commit-Position: refs/heads/master@{#296831}

BUG=415166

Review URL: https://codereview.chromium.org/604563003

Cr-Commit-Position: refs/heads/master@{#296830}

Review URL: https://codereview.chromium.org/600363002

Cr-Commit-Position: refs/heads/master@{#296829}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/7214955..f1db071

TBR=ager@chromium.org,zmo@chromium.org

Review URL: https://codereview.chromium.org/610433004

Cr-Commit-Position: refs/heads/master@{#296828}

BUG=chromedriver:908

Review URL: https://codereview.chromium.org/605863002

Cr-Commit-Position: refs/heads/master@{#296827}

The problem is that calling hide/cancel on a popup is async.

With two popups, you would get

Previously:
SHOW A
HIDE A
SHOW B
NOTIFY SELECTION A
HIDE B
NOTIFY SELECTION B

When notifying the selection of A in the above scenario, it was
clearing the frame pointer (and actually updating the wrong popup).
Then when hiding the second popup, it's reference was actually gone
already.

Fix this by making the call to hide sync instead of async.

BUG=416354

Review URL: https://codereview.chromium.org/603303002

Cr-Commit-Position: refs/heads/master@{#296826}

We won't launch this feature for M39, so disable public opt-in flag.

BUG=416664

Review URL: https://codereview.chromium.org/610443002

Cr-Commit-Position: refs/heads/master@{#296825}

I missed this one when I was reformatting the controlled settings.

BUG=417624

Review URL: https://codereview.chromium.org/601283004

Cr-Commit-Position: refs/heads/master@{#296824}

It is impossible to whitelist an extension which is note a component, when the
api is restricted to components. Since we need to whitelist such regular
extension, this patch removes the location restriction and whitelists all
clients of this API.

Along the way, id testing extensions using this api has been changed to a new
one, since it was reused with other tests not related to the api.

TEST=browser_tests: *FileBrowser*, *FileManager*
BUG=None

Review URL: https://codereview.chromium.org/601103002

Cr-Commit-Position: refs/heads/master@{#296823}

Move initializeChildFrame from RenderFrameImpl to RenderFrameProxy.  Replace RenderFrameImpl::OnSwapOut's call to setIsRemote with WebRemoteFrame::initializeFromFrame that was introduced in part 1 of this change (https://codereview.chromium.org/591413004/)

This will be followed by another CL that will remove setIsRemote on the blink side.

BUG=416102

Review URL: https://codereview.chromium.org/592263002

Cr-Commit-Position: refs/heads/master@{#296822}