GitLab

We also want the "classical" NTP to coexist with the new design, and
switchable via Finch. Key changes:
- Adding classes to #ntp-content, then using CSS to control styling.
  - .classical for old NTP
  - .md for Material Design
  - .default-theme for values that will be overriden by themes. This is
    needed so dynamic CSS changes in setCustomThemeStyle() are not
    subsumed by specialized .classical and .md.
  - .dark: for dark background.
- The .mv-mask <div> is promoted to handle tile effects, including
  border, shadow, and background. Borders are drawn differently now.
  This lead to fix of 1px offset of .mv-title .

Some new designs are still in flux. Our goal is to get a reasonable bulk
committed for m38, then worry about some tune-ups later.

TEST=Run "chrome.exe --force-fieldtrials=MaterialDesignNTP/Enabled/",
visit chrome-search://local-ntp/local-ntp.html . If chrome.exe is run
without the switch, the local NTP should be identical to before, except
for 1px .mv-title shift.

BUG=400332

Review URL: https://codereview.chromium.org/473583002

Cr-Commit-Position: refs/heads/master@{#289682}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289682 0039d316-1c4b-4281-b951-d872f2087c98

Remove the dependency on a NavigationEntry in
RenderFrameHostManager::ShouldSwapBrowsingInstancesForNavigation and
RenderFrameHostManager::GetSiteInstanceForEntry. This is needed by the
PlzNavigate project, as the SiteInstance to use is determined when the
navigation is about to commit. At this time, there is no convinient access to a
SNavigationEntry.

BUG=376082

Review URL: https://codereview.chromium.org/471603002

Cr-Commit-Position: refs/heads/master@{#289681}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289681 0039d316-1c4b-4281-b951-d872f2087c98

We build with 3rd-party code with -Wall on the clang/win bot. This has already
been fixed upstream, so there's no need to keep this around

https://github.com/grate-driver/mesa/commit/68a4bb553bcd51d336d95b0c845c0dc63adc3bd3

BUG=none
R=hans@chromium.org
TBR=cpu

Review URL: https://codereview.chromium.org/470313002

Cr-Commit-Position: refs/heads/master@{#289680}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289680 0039d316-1c4b-4281-b951-d872f2087c98

The log of the flaky LayerTreeHostTestBreakSwapPromise test shows that

[ RUN      ] LayerTreeHostTestBreakSwapPromise.RunMultiThread_DelegatingRenderer_ImplSidePaint
c:\b\build\slave\win_x64_builder\build\src\cc\trees\layer_tree_host_unittest.cc(4500): error: Value of: swap_promise_result_[2].did_not_swap_called
Actual: false
Expected: true
c:\b\build\slave\win_x64_builder\build\src\cc\trees\layer_tree_host_unittest.cc(4501): error: Value of: swap_promise_result_[2].reason
Actual: 0
Expected: SwapPromise::SWAP_FAILS
Which is: 1
[  FAILED  ] LayerTreeHostTestBreakSwapPromise.RunMultiThread_DelegatingRenderer_ImplSidePaint (3 ms)

We have a EXPECT_TRUE on the SwapPromise's dtor to be called which passes.
It means we destroyed the SwapPromise without ever breaking them first
(no DidNotSwap ever called). I checked all the code path where we destroy
SwapPromise and we always break them first. So I can't figure out why it
is the case. I am adding some logging information and turn this test back
on, hopefully next time it fails on the bot I can use the extra logging
information to debug.

BUG=393995

Review URL: https://codereview.chromium.org/477443002

Cr-Commit-Position: refs/heads/master@{#289679}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289679 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://codereview.chromium.org/477633002

Cr-Commit-Position: refs/heads/master@{#289678}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289678 0039d316-1c4b-4281-b951-d872f2087c98

Loaded app comes first, and then mock icons follow. Otherwise,
the loaded calculator app icon won't appear in the new design
because the new design has three icons for apps.

BUG=None
R=oshima@chromium.org
TEST=manually

Review URL: https://codereview.chromium.org/470713002

Cr-Commit-Position: refs/heads/master@{#289677}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289677 0039d316-1c4b-4281-b951-d872f2087c98

Cr-Commit-Position: refs/heads/master@{#289676}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289676 0039d316-1c4b-4281-b951-d872f2087c98

This pulls in the following Native Client changes:

r13612: (mseaborn) Disable the imc_mem_obj_create() syscall by default
r13613: (dschuff) Update revision for PNaCl r13557->r13609

BUG=none
TEST=browser_tests and nacl_integration
CQ_EXTRA_TRYBOTS=tryserver.chromium.linux:linux_rel_precise32,linux_nacl_sdk_build

Review URL: https://codereview.chromium.org/472893002

Cr-Commit-Position: refs/heads/master@{#289675}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289675 0039d316-1c4b-4281-b951-d872f2087c98

BUG=395729

Review URL: https://codereview.chromium.org/464463003

Cr-Commit-Position: refs/heads/master@{#289673}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289673 0039d316-1c4b-4281-b951-d872f2087c98

BUG=403776

Review URL: https://codereview.chromium.org/464213004

Cr-Commit-Position: refs/heads/master@{#289671}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289671 0039d316-1c4b-4281-b951-d872f2087c98

When in the task runner MessageLoop::current() returns NULL. Use the task runner instead of MessageLoop::current() to PostTask.

BUG=392621

Review URL: https://codereview.chromium.org/472843002

Cr-Commit-Position: refs/heads/master@{#289670}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289670 0039d316-1c4b-4281-b951-d872f2087c98

Note: clang-format misformats some C-style casts (in the C-compilable
parts of header files; e.g., it formats |(Foo)-5| as |(Foo) - 5|), but I
can live with that.

R=jamesr@chromium.org

Review URL: https://codereview.chromium.org/469393002

Cr-Commit-Position: refs/heads/master@{#289669}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289669 0039d316-1c4b-4281-b951-d872f2087c98

Due to promise chaining, the rejecting promise may not necessarily be the
topmost method returning the promise.

In this specific case, getAuth() is chained in by requestFromServer, and
getAuth can reject with no arguments.

The fix here is to ensure the argument is always an object.

BUG=402729

Review URL: https://codereview.chromium.org/469943002

Cr-Commit-Position: refs/heads/master@{#289668}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289668 0039d316-1c4b-4281-b951-d872f2087c98

Make the existing RawSharedBuffer an implementation of that. Later,
we'll plumb in a "PlatformSupport" that'll produce these things, but for
now keep the (static) creation methods where they are.

R=darin@chromium.org

Review URL: https://codereview.chromium.org/471773002

Cr-Commit-Position: refs/heads/master@{#289667}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289667 0039d316-1c4b-4281-b951-d872f2087c98

view_manager.mojom imports service_provider.mojom, so the bindings
target for view_manager.mojom must depend on the bindings target for
service_provider.mojom

R=viettrungluu@chromium.org
BUG=403719

Review URL: https://codereview.chromium.org/469583004

Cr-Commit-Position: refs/heads/master@{#289666}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289666 0039d316-1c4b-4281-b951-d872f2087c98

Cr-Commit-Position: refs/heads/master@{#289664}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289664 0039d316-1c4b-4281-b951-d872f2087c98

https://skia.googlesource.com/skia/+log/c773390eefe70d13faab147a70bd9319b9ac3afb..5c4b137592a7937b4d02d72a976fa783f8d3675a

CQ_EXTRA_TRYBOTS=tryserver.chromium.linux:linux_layout_rel
TBR=bsalomon@google.com
BUG=

Review URL: https://codereview.chromium.org/474933003

Cr-Commit-Position: refs/heads/master@{#289662}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289662 0039d316-1c4b-4281-b951-d872f2087c98

Params "user_id" and "profile_picture_enabled" were not used so I
removed them and related code.
Also one usage of GetProfileByUserUnsafe eliminated.

BUG=361528
TEST=browser_tests

Review URL: https://codereview.chromium.org/466793002

Cr-Commit-Position: refs/heads/master@{#289661}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289661 0039d316-1c4b-4281-b951-d872f2087c98

The previous assets had an extra 1px on each side.

BUG=399696
R=oshima@chromium.org

Review URL: https://codereview.chromium.org/473933002

Cr-Commit-Position: refs/heads/master@{#289660}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289660 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None

Review URL: https://codereview.chromium.org/474553002

Cr-Commit-Position: refs/heads/master@{#289659}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289659 0039d316-1c4b-4281-b951-d872f2087c98

(This matches the gyp build)

R=brettw@chromium.org
TBR=brettw@chromium.org
BUG=354261

Review URL: https://codereview.chromium.org/467413002

Cr-Commit-Position: refs/heads/master@{#289657}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289657 0039d316-1c4b-4281-b951-d872f2087c98

BUG=304309,372381

Review URL: https://codereview.chromium.org/475743002

Cr-Commit-Position: refs/heads/master@{#289656}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289656 0039d316-1c4b-4281-b951-d872f2087c98

BUG=397741

Review URL: https://codereview.chromium.org/473553002

Cr-Commit-Position: refs/heads/master@{#289655}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289655 0039d316-1c4b-4281-b951-d872f2087c98

BUG=403083

Review URL: https://codereview.chromium.org/454053005

Cr-Commit-Position: refs/heads/master@{#289654}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289654 0039d316-1c4b-4281-b951-d872f2087c98

Format conversions are now handled internally by AP, allowing us to
to tear out the conversion code from MSAP. Instead, we use a simple
FIFO to buffer the incoming audio into the 10 ms chunks AP requires.

Add an AudioBus wrapper to get an array of channel pointers, since this
is the type AP deals in.

Add a few TODOs around two suspect areas: i) there appears to be an
assumption on the sink type based on the input format, and ii) don't
we want ProcessAndConsumeData() to output float data?

BUG=400933
TESTED=Audio processing behaves as expected on a MacBook, using both
44.1 and 48 kHz as the render and capture rates (set through "Audio
Midi Setup"). Audio sounds fine when processing is disabled through
constraints.

Review URL: https://codereview.chromium.org/420603004

Cr-Commit-Position: refs/heads/master@{#289653}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289653 0039d316-1c4b-4281-b951-d872f2087c98

Cr-Commit-Position: refs/heads/master@{#289652}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289652 0039d316-1c4b-4281-b951-d872f2087c98

There is no active profile in guest mode, thus we cannot call
avatar_menu->GetItemAt(avatar_menu->GetActiveProfileIndex()) in ProfileChooserView::ShowView unconditionally. Instead the call should be moved to CreateSwitchUserView where it is guaranteed that the current profile is signed in.

BUG=402119
R=noms@chromium.org, sky@chromium.org

Review URL: https://codereview.chromium.org/458893003

Cr-Commit-Position: refs/heads/master@{#289651}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289651 0039d316-1c4b-4281-b951-d872f2087c98

TBR=iannucci@chromium.org
NOTRY=true
BUG=402543

Review URL: https://codereview.chromium.org/458353003

Cr-Commit-Position: refs/heads/master@{#289650}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289650 0039d316-1c4b-4281-b951-d872f2087c98

BUG=399660
R=oshima@chromium.org

Review URL: https://codereview.chromium.org/475813002

Cr-Commit-Position: refs/heads/master@{#289649}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289649 0039d316-1c4b-4281-b951-d872f2087c98

Revert of Remove unnecessary static function (patchset #2 of https://codereview.chromium.org/444983002/)

Reason for revert:
Perf regression, due to the method no longer being inline, and using doubles instead of floats.

Original issue's description:
> Remove unnecessary static function
> 
> Use gfx::CrossProduct() instead of defining a new static function.
> 
> BUG=
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=288372

TBR=enne@chromium.org,heejin.r.chung@samsung.com
NOTREECHECKS=true
NOTRY=true
BUG=402552

Review URL: https://codereview.chromium.org/471113002

Cr-Commit-Position: refs/heads/master@{#289645}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289645 0039d316-1c4b-4281-b951-d872f2087c98

Cr-Commit-Position: refs/heads/master@{#289644}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289644 0039d316-1c4b-4281-b951-d872f2087c98

TBR=nduca@chromium.org

Review URL: https://codereview.chromium.org/469303002

Cr-Commit-Position: refs/heads/master@{#289643}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289643 0039d316-1c4b-4281-b951-d872f2087c98

Fixes:
base/android/linker/linker_jni.cc:300:  Single-parameter constructors should be marked explicit.  [runtime/explicit] [5]
base/android/linker/linker_jni.cc:313:  Weird number of spaces at line-start.  Are you using a 2-space indent?  [whitespace/indent] [3]
base/android/linker/linker_jni.cc:319:  Weird number of spaces at line-start.  Are you using a 2-space indent?  [whitespace/indent] [3]
base/android/linker/linker_jni.cc:320:  Weird number of spaces at line-start.  Are you using a 2-space indent?  [whitespace/indent] [3]

BUG=

Review URL: https://codereview.chromium.org/470943002

Cr-Commit-Position: refs/heads/master@{#289642}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289642 0039d316-1c4b-4281-b951-d872f2087c98

Cr-Commit-Position: refs/heads/master@{#289641}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289641 0039d316-1c4b-4281-b951-d872f2087c98

http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog_blink.html?url=/trunk&range=180192:180219&mode=html
TBR=zerny@chromium.org,adamk@chromium.org
BUG=

Review URL: https://codereview.chromium.org/477503002

Cr-Commit-Position: refs/heads/master@{#289640}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289640 0039d316-1c4b-4281-b951-d872f2087c98

GetProfileByUserUnsafe was called to yearly sometimes, when profile was not
created yet. I replaced GetProfileByUserUnsafe with GetProfileByUser and added
check that it is not NULL.

BUG=361528

Review URL: https://codereview.chromium.org/460233004

Cr-Commit-Position: refs/heads/master@{#289639}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289639 0039d316-1c4b-4281-b951-d872f2087c98

I see it sometimes failing on try jobs even with reruns. i.e.
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_swarming/builds/1619
http://build.chromium.org/p/tryserver.chromium.win/builders/win_chromium_rel_swarming/builds/1672

BUG=128961
TBR=sky@chromium.org

Review URL: https://codereview.chromium.org/473733004

Cr-Commit-Position: refs/heads/master@{#289638}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289638 0039d316-1c4b-4281-b951-d872f2087c98

the cookies.
Before, in the password manager python tests, each website had its
own way to logout (ex: click on a button or follow a link). This cl
allows to have the same logout mechanism for all the websites
thought cleaning cookies because this makes creating and maintaining
tests simpler. It makes also the tests more robust.

TBR=vabr@chromium.org
BUG=403746

Review URL: https://codereview.chromium.org/472823002

Cr-Commit-Position: refs/heads/master@{#289637}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289637 0039d316-1c4b-4281-b951-d872f2087c98

- remove the profile preference that used to keep track of showing the
tutorial any time the user manager was shown for a particular profile.

- show the upgrade bubble for local profiles as well (so that local profiles
have a chance to see the tutorial)

- for the local profiles upgrade bubble, don't show the "Not you?" link as
it doesn't really make sense.

BUG=399679
TBR=rpetterson@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=288817

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=289005

Review URL: https://codereview.chromium.org/454153002

Cr-Commit-Position: refs/heads/master@{#289636}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289636 0039d316-1c4b-4281-b951-d872f2087c98

Implementation of the Battery Status API for the Windows platform.
Implementation uses a message window to receive battery notifications.
On versions prior to Vista there is limited support as the
RegisterPowerSettingNotification function is not available.

BUG=122593
TEST=http://jsbin.com/battery-status-test (manual)

Review URL: https://codereview.chromium.org/447853002

Cr-Commit-Position: refs/heads/master@{#289634}
git-svn-id: svn://svn.chromium.org/chrome/trunk/src@289634 0039d316-1c4b-4281-b951-d872f2087c98