GitLab

BUG=304341
R=bulach@chromium.org

Review URL: https://codereview.chromium.org/303503008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274339 0039d316-1c4b-4281-b951-d872f2087c98

Already marked as retired on bug, suppressions query has no results for
this either.

BUG=258466
R=thestig@chromium.org
TBR=thestig@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/309533015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274338 0039d316-1c4b-4281-b951-d872f2087c98

Allows skia to remove SK_SUPPORT_LEGACY_INSTALLPIXELSPARAMS

NOTRY=True
failures unrelated to this CL

Review URL: https://codereview.chromium.org/308833004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274337 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274336 0039d316-1c4b-4281-b951-d872f2087c98

http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog_blink.html?url=/trunk&range=175159:175308&mode=html

TBR=mvanouwerkerk@chromium.org

Review URL: https://codereview.chromium.org/301173008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274335 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274334 0039d316-1c4b-4281-b951-d872f2087c98

This pulls in the following Native Client changes:

r13252: (sehr) Make Android scons build align better with gyp android build
r13253: (mseaborn) Add test for PNaCl setjmp() bug in which stack slots get reused wrongly
r13254: (jvoung) Remove -sfi-load, etc. from the pnacl-translate driver (always set for nacl)
r13255: (dyen) Added a way for toolchain_build to update known git mirrors.
r13256: (hamaji) Non-SFI Mode: Build newlib-based Linux static executables for ARM
r13257: (mcgrathr) Include address in ReserveBottomPages failure message
r13258: (sbc) Add {TOOLCHAIN}/usr/local/lib to pnacl library search path.
r13259: (sehr) Fix stlport issues in scons build
r13260: (mcgrathr) Accept mmap EINVAL failure for any <64k page, not just 0
r13261: (jvoung) PNaCl: Update LLVM revision in pnacl/COMPONENT_REVISIONS

BUG=none
TEST=browser_tests and nacl_integration

Review URL: https://codereview.chromium.org/304323011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274333 0039d316-1c4b-4281-b951-d872f2087c98

As a first step to getting a separate target for the display unittests, this CL is adding a simple target with one unittest. All display_unittests currently running under ui_unittests will be migrated under this target when display_unittests is enabled on the main waterfall.

BUG=333413

Review URL: https://codereview.chromium.org/304183010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274332 0039d316-1c4b-4281-b951-d872f2087c98

This test is currently disabled on Linux Aura because it was running too slowly
and caused flakes. I'm hoping that the test will run faster in
content_browsertests, so we won't need to disable the test in that case.

I'm making a commit that only moves this test so it can be easily reverted if
necessary.

BUG=371873
R=dmichael@chromium.org

Review URL: https://codereview.chromium.org/306393002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274331 0039d316-1c4b-4281-b951-d872f2087c98

When no device reporting policy is present, we should assume that
reporting (with the exception of device location) is enabled.

BUG=379284

Review URL: https://codereview.chromium.org/308643006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274330 0039d316-1c4b-4281-b951-d872f2087c98

permissions.

This is so we can check if the user has opted into stats before enabling
initializing analytics in our app.

BUG=378067

Review URL: https://codereview.chromium.org/305663002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274329 0039d316-1c4b-4281-b951-d872f2087c98

> Implement a demo that shows one app embedding rendering in another.
> 
> launcher inits the view manager, running the "window manager" app @ root.
> window manager embeds another app.
> 
> R=sky@chromium.org
> TEST=mojo_shell --origin=http://127.0.0.1:4444 --disable-cache mojo:mojo_demo_launcher
> http://crbug.com/365012
> 
> Review URL: https://codereview.chromium.org/303163005

TBR=ben@chromium.org

Review URL: https://codereview.chromium.org/306403003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274328 0039d316-1c4b-4281-b951-d872f2087c98

This works around missing linkage between mesa's DRI drivers (e.g.
i965_dri.so) and libglapi.so.0. This causes problems under gbm.

TEST=chrome --ozone-platform=gbm on link_freon
BUG=377497
NOTRY=true

Review URL: https://codereview.chromium.org/310673002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274327 0039d316-1c4b-4281-b951-d872f2087c98

On desktop, always set the memory limit to 256MB, and ask that the renderer
only use that memory to draw content that is near the viewport. The effective
limit prior to this patch was 384MB, but would quickly get cut down as multiple
windows are opened, but would not fall too much lower than 256MB (e.g, 128
on Mac).

On mobile, use all available memory for the current, only, renderer.

Do not take into account unmanaged (e.g, WebGL) memory usage.

BUG=377065

Review URL: https://codereview.chromium.org/308743005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274326 0039d316-1c4b-4281-b951-d872f2087c98

Original CL: https://codereview.chromium.org/300853008/
Original Description:
Refactors ExternalInstallUI to have determinate lifetimes.
This is likely the first of multiple patches targeting this area of code; the
next one will concentrate on pulling the related logic out of ExtensionService.

----------------------------------------------------

The original patch contained a cycle in that when an external install warning
was clicked, it would accept or remove the extension, and then remove the error.
Since the manager listened for the extension to be loaded or unloaded, this
resulted in a double-removal.  Fix is straightforward.

We also didn't have tests for actually interacting with the dialog, so two have
been added.

TBR=yoz@chromium.org
BUG=378042

Review URL: https://codereview.chromium.org/309643007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274325 0039d316-1c4b-4281-b951-d872f2087c98

Adds the public headers and visibility to the desc command.

Improve formatting of visibility printing.

Don't duplicate group deps in the tree output of "desc"

R=scottmg@chromium.org

Review URL: https://codereview.chromium.org/306013007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274324 0039d316-1c4b-4281-b951-d872f2087c98

BUG=379798
TBR=tommi@chromium.org

Review URL: https://codereview.chromium.org/307253002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274323 0039d316-1c4b-4281-b951-d872f2087c98

launcher inits the view manager, running the "window manager" app @ root.
window manager embeds another app.

R=sky@chromium.org
TEST=mojo_shell --origin=http://127.0.0.1:4444 --disable-cache mojo:mojo_demo_launcher
http://crbug.com/365012

Review URL: https://codereview.chromium.org/303163005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274322 0039d316-1c4b-4281-b951-d872f2087c98

This blocks the DLL from loading in sandbox processes.  It was causing sandbox
startup failure on Win64.

TBR=jschuh@chromium.org
BUG=379730

Review URL: https://codereview.chromium.org/308993003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274319 0039d316-1c4b-4281-b951-d872f2087c98

With each TrayBackgroundView now backed by its own layer, calling SchedulePaint on StatusAreaWidgetDelegate no longer notifies them of the need to paint.

When the shelf changes its dimming state it triggers painting. Add support for StatusAreaWidget to notify each of the layer backed trays that painting is needed

TEST=Launch a non-maximized browser. The system tray will be a dark grey. Maximize the browser. The tray should be a lighter grey. This will mimic the colour changes of the application launcher on the shelf.
BUG=378115

Review URL: https://codereview.chromium.org/302153002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274318 0039d316-1c4b-4281-b951-d872f2087c98

E.g., something like

  out/Debug/mojo_shell file://$PWD/out/Debug/lib/libmojo_sample_app.so

should work.

R=sky@chromium.org

Review URL: https://codereview.chromium.org/302173003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274317 0039d316-1c4b-4281-b951-d872f2087c98

BUG=379170
TBR=asargent

Review URL: https://codereview.chromium.org/309533013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274316 0039d316-1c4b-4281-b951-d872f2087c98

Broke Windows compile:
http://build.chromium.org/p/chromium.win/buildstatus?builder=Win%20x64%20Builder&number=180

FAILED: ninja -t msvc -e environment.x64 -- C:\b\build\goma/gomacc "C:\b\depot_tools\win_toolchain\vs2013_files\VC\bin\amd64\cl.exe" /nologo /showIncludes /FC @obj\remoting\host\win\remoting_core.wts_session_process_delegate.obj.rsp /c ..\..\remoting\host\win\wts_session_process_delegate.cc /Foobj\remoting\host\win\remoting_core.wts_session_process_delegate.obj /Fdobj\remoting\remoting_core.cc.pdb 
c:\b\build\slave\win_x64_builder\build\src\remoting\host\win\wts_session_process_delegate.cc(386) : error C2661: 'IPC::ChannelProxy::ChannelProxy' : no overloaded function takes 4 arguments
ninja: build stopped: subcommand failed.


> Introduce IPC::ChannelProxy::Create*() and IPC::SynChannel::Create*()
> 
> This change hides constructors of these classes so that we can turn
> them polymorphic classes.
> 
> Note that having almost identical ChannelProxy::Init*() isn't great
> and they will be replaced by a factory-like abstraction in coming
> changes.
> 
> TEST=none
> R=darin,cpu
> BUG=377980
> 
> Review URL: https://codereview.chromium.org/301973003

TBR=morrita@chromium.org

Review URL: https://codereview.chromium.org/312553004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274315 0039d316-1c4b-4281-b951-d872f2087c98

Expose client_id in chromeosInforPrivate for enterprise
kiosk.

BUG=328437

Review URL: https://codereview.chromium.org/308913004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274313 0039d316-1c4b-4281-b951-d872f2087c98

Currently log lines from the plugin get 2 newlines, which wastes space in
the console output. This strips off the extra newline.

Review URL: https://codereview.chromium.org/308183004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274312 0039d316-1c4b-4281-b951-d872f2087c98

Enable a Shell for tests that exist in Linux with Ash.

TEST=TabDragControllerInteractiveUiTest
BUG=330429

Review URL: https://codereview.chromium.org/301613002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274311 0039d316-1c4b-4281-b951-d872f2087c98

This change hides constructors of these classes so that we can turn
them polymorphic classes.

Note that having almost identical ChannelProxy::Init*() isn't great
and they will be replaced by a factory-like abstraction in coming
changes.

TEST=none
R=darin,cpu
BUG=377980

Review URL: https://codereview.chromium.org/301973003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274310 0039d316-1c4b-4281-b951-d872f2087c98

TBR=jyasskin@chromium.org
BUG=379204

Review URL: https://codereview.chromium.org/304423005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274307 0039d316-1c4b-4281-b951-d872f2087c98

Before the CL that introduced NetMetricsLogUploader, Start()
was being called from MetricsService::SendStagedLog().

However, it got lost in the refactoring. This re-adds it.

BUG=375771
R=isherman@chromium.org

Review URL: https://codereview.chromium.org/309793006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274306 0039d316-1c4b-4281-b951-d872f2087c98

When the chip is hidden, we instantly show the entire omnibox text, aligned so
that the hostname is in the same place it was in the chip, and clipped to just
have the hostname visible.  Then we animate moving the alignment over to the
normal omnibox text position (i.e. just after the location icon) while
expanding the clip to gradually reveal the whole text.  We also crossfade the
selection colors from the chip colors to the standard omnibox ones.

This also:
* Animates correctly when the hostname is not at the beginning of the origin
  chip (i.e. for EV certs)
* Properly reverses the animation if the user changes state in the middle (hard
  to see with quick animations, but looks nicer if we slow them down or with
  very large animation distances)
* Fixes dropping characters during animation (because the real omnibox is
  visible and focused, and can respond correctly to keypresses)
* Fixes animations improperly running across tab switches
* Cleans up some unused origin chip code

Also tested in RTL mode.

BUG=363681,365818,374251
TEST=With the origin chip on, showing and hiding the URL should animate much more quickly and cleanly

Review URL: https://codereview.chromium.org/302453002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274305 0039d316-1c4b-4281-b951-d872f2087c98

view-souce feature and the XSSAuditor in blink. The current technique for testing
view-source pages is to use a hack in the content_shell's dump-render-tree mode
to turn on view-source for the main page when the path /viewsource/ is 
present in the URL.

To test XSS reflections, however, there needs to be a query string containing
the reflection (because putting it in the path would lead to horribly named 
files). To get a query string, we traditionally load a main page which then
loads the more complex URL in a iframe.  Presently, there's no way to make
that iframe a view-source frame (since some guy removed the viewsource="true"
attribute from HTML because it was dangerous).

There are also two other methods that are exposed to javascript: dumpAsMarkup()
and dumpChildFramesAsMarkup().  These give the test scripts the ability to control
the format of the output.  This is needed to simplify diffing the results 
of these kinds of tests.

CL https://codereview.chromium.org/301813002/ is blocked on this.

Review URL: https://codereview.chromium.org/302043002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274304 0039d316-1c4b-4281-b951-d872f2087c98

This API is controlled by the app.window.shape permission which has its own whitelist, so there's no need to duplicate in the SetShape implementation.

Review URL: https://codereview.chromium.org/282293005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274303 0039d316-1c4b-4281-b951-d872f2087c98

- This strips conditions inside a variables in addition to the toplevel conditions.

- Removes percent signs from the ends of variable names.

Both of these are required to import some of the Blink .gypis.

Make net forward the crypto config. This appears to be necessary to use net's public API.

R=dpranke@chromium.org

Review URL: https://codereview.chromium.org/305013009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274302 0039d316-1c4b-4281-b951-d872f2087c98

BUG=373252
TBR=grunell@chromium.org

Review URL: https://codereview.chromium.org/302173009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274301 0039d316-1c4b-4281-b951-d872f2087c98

This is to share the same code path with WebView, which currently
has its own copy/paste version of the code, and keep its behavior
in sync with it.
This change is a pure refactoring and should have no effect on
Chrome.

BUG=b/14438911

Review URL: https://codereview.chromium.org/308313004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274300 0039d316-1c4b-4281-b951-d872f2087c98

BUG=379432
TEST=None

Review URL: https://codereview.chromium.org/312453002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274299 0039d316-1c4b-4281-b951-d872f2087c98

Some ICU settings were incorrect in the non-component build that caused functions not to get linked properly.

We set the wrong exception handling flag in non-component mode.

Add rlz to the GN build.

Some NSS files were missing.

R=scottmg@chromium.org

Review URL: https://codereview.chromium.org/307993010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274298 0039d316-1c4b-4281-b951-d872f2087c98

We will limit the enhanced bookmark image store database size
to ensure that the db file size is reasonable. So add a function
to query the actual file size.

BUG=378853

Review URL: https://codereview.chromium.org/305963004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274297 0039d316-1c4b-4281-b951-d872f2087c98

Reverting as we think it caused checkdeps failure on Linux:
http://build.chromium.org/p/chromium.linux/builders/Linux%20Builder%20%28dbg%29%2832%29/builds/24742/steps/checkdeps/logs/stdio

ERROR in /b/build/slave/Linux_Builder__dbg__32_/build/src/mojo/android/javatests/src/org/chromium/mojo/bindings/BindingsTest.java
  Illegal include: "out/Debug/java_mojo/mojo_public_test_interfaces/src/org/chromium/mojo/bindings/test/sample/SampleServiceConstants.java"
    Because of no rule applying.


> Generate java bindings for constants.
> 
> This CL is the first CL introducing java bindings. It only generates constants.
> 
> This is a reland of https://codereview.chromium.org/291903003
> 
> TBR=viettrungluu@chromium.org,rmcilroy@chromium.org
> 
> Review URL: https://codereview.chromium.org/305963009

TBR=qsr@chromium.org

Review URL: https://codereview.chromium.org/311643002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274296 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none

Review URL: https://codereview.chromium.org/309053003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@274288 0039d316-1c4b-4281-b951-d872f2087c98