GitLab

BUG=268946
TBR=szym@chromium.org

Review URL: https://codereview.chromium.org/135693002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244359 0039d316-1c4b-4281-b951-d872f2087c98

Caused lots of races on the tsan bot:
http://build.chromium.org/p/chromium.memory.fyi/builders/Linux%20Tests%20%28tsan%29%283%29?numbuilds=200
Example:

ThreadSanitizer:Race
fun:std::_Rb_tree::_M_begin
fun:std::_Rb_tree::~_Rb_tree
fun:std::map::~map
fun:content::VideoCaptureImpl::~VideoCaptureImpl
fun:content::MockVideoCaptureImpl::~MockVideoCaptureImpl
fun:content::MockVideoCaptureImpl::~MockVideoCaptureImpl
fun:linked_ptr::depart
fun:linked_ptr::~linked_ptr
fun:std::pair::~pair
fun:std::pair::~pair
fun:__gnu_cxx::new_allocator::destroy
fun:std::_Rb_tree::_M_destroy_node
fun:std::_Rb_tree::_M_erase
fun:std::_Rb_tree::~_Rb_tree
fun:std::map::~map
fun:content::VideoCaptureImplManager::~VideoCaptureImplManager

ThreadSanitizer:Race
fun:std::_Rb_tree::_M_begin
fun:std::_Rb_tree::~_Rb_tree
fun:std::map::~map
fun:content::VideoCaptureImpl::~VideoCaptureImpl
fun:content::VideoCaptureImplTest::MockVideoCaptureImpl::~MockVideoCaptureImpl
fun:content::VideoCaptureImplTest::MockVideoCaptureImpl::~MockVideoCaptureImpl
fun:content::VideoCaptureImplTest::~VideoCaptureImplTest
fun:content::VideoCaptureImplTest_TwoClientsInSequence_Test::~VideoCaptureImplTest_TwoClientsInSequence_Test
fun:content::VideoCaptureImplTest_TwoClientsInSequence_Test::~VideoCaptureImplTest_TwoClientsInSequence_Test
fun:testing::Test::DeleteSelf_

> Eliminate video capture thread in renderer
> 
> The main motivation of this change is to remove the video capture thread
> in the renderer. All users of a video capture device already handles the
> video frame on their thread. There is no need to call the clients with
> an additional thread.
> 
> Summary of this change:
> * Video capture thread eliminated
>   VideoCaptureImpl now runs on the IO thread. Clients are called on the
>   IO thread.
> * Simplified VideoCaptureImplManager
>   We still need to keep this object for the purpose of sharing a
>   VideoCaptureImpl object with multiple clients. It should own these
>   objects and maintain the usage count. A couple clean up items are done
>   on this class:
>   * It doesn't own the video capture thread now.
>   * It is now a render thread only object.
>   * It maintains refcount of a VideoCaptureImpl explicitly.
>   * It is no longer refcounted.
>   * Clients access it through RenderThreadImpl. Which ensures usage is
>     on the render thread.
> * New VideoCaptureHandle class
>   Object of this class is returned by VideoCaptureImplManager to give
>   access to a media::VideoCapture object. It is purely a wrapper and
>   helps to do refcounting on the render thread.
> 
> Testing:
> Added unit tests for VideoCaptureImplManager to test refcounting.
> Also updated unit test for VideoCaptureImpl due to the threading
> changes.
> 
> Review URL: https://codereview.chromium.org/120893002

TBR=hclam@chromium.org

Review URL: https://codereview.chromium.org/135683002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244358 0039d316-1c4b-4281-b951-d872f2087c98

BUG=333564
TBR=jchaffraix@chromium.org

Review URL: https://codereview.chromium.org/135643003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244357 0039d316-1c4b-4281-b951-d872f2087c98

R=dtu@chromium.org,grt@chromium.org
TBR=dtu@chromium.org

Review URL: https://codereview.chromium.org/135353003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244356 0039d316-1c4b-4281-b951-d872f2087c98

This ports the PrepareDragForDownload path over from the native windows
code path. This makes dragging files to explorer from, for example, gmail
save the content of the file, rather than a link, or the thumbnail
image.

(This needs to be merged to 32, so attempted to keep it semi-localized.)

BUG=332579,333320
R=ben@chromium.org

Review URL: https://codereview.chromium.org/135393002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244355 0039d316-1c4b-4281-b951-d872f2087c98

BUG=333554
TBR=pfeldman@chromium.org

Review URL: https://codereview.chromium.org/135553002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244351 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244340 0039d316-1c4b-4281-b951-d872f2087c98

These are generally needed by aura and other parts of the code.

BUG=none
TEST=none
R=ben@chromium.org

Review URL: https://codereview.chromium.org/134523004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244336 0039d316-1c4b-4281-b951-d872f2087c98

Reverting since there is other noise on the perf-bots and I want to be sure this didn't cause it.

This reverts commit 074821be1ec9c73a25e23e6dbcbd8ccd0aec8632.

TBR=zmo@chromium.org
BUG=329439
NOTRY=true

Review URL: https://codereview.chromium.org/134383003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244335 0039d316-1c4b-4281-b951-d872f2087c98

This entry point is only used by the compositor bindings which should use
gpu::gles2::GLES2Interface to talk to the command buffer.

BUG=181120

Review URL: https://codereview.chromium.org/134343003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244334 0039d316-1c4b-4281-b951-d872f2087c98

It is obsolete now that the code is using a MessageLoopProxy.

Review URL: https://codereview.chromium.org/133903006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244333 0039d316-1c4b-4281-b951-d872f2087c98

This allows me to mention the ImageData type without it being transformed to
"imagedata".  It may have some side-effects where we've been sloppy with
typenames elsewhere, but let me know if you see any, and I'll fix them.

Review URL: https://codereview.chromium.org/132203009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244332 0039d316-1c4b-4281-b951-d872f2087c98

The previous patch removed unwind support on x64 official builds,
including Debug builds. That may have been a bit too far. Add it back to
Debug builds (but not Release builds).

BUG=180975

Review URL: https://codereview.chromium.org/131193005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244331 0039d316-1c4b-4281-b951-d872f2087c98

Metro -> Metro restarts can happen as a result of changing about:flags, or a chrome update.  In order for the metro viewer to exit, it is necessary for the browser to notify the metro viewer that it needs to close itself.  This was happening for Metro -> Desktop restarts, but not for Metro -> Metro restarts.

BUG=332063

Review URL: https://codereview.chromium.org/130223002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244330 0039d316-1c4b-4281-b951-d872f2087c98

Certain interfaces in sync had signatures that included vectors of raw
pointers to ModelSafeWorkers.  The use of raw pointers to refcounted
objects looks a lot like a bug waiting to happen.

This CL replaces any use of std::vector<ModelSafeWorker*> with
std::vector<scoped_refptr<ModelSafeWorker> >.  This is not expected to
alter sync behavior in any significant way.

BUG=332251

Review URL: https://codereview.chromium.org/130193002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244329 0039d316-1c4b-4281-b951-d872f2087c98

BUG=330235

Review URL: https://codereview.chromium.org/113963003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244328 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244327 0039d316-1c4b-4281-b951-d872f2087c98

Pulls in the following patches:
172513a Linux: upstream Gentoo Linux patches:
cfe978e Remove trailing periods from rule messages.

BUG=none
TEST=none
TBR=thakis

Review URL: https://codereview.chromium.org/134333004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244326 0039d316-1c4b-4281-b951-d872f2087c98

We were using GetOobe to figure out if the oobe page existed. This creates an InspectorBackend, which creates an InspectorPage, which calls _EnablePageNotifications, which calls SyncRequest with a timeout of 60 sec. If the page goes away right before SyncRequest is called, we could end up waiting 60 sec.

In general, GetOobe is too expensive if all we're interested in is checking the existence of oobe, so fix all instances of such checks to use OobeExists instead.

BUG=331940
TEST=manual
NOTRY=True

Review URL: https://codereview.chromium.org/134043005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244325 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244323 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none
R=ben@chromium.org

Review URL: https://codereview.chromium.org/134043003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244322 0039d316-1c4b-4281-b951-d872f2087c98

We'll use this in-tree fontconfig when "embedded=1" is set in GYP_DEFINES.

This was reverted pending license checker changes (landed as r127213003).

Relanding with the following changes from the third_party HOWTO:
  - Added OWNERS file
  - Updated .gitignore

BUG=318315

Review URL: https://codereview.chromium.org/133903005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244318 0039d316-1c4b-4281-b951-d872f2087c98

BUG=333179

Review URL: https://codereview.chromium.org/133623002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244317 0039d316-1c4b-4281-b951-d872f2087c98

The code only works as is because GRIT does not properly respect an 'is_win' conditional.

Review URL: https://codereview.chromium.org/133563002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244316 0039d316-1c4b-4281-b951-d872f2087c98

Some Linux file managers will set both URI list representation and the
plain text representation when dragging files. Chrome interprets this as
a chunk of text and a file, and offers both to the renderer. This broke
the javascript on the Dropbox website.

This ports the hack in WebDragDestGtk::OnDragDataReceived() to
linux_aura. This hack prevents us from offering the String type when we
also have a URI list.

BUG=318796

Review URL: https://codereview.chromium.org/133053003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244315 0039d316-1c4b-4281-b951-d872f2087c98

 Touch UI has been removed long time ago, and this CL removes
the code that used to load resources for Touch UI.

 I'll remove the assets in separate CL.

BUG=none
R=girard@chromium.org

Review URL: https://codereview.chromium.org/133053002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244314 0039d316-1c4b-4281-b951-d872f2087c98

fgets was refactored to read because fdopen was returning null in glibc.
These tests were disabled for glibc without explanation, so working around this
bug loses no coverage.

BUG=154400

Review URL: https://codereview.chromium.org/133033002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244313 0039d316-1c4b-4281-b951-d872f2087c98

This replaces uses of blink's WebGraphicsContext3D.h typedefs for GL numeric
types with the types from the khronos headers.

BUG=181120

Review URL: https://codereview.chromium.org/133023002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244312 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None

Review URL: https://codereview.chromium.org/132983002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244311 0039d316-1c4b-4281-b951-d872f2087c98

BUG=318450

Review URL: https://codereview.chromium.org/103283003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244310 0039d316-1c4b-4281-b951-d872f2087c98

Some of these tests need to be fixed and others should probably stay disabled.

BUG=331940
TEST=manual

Review URL: https://codereview.chromium.org/132923002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244307 0039d316-1c4b-4281-b951-d872f2087c98

Allows OSX to handle remapping channels according to a user's
preferences instead of forcing us to handle it.

BUG=266674
TEST=Multichannel users report proper output.
NOTRY=true

Review URL: https://codereview.chromium.org/132873002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244306 0039d316-1c4b-4281-b951-d872f2087c98

BUG=316496

Review URL: https://codereview.chromium.org/132583002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244305 0039d316-1c4b-4281-b951-d872f2087c98

Some of the ifndef's were missing GL_.

BUG=none
TEST=none

Review URL: https://codereview.chromium.org/107543005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244304 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://codereview.chromium.org/132473005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244303 0039d316-1c4b-4281-b951-d872f2087c98

BUG=143619
R=erg@chromium.org

Review URL: https://codereview.chromium.org/132183004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244302 0039d316-1c4b-4281-b951-d872f2087c98

BUG=170921
TEST=everything still works

Review URL: https://codereview.chromium.org/129743003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244301 0039d316-1c4b-4281-b951-d872f2087c98

BUG=170921
TEST=everything still works

Review URL: https://codereview.chromium.org/126613002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244300 0039d316-1c4b-4281-b951-d872f2087c98

Ensure that trackpad and trackpoint scrolling works on Windows desktop Aura with legacy trackpad drivers.

Legacy trackpad drivers like the older synaptics drivers have bugs where in they don't generate the WM_MOUSEWHEEL
messages for applications unless there is an entry for the application with the class name of the window in their
config file. If an entry does not exist here, these drivers expect the window to have the WS_VSCROLL/WS_HSCROLL style
to generate the legacy WM_VSCROLL/WM_HSCROLL messages.

This patch adds these styles to top level windows for Windows Aura. To ensure that Windows does not draw these
scrollbars, we hide them when the window size changes and readd them in a posted task which works correctly in our
testing.

Changes in this patch are as below:-
1. Add handlers to the HWNDMessageHandler class for the WM_VSCROLL/WM_HSCROLL messages.
2. Forward the scroll events to the HWNDMessageHandlerDelegate which then forwards it off to its delegate.
3. Miscellaneous fixes in the events code to allow the WM_VSCROLL/WM_HSCROLL messages to go through.

BUG=320849
R=cpu@chromium.org, sky@chromium.org, cpu, sky

Review URL: https://codereview.chromium.org/133053004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244295 0039d316-1c4b-4281-b951-d872f2087c98

> Password manager: Gnome support for Public Suffix List matching
> 
> This CL:
> 
> * Enables Public Suffix List (PSL) general utilities on Linux.
> * Enables PSL flag for Linux.
> * Adds Gnome Keyring support for searching and filtering multiple origins, including UMA metrics and a unit test.
> 
> BUG=324291
> 
> Review URL: https://codereview.chromium.org/123223003

TBR=vabr@chromium.org

Review URL: https://codereview.chromium.org/134923004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@244292 0039d316-1c4b-4281-b951-d872f2087c98