GitLab

TEST=none
BUG=none

R=jbates@chromium.org


Review URL: http://codereview.chromium.org/8759001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112782 0039d316-1c4b-4281-b951-d872f2087c98

BUG=88782
TEST=X509UtilNSSTest

Review URL: http://codereview.chromium.org/8537025

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112780 0039d316-1c4b-4281-b951-d872f2087c98

Dues to link issues on chromeos clang.

This reverts commit 2c6c8fc139bae1b64a184c13c979f408721316a3.

TBR=dennisjeffrey@chromium.org

BUG=
TEST=

Review URL: http://codereview.chromium.org/8774053

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112779 0039d316-1c4b-4281-b951-d872f2087c98

CairoCachedSurfaces are representations of a GdkPixbuf that live on the display
server instead of in process. GtkThemeService currently returns a
CairoCachedSurface for a (IDR#, Display). Instead, make gfx::Image keep
CairoCachedSurfaces as a representation and make CairoCachedSurface work on any
display, keeping the mapping to display server resources as an implementation
detail.

This has the benefit of:

- You don't need to go through GtkThemeService to access all resources,
  simplifying the code in some places and allowing sharing of static images
  across profiles.
- This will let us (in the future) remove several image conversions in the GTK
  port.

This is part 1. Part 2 will remove the custom CairoCachedSurface calls from
GtkThemeService.

BUG=106060
TEST=none
R=rsesek@chromium.org
TBR=ben@chromium.org

Review URL: http://codereview.chromium.org/8769017

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112778 0039d316-1c4b-4281-b951-d872f2087c98

WebKit side (which needs to land first) is: https://bugs.webkit.org/show_bug.cgi?id=73232 which landed as WebKit r101602

BUG=103578
TEST=

Review URL: http://codereview.chromium.org/8743001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112777 0039d316-1c4b-4281-b951-d872f2087c98

We've had lots of trouble over the last six months with the ChromeOS extension-based file picker.  In particular, for the Windows/Mac/GTK file pickers selecting a file is synchronous with closing the dialog.  On ChromeOS, with the extension picker, it's not.  We've historically gotten into trouble with the order of deletion of all the associated objects.  This patch defers the SelectFileDialog::Listener callback until the dialog closes, to match the semantics of Win/Mac/GTK.  This allows us to use a more natural window.close() call in the file manager JavaScript, though sadly we still have to call the unload handler manually.  I also renamed ExtensionDialogIsClosing() to ExtensionDialogClosing() to match WindowClosing().

This also fixed a problem on Aura where closing the file picker by clicking the dialog "X" button resulted in the inability to reopen it.

BUG=105311
TEST=browser_test SelectFileDialog*, also navigate to a web page with <input type='file'> and be sure you can cancel with the 'X' button on Aura and reopen the dialog.


Review URL: http://codereview.chromium.org/8770028

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112776 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/8771026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112775 0039d316-1c4b-4281-b951-d872f2087c98

Store frame intervals in an array rather than throwing them away. This is to
enable more-detailed study of frame rates.

Report error as sigma directly, instead of 60 - 1000/sigma. This should help
understand whether standard deviation is even the right metric to use for frame
rate instability.


Review URL: http://codereview.chromium.org/8536008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112774 0039d316-1c4b-4281-b951-d872f2087c98

Added a unit test to check such a scenario.

BUG=105391
TEST=Ensure that all TemplateURLServiceSyncTest unit tests succeed.

Review URL: http://codereview.chromium.org/8689007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112773 0039d316-1c4b-4281-b951-d872f2087c98

BUG=106114

Review URL: http://codereview.chromium.org/8770038

TBR=pfeldman@chromium.org
Review URL: http://codereview.chromium.org/8775068

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112772 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/8771032

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112771 0039d316-1c4b-4281-b951-d872f2087c98

It is possible to try and close the window now before the TemplateURLService is loaded. In this case we will not have a fallback search engine to select by default. So we will just reject attempts to close the window in this circumstance until it is loaded and a fallback choice has been set. It should be an edge case.

http://crbug.com/106078
TEST=unit test
Review URL: http://codereview.chromium.org/8771018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112770 0039d316-1c4b-4281-b951-d872f2087c98

Don't make classes derive from ChildProcessHost, and instead have them use it through composition. This cleans up the code and makes it easier to understand (as well as more closely conform to the Google C++ style guide). It also makes it possible to add an interface around ChildProcessHost in a future change.

BUG=98716
Review URL: http://codereview.chromium.org/8774040

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112769 0039d316-1c4b-4281-b951-d872f2087c98

BUG=104749
TEST=valgrind bots

Review URL: http://codereview.chromium.org/8775065

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112768 0039d316-1c4b-4281-b951-d872f2087c98

Fix use-after-free. OnChildDied deletes the current object, so we shouldn't call member functions afterwards. I'm refactoring this code in another cl, but figured I'd fix this separately to make things clearer.
Review URL: http://codereview.chromium.org/8776041

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112767 0039d316-1c4b-4281-b951-d872f2087c98

BUG=101992, 104103
TEST=existing, ManifestTest

Review URL: http://codereview.chromium.org/8654001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112764 0039d316-1c4b-4281-b951-d872f2087c98

Changed the accelerated html pages to use raf to draw multiple frames. Changed the webgl and canvas2d pages to draw onscreen rather than offscreen to make sure the compositor draws the page.

BUG=105785,104142


Review URL: http://codereview.chromium.org/8692013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112763 0039d316-1c4b-4281-b951-d872f2087c98

Also, remove a couple of unused functions and rename some to be more meaningful following the recent refactoring.

BUG=102441
TEST=Manual


Review URL: http://codereview.chromium.org/8775021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112762 0039d316-1c4b-4281-b951-d872f2087c98

TBR=glider@chromium.org
BUG=99304
TEST=less flakiness on valgrind bots

Review URL: http://codereview.chromium.org/8776062

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112761 0039d316-1c4b-4281-b951-d872f2087c98

BUG=105968
TEST=python ~/depot_tools/watchlists.py ui/views/
R=nirnimesh@chromium.org
TBR=sky@chromium.org

Review URL: http://codereview.chromium.org/8776027

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112760 0039d316-1c4b-4281-b951-d872f2087c98

Add abstract GetGroupsToChange() method to ModelChangingSyncerCommand.
Use that to figure out which worker threads to post work on (instead of
posting on all of them).

Implement GetGroupsToChange() for each ModelChangingSyncerCommand.

Add GetEnabledGroups() and GetEnabledGroupsWithConflicts() functions to SyncSession.

Key unapplied updates index by type for ApplyUpdatesCommand.

Make the abstract methods of ModelChangingSyncerCommand protected.

This patch includes a speculative fix for the perf regression introduced by the last time this was landed (112178).

BUG=97832
TEST=

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=112178

Review URL: http://codereview.chromium.org/8637006

TBR=akalin@chromium.org
Review URL: http://codereview.chromium.org/8771044

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112759 0039d316-1c4b-4281-b951-d872f2087c98

TBR=simonjam@chromium.org
BUG=106117
TEST=mac valgrind bots go greener

Review URL: http://codereview.chromium.org/8776061

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112756 0039d316-1c4b-4281-b951-d872f2087c98

BUG=102731
TEST=Manual steps in the bug report.

Review URL: http://codereview.chromium.org/8686012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112755 0039d316-1c4b-4281-b951-d872f2087c98

Revert 112693 - committed before review was done - Landing 8688002: PPB_TCPSocket_Private/PPB_UDPSocket_Private are exposed to Browser

Review URL: http://codereview.chromium.org/8688002
Patch from Yuri Gorshenin <ygorshenin@chromium.org>.

TBR=dpolukhin@chromium.org
Review URL: http://codereview.chromium.org/8775063

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112754 0039d316-1c4b-4281-b951-d872f2087c98

TBR=mcgrathr@chromium.org
BUG=106197
TEST=tsan bots go greener

Review URL: http://codereview.chromium.org/8776056

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112753 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=Run with no dump.json around.

Review URL: http://codereview.chromium.org/8770056

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112752 0039d316-1c4b-4281-b951-d872f2087c98

BUG=chromium:104530
TEST=PyAuto tests

Review URL: http://codereview.chromium.org/8587004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112751 0039d316-1c4b-4281-b951-d872f2087c98

Remove code. We can add this back when we actually need it.

Review URL: http://codereview.chromium.org/8775038

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112750 0039d316-1c4b-4281-b951-d872f2087c98

BUG=106016
TEST=manual


Review URL: http://codereview.chromium.org/8772006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112749 0039d316-1c4b-4281-b951-d872f2087c98

This adds WEBKIT_GLUE_EXPORT, WEBKIT_PLUGINS_EXPORT, and WEBKIT_EXTENSIONS_EXPORT macros that are used by the files in
webkit/{glue,plugins,extensions} to show what needs to be exported from a DLL.

R=darin@chromium.org
BUG=98755
TEST=waterfall stays green.

Review URL: http://codereview.chromium.org/8741006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112748 0039d316-1c4b-4281-b951-d872f2087c98

ActiveLoader encapsulates an active WebURLLoader and takes care of maintaining deferred status, references to parent object, and automatic cancelation during teardown.

As a result of fixing the imbalanced reference counts to BufferedResourceLoader there were a few use-after-free bugs due to doing work after executing callbacks. The ordering has been updated to ensure that no more work is done after executing callbacks.

BUG=100914

Review URL: http://codereview.chromium.org/8667002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112747 0039d316-1c4b-4281-b951-d872f2087c98

Fast close to titlebar only -> pause on titlebar-only -> slow minimize.

Also, moved close button 1 px lower so it doesn't show up when panel is minimized.

BUG=104645

Review URL: http://codereview.chromium.org/8773030

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112746 0039d316-1c4b-4281-b951-d872f2087c98

BUG=105130
TEST=Open Task Manager. Enable Instant. Type a query into the omnibox. Observe that the Instant preview is listed as "Instant: ..." instead of "Tab: ...".


Review URL: http://codereview.chromium.org/8759009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112745 0039d316-1c4b-4281-b951-d872f2087c98

The pyauto-based perf tests output perf results in a special format, which
are extracted from stdout and stored for graphing purposes.  I encountered
a case where one of these specially-formatted perf output lines was broken
up by other, interleaved output, causing the perf value to be missed when
the output was parsed to search for perf values.  I believe that flushing
stdout immediately after printing each perf output line should fix this
problem.

BUG=None
TEST=None

Review URL: http://codereview.chromium.org/8772022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112744 0039d316-1c4b-4281-b951-d872f2087c98

Add abstract GetGroupsToChange() method to ModelChangingSyncerCommand.
Use that to figure out which worker threads to post work on (instead of
posting on all of them).

Implement GetGroupsToChange() for each ModelChangingSyncerCommand.

Add GetEnabledGroups() and GetEnabledGroupsWithConflicts() functions to SyncSession.

Key unapplied updates index by type for ApplyUpdatesCommand.

Make the abstract methods of ModelChangingSyncerCommand protected.

This patch includes a speculative fix for the perf regression introduced by the last time this was landed (112178).

BUG=97832
TEST=

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=112178

Review URL: http://codereview.chromium.org/8637006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112743 0039d316-1c4b-4281-b951-d872f2087c98

Was causing confusion with code within the file calling the helper functions directly.

Also put back accidentally deleted line and re-enabled NoOverlapping test.

BUG=None
TEST=Panel*Test.*

Review URL: http://codereview.chromium.org/8772029

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112742 0039d316-1c4b-4281-b951-d872f2087c98

TBR=rtenneti@chromium.org
BUG=106196
TEST=greener tsan bots

Review URL: http://codereview.chromium.org/8769047

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112741 0039d316-1c4b-4281-b951-d872f2087c98

This removes some manual registration code. It should also make FileRef private
work in process, it didn't look like it was being registered for in-process use
before.

Review URL: http://codereview.chromium.org/8772027

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112740 0039d316-1c4b-4281-b951-d872f2087c98

- This CL makes media_unittests linked.
- The audio_track_output_stub_android.cc is in place of
  audio_track_output_android.cc because the Java environment is not
  avaliable yet.
- Also changed to dependence on '../third_party/ffmpeg/ffmpeg.gyp'
  only if os is not Android in media.gyp.

BUG=
TEST=


Review URL: http://codereview.chromium.org/8718014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112739 0039d316-1c4b-4281-b951-d872f2087c98

Small changes to WGC3DInProcessImpl to pass the extension up to WebKit.

BUG=none
TEST=with https://bugs.webkit.org/show_bug.cgi?id=67341 and --use-gl=osmesa


Review URL: http://codereview.chromium.org/8772021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@112738 0039d316-1c4b-4281-b951-d872f2087c98