BUG=63073
TEST=see bug

Review URL: http://codereview.chromium.org/4973001

TBR=tfarina@chromium.org
Review URL: http://codereview.chromium.org/4953003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66195 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/4967003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66194 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/4983004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66193 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=None
TBR=bulach@chromium.org

Review URL: http://codereview.chromium.org/5040001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66192 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/4303003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66191 0039d316-1c4b-4281-b951-d872f2087c98

BUG=63073
TEST=see bug

Review URL: http://codereview.chromium.org/4973001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66190 0039d316-1c4b-4281-b951-d872f2087c98

Instead, always show it in a normal window (creating one if necessary).

The change is made in BrowserNavigator because that is the control flow bottleneck.

BUG=61813
TEST=manual

Review URL: http://codereview.chromium.org/4655002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66189 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=None
TBR=bulach@chromium.org

Review URL: http://codereview.chromium.org/5033001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66188 0039d316-1c4b-4281-b951-d872f2087c98

. Makes the code for running the promo more real. We're not ready to
  flip it on yet, so it isn't enabled.
. Nukes verbatim from about:flags and instead randomly chooses
  verbatim/predictive when the user opts in. We're doing this to try
  and see if we get feedback as to which one folks like more. Windows
  preferences shows you which one you get (by way of an obscure
  label). I'll change gtk do to that shortly too.
. Adds metrics.

And a couple of random style issues I encountered in some files.

BUG=NONE
TEST=NONE

Review URL: http://codereview.chromium.org/5023001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66187 0039d316-1c4b-4281-b951-d872f2087c98

BUG=63151
TEST=see bug

Review URL: http://codereview.chromium.org/4978002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66186 0039d316-1c4b-4281-b951-d872f2087c98

Xib changes: Removed the close button and tear image.

BUG=59068
TEST=Tear image should be gone.
Review URL: http://codereview.chromium.org/5031001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66185 0039d316-1c4b-4281-b951-d872f2087c98

Collecting this can take a couple of seconds. I put the code onto a worker thread. The about:gpuinfo handler polls for it until it is available, initially displaying only the subset of informtation that can be retreived quickly.

This makes the startup time for accelerated compositing, WebGL, etc significantly lower on Windows. None of the other platforms have this issue.

TEST=go to about:gpuinfo, try
BUG=59711

Review URL: http://codereview.chromium.org/4860001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66184 0039d316-1c4b-4281-b951-d872f2087c98

MediaFilterCollection is the only code that uses FilterType now so
I'm moving the enum into that class and removing all references from
the MediaFilter & derived classes.

BUG=61823
TEST=No new functionality. Existing tests pass.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=66125

Review URL: http://codereview.chromium.org/4653005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66182 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=trybots

Review URL: http://codereview.chromium.org/5004002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66181 0039d316-1c4b-4281-b951-d872f2087c98

we should attempt re-install. This helps robustness e.g. in the
case where we change profile directories (as we recently did).

Mark a test that I noticed was flaky (and open a bug for it).

BUG=none
TEST=none


Review URL: http://codereview.chromium.org/5027001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66180 0039d316-1c4b-4281-b951-d872f2087c98

TEST=none
BUG=none


git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66178 0039d316-1c4b-4281-b951-d872f2087c98

Some tests require regular GL and appear not to be flaky.

This command line switch is now only passed to tests that are flaky on bots using regular GL.

TEST=run affected tests locally, try
BUG=55477, 62602

Review URL: http://codereview.chromium.org/4784001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66177 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none
Review URL: http://codereview.chromium.org/4986002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66176 0039d316-1c4b-4281-b951-d872f2087c98

Fixed a hang in the renderer process. If the renderer process requests for a channel synchronously and the channel could not be established, the renderer thread would keep waiting for the reply to synchronization request.
BUG=61771
TEST=Manually tested pages that require hardware acceleration to make sure they still render correctly
Review URL: http://codereview.chromium.org/4884002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66175 0039d316-1c4b-4281-b951-d872f2087c98

Aaron's new test in http://codereview.chromium.org/4471001 to pass.

BUG=49233
TEST=base_unittests Time.FromExplodedWithMilliseconds from
     http://codereview.chromium.org/4471001
Review URL: http://codereview.chromium.org/5042001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66174 0039d316-1c4b-4281-b951-d872f2087c98

BUG=57799
TEST=none
TBR=bulach
Review URL: http://codereview.chromium.org/5037001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66173 0039d316-1c4b-4281-b951-d872f2087c98

No need to include all of carbon, just need CoreFoundation, and impl gets that from the header including it.

BUG=none
TEST=none
Review URL: http://codereview.chromium.org/5022002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66172 0039d316-1c4b-4281-b951-d872f2087c98

unit_tests fails to link due to requirement on libXi. This is implicitly
provided by 'gdk', but unit_tests does not depend on this target. We add
a 'xi' dependency for any target depending on 'x11'.

BUG=63209
TEST=None
(Landing patch for jknotten@chromium.org.)

Review URL: http://codereview.chromium.org/5016001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66169 0039d316-1c4b-4281-b951-d872f2087c98

Some deps in the chrome tree and support classes on the Chrome OS side
have changed, so update our autotest wrappers to work again.  The new
code also takes advantage of some new features in the ChromeTestBase class.

I've also split the control files to each run 1/4 of the browser tests and ui tests.  I tried to come up with a way to divide them more cleanly, but this seems the best way to split them across multiple VMs with the infrastructure on the bots.  If we used autoserv to run smoke suite, we could be more clever here, but that's a big hammer for this problem.

BUG=chromium-os:8544
TEST=install chrome with tests, then run desktopui_BrowserTest and friends.  They should be able to run, though not necessarily succeed.

Review URL: http://codereview.chromium.org/4698004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66168 0039d316-1c4b-4281-b951-d872f2087c98

the 'Get Themes' link.

BUG=62887
TEST=none

Review URL: http://codereview.chromium.org/4821003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66166 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=None
TBR=bulach@chromium.org

Review URL: http://codereview.chromium.org/4967002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66165 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/4932001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66164 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=None
TBR=eglaysher@chromium.org

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66163 0039d316-1c4b-4281-b951-d872f2087c98

BUG=50121

Review URL: http://codereview.chromium.org/5017002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66162 0039d316-1c4b-4281-b951-d872f2087c98

BUG=63239
TEST=None
TBR=brettw@chromium.org

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66160 0039d316-1c4b-4281-b951-d872f2087c98

The patch makes EventSendingController send touch events more like
how it currently handles mouse events: give them valid screenPosition
fields and do layout before dispatching them.

BUG=none
TEST=via layout tests

Review URL: http://codereview.chromium.org/4703004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66159 0039d316-1c4b-4281-b951-d872f2087c98

the tests to completion and hopefully turn green.

1. CF_DownloadFileFromPost. This test expects to see a File save dialog. On IE9
   the prompt to save the file is displayed as a link on the frame window.

2. WidgetModeIE_CFInstancePost, WidgetModeIE_CFInstanceRPC, WidgetModeIE_CFInstanceRPCInterna
   These tests fail on IE9. Needs more investigation.

TBR=amit


Review URL: http://codereview.chromium.org/4998003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66158 0039d316-1c4b-4281-b951-d872f2087c98

BUG=62943
TEST=ClientSideDetectionServiceTest

Review URL: http://codereview.chromium.org/4808003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66157 0039d316-1c4b-4281-b951-d872f2087c98

Since out-of-process tests override the EXE path to look like the bundled app, it makes sense to override AmIBundled() as well.

This is important because the renderer process started from browser_tests runs as bundled, and if browser and renderer process don't agree on bundled-ness, the "load plugin" requests for internal plugins from the renderer have the wrong plugin path, causing the plugin load to fail.

Also add a DCHECK that makes sure that AmIBundled() doesn't flip-flop.

This makes PDFBrowserTest work on mac, so enable it.

It looks like even unit_tests uses the out-of-process test runner, so this change is a bit hairy :-/

BUG=61258,63183
TEST=all existing tests still pass, PDFBrowserTest.* passes.

Review URL: http://codereview.chromium.org/4947002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66156 0039d316-1c4b-4281-b951-d872f2087c98

BUG=63228
TEST=None
TBR=cmasone@chromium.org

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66155 0039d316-1c4b-4281-b951-d872f2087c98

Instead of forcing all linux devs to have an acceptable version of git
installed, mitigate the problem of the cros_deps/DEPS file living outside
the main chrome checkout by moving our extra DEPS file inside the main chrome
tree.

Also change our naming convention to be more like that which is in use
elsewhere in the project.

In a week, cros_deps/DEPS will be end-of-lifed.

BUG=None
TEST=None

Review URL: http://codereview.chromium.org/4970003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66153 0039d316-1c4b-4281-b951-d872f2087c98

BUG=63228
TEST=None
TBR=cmasone@chromium.org

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66150 0039d316-1c4b-4281-b951-d872f2087c98

There was a problem with pointer arithmetic for 64 bit systems
so invalid data was not properly detected. Now we do explicit tests.

BUG=56449
TEST=base_unittests

Review URL: http://codereview.chromium.org/4716006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66149 0039d316-1c4b-4281-b951-d872f2087c98

BUG=63065
TEST=none
TBR=isherman
Review URL: http://codereview.chromium.org/4996003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66148 0039d316-1c4b-4281-b951-d872f2087c98

The sequence of events needed to trigger the assertion is a little complicated (see unittest for specifics).

The general pattern goes something like this:
  (1) Add a dependency between two sources of different types. (This increments the refcount on the dependency.)
  (2) Saturate the "graveyard" containing the dependency causing it to be nuked.
  (3) Add a new entry to the original dependency, causing it to re-create the source (but now with 0 refcount).
  (4) Force the buffer containing the source in step (1) to be nuked
  (5) In releasing the dependencies, it will now try to release reference on non-existent source.

While this was crashing debug mode, I don't believe there would have been an impact on release builds (since DeleteSourceInfo() early returns if it can't find the source).

BUG=58847
TEST=PassiveLogCollectorTest.ReleaseDependencyToUnreferencedSource
  
Review URL: http://codereview.chromium.org/4960001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@66147 0039d316-1c4b-4281-b951-d872f2087c98