GitLab

TEST=demos run
BUG=none


Review URL: http://codereview.chromium.org/8947016

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114697 0039d316-1c4b-4281-b951-d872f2087c98

Now that we've tracked down the root cause for the perf regressions
triggered by custom groups, we can remove this function now.

BUG=97832,107040
TEST=


Review URL: http://codereview.chromium.org/8957001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114696 0039d316-1c4b-4281-b951-d872f2087c98

concurently with the extension garbage collection to sometimes get marked for
deletion.  The concurrency is between the extension installation in the browser
process and the garbage collection in the import process.

BUG=101935
TEST=Install google chrome and look at the default apps installed in the first
profile.  The icons for all the apps should be correct.  This does not work
with chromium builds since default apps are not installed there.

Review URL: http://codereview.chromium.org/8872031

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114695 0039d316-1c4b-4281-b951-d872f2087c98

Add a command-line switch --enable-per-tile-painting and make it consistent by changing --enable-accelerated-drawing to --enable-accelerated-painting.

With this switch, webkit compositor will paint tiles separately.

BUG=none
TEST=no behavior change in existing browser tests and ui tests

Review URL: http://codereview.chromium.org/8771056

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114694 0039d316-1c4b-4281-b951-d872f2087c98

BUG=107047
TEST=crypto_unittests
TBR=wtc@chromium.org

Review URL: http://codereview.chromium.org/8956003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114693 0039d316-1c4b-4281-b951-d872f2087c98

extension popups.

Also fix a leak that happens if the renderer that initiated a file chooser is
closed before the file chooser appears.

BUG=104222
TEST=see bug for repro

Review URL: http://codereview.chromium.org/8909015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114692 0039d316-1c4b-4281-b951-d872f2087c98

BUG=106856
TEST=DownloadTest.DownloadUrlToFile should pass on ChromeOS.

Review URL: http://codereview.chromium.org/8929026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114691 0039d316-1c4b-4281-b951-d872f2087c98

TBR=sky@chromium.org
BUG=none
TEST=check_perms
Review URL: http://codereview.chromium.org/8967002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114689 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/8956001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114688 0039d316-1c4b-4281-b951-d872f2087c98

BUG=105280
TEST=Lock the screen and enable caps lock. No bubble indicators are shown.


Review URL: http://codereview.chromium.org/8869002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114687 0039d316-1c4b-4281-b951-d872f2087c98

We were hitting the MAXPATH limit on the build.webkit.org win bots.

BUG=webkit.org/b/74320


Review URL: http://codereview.chromium.org/8930019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114686 0039d316-1c4b-4281-b951-d872f2087c98

We should consistently:
#include "base/gtest_prod_util.h"
which does the gtest include (from the right place, in a
second repo, hidden in our tree).


This is critical to being able to supplant the FRIEND_TEST
macro, which we're experimenting with to optimize builds.

r=brettw
TBR=other owners (this is a trivial change)
BUG=107727
Review URL: http://codereview.chromium.org/8954017

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114685 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=zip_reader_unittest.cc (compiles into 'unit_tests' binary)


Review URL: http://codereview.chromium.org/8873039

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114684 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/8905024

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114683 0039d316-1c4b-4281-b951-d872f2087c98

Friends may access methods that are otherwise unused.
Removing friends may cause dead-code elimination, and
also reduce the need for copies of entry points (allowing
more inlining).

r=laforge

TBR=jar@chromium.org
Review URL: http://codereview.chromium.org/8970002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114682 0039d316-1c4b-4281-b951-d872f2087c98

BUG=107547
TEST=Manual

Review URL: http://codereview.chromium.org/8936015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114681 0039d316-1c4b-4281-b951-d872f2087c98

Friends may access methods that are otherwise unused.
Removing friends may cause dead-code elimination, and
also reduce the need for copies of entry points (allowing
more inlining).

r=laforge

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114680 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=trybots
TBR=brettw

Review URL: http://codereview.chromium.org/8898019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114679 0039d316-1c4b-4281-b951-d872f2087c98

This test is crashing on linux and windows sync bots.

BUG=107743
TEST=

Review URL: http://codereview.chromium.org/8972001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114678 0039d316-1c4b-4281-b951-d872f2087c98

TBR=acolwell
BUG=106735

Review URL: http://codereview.chromium.org/8966002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114675 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none
R=dpapad

Review URL: http://codereview.chromium.org/8956014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114674 0039d316-1c4b-4281-b951-d872f2087c98

MigrationTwoClientTest.MigrationHellWithoutNigori and MigrationTwoClientTest.MigrationHellWithNigori have been crashing on various bots. Backtrace attached to bug.

BUG=107205
TEST=

Review URL: http://codereview.chromium.org/8966001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114673 0039d316-1c4b-4281-b951-d872f2087c98

[This change was shipping to dev/canary over the past
weekend, and had no impact on buildability.  I'm landing
this yet again since the bug has a large impact on XP
users, and we need it tested in the field if we are
going to push it to M16 in the new year]


This code should only have much of an impact on the XP
platform, where thread teardown notification are not
currently made.

Provide an automatic fallback scan of the linker
list of notifcation callbacks via DllMain.

This fallback is ONLY used if there are no notifications
coming via the linker list. This allows all existing
code to continue working as it does today, and provides
thread teardown notifcatino on XP even though we use
the LoadLibrary call.

It is a minimal change, suitable for pushing to other
channels (beta? stable?).

r=cpu
BUG=103209
Review URL: http://codereview.chromium.org/8836003

TBR=jar@chromium.org
Review URL: http://codereview.chromium.org/8934006

TBR=jar@chromium.org
Review URL: http://codereview.chromium.org/8964002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114672 0039d316-1c4b-4281-b951-d872f2087c98

Now that Chrome 16 is out the door, the blacklist erosion process is resuming.

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114671 0039d316-1c4b-4281-b951-d872f2087c98

- As promised.

Strange microsoft thing: there is a function to intialize the cv but no function to unititialize it.

BUG=none
TEST=current tests suffice
Review URL: http://codereview.chromium.org/8953003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114670 0039d316-1c4b-4281-b951-d872f2087c98

The sync integration test for auth errors was failing on the mac buildbots. It turns out that this was due to keychain prompts on mac, as we were trying to add a password form, due to a known issue with mac keychain.

This patch switches the datatype used by the auth error test to bookmarks, so that this test can be reenabled. The underlying issue with keychain will be fixed in another patch.

BUG=107611
TEST=SyncErrorTest.AuthErrorTest

Review URL: http://codereview.chromium.org/8954018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114669 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none


Review URL: http://codereview.chromium.org/8914021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114668 0039d316-1c4b-4281-b951-d872f2087c98

app list, hence code changes to shuffle things around.

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

Review URL: http://codereview.chromium.org/8957012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114666 0039d316-1c4b-4281-b951-d872f2087c98

Now with ChromeOS fix.

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114664 0039d316-1c4b-4281-b951-d872f2087c98

BUG=107469
TEST=Verify that chrome://settings-frame still functions sans extensions.
Review URL: http://codereview.chromium.org/8949006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114663 0039d316-1c4b-4281-b951-d872f2087c98

This was accidentally removed in http://codereview.chromium.org/8491043/diff/19001/content/browser/gpu/gpu_process_host.cc

BUG=None
TEST=start browser with --trace-startup


Review URL: http://codereview.chromium.org/8912005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114662 0039d316-1c4b-4281-b951-d872f2087c98

BUG=
TEST=

Review URL: http://codereview.chromium.org/8949018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114661 0039d316-1c4b-4281-b951-d872f2087c98

CMT can now generate motion events w/ valuators instead of a sequence of button events. This supports them

BUG=None
TEST=None

Review URL: http://codereview.chromium.org/8907005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114660 0039d316-1c4b-4281-b951-d872f2087c98

*) When old and new IDs are equal, ignores old ID (fallback to the prepopulated provider).
*) When new and fallback providers have the same name, doesn't suggest to 'Change to <new provider>'.
*) Ignores logo_id when comparing providers.


BUG=107516
TEST=Manual: see bug description


Review URL: http://codereview.chromium.org/8954014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114659 0039d316-1c4b-4281-b951-d872f2087c98

to tell it the size of the launcher so that it doesn't layout on top
of the status.

BUG=106432
TEST=covered by tests.
R=ben@chromium.org


Review URL: http://codereview.chromium.org/8953004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114658 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=existing automation
TBR=sky
Review URL: http://codereview.chromium.org/8956004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114657 0039d316-1c4b-4281-b951-d872f2087c98

ninja doesn't like there being several targets with the name 'ssl', so I
renamed NSS's 'ssl' target to 'libssl'.

BUG=none
TEST=compiles


Review URL: http://codereview.chromium.org/8907043

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114656 0039d316-1c4b-4281-b951-d872f2087c98

because there is an existing TabDragController when a new one is
created. Right after we create the new TabDragController it replaces
the old, but if we create the new one before the old is destroyed it
means the old is still set as a delegate on the TabContents so that
when the new restores the delegates it restores the old
TabDragController, which has now been deleted.

BUG=107172
TEST=none
R=ben@chromium.org


Review URL: http://codereview.chromium.org/8915026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114653 0039d316-1c4b-4281-b951-d872f2087c98

BUG=104399
TEST=TwoClientExtensionsSyncTest.UninstallPreinstalledExtensions

Review URL: http://codereview.chromium.org/8889048

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114652 0039d316-1c4b-4281-b951-d872f2087c98

R=avi@chromium.org
TBR=rlp@chromium.org
BUG=98716
TEST=

Review URL: http://codereview.chromium.org/8921021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@114651 0039d316-1c4b-4281-b951-d872f2087c98