GitLab

This change attempts to reduce the load time of the file manager (especially on slow Alex devices) by trimming unnecessary work done in javascript and reducing layouts.

- Enable batch updating in cr.ui.Table (exactly how it's done in cr.ui.List).
- Add more 'on complete' callbacks to some of the FileManager infrastructure so we know when to stop batch UI updates.
- Use batch updates for some operations which profiling indicates causes non-trivial amounts of duplicated work.  In particular, in my testing this reduces the number of (sometimes expensive) List.redraw() calls on startup for the table from 6 down to 1, and for the roots list from 4 down to 2.

Measurements on alex are quite variable, but these changes result in about 70ms savings on startup (or about 17% of the time spent under 'v8.callChromeHiddenMethod' - i.e. JS callbacks through the extension system, which itself is about 1/3rd of total load time).

The majority of file manager load time is spent inside of v8, and there are many more opportunities like these to trim various code paths.  But it seems clear that major improvements are going to require more drastic approaches (eg. I'm experimenting with painting the initial UI after parsing/running a small fraction of the JS).

BUG=105181
TEST=


Review URL: http://codereview.chromium.org/9379023

TBR=rbyers@chromium.org
Review URL: https://chromiumcodereview.appspot.com/9580035

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124786 0039d316-1c4b-4281-b951-d872f2087c98

Also re-introduces a missing life check in case a request callback deletes the HostResolver.

R=eroman@chromium.org,cbentzel@chromium.org
BUG=115399
TEST=./net_unittest --gtest_filter=HostResolverImplTest.CanceledRequestsReleaseJobSlots


Review URL: http://codereview.chromium.org/9572018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124783 0039d316-1c4b-4281-b951-d872f2087c98

There were two problems.
1) We had started to write into profile/IndexedDB/IndexedDB/origin.leveldb/ (note the double IndexedDB).
2) We were writing into file__0.leveldb but checking for quota in __0.leveldb.  Now we always use __0.leveldb.

BUG=104748
TEST=


Review URL: http://codereview.chromium.org/9567030

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124782 0039d316-1c4b-4281-b951-d872f2087c98

depends on http://codereview.chromium.org/9465018/

BUG=110050
TEST=No user visible changes, see unit tests.

Review URL: http://codereview.chromium.org/9466022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124781 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none


Review URL: http://codereview.chromium.org/9566046

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124780 0039d316-1c4b-4281-b951-d872f2087c98

-- limit visit history considered to 100k entries to avoid out of memory crashes
-- exclude backforward/homepage/intermediate pages in redirect chain
-- reduce minimum spacing of pages considered from 1s to 500ms
-- collect histogram of core transition type of qualifying page views
R=brettw, dominich
Review URL: https://chromiumcodereview.appspot.com/9572021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124779 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=happy heapcheck bots
TBR=mdm@chromium.org

Review URL: https://chromiumcodereview.appspot.com/9586023

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124778 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/9463026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124777 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124776 0039d316-1c4b-4281-b951-d872f2087c98

BUG=chromium-os:27223
TEST=GDataFileSystemTest.SearchExistingDocument
Review URL: https://chromiumcodereview.appspot.com/9581024

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124775 0039d316-1c4b-4281-b951-d872f2087c98

TBR=haraken@chromium.org

Review URL: https://chromiumcodereview.appspot.com/9580028

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124774 0039d316-1c4b-4281-b951-d872f2087c98

Unit tests to catch this will come in another CL.

BUG=116607
TEST=

Review URL: https://chromiumcodereview.appspot.com/9585025

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124772 0039d316-1c4b-4281-b951-d872f2087c98

"--use-system-ssl" usage. The flag forces the user to use SSLClientSocket[Mac,
  Win], instead of the default NSS SSL implementation.

I also changed the order of initialize callings in
BrowserMainRunnerImpl::Initialize(), because originally the StatisticsRecorder
was created after main loop flag parsing so the new histogram can not be
recorded. Since StatisticsRecorder has few dependencies, I think this will
not have negative effect.

BUG=102003

Review URL: http://codereview.chromium.org/9570052

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124770 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
TEST=Manual


Review URL: http://codereview.chromium.org/9582022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124769 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none
Review URL: https://chromiumcodereview.appspot.com/9581028

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124768 0039d316-1c4b-4281-b951-d872f2087c98

This is a minimal starting point for merging our upstream android
ntp resource changes into a somewhat clean slate.  Based on how little of the
shared resource cache code we used, it seems cleaner to start our own and
not muddle that file with a ton of ifdefs.

This is all in the hopes of getting sync_unit_tests linking upstream.

BUG=113487
TEST=


Review URL: http://codereview.chromium.org/9558014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124767 0039d316-1c4b-4281-b951-d872f2087c98

Page content often isn't completely loaded or laid out when
the tab is first rendered.  This makes the class
additionally notify about subsequent renders so the delegate
can hold off until it sees one after it's e.g. received
notification via a WebUI message that the page is truly
ready.

BUG=chromium-os:26210
TEST=manual; also un-disabled the tests and checked that they still pass
TBR=ben@chromium.org

Review URL: http://codereview.chromium.org/9582012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124765 0039d316-1c4b-4281-b951-d872f2087c98

Update strings.


BUG=106636
TEST=


Review URL: http://codereview.chromium.org/9433054

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124764 0039d316-1c4b-4281-b951-d872f2087c98

Split out from https://chromiumcodereview.appspot.com/9363008/

BUG=110400
TEST=NONE


Review URL: http://codereview.chromium.org/9580018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124763 0039d316-1c4b-4281-b951-d872f2087c98

This is addition to http://codereview.chromium.org/9477021/

BUG=chromium-os:26646
TEST=none

Review URL: https://chromiumcodereview.appspot.com/9584026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124762 0039d316-1c4b-4281-b951-d872f2087c98

Changing some highlight / autocomplete colors


BUG=115861
TEST=Visually tested


Review URL: http://codereview.chromium.org/9585015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124761 0039d316-1c4b-4281-b951-d872f2087c98

This removes an animation that adds a bit of jank at
startup.  It doesn't have a significant effect on boot time
in either direction, from my measurements.

BUG=chromium-os:26210
TEST=manual: booted and checked that animation is gone


Review URL: http://codereview.chromium.org/9568044

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124760 0039d316-1c4b-4281-b951-d872f2087c98

Adding this message all the time prevents us from processing other command buffers when there really isn't anything left to do.

BUG=112349
TEST=


Review URL: http://codereview.chromium.org/9583021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124759 0039d316-1c4b-4281-b951-d872f2087c98

Make everything use this instead of rolling their own.

BUG=113723
TEST=

Review URL: https://chromiumcodereview.appspot.com/9562037

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124758 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: http://codereview.chromium.org/9569032

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124757 0039d316-1c4b-4281-b951-d872f2087c98

BUG=chromium-os:26939
TEST=--allow-webui-compositing no longer needed to get HW-accelerated decode & render on ARM crosbook.


Review URL: http://codereview.chromium.org/9583022

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124756 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none


Review URL: http://codereview.chromium.org/9569029

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124755 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=build using the command line on Windows, piping through sed, see less extraneous output


Review URL: http://codereview.chromium.org/9423018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124754 0039d316-1c4b-4281-b951-d872f2087c98

TBR=jochen
BUG=
TEST=cookies.CookiesTest.testCookiesHttps

Review URL: https://chromiumcodereview.appspot.com/9584029

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124753 0039d316-1c4b-4281-b951-d872f2087c98

BUG=116458
R=ben
TBR=davemoore

Review URL: http://codereview.chromium.org/9570044

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124752 0039d316-1c4b-4281-b951-d872f2087c98

BUG=116501

Review URL: http://codereview.chromium.org/9582020

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124751 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124748 0039d316-1c4b-4281-b951-d872f2087c98

TBR=haraken@chromium.org

Review URL: https://chromiumcodereview.appspot.com/9583028

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124747 0039d316-1c4b-4281-b951-d872f2087c98

It includes:
* intoducing -z commandline option to show/not-show issue details
* fix the issues discovered by gpylint
* adding bug link and flakiness dashboard link to test expection file display

BUG=109008,107773
TEST= unit test passes and run script locally.

Review URL: https://chromiumcodereview.appspot.com/9476021

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124746 0039d316-1c4b-4281-b951-d872f2087c98

(1) User types something that doesn't look at all like a keyword (e.g. "go om") and gets a non-default dropdown result that does look like a keyword ("google.com").  Arrowing down to this result, the user then presses space after the keyword name.
(2) Similar to the case above, but this time the non-default result in question looks like a keyword + some more text, with no whitespace between ("google.comxxx").  The user arrows to this result, arrows back to just after the keyword, and presses space.

Item (2) was happening because ShouldAllowExactKeywordMatch() (which has now been renamed) was always looking at the old |user_text_| as the "before change" text, which was wrong if there was temporary text, since the user text was not visible at that point.  Fixed by getting the actual displayed text in that case.

Item (1) was happening for similar reasons, but in MaybeAcceptKeywordBySpace() instead.  This could have been fixed by the same change as above.  However, I elected to change this function to look at the |keyword_| instead of the previously-displayed text, as in https://chromiumcodereview.appspot.com/9289034/ patch set 1.  Not only is this slightly simpler, it is more robust against future changes.  Let's say that someday we want to allow a provider to show a keyword hint for keyword "foobar" on input of "foo ".  By checking against |keyword_| in this function, we ensure that in that future case, the space after "foo" will never trigger keyword search mode for "foobar".  (If we ever bother to fix our bugs with keywords containing spaces, this might apply today for those as well.)

I also tried to change the naming and comments of some things for clarity.

BUG=none
TEST=none
Review URL: https://chromiumcodereview.appspot.com/9570064

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124745 0039d316-1c4b-4281-b951-d872f2087c98

* New window frames, still behind a flag as there's more refinement to do for incognito, popups, and maximized windows.
* Translucent frames now only require a single layer the size of the window instead of two.  We may be able to cut the layer size in the future to the area of the window header + toolbar
* Rewrite BrowserNonClientFrameViewAura
* Use updated art assets from crrev.com/124602
* Allow windows optionally to specify that hit tests should include a region a few pixels outside the window. For Aura browser windows this area acts as non-client area resize handles which we need because the window frame borders are now very thin or not present at all.

BUG=113075
TEST=manual, visual inspection of browser active, inactive, incognito, popup, hosted, and app windows.  Updated aura_unittest for Window to check the expanded bounds.

Review URL: https://chromiumcodereview.appspot.com/9580001

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124743 0039d316-1c4b-4281-b951-d872f2087c98

BUG=chromium-os:26961
TEST=none
Review URL: https://chromiumcodereview.appspot.com/9580016

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124742 0039d316-1c4b-4281-b951-d872f2087c98

Bug 111139: Deprecate HOMEPAGE option, remove the UI for it, migrate users away from it

BUG=111139
TEST=SessionStartupPrefTest.HomePageMigration
TBR=ben@chromium.org

Review URL: https://chromiumcodereview.appspot.com/9559002

TBR=tbreisacher@chromium.org
Review URL: https://chromiumcodereview.appspot.com/9583025

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124740 0039d316-1c4b-4281-b951-d872f2087c98

Before this CL the contract of SetMaxTime was unsatisfiable; a clock client
could check that the max_time it was about to set wasn't greater than Elapsed()
before calling SetMaxTime(), but between the time that check was done and the
DCHECK inside SetMaxTime, time could move forward!
After this CL, SetMaxTime() is allowed to move the media_time_ backward and set
the underflow_ bit if the pipeline is slow enough.

BUG=113037,chromium-os:26939
TEST=A Debug binary on an underpowered ARM crosbook with no audio device plays back correctly.


Review URL: http://codereview.chromium.org/9582017

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124739 0039d316-1c4b-4281-b951-d872f2087c98

Also make it possible to run the test against other hosts by specifying the environment variables COOKIES_TEST_HOST

BUG=none
TEST=pyauto


Review URL: http://codereview.chromium.org/9584012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@124738 0039d316-1c4b-4281-b951-d872f2087c98