GitLab

Allows workaround to work without precompiled headers (as in
buildtype=Official)

R=cpu@chromium.org
BUG=225822

Review URL: https://codereview.chromium.org/14201003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194506 0039d316-1c4b-4281-b951-d872f2087c98

> Add the LoggedIn Predictor, to detect which websites a user is likely
> logged into (or has been logged into).
> R=shishir@chromium.org, bauerb@chromium.org
> 
> Review URL: https://codereview.chromium.org/13903018

TBR=tburkard@chromium.org

Review URL: https://codereview.chromium.org/13995007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194505 0039d316-1c4b-4281-b951-d872f2087c98

R=shishir

Review URL: https://codereview.chromium.org/14307004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194504 0039d316-1c4b-4281-b951-d872f2087c98

Reverting seems to fix memory_test test locally on linux.

BUG=230085

> Revert r192949 "Disable shader disk cache by default."
> 
> This reverts commit r192949. The disable has been merged into
> M27 so re-enabling on trunk.
> 
> BUG=228989
> 
> Review URL: https://chromiumcodereview.appspot.com/13872012

TBR=dsinclair@chromium.org

Review URL: https://codereview.chromium.org/13924005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194503 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194500 0039d316-1c4b-4281-b951-d872f2087c98

BUG=160311
TBR=ben

Review URL: https://codereview.chromium.org/13998026

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194499 0039d316-1c4b-4281-b951-d872f2087c98

These were introduced in https://codereview.chromium.org/13932025/.

R=pkasting@chromium.org

Review URL: https://codereview.chromium.org/14081007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194498 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none

Review URL: https://codereview.chromium.org/14148004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194497 0039d316-1c4b-4281-b951-d872f2087c98

Add ChangeListLoader::CheckForUpdates()
Make ChangeListLoader::LoadFromServerIfNeeded() private.

BUG=193417
TEST=Open Drive in Files.app; Ctrl-click-on the gear menu and select 'Reload'. Confirm that the files appear shortly.

Review URL: https://codereview.chromium.org/14292002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194496 0039d316-1c4b-4281-b951-d872f2087c98

> Re-enable various ExtensionManagementApiTest tests.
> 
> These had been disabled due to flakes on Windows. The Win builders now
> haves symbols, so the stacks generated by flakes should be actionable.
> 
> BUG=174411,176023,179430,230165
> 
> Review URL: https://codereview.chromium.org/14232006

TBR=grt@chromium.org

Review URL: https://codereview.chromium.org/14318002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194495 0039d316-1c4b-4281-b951-d872f2087c98

The UMA histogram name is appended with library names. Since we only have limited number of ppapi plugins this shouldn't be an issue for the server to handle.

BUG=226107
TEST=Tested Widevine CDM loading.

Review URL: https://codereview.chromium.org/13548005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194494 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none

Review URL: https://codereview.chromium.org/14096008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194493 0039d316-1c4b-4281-b951-d872f2087c98

Previously, Telemetry would not output any summary ("RESULT=") lines if
at least one page in a page set failed.  Now, if a failed page exists,
Telemetry will output the individual page results from the passing pages,
but it will not output any average data or any overall results that are
not associated with a page.

Also added a few unit tests for this change.

BUG=230998
TEST=Verified with the scrolling_benchmark on a local chromeOS device
(which currently has at least one failing page) that RESULT= lines are
now outputted only for the passing pages.  Also verified that all unit
tests for page_benchmark_results.py pass.

Review URL: https://codereview.chromium.org/14172017

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194491 0039d316-1c4b-4281-b951-d872f2087c98

Still hanging on the bots

> Revert 194260 "Revert 194220 "[Telemetry] Make page cyclers more..."
> 
> Re-landing. Should work now that a race has been fixed by r194470.
> 
> > Revert 194220 "[Telemetry] Make page cyclers more telemetric."
> > 
> > Caused page cycler to hang on bots.
> > 
> > > [Telemetry] Make page cyclers more telemetric.
> > > 
> > > This causes the page cyclers to use page_sets instead of just navigating to the
> > > start page and letting the pages cycle themselves. This means we'll be using
> > > Telemetry "properly" which will allow us to record new page sets to update the
> > > page cyclers and it means that page cycler page sets are interchangeable with
> > > other benchmarks.
> > > 
> > > The page cyclers still measure the same thing as they measured previously: the
> > > time to load the page and perform a layout. However, they measure it differently
> > > now. Instead of setting a new Date().getTime() in the cookie immediately before
> > > navigating and then grabbing the finish Date in the onload handler, this now
> > > just grabs performance.now() in the onload handler. This returns the number of
> > > milliseconds since navigationStart where navigation start is the same as the
> > > immediately before navigation time marked before. I've verified locally that the
> > > times reported are consistent, and will monitor the bots to verify.
> > > 
> > > This involved moving the serving_dirs property up from the page to the page_set.
> > > 
> > > BUG=None
> > > TEST=All page cyclers on linux and morejs on windows
> > > NOTRY=True
> > > 
> > > Review URL: https://codereview.chromium.org/13817009
> > 
> > TBR=tonyg@chromium.org
> > 
> > Review URL: https://codereview.chromium.org/14048008
> 
> TBR=tonyg@chromium.org
> 
> Review URL: https://codereview.chromium.org/14119007

TBR=tonyg@chromium.org

Review URL: https://codereview.chromium.org/14222013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194489 0039d316-1c4b-4281-b951-d872f2087c98

Previously, the state stansition was CONNECTING->CONNECTED, but r194079 introduced the AUTHENTICATED state without updating the expectation in ConnectionToHost::NotifyIfChannelsReady

BUG=231997

Review URL: https://codereview.chromium.org/14308012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194488 0039d316-1c4b-4281-b951-d872f2087c98

BUG=225975
TBR=oshima@chromium.org

Review URL: https://codereview.chromium.org/13813004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194487 0039d316-1c4b-4281-b951-d872f2087c98

BUG=174002
This extension is not needed anymore.

Review URL: https://codereview.chromium.org/14301007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194486 0039d316-1c4b-4281-b951-d872f2087c98

logged into (or has been logged into).
R=shishir@chromium.org, bauerb@chromium.org

Review URL: https://codereview.chromium.org/13903018

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194484 0039d316-1c4b-4281-b951-d872f2087c98

It was missing en-US.pak, causing language string to be missing.

TBR=csharp@chromium.org
BUG=

Review URL: https://codereview.chromium.org/13921003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194483 0039d316-1c4b-4281-b951-d872f2087c98

In https://codereview.chromium.org/14058002, the function named
_IsWebkitVersionNotOlderThan was changed to
_IsContentVersionNotOlderThan.  There was one more reference to the
old function name in the file that needed to be updated.  That is
done here.  Without this fix, the Chrome Endure tests are broken.

BUG=None
TEST=Verified that a basic run of a Chrome Endure test failed before
this change, but passed after this change.

Review URL: https://codereview.chromium.org/14272005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194482 0039d316-1c4b-4281-b951-d872f2087c98

BUG=229660
R=thakis@chromium.org

Review URL: https://codereview.chromium.org/14064011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194481 0039d316-1c4b-4281-b951-d872f2087c98

Be more consistent about respecting link click dispositions in infobars: for most infobars, we should open with the desired disposition, unless it would overwrite the current window, in which case we use a new foreground tab.

BUG=none
TEST=none
Review URL: https://codereview.chromium.org/11778029

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194480 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none
TEST=none
Review URL: https://codereview.chromium.org/11773036

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194479 0039d316-1c4b-4281-b951-d872f2087c98

GeolocationConfirmInfoBarDelegate* -> GeolocationInfoBarDelegate*, for consistency with other infobars and brevity.

BUG=none
TEST=none
Review URL: https://codereview.chromium.org/11785030

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194478 0039d316-1c4b-4281-b951-d872f2087c98

Use of this animation was originally enabled during the initial constrained
window styling changes but was removed along with those changes.  Re-enable it
now for the web contents modal dialog.

Original review: http://crrev.com/11275028
Original bug: http://crbug.com/157796

BUG=166075

Review URL: https://codereview.chromium.org/13932019

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194477 0039d316-1c4b-4281-b951-d872f2087c98

BUG=224999

Review URL: https://codereview.chromium.org/14081009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194476 0039d316-1c4b-4281-b951-d872f2087c98

Re-landing. Should work now that a race has been fixed by r194470.

> Revert 194220 "[Telemetry] Make page cyclers more telemetric."
> 
> Caused page cycler to hang on bots.
> 
> > [Telemetry] Make page cyclers more telemetric.
> > 
> > This causes the page cyclers to use page_sets instead of just navigating to the
> > start page and letting the pages cycle themselves. This means we'll be using
> > Telemetry "properly" which will allow us to record new page sets to update the
> > page cyclers and it means that page cycler page sets are interchangeable with
> > other benchmarks.
> > 
> > The page cyclers still measure the same thing as they measured previously: the
> > time to load the page and perform a layout. However, they measure it differently
> > now. Instead of setting a new Date().getTime() in the cookie immediately before
> > navigating and then grabbing the finish Date in the onload handler, this now
> > just grabs performance.now() in the onload handler. This returns the number of
> > milliseconds since navigationStart where navigation start is the same as the
> > immediately before navigation time marked before. I've verified locally that the
> > times reported are consistent, and will monitor the bots to verify.
> > 
> > This involved moving the serving_dirs property up from the page to the page_set.
> > 
> > BUG=None
> > TEST=All page cyclers on linux and morejs on windows
> > NOTRY=True
> > 
> > Review URL: https://codereview.chromium.org/13817009
> 
> TBR=tonyg@chromium.org
> 
> Review URL: https://codereview.chromium.org/14048008

TBR=tonyg@chromium.org

Review URL: https://codereview.chromium.org/14119007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194475 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none

Review URL: https://codereview.chromium.org/12825014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194474 0039d316-1c4b-4281-b951-d872f2087c98

Sharing the same socket name causes tests to crash if
content_browsertests process is still running while running
content_shell instrumentation tests.

Examples failures can be seen on the tryjobs for: https://codereview.chromium.org/13445005/

BUG=230113

Review URL: https://codereview.chromium.org/14301005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194473 0039d316-1c4b-4281-b951-d872f2087c98

Previously we were only waiting for DOMContentLoaded instead of Load.

BUG=None
TEST=Recorded new page_sets on linux
NOTRY=True

Review URL: https://codereview.chromium.org/13991015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194472 0039d316-1c4b-4281-b951-d872f2087c98

BUG=223859

Review URL: https://codereview.chromium.org/14295015

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194471 0039d316-1c4b-4281-b951-d872f2087c98

I believe this is what was causing the page cyclers to hang on the bots. We
must add the script to evaluate on load prior to performing the navigation. If
we add it after, it is racy.

The reason it wasn't like this before is that Page.addScriptToEvaluateOnLoad
didn't work on cross-renderer navigations. But that problem is being fixed by
https://codereview.chromium.org/13949006/

The current file-based page cyclers don't do cross renderer navigations, so
will be fine even without the above fix. However, the above fix will be
necessary for the new WPR-based page sets.

BUG=None
TEST=page set based page cyclers on linux
NOTRY=True

Review URL: https://codereview.chromium.org/13926007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194470 0039d316-1c4b-4281-b951-d872f2087c98

Broke linux compilation http://build.chromium.org/p/chromium.linux/buildstatus?builder=Linux%20Builder%20%28dbg%29&number=47451
> media: Add support for playback for VP8 Alpha video streams.
> 
> BUG=147355
> TEST=VP8 Alpha video streams play
> 
> Review URL: https://codereview.chromium.org/12263013
> 
> Patch from Vignesh Venkatasubramanian <vigneshv@chromium.org>.

TBR=tomfinegan@chromium.org

Review URL: https://codereview.chromium.org/13972014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194469 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none

Review URL: https://codereview.chromium.org/13925014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194468 0039d316-1c4b-4281-b951-d872f2087c98

Using the new AudioBusTest::InterleaveBench on my Z600 over 5 runs:

Old:
ToInterleaved uint8 took 57.95ms.
FromInterleaved uint8 took 24.76ms.
ToInterleaved int16 took 47.67ms.
FromInterleaved int16 took 23.06ms.
ToInterleaved int32 took 59.64ms.
FromInterleaved int32 took 26.53ms.

New:
ToInterleaved uint8 took 56.09ms.
FromInterleaved uint8 took 21.91ms.
ToInterleaved int16 took 32.68ms.
FromInterleaved int16 took 22.96ms.
ToInterleaved int32 took 42.30ms.
FromInterleaved int32 took 33.45ms.

Results are faster across the board except for FromInterleaved int32
which lost ~10ms with the removal of the static const versus an inline
constant.  I'm writing it off to helpful pipelining.

Overall the new methods are ~13% faster. Benchmarking 10000 calls of
each data type over 480 frames instead of a single block of 120 seconds
of audio yields an even more impressive 2.7x speedup on OSX:

Old:
[       OK ] AudioBusTest.DISABLED_InterleaveBench (3157 ms)

New:
[       OK ] AudioBusTest.DISABLED_InterleaveBench (1154 ms)

Linux saw only a 10% improvement though -- the discrepancy likely has
to do with OSX's usage of SSE2 instructions (versus no-SSE for Linux).

The new methods are ~15% faster in aggregate on Windows.  It's ~ the same
on ToInterleaved() except in the int32 case where the new method is ~40% faster.
FromInterleaved() is ~ the same except for the uint8 case which is ~70% faster.

BUG=224662
TEST=media_unittests

Review URL: https://codereview.chromium.org/13730004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194467 0039d316-1c4b-4281-b951-d872f2087c98

The Perf Blink bots are failing with the following error:

File "/b/build/slave/Linux_Perf/build/src/tools/telemetry/telemetry/core/chrome/desktop_browser_finder.py", line 129, in AddCrOSIfFound
    ldd_out = subprocess.check_output([ldd_path, app])
AttributeError: 'module' object has no attribute 'check_output'

> Better support for chrome for cros local builds.
> 
> * Use ldd to detect chrome for cros local builds.
> * Add browser types debug-cros and release-cros for chrome for cros on linux.
> * Remove browser option --cros-desktop
> 
> BUG=229692
> TEST=run_tests works with cros-debug going through the login flow.
> NOTRY=True
> Review URL: https://codereview.chromium.org/14102002

TBR=achuith@chromium.org

Review URL: https://codereview.chromium.org/14298014

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194466 0039d316-1c4b-4281-b951-d872f2087c98

BUG=147355
TEST=VP8 Alpha video streams play

Review URL: https://codereview.chromium.org/12263013

Patch from Vignesh Venkatasubramanian <vigneshv@chromium.org>.

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194465 0039d316-1c4b-4281-b951-d872f2087c98

These had been disabled due to flakes on Windows. The Win builders now
haves symbols, so the stacks generated by flakes should be actionable.

BUG=174411,176023,179430,230165

Review URL: https://codereview.chromium.org/14232006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194464 0039d316-1c4b-4281-b951-d872f2087c98

If the ResourceBundle is not yet initialized, l10n_util::IsLocaleAvailable() should
return false to avoid the CHECK failure in ResourceBundle::GetSharedInstance().

BUG=230432
TEST=CQ

Review URL: https://codereview.chromium.org/14230008

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194463 0039d316-1c4b-4281-b951-d872f2087c98

When the notification views were resized the bindings that minimize
the amount of data sent across to the browser process weren't updated.

BUG=231015

Review URL: https://codereview.chromium.org/14181007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@194462 0039d316-1c4b-4281-b951-d872f2087c98