GitLab

1. Make sure chrome_elf.dll imports nothing besides kernel32, advapi32, and some msvc libs (DEBUG builds)
2. Add gyp action and test to ensure chrome_elf.dll is always the first entry in chrome.exe's import table.

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=242834

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=243048

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=245197

Review URL: https://codereview.chromium.org/109483003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245312 0039d316-1c4b-4281-b951-d872f2087c98

Reason for revert:
Causes crashes on Android on loading a web page, more details in https://code.google.com/p/chromium/issues/detail?id=335258 

Original issue's description:
> Enable OrientationEvent only for android.
> 
> BUG=181136, 331337
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=245210

TBR=darin@chromium.org,mahesh.kk@samsung.com
NOTREECHECKS=true
NOTRY=true
BUG=181136, 331337

Review URL: https://codereview.chromium.org/130053005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245310 0039d316-1c4b-4281-b951-d872f2087c98

Changes in events can affect a number of different modules, so make sure
the unit-tests for at least ash, aura, events, keyboard, and views are
tried on the appropriate bots (linux_aura, win, win_rel, linux_chromeos,
linux_chromeos_asan).

BUG=318879
R=stip@chromium.org

Review URL: https://codereview.chromium.org/136853006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245309 0039d316-1c4b-4281-b951-d872f2087c98

After this change, DevToolsWindow waits for frontend to decide
on dock side, and only then shows it either docked or undocked.

This also cleans up DevToolsWindow public API, so clients don't use
test function, and test clients are explicit with their docking expectations.

Note: follow up patch on frontend side is require to switch
from SetDockSide to SetIsDocked call.

BUG=318751

Review URL: https://codereview.chromium.org/108213012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245308 0039d316-1c4b-4281-b951-d872f2087c98

This code is not chrome_elf-specific, rather, chrome_elf
is the only current user of it. Move the manifest building
logic to a more generic location.

BUG=324863

Review URL: https://codereview.chromium.org/137373002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245307 0039d316-1c4b-4281-b951-d872f2087c98

This was used by the make generator. The ninja generator uses ninja's built-in
"pools" feature for limiting ld parallelism, so this is no longer needed.

BUG=330631
R=torne@chromium.org

Review URL: https://codereview.chromium.org/139263010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245304 0039d316-1c4b-4281-b951-d872f2087c98

These things are exposed through the embedder API (and/or to be used by
the embedder).

R=sky@chromium.org

Review URL: https://codereview.chromium.org/137063010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245303 0039d316-1c4b-4281-b951-d872f2087c98

Added PrivetJSONOperation to call /privet/storage/list.

BUG=332182

Review URL: https://codereview.chromium.org/138983002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245302 0039d316-1c4b-4281-b951-d872f2087c98

This changes our unit from percentage-busy to ms-per-frame.
This should be less noisy to stalls where we are idle not
producing frames.

NOTRY=true
BUG=334063

No-try since this obviously doesn't break net-unittests.

Review URL: https://codereview.chromium.org/138923004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245301 0039d316-1c4b-4281-b951-d872f2087c98

BUG=335051
TEST=open create profile dialog, observe spacing above create button.

Review URL: https://codereview.chromium.org/140863002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245300 0039d316-1c4b-4281-b951-d872f2087c98

Adding the content_gl_tests unit tests to the
gtest suite list supported by test_runner.py for Android platform.

BUG=none

R=bulach

TEST= Command to execute the content_gl_tests on Android:
path/to/build/android/test_runner.py gtest -s content_gl_tests <--gtest_filter=[TestFilter]>

Review URL: https://codereview.chromium.org/140623002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245299 0039d316-1c4b-4281-b951-d872f2087c98

Avoid unneccessary copy of structures gfx::SizeF
by passing them by const ref rather than value.

Any struct of size > 4 bytes should be passed by const ref.
Passing by ref for these structs is faster than passing
by value, especially when invoking function has multiple parameters.

Pass by value creates unneccessary overhead which should be avoided.

BUG=159273

R=danakj, enne

Review URL: https://codereview.chromium.org/139173004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245297 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245296 0039d316-1c4b-4281-b951-d872f2087c98

The status code is consistent with other methods like Files.delete,
so we should better align to it. This CL fixes the return code
assumption in DriveAppRegistry that previously expected HTTP_SUCCESS.

BUG=324166

Review URL: https://codereview.chromium.org/140533002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245295 0039d316-1c4b-4281-b951-d872f2087c98

BUG=None
NOTRY=true

Review URL: https://codereview.chromium.org/140383002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245294 0039d316-1c4b-4281-b951-d872f2087c98

Ensure there is a) a focus ring around the entirety of the account
chooser, not just the image, if Chrome is running on 10.6. And b) use
the SDK's focus ring drawing from 10.7 on instead. (Before this CL,
10.7+ systems get _two_ focus rings for ImageButtonCells.)

BUG=308192
R=thakis@chromium.org

Review URL: https://codereview.chromium.org/140353002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245293 0039d316-1c4b-4281-b951-d872f2087c98

BUG=326250
R=mkearney@chromium.org
TBR=yoz@chromium.org

Review URL: https://codereview.chromium.org/134533004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245292 0039d316-1c4b-4281-b951-d872f2087c98

BUG=240309
R=scottmg@chromium.org

Review URL: https://codereview.chromium.org/140513005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245291 0039d316-1c4b-4281-b951-d872f2087c98

- make string translations work
- break libaddressinput standalone build

BUG=333405

Review URL: https://codereview.chromium.org/134903003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245290 0039d316-1c4b-4281-b951-d872f2087c98

Relanding this as this caused layout tests failures on Win7 due to the call to ShowScrollBars being incorrectly deleted
in the HWNDMessageHandler::OnSize for a regular WM_SIZE. Added that call back. The rest of the CL is exactly the same as
the previous one. TBR'ing sky.

Don't set the scroll styles (WS_VSCROLL and WS_HSCROLL) for WS_POPUP windows.

This causes issues with select boxes on Windows 7 where hovering at the end of the window returns the scroll WM_NCHITTEST
codes. Works fine on Windows 8. In any case we don't want the scrolling styles set on windows other than the main Chrome window
which is the only window which should be receive mousewheel messages.

I moved the scroll style setting code from the HWNDMessageHandler::OnCreate function to HWNDMessageHandler::Init function as that
would prevent the initial WM_SIZE message from hiding the scrollbar. The other change is to hide the scrollbars and readd the
scroll styles if we are sizing the window, when the sizing completes. Basically when we receive the WM_EXITSIZEMOVE message.
For normal sizing operations we continue to do this in WM_SIZE as before.

From testing on my thinkpad with Win7, desktop with Win7 and Win8 this works well

BUG=334454, 334541
TBR=sky

Review URL: https://codereview.chromium.org/133273020

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245289 0039d316-1c4b-4281-b951-d872f2087c98

BUG=332183

Review URL: https://codereview.chromium.org/140323002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245288 0039d316-1c4b-4281-b951-d872f2087c98

There's already a robust check in page_runner. The extra check is messing up record_wpr.


BUG=333609
TEST=mv ~/.boto ~/.boto.bak; rm tools/perf/page_sets/data/top_10_000.wpr; tools/perf/record_wpr page_sets/top_10.json; mv ~/.boto.bak ~/.boto

Review URL: https://codereview.chromium.org/140293002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245287 0039d316-1c4b-4281-b951-d872f2087c98

Implement a test fixture in DeviceMotionEventPumpTest to consistently
test DeviceMotionEventPump. The new implementation also avoids calls
to PlatformThread::Sleep() since it is generally undesirable in tests.

BUG=333928

Review URL: https://codereview.chromium.org/134073002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245286 0039d316-1c4b-4281-b951-d872f2087c98

1) To match the gyp dependencies, move clearkeycdmadapter binaries, plugins/,
   lib32/ and lib64/ from chrome.isolate to browser_tests.isolate.

2) On Windows, include <(version_full).manifest in chrome.isolate to be able
   to execute chrome.exe. Add needed chrome_child.dll and chrome_elf.dll.

3) Add libEGL.dll and libGLESv2.dll dependencies to gpu.gyp:gl_tests and a
   new angle.isolate; refer to it from gl_tests.isolate and
   gles2_conform_test.isolate.

4) Copy run_gpu_test to run_gpu_test.py, and invoke that instead in
   telemetry_gpu_test.isolate, to make this isolate work on Windows.

BUG=321878,334539

Review URL: https://codereview.chromium.org/140263002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245283 0039d316-1c4b-4281-b951-d872f2087c98

FakeDB::EntryMap wasn't getting cleaned up in the test. Changed the API
to make it clearer that ownership isn't taken.
BUG=334682

Review URL: https://codereview.chromium.org/140113002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245282 0039d316-1c4b-4281-b951-d872f2087c98

We need the file counts in order to send the completion event if we return a cached scan.  Also, in the scan result dialog we want to present the most interesting results first.

BUG=161119

Review URL: https://codereview.chromium.org/140033003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245281 0039d316-1c4b-4281-b951-d872f2087c98

Transitions does not work if change is 'auto' -> '0'. As work around code sets
height to offsetHeight before changing height to '0'. However Blink still uses
computed style and does not starts transition. So we have to call el.offsetHeight
to force update computed style.

Removed unused argument from onFadeOutTransitionEnd.

BUG=330604

Review URL: https://codereview.chromium.org/139963002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245280 0039d316-1c4b-4281-b951-d872f2087c98

Keeping the cache entry dirty to make sure that the edited file will eventually be uploaded.

BUG=326054
TEST=unit_tests

Review URL: https://codereview.chromium.org/139783004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245277 0039d316-1c4b-4281-b951-d872f2087c98

BUG=334711
TEST=Build on Chrome OS, check about:plugins, O3D is not there.

Review URL: https://codereview.chromium.org/139733002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245276 0039d316-1c4b-4281-b951-d872f2087c98

This CL also does a bit of cleanup moving the WeakPtrFactory to the bottom of the member variables of a couple of BrowserPlugin/webview classes.

BUG=334606

Review URL: https://codereview.chromium.org/139603002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245275 0039d316-1c4b-4281-b951-d872f2087c98

This message indicates to the recipient that the sender believes that it
is flow-control blocked but otherwise ready to send data. The BLOCKED frame
is purely advisory and optional.

This lands server change 56838900 by ygi.

R=rch@chromium.org

Review URL: https://codereview.chromium.org/139573002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245274 0039d316-1c4b-4281-b951-d872f2087c98

Revert of Convert the re2 GYP targets to be generated by GN. (https://codereview.chromium.org/68853015/)

Reason for revert:
Broke Blink compile

Original issue's description:
> Convert the re2 GYP targets to be generated by GN.
> 
> BUG=288991
> R=bradnelson@google.com
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=245254

TBR=bradnelson@chromium.org,bradnelson@google.com
NOTREECHECKS=true
NOTRY=true
BUG=288991

Review URL: https://codereview.chromium.org/141063002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245271 0039d316-1c4b-4281-b951-d872f2087c98

Reason for revert:
The indexeddb_perf tests on all Windows testers regressed in performance significantly. Don't yet know the exact cause, but because the performance drop is significant will revert this change first, and then proceed with the investigation.

Original issue's description:
> Created new Win32 LevelDB environment.
>     
> This change splits ChromiumEnv into two classes: ChromiumEnv/ChromiumEnvPosix.
> ChromiumEnvWin32 was also added which uses the Win32 API - except for logging.
> 
> The default is to create POSIX environments for all platforms for non-IDB
> LevelDB users. For IDB we create a Windows environment on Windows, else a
> POSIX environment like before.
> 
> BUG=222623
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=245135

TBR=dgrogan@chromium.org,alecflett@chromium.org,jsbell@chromium.org,iannucci@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=222623

Review URL: https://codereview.chromium.org/140923005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245270 0039d316-1c4b-4281-b951-d872f2087c98

Under non-impl-side painting, the thread proxy ends up releasing the
main thread then setting commit_waits_for_activation_ to false, which
is the wrong way around. This puts the order right.

R=enne
BUG=334611

Review URL: https://codereview.chromium.org/139553003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245269 0039d316-1c4b-4281-b951-d872f2087c98

TBR=szym@chromium.org
BUG=334837

Review URL: https://codereview.chromium.org/139543006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245268 0039d316-1c4b-4281-b951-d872f2087c98

BUG=

Review URL: https://codereview.chromium.org/139923002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245267 0039d316-1c4b-4281-b951-d872f2087c98

Review URL: https://codereview.chromium.org/139903003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245266 0039d316-1c4b-4281-b951-d872f2087c98

BUG=

Review URL: https://codereview.chromium.org/139883003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245265 0039d316-1c4b-4281-b951-d872f2087c98

Simplify the code by using DOM Promises.

BUG=282464
TEST=browser_tests

Review URL: https://codereview.chromium.org/132013002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245264 0039d316-1c4b-4281-b951-d872f2087c98

This CL updates the inbox_app in the Silk benchmark to use the newest
dynamics.js. The main difference is that we now do more work inside
requestAnimationFrame.

TBR=nduca

Review URL: https://codereview.chromium.org/140173002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@245263 0039d316-1c4b-4281-b951-d872f2087c98