GitLab

leak was fixed in r213560

BUG=175815
NOTRY=true

Review URL: https://codereview.chromium.org/383333002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282724 0039d316-1c4b-4281-b951-d872f2087c98

DeviceMonitorMac: move CrAVFoundationDeviceObserver and most of SuspendObserverDelegate to UI Thread

(tl; dr : Move all operations to UI thread except device enumerations.)

CrAVFoundationObserver was located in Device Thread based 
on the assumption that OS KVO callbacks would come on that 
thread too, but instead they come from UI thread.
-observeValueForKeyPath:... is then called in UI thread, and since
the rest of the actions of the class are small, this CL moves the
whole class to UI thread. 

Its overlord SuspendObserverDelegate (best not use acronyms 
for its name :) ), however, lives a mixed life in UI and Device
 threads. The model is simplified by making it work always in
UI thread _except_ for device enumerations (done via
[AVCaptureDeviceglue devices]).

AVFoundationMonitorImpl will destroy SuspendObserverDelegate
in UI thread and that in turn destroys CrAVFoundationObserver
in that very thread, thus cleaning up the multi threading and
hopefully addressing the bug reports of a small but consistent
crash rate (~2 crashes every four canaries or so). 

UI Thread checks are added everywhere.

The code around CrAVFoundationDeviceObserver::dealloc and 
-stopObserving is refactored in order to avoid a redundant 
search-for-device in |monitoredDevices_|.

BUG=366087

Review URL: https://codereview.chromium.org/368613002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282723 0039d316-1c4b-4281-b951-d872f2087c98

IDs can't begin with a digit. However, this syntax allows it.

BUG=361809
TBR=dtu@chromium.org,chrishenry@google.com
NOTRY=True

Review URL: https://codereview.chromium.org/383263004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282721 0039d316-1c4b-4281-b951-d872f2087c98

R=estade@chromium.org
BUG=none

Review URL: https://codereview.chromium.org/388723002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282720 0039d316-1c4b-4281-b951-d872f2087c98

BUG=337663

Review URL: https://codereview.chromium.org/383903004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282718 0039d316-1c4b-4281-b951-d872f2087c98

Patch set 1: Use XInput2 events for keyboard events.

XI2 keyboard events are generated, rewritten to core events, and consumed. EventRewriter tests pass with XI2 key events.

Patch set 2: Move source_device_id_ up to |ui::Event|.

Patch set 3: EventRewriter is no longer a PlatformEventObserver or DeviceHierarchyObserver.

|chromeos::EventRewriter| now uses |ui::KeyEvent::source_device_id()| instead of tracking XI2 key events itself as a chromeos::DeviceHierarchyObserver and |ui::PlatformEventObserver|.

Patch set 4: Convert Alt + Left Button rewriter from XI2 events to|ui::Event|s.

BUG=368750

Review URL: https://codereview.chromium.org/336403005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282717 0039d316-1c4b-4281-b951-d872f2087c98

BUG=391943
TBR=nduca@chromium.org

Review URL: https://codereview.chromium.org/387893004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282715 0039d316-1c4b-4281-b951-d872f2087c98

Fixes r282703.

BUG=none
R=brettw@chromium.org
TBR=msw@chromium.org

Review URL: https://codereview.chromium.org/387993006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282714 0039d316-1c4b-4281-b951-d872f2087c98

The current implementation assumes the Android system OpenSSL and our copy have
compatible ABIs. This will be a problem when switching to BoringSSL which has
already changed from ABIs. Moreoever, it's already slightly off now as any
globals (error queue and locks) have different instances between the two.

Rework the code to never mix the two ABIs. We replicate the subset of OpenSSL
ABI we care about and use it to find the rsa_priv_enc implementation. This is
then mapped to Chromium OpenSSL via the custom RSA_METHOD. In addition, because
we cannot safely increase reference counts, retain references to Java wrappers
where appropriate in lieu of the C structures.

Also fix AndroidKeyStore unit tests on 4.1; they broke after an overzealous
NOTREACHED() in r278305.

BUG=389414

Review URL: https://codereview.chromium.org/365503007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282713 0039d316-1c4b-4281-b951-d872f2087c98

BUG=384640

Review URL: https://codereview.chromium.org/347183005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282712 0039d316-1c4b-4281-b951-d872f2087c98

The crash is caused when a guest is attached, a parameter "repaint" is passed in. For most device, their scale factor is 1, so we don't need to repaint and "repaint" is set to  false. However, for high dpi devices like Mac, the scale factor is 2, rather than 1. Therefore, a DECHECK of the scale factor == 1 is failed. 

So we use GetDeviceScaleFactor() to initialize last_device_scale_factor as well.

BUG=393278

Review URL: https://codereview.chromium.org/385173003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282711 0039d316-1c4b-4281-b951-d872f2087c98

If an autofill popup is supposed to display a multi-line address,
the address needs to be converted into a single-line format with
separators.

BUG=362373
R=estade@chromium.org
TBR=isherman@chromium.org

Review URL: https://codereview.chromium.org/264053007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282710 0039d316-1c4b-4281-b951-d872f2087c98

BUG=349436

Review URL: https://codereview.chromium.org/384893002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282709 0039d316-1c4b-4281-b951-d872f2087c98

As of current code, HiDPI should be enabled by default for both Desktop and Metro mode unless someone overrides with registry key.

BUG=393050
R=cpu,ananta

Review URL: https://codereview.chromium.org/385023002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282706 0039d316-1c4b-4281-b951-d872f2087c98

BUG=375333

Review URL: https://codereview.chromium.org/386933007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282705 0039d316-1c4b-4281-b951-d872f2087c98

NSS gets dlopen'd on Windows and Mac, so it can be built
in the OpenSSL port.

BUG=38883

Review URL: https://codereview.chromium.org/385873002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282704 0039d316-1c4b-4281-b951-d872f2087c98

Make SetFontRenderParams() pass the hinting setting through
to SkPaint instead of exposing a separate SetFontHinting()
method. Also add a default c'tor for FontRenderParams and
pass its autohinter setting through to SkPaint as well.

BUG=125235,375824

Review URL: https://codereview.chromium.org/387743002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282703 0039d316-1c4b-4281-b951-d872f2087c98

This broke LSAN.

http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Builder/builds/9737/steps/compile/logs/stdio


> Non-SFI NaCl: Fix browser_tests based on libc_free.c
> 
> There were two issues with clang:
> 
> - Clang emits .data.rel.ro.local for local struct values with
>   an initializer, which lets the linker to emit a few
>   relocation info.
> - In debug build, clang uses memcpy to copy a structure with
>   five members. Neither -fno-builtin nor -ffreestanding did
>   not prevent this issue.
> - In release build, clang translates for-loop based zero copy
>   to memset.
> 
> This patch initializes all structures without initializers
> or copy. This patch works with GYP_DEFINES=clang=0.
> 
> To make sure we will not add memcpy or something in future,
> we will build libc_free.nexe with -Wl,--no-undefined.
> 
> This also reverts
> 
> https://codereview.chromium.org/386543002
> 
> to enable the disabled tests.
> 
> BUG=392768
> TEST=./out/Debug/browser_tests --gtest_filter='*NonSfi*Messaging*'
> 
> Review URL: https://codereview.chromium.org/381883002

TBR=hamaji@chromium.org

Review URL: https://codereview.chromium.org/385073010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282702 0039d316-1c4b-4281-b951-d872f2087c98

BUG=383941

Review URL: https://codereview.chromium.org/382643002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282701 0039d316-1c4b-4281-b951-d872f2087c98

The non-testing code in src/mojo/public should be buildable without
depending on anything outside of that directory. It's fine for test
code to depend on base/ and gtest/, though. Due to the way GYP works,
having targets for both production and test code in the same .gypi
file winds up trying to resolve the dependencies for both kinds of
targets, breaking the ability to build the public mojo code
hermetically.

Splitting the targets out will enable consumers of the mojo code to
build the public code standalone using GYP.

BUG=chromium:388412
TEST=build and run all mojo unittests

Review URL: https://codereview.chromium.org/385603003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282699 0039d316-1c4b-4281-b951-d872f2087c98

BUG=392745
R=rsleevi@chromium.org

Review URL: https://codereview.chromium.org/385983006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282698 0039d316-1c4b-4281-b951-d872f2087c98

There were two issues with clang:

- Clang emits .data.rel.ro.local for local struct values with
  an initializer, which lets the linker to emit a few
  relocation info.
- In debug build, clang uses memcpy to copy a structure with
  five members. Neither -fno-builtin nor -ffreestanding did
  not prevent this issue.
- In release build, clang translates for-loop based zero copy
  to memset.

This patch initializes all structures without initializers
or copy. This patch works with GYP_DEFINES=clang=0.

To make sure we will not add memcpy or something in future,
we will build libc_free.nexe with -Wl,--no-undefined.

This also reverts

https://codereview.chromium.org/386543002

to enable the disabled tests.

BUG=392768
TEST=./out/Debug/browser_tests --gtest_filter='*NonSfi*Messaging*'

Review URL: https://codereview.chromium.org/381883002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282697 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282696 0039d316-1c4b-4281-b951-d872f2087c98

http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog_blink.html?url=/trunk&range=177948:177973&mode=html
TBR=junov@chromium.org,falken@chromium.org
BUG=

Review URL: https://codereview.chromium.org/387783003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282695 0039d316-1c4b-4281-b951-d872f2087c98

BUG=391943
TBR=nednguyen@chromium.org,nduca@chromium.org
NOTRY=True

Review URL: https://codereview.chromium.org/387093002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282693 0039d316-1c4b-4281-b951-d872f2087c98

Change PPB_ImageData_Proxy and PPB_Instance_Proxy to always initialize
'out' params, to avoid sending uninitialized memory from renderer to
plugin.

BUG=391023

Review URL: https://codereview.chromium.org/383143003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282692 0039d316-1c4b-4281-b951-d872f2087c98

It hangs on reliably audio_playback.html?id=mixed_audio on a Nexus 7 v2.

BUG=389362
NOTRY=True
TBR=shadi@chromium.org

Review URL: https://codereview.chromium.org/388883003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282691 0039d316-1c4b-4281-b951-d872f2087c98

Blink closes it when a page is disposed. However, Blink can do nothing
when the renderer process is terminated. Chromium needs to close it in
such case.

BUG=355564

Review URL: https://codereview.chromium.org/385043002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282690 0039d316-1c4b-4281-b951-d872f2087c98

SetViewportSize previously relied on UpdateInnerViewportContainerSize to
resize the inner viewport container layer; however, as of r253909, this
method is a no-op if the container layer has masks_to_bounds set (i.e.
non-Android platforms).

This fix explicitly resizes the inner viewport container layer to be the
same size as the newly resized device viewport before calling
UpdateInnerViewportContainerSize. This fixes the situation in the bug
where the viewport size is changed but nothing in the layer tree changes
so no new frame is committed.

BUG=387678

Review URL: https://codereview.chromium.org/381703002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282687 0039d316-1c4b-4281-b951-d872f2087c98

This is a preparation of crbug.com/387227.

BUG=387227
R=oshima@chromium.org
TEST=manually check for no regression

Review URL: https://codereview.chromium.org/388703002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282686 0039d316-1c4b-4281-b951-d872f2087c98

The problem was that the toolchain definitions had an is_clang block in them, but this is only evaluated in the context of the default toolchain (so false for and Android build). But then when we were re-evaluating the build config for the host compile, it was forcing clang to true which was causing the clang flags to be set.

This changes the way we default the is_clang value. Forcing it to true prevents it from being overridden manually. In some cases, like mac, this is pointless. In other cases like Linux, we may want to turn it on and off. Changing this definition allows us to toggle it on and off for different toolchains as we desire.

Ideally I think the way of defining toolchains would be different to make this a bit more flexible, I'll think about this. But this will work for now.

R=thakis@chromium.org

Review URL: https://codereview.chromium.org/385823002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282685 0039d316-1c4b-4281-b951-d872f2087c98

BUG=none

Review URL: https://codereview.chromium.org/388683002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282684 0039d316-1c4b-4281-b951-d872f2087c98

1. Defers command buffer destruction until after the native viewport client has had a chance to cleanup.
2. Destroy the compositor before quitting the view manager message loop.
3. Make various NodeObservers properly unhook themselves as node/view observers during destruction and view swapping.
4. Adds a "ViewManagerDisconnected" method to ViewManagerDelegate to allow the application to exit when the connection to the view manager is closed. This is necessary in addition to root monitoring because for some connections (e.g. the root node->window manager) the root node is not destroyed. I'm not entirely happy with imposing this burden on application developers but I will try and tidy this up later.
5. Instantiate nested AtExitManagers only in a static build. In the component build we can use the shell's.
6. Adds a callback to notify the ViewManagerInitServiceImpl when the NativeViewport is destroyed. The VMISI uses this callback to close its connection to the view manager, which triggers the view manager to shut down.

Since this still crashes, I can't test this yet.

R=sky@chromium.org
http://crbug.com/325901

Review URL: https://codereview.chromium.org/372273004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282683 0039d316-1c4b-4281-b951-d872f2087c98

Fix the black patch which would appear at the top of app windows in Chrome desktop windows with HiDPI

The bug occurs because the non client insets which account for the window frame are retrieved via the
GetSystemMetrics API which returns values in pixels.

Changes as below:-
1. Renamed the GlassAppWindowFrameViewWin::GetGlassInsets function to GetGlassInsetsInDIP.
2. Replaced calls to the GetSystemMetrics API in the GetGlassInsetsInDIP function with the dpi helper
   GetSystemMetricsInDIP.
3. Added code to convert the insets to pixels in the AppWindowDesktopWindowTreeHostWin::UpdateDWMFrame
   function.

BUG=392603
TBR=benwells

Review URL: https://codereview.chromium.org/382013004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282682 0039d316-1c4b-4281-b951-d872f2087c98

The switch will force use auxiliary bitmap path for all
rendering. This path is known to be slow.

This switch is temporary until Android L is released to AOSP.

BUG=

Review URL: https://codereview.chromium.org/388633003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282681 0039d316-1c4b-4281-b951-d872f2087c98

This can happen when an iframe is opened, then popualted via a document.write()
from its parent. This will cause the url to change to the parent's url, but the
browser process will not be notified of this url change. If the iframe then
attempts a fragment navigation, it looks like a cross-origin navigation from
about:blank.

BUG=390798
TEST=Added case to NavigationControllerTest.IsInPageNavigation

Review URL: https://codereview.chromium.org/372403002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282679 0039d316-1c4b-4281-b951-d872f2087c98

I meant to revert r281866, not this ...

> Revert 281692 "Mac: Enable delegated renderer by default"
> 
> This is causing tabs to get stuck white. Removing before today's
> Canary gets cut. The fix for the stuck white issue is being worked
> on at
> https://codereview.chromium.org/388793003/
> 
> > Mac: Enable delegated renderer by default
> > 
> > Delegated renderer disabled for layout tests (as on other platforms).
> > 
> > BUG=314190
> > TBR=piman
> > 
> > Review URL: https://codereview.chromium.org/374643002
> 
> TBR=ccameron@chromium.org
> 
> Review URL: https://codereview.chromium.org/386943008

TBR=ccameron@chromium.org

Review URL: https://codereview.chromium.org/388973002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282678 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=True

Review URL: https://codereview.chromium.org/389433002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282677 0039d316-1c4b-4281-b951-d872f2087c98

BUG=391943
TBR=nduca@chromium.org

Review URL: https://codereview.chromium.org/387003002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282676 0039d316-1c4b-4281-b951-d872f2087c98

This is causing tabs to get stuck white. Removing before today's
Canary gets cut. The fix for the stuck white issue is being worked
on at
https://codereview.chromium.org/388793003/

> Mac ÜC: Make resize smooth
> 
> Add helpers to RenderWidgetHelper to associate a NSView with
> a given RenderWidgetHelper (process and widget routing id), so
> that browser compositor swaps can be received while pausing
> waiting for correctly sized frames.
> 
> BUG=388005
> TBR=cevans
> 
> Review URL: https://codereview.chromium.org/370513002

TBR=ccameron@chromium.org

Review URL: https://codereview.chromium.org/387863004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@282675 0039d316-1c4b-4281-b951-d872f2087c98