GitLab

Previously, the script killed only the Xvfb process, relying on the
broken X connection to terminate other processes. This CL ensures that
all managed processes are explicitly killed.

BUG=388957
NOTRY=true

Review URL: https://codereview.chromium.org/635953002

Cr-Commit-Position: refs/heads/master@{#299206}

For certain unittest failures, these long timeouts were causing cycle
time problems for the bots. This CL mitigates those delays.

BUG=418577

Review URL: https://codereview.chromium.org/646063002

Cr-Commit-Position: refs/heads/master@{#299205}

Send the CGL renderer ID for the context being used by the
CAOpenGLLayer.

Tested with WebGL apps, gfxCardStatus switching.

BUG=312462

Review URL: https://codereview.chromium.org/647053002

Cr-Commit-Position: refs/heads/master@{#299204}

IME support in Chrome OS mode/Windows 8 mode is currently implemented in two parts.
1. Metro driver (Viewer process):
   Here we rely on the Text services framework for IME notifications and relay those
   to the browser process.

2. Chrome Browser:
   The browser initializes the input method via the RemoteInputMethodWin class which initializes
   the input method if the viewer is an immersive process.

Fixes as below:-
1. The metro driver initializes the text services framework by instantiating the CLSID_TF_ThreadMgr
   COM object and requests the ITfThreadMgr2 interface.
   This interface is not implemented for Windows 7. Fix is to use the ITfThreadMgr interface instead
   for Windows 7 and up. This provides all the functionality we need.

2. The metro driver was instantiating a MTA COM apartment. The text services COM objects expect to be
   instantiated in a STA.

3. The AppListService object on Windows attempts to bring up the app list bubble in ASH mode even if we are in desktop
mode. Fix is to avoid that.

4. The IsRemoteInputMethodWinRequired function initialized the remote input method if the viewer was an immersive process.
To ensure that it also works on Windows 7 we check whether the browser process is launched with the kViewerConnect switch.

BUG=421980

Review URL: https://codereview.chromium.org/645683002

Cr-Commit-Position: refs/heads/master@{#299203}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/7bb9ed7..93bee71

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=reed@google.com

Review URL: https://codereview.chromium.org/644113002

Cr-Commit-Position: refs/heads/master@{#299202}

The previous logic doesn't account for divide truncation and
border texels, resulting in an additional visible tile with
only a couple of rows visible. This is hidden by the URL bar
sometimes, but it still affects full-screen, web-apps and
desktop browsers.

BUG=365877

Review URL: https://codereview.chromium.org/626113004

Cr-Commit-Position: refs/heads/master@{#299201}

This change pulls in the following from BoringSSL:

7ea8481 Add generic OPERATION_NOT_SUPPORTED error code.
bc8d737 Add tests for AlgorithmIdentifier support in EVP.
8f160a6 Add EVP_DigestVerifyInitFromAlgorithm and EVP_DigestSignAlgorithm.
2e1594d Remove #if-0'd ASN1_sign.
75a1dd0 Remove EVP_PKEY_CTX case in rsa_pss_to_ctx.
f4b4952 Fix rsa_pss_to_ctx return value.
5db5d4e RSA_encrypt and RSA_sign_raw aren't the same.
a87941a Remove misspelled error code.
b1659f5 Fix err_test.
57e52f2 Fix invariant broken in 5f1374e2.
5f1374e Retain ownership of malloced error data.
622a6db Fix SHA*_Final documentation.
83e52bc HASH_UPDATE always succeeds.
bce495c Fix a_{digest,verify}.c error codepaths.

BUG=none

Review URL: https://codereview.chromium.org/649503002

Cr-Commit-Position: refs/heads/master@{#299200}

BUG=419876

Review URL: https://codereview.chromium.org/649773002

Cr-Commit-Position: refs/heads/master@{#299199}

Mechanical change that adds instrumentation required to locate the source of jankiness (i.e. a long-running fragment of code executed as a part of the task that causes jank) in the code. See the bug for details on what kind of jank we are after.
A number of similar CLs were landed, and none of them caused issues. The code of the instrumentation is highly optimized and is not expected to affect performance. The code simply creates a diagnostic task which is identical to ones created by PostTask or IPC message handlers.
Landing as TBR since this is a mechanical, safe and temporary change.

BUG=422460
TBR=isherman@chromium.org, caitkp@chromium.org

Review URL: https://codereview.chromium.org/647123002

Cr-Commit-Position: refs/heads/master@{#299198}

This is the final step towards the goals listed in crbug.com/419160. It builds on https://codereview.chromium.org/628763002. The trivial NativeViewport application listed there can now be written relatively simply:

  define("test", [
    "mojo/services/public/interfaces/geometry/geometry.mojom",
    "mojo/services/public/interfaces/native_viewport/native_viewport.mojom",
    "mojo/apps/js/mojo",
    "console",
  ], function(geometry, viewport, mojo, console) {

    var client = {
      onDestroyed: mojo.quit,
      onEvent: function(event) {
        console.log("event.type=" + event.action);
        return Promise.resolve(); // This just gates the next event delivery
      },
    };
    var viewport = mojo.connectToService(
      "mojo:mojo_native_viewport_service", viewport.NativeViewport, client);
    viewport.create(new geometry.Size({width: 256, height: 256}));
    viewport.show();
  });

The mojo connectToService() function now just returns a proxy to the remote interface. The 3rd (optional) connectToService() parameter is an object that implements the client interface for the remote interface.

In this version the NativeViewportClient interface implementation is just an object with function properties whose names match the interface. Subclassing isn't needed and it's not necessary to implement all of the client functions.

The Mojo JS bindings now generate a subclass of each FooStub class called DelegatingFooStub. The Delegating version just forwards the Stub methods to the value of |this.delegate$|. The ugly "$" suffix avoids collisions with mojom function names. The JS bindings' interface Foo object now includes a delegatingStubClass property whose value is FooDelegatingStub.

|mojo.connectToService(url, Foo, client)| only uses Foo.client.delegatingStubClass if a client is specified.

The mojo_module has been split into two: a native part, called MojoInternalsModule, which integrates with JSApp, and a JavaScript part that depends on the internals. Apps must now import "mojo/apps/js/mojo", rather than just "mojo". For now, quit and connectToService() are the only functions in the JS mojo module.

BUG=419160

Review URL: https://codereview.chromium.org/637373002

Cr-Commit-Position: refs/heads/master@{#299197}

Depends on https://codereview.chromium.org/642203002.

BUG=346764
TEST=Input events work in OOPIFs with --site-per-process.

Review URL: https://codereview.chromium.org/647653002

Cr-Commit-Position: refs/heads/master@{#299196}

This allows us to query version information (e.g. the build version,
whether it's an offical build) whether or not the native library is
loaded.

Future work: put channel information in this file too, and delete the
ChromeVersionInfo.init() method.

TBR=mmenke@chromium.org

Review URL: https://codereview.chromium.org/623403002

Cr-Commit-Position: refs/heads/master@{#299195}

TESTED=Build valgrind on Linux/Mac, tests pass 100/100 times.
BUG=159234,227278,222532

Review URL: https://codereview.chromium.org/647103002

Cr-Commit-Position: refs/heads/master@{#299194}

BUG=413855

Review URL: https://codereview.chromium.org/639183003

Cr-Commit-Position: refs/heads/master@{#299193}

This moves the ExtensionCache interface out to extensions and
delegates instance provisioning out to ExtensionsBrowserClient
implementations.

BUG=398671

Review URL: https://codereview.chromium.org/612423003

Cr-Commit-Position: refs/heads/master@{#299192}

SigninManager.Observer was renamed to SignInFlowObserver, which should now
be used instead.

BUG=421058

Review URL: https://codereview.chromium.org/648843002

Cr-Commit-Position: refs/heads/master@{#299191}

BUG=422386
NOTRY=true

Review URL: https://codereview.chromium.org/648783002

Cr-Commit-Position: refs/heads/master@{#299190}

R=tonyg@chromium.org,qyearsley@chromium.org,nednguyen@chromium.org
BUG=422093

Review URL: https://codereview.chromium.org/645753002

Cr-Commit-Position: refs/heads/master@{#299189}

BUG=None

Review URL: https://codereview.chromium.org/649793002

Cr-Commit-Position: refs/heads/master@{#299188}

BUG=None
TBR=oshima@chromium.org
TEST=None

Review URL: https://codereview.chromium.org/649803002

Cr-Commit-Position: refs/heads/master@{#299187}

BUG=380875, 421444
R=oshima@chromium.org, xiyuan@chromium.org
TBR=pkasting@chromium.org, jar@chromium.org
TEST=manually

Review URL: https://codereview.chromium.org/640103002

Cr-Commit-Position: refs/heads/master@{#299186}

BUG=421208
R=zea@chromium.org

Review URL: https://codereview.chromium.org/635073002

Cr-Commit-Position: refs/heads/master@{#299185}

Summary of changes available at:
https://chromium.googlesource.com/chromium/blink/+log/853a4b9..6f078f3f

TBR=eseidel@chromium.org,caseq@chromium.org,apavlov@chromium.org

Review URL: https://codereview.chromium.org/648853002

Cr-Commit-Position: refs/heads/master@{#299184}

BUG=None
R=rockot@chromium.org
TEST=None

Review URL: https://codereview.chromium.org/642353003

Cr-Commit-Position: refs/heads/master@{#299183}

This reverts commit 6b730574, except for the
AUTHORS change from that commit.

BUG=398345,420226
TBR=garykac

Review URL: https://codereview.chromium.org/643243002

Cr-Commit-Position: refs/heads/master@{#299182}

Texture::Update() was doing a lot of calculations that did not change
much per frame. Some of the calculations such as whether or not has
non-power of 2 dimensions has been moved to when the texture
levels are set since they do not change. Others are only recalculated
when a dirty bit is set.

BUG=420152
TEST=trybots

Review URL: https://codereview.chromium.org/633773002

Cr-Commit-Position: refs/heads/master@{#299181}

BUG=401311,421705
TEST=gpu bots
NOTRY=true
R=kbr@chromium.org

Review URL: https://codereview.chromium.org/641173002

Cr-Commit-Position: refs/heads/master@{#299180}

NOTRY=true
TBR=benwells@chromium.org, halcanary@chromium.org
BUG=422485

Review URL: https://codereview.chromium.org/642353004

Cr-Commit-Position: refs/heads/master@{#299179}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/71856d5..7bb9ed7

CQ_EXTRA_TRYBOTS=tryserver.blink:linux_blink_rel,linux_blink_dbg
TBR=reed@google.com

Review URL: https://codereview.chromium.org/647073002

Cr-Commit-Position: refs/heads/master@{#299178}

We can enable those tests now that crbug.com/419020 is fixed.

BUG=352293

Committed: https://crrev.com/9b10e4b077ee7106ab57d4728c5199a03638701a
Cr-Commit-Position: refs/heads/master@{#298062}

Review URL: https://codereview.chromium.org/623193002

Cr-Commit-Position: refs/heads/master@{#299177}

See the bug.

I was told that one of the pieces of code that could cause jank is the callback invoked in IO thread once the read on a socket is complete. I'm not familiar with the code, and I instrumented places that look relevant; feel free to suggest adding or removing the instrumentation.

BUG=418183
TBR=asanka@chromium.org

Review URL: https://codereview.chromium.org/637733002

Cr-Commit-Position: refs/heads/master@{#299176}

NOTRY=true
TBR=benwells@chromium.org, burnik@chromium.org
BUG=422472

Review URL: https://codereview.chromium.org/647153002

Cr-Commit-Position: refs/heads/master@{#299175}

NOTRY=true
TBR=benwells@chromium.org, burnik@chromium.org
BUG=422472

Review URL: https://codereview.chromium.org/643443005

Cr-Commit-Position: refs/heads/master@{#299174}

The current implementation is not used in production code
and there is no reasonable common denominator anymore between
the different implementations.

Also removing the compositor example app (which was using
cc::OutputSurface) since James says it's not incredibly useful anymore.

Review URL: https://codereview.chromium.org/640023004

Cr-Commit-Position: refs/heads/master@{#299173}

This class is the single entry point for all Geofencing related API calls, and will be responsible for storing all geofence registrations, and registering them with the underlying platform specific GeofencingProvider.

BUG=383125

Review URL: https://codereview.chromium.org/586163003

Cr-Commit-Position: refs/heads/master@{#299172}

Plumb LTHI::OnRootLayerDelegatedScrollOffsetChanged to the affected
layers and properly set damage on those layers as if they have been
scrolled.

BUG=422061

Review URL: https://codereview.chromium.org/642113004

Cr-Commit-Position: refs/heads/master@{#299171}

For supporting sending browser-hosted resources (like FileSystem and FileRef) in postMessageAndAwaitResponse, the ResourceConverter is going to need to save some state across conversions. Therefore, MessageChannel will have to use one V8VarConverter instead of using it entirely on the stack. This also means PepperTryCatch* needs to be passed a V8VarConverter instead of making its own.

BUG=417316

Review URL: https://codereview.chromium.org/635593004

Cr-Commit-Position: refs/heads/master@{#299170}

This message fills up about:gpu and stdout, and doesn't provide any
value.

BUG=

Review URL: https://codereview.chromium.org/645143002

Cr-Commit-Position: refs/heads/master@{#299169}

The original attempt to land the CL incorrectly interacted with another
recently landed CL that added the tabStripBackgroundView
(https://codereview.chromium.org/611453004). This CL was causing that view to
become layer backed, which caused it to draw on top of non-layer backed window
content, including the window controls.

This reland reworks the logic so that the tabStripBackgroundView is added to
the root view and does not become layer backed. This reland also updates
browser_tests to catch similar such errors in the future.

--------------Original CL Description---------------
Make the window's contentView have the same size as the window itself. Adjust
the view hierarchy so that Chrome does not add subviews directly to the root
view.

By default, the contentView does not occupy the full size of a framed window.
Chrome still wants to draw in the title bar. Historically, Chrome has done this
by adding subviews directly to the root view. This causes several problems. The
most egregious is related to layer ordering when the root view does not have a
layer. By making the window's contentView full-sized, there is no longer any
need to add subviews to the root view.

I deleted the tests in presentation_mode_controller_unittest.mm since they were
layout tests for the browser_window_controller, but the tests in
browser_window_layout_unittest.mm are both more robust and more comprehensive.

I fixed a bug in moveViewsForImmersiveFullscreen:... where the tabStripView was
being added to the source window rather than the destination window.

This CL is mostly a reland of https://codereview.chromium.org/399803002/.

Original CL Committed: https://crrev.com/746dbb9cfefaac243704956db431ff9bb9b0485b
Original CL Cr-Commit-Position: refs/heads/master@{#298584}

BUG=417923

Review URL: https://codereview.chromium.org/646703002

Cr-Commit-Position: refs/heads/master@{#299168}

BUG=422450
TBR=govind@chromium.org

Review URL: https://codereview.chromium.org/641183007

Cr-Commit-Position: refs/heads/master@{#299167}