GitLab

Previously LibjingleTransport was waiting for channel to be become
writable before considering it to be connected. As result it was loosing
all data that's received before the channel is marked as writable.

BUG=477142

Review URL: https://codereview.chromium.org/1080633005

Cr-Commit-Position: refs/heads/master@{#325347}

When we skip shared tiles in raster iterators, we need to ensure
that we consider resolution as a factor, since the twin will not be
returning tiles if they are of non-ideal resolution (so we shouldn't
be skipping the tile if the twin has higher, but non ideal, priority).

BUG=477136
R=sunnyps, vmiura

Review URL: https://codereview.chromium.org/1064743003

Cr-Commit-Position: refs/heads/master@{#325346}

This bug is introduced in r324679. When Initialize/ReleaseHardware was
removed, the call to the factory the kept a count on the number of
compositors with hardware acceleration was dropped. This means the
factory refuses to create a context for video.

Add those callbacks, and emulate the original call by checking if the
memory policy is zero.

Review URL: https://codereview.chromium.org/1073893004

Cr-Commit-Position: refs/heads/master@{#325345}

BUG=434572

Review URL: https://codereview.chromium.org/1072403004

Cr-Commit-Position: refs/heads/master@{#325344}

"Merge apps and tabs" feature promotion is moving to the new tab page
so it is no longer needed in first run.

BUG=467437

Review URL: https://codereview.chromium.org/1071303003

Cr-Commit-Position: refs/heads/master@{#325343}

BUG=NONE
TEST=Building html_viewer from scratch also builds the network service.
R=sky@chromium.org

Review URL: https://codereview.chromium.org/1090753002

Cr-Commit-Position: refs/heads/master@{#325342}

The Chrome-Proxy integration tests should not be dependent on the Chrome
404 error page. Use the new chrome.googlezip.net/chrome-proxy-header/c_*
urls, which have their own resources.

BUG=472298

Review URL: https://codereview.chromium.org/1088833006

Cr-Commit-Position: refs/heads/master@{#325341}

Summary of changes available at:
https://chromium.googlesource.com/chromiumos/platform/system_api/+log/b5515b1..beb5547

BUG=345072
R=rkc@chromium.org

Review URL: https://codereview.chromium.org/1073983009

Cr-Commit-Position: refs/heads/master@{#325340}

CrxInstaller currently just guarantees that its ExtensionInstallPrompt
|client_| data member is destroyed on the UI Thread. But this is not
enough -- CrxInstaller is the InstallPrompt's delegate via a weak
reference, so it can't be destroyed first.

This changes CrxInstaller from a RefCountedThreadSafe to a
RefCountedDeleteOnMessageLoop to ensure the whole thing is destroyed on
the UI thread when all references are gone.

Note that CrxInstaller also owns a ChromeRequirementsChecker which is a
WeakReferenceOwner. WeakReferenceOwners must be destroyed on the thread
that created them, so CrxInstaller being destroyed on the FILE thread
also hits a DCHECK at weak_ptr.cc:20.

Since SandboxedUnpackerClient is the base class that "is" the
RefCountedThreadSafe other classes derived from SandboxedUnpackerClient
are affected. To keep things simple and consistent, they all become
RefCountedDeleteOnMessageLoop.

BUG=410415
TEST=At the webstore: {show an install dialog, then cancel it}. Repeat
many times. Chrome should not crash. Previously a race condition could
be triggered causing Chrome to crash (maybe a 1-in-20 chance).

Review URL: https://codereview.chromium.org/1080453002

Cr-Commit-Position: refs/heads/master@{#325339}

When the requester of a host resolution doesn't specify IPv4-only
(i.e. when it's safe to do so) HostResolverImpl should be
successfully resolving IPv6 literals.

BUG=428280

Review URL: https://codereview.chromium.org/1088513004

Cr-Commit-Position: refs/heads/master@{#325338}

If this stat is strongly bimodal it will be used to distinguish warm from cold for various startup metrics.

BUG=476923
TBR=jeremy@chromium.org

Review URL: https://codereview.chromium.org/1084943002

Cr-Commit-Position: refs/heads/master@{#325337}

WebRTC 9005:9013
Changes: https://chromium.googlesource.com/external/webrtc/trunk/webrtc.git/+log/c49056d..c0075f6

Libjingle 9001:9009
Changes: https://chromium.googlesource.com/external/webrtc/trunk/talk.git/+log/1d4447b..e8df6da

TBR=niklase@chromium.org

Review URL: https://codereview.chromium.org/1089293003

Cr-Commit-Position: refs/heads/master@{#325336}

Revert of Split spoken feedback up on a per format rule basis. (patchset #20 id:440001 of https://codereview.chromium.org/1035983003/)

Reason for revert:
Looks like this is causing test timeouts on linux asan.  For details, see https://code.google.com/p/chromium/issues/detail?id=477492

Original issue's description:
> Split spoken feedback up on a per format rule basis.
>
> Committed: https://crrev.com/245c79e93e9c0ae0261e441537ecc35c78fcb134
> Cr-Commit-Position: refs/heads/master@{#325301}

TBR=plundblad@chromium.org,dmazzoni@chromium.org,dtseng@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1060913004

Cr-Commit-Position: refs/heads/master@{#325335}

The bundled gold contains a number of bug fixes (mostly LTO-related, but also
http://crbug.com/161942), so use it instead of the NDK gold. Both versions
of gold are based on the same version of binutils (2.24), so this should be
relatively safe.

BUG=161942,469376
R=thakis@chromium.org
TEST=Built libchromeshell.so with default GYP_DEFINES with and without this change, verified that binaries are equivalent

Review URL: https://codereview.chromium.org/1084133002

Cr-Commit-Position: refs/heads/master@{#325334}

hclam is not actively working on the project anymore

BUG=None

Review URL: https://codereview.chromium.org/1055933008

Cr-Commit-Position: refs/heads/master@{#325333}

BUG=

Review URL: https://codereview.chromium.org/1088293002

Cr-Commit-Position: refs/heads/master@{#325332}

We now render the Bing frontpage without any NOTIMPLEMENTED() spam.

BUG=none

Review URL: https://codereview.chromium.org/1088363002

Cr-Commit-Position: refs/heads/master@{#325331}

VC++ 2015 gives the following warning when compiling jsoncpp:

  json_value.cpp(480): warning C4463: overflow; assigning 1 to bit-field
  that can only hold values from -1 to 0

This warning is correct. A signed one-bit value can only hold 0 and -1.
This is true on gcc and clang, but only VC++ 2015 warns about it.

The code works because the stored value is never compared against 1 or
printed, but the current code is likely to cause surprises or bugs in the
future and should be fixed.

R=cpu@chromium.org
BUG=440500

Review URL: https://codereview.chromium.org/1060383002

Cr-Commit-Position: refs/heads/master@{#325330}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/ddb714b..221360a

CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
TBR=jcgregorio@google.com

Review URL: https://codereview.chromium.org/1084263002

Cr-Commit-Position: refs/heads/master@{#325329}

R=sheyang@chromium.org
BUG=475701

Review URL: https://codereview.chromium.org/1067633004

Cr-Commit-Position: refs/heads/master@{#325328}

The ability to show ConnectionStats is accidentally removed in a
previous revision (https://crrev.com/1050ee48b1d8fe353bfc7df3b0ea4e5300260f5c).

This CL adds it back with the following additional changes:
1. Instead of having a global remoting.stats, remoting.ConnectionStats is now
   owned by ApplicationContextMenu in app remoting and DesktopConnectedView in
   deskop remoting.
2. Adds the show statistics functionality to desktop remoting.

BUG=476132

Review URL: https://codereview.chromium.org/1089493002

Cr-Commit-Position: refs/heads/master@{#325327}

[ozone] Keep track of pending frames on drm_window when hardware display controller is disabled or when in headless mode.

When display controller is disabled, it used to stop tracking which buffer was the considered front. Surface would fast ack all incoming buffers without presenting them, which means that any one of them could be getting painted into while display controller is off. Upon wakeup, the controller would grab what it thought was the last front buffer, but in practice it was not because of the fast acks. Net result is that it may try to modeset with the buffer GL is painting into, making it hang on a fence in the driver.

This patch removes all knowledge of front buffers from the display controller and explicitly sets it on the window, which tracks it while the controller is off.

BUG=473831

Review URL: https://codereview.chromium.org/1078183003

Cr-Commit-Position: refs/heads/master@{#325326}

It appears that at least in some cases, the linker uses a stricter set of
rules when linking LTO objects, which causes the link to fail in some cases
where an unused function depends on an undefined symbol. This change fixes
all such cases where the error caused the LTO build to fail, by either
moving source files between lists, or by selectively disabling parts of
source files depending on the configuration.

There are more errors like this, which for some reason do not affect LTO. They
can be inspected like this (with a regular Chromium/android build):

$ ninja -C out/Release chrome_shell_apk
$ nm out/Release/lib/libchromeshell.so | grep " U _Z"

BUG=407544
R=thakis@chromium.org

Review URL: https://codereview.chromium.org/1086973003

Cr-Commit-Position: refs/heads/master@{#325325}

This CL adds a new page set ProfileSafeUrlsPageSet, which is used by the
CookieProfileExtender. This CL also hooks up WPR to the CookieProfileExtender.

BUG=472745

Review URL: https://codereview.chromium.org/1070053004

Cr-Commit-Position: refs/heads/master@{#325324}

BUG=

Review URL: https://codereview.chromium.org/1086223002

Cr-Commit-Position: refs/heads/master@{#325323}

Chromium builds its own mojo_shell from forked source.

This logic was originally changed incorrectly:
  https://codereview.chromium.org/1049993002/diff/170001/third_party/mojo/src/mojo/public/mojo.gni
That was clobbered in the latest Mojo->Chromium roll:
  https://codereview.chromium.org/1082343002/diff/40001/third_party/mojo/src/mojo/public/mojo.gni

BUG=NONE
TEST=mojo_shell binary built from chromium repo.
R=jam@chromium.org
TBR=scottmg@chromium.org

Review URL: https://codereview.chromium.org/1087333002

Cr-Commit-Position: refs/heads/master@{#325322}

If there is a slow pending load for a page, and in the meantime the page uses replaceState, the existence of the pending entry could cause a misclassification of the replaceState as SAME_PAGE.

Ensure that SAME_PAGE classification happens only if the pages bear some relationship.

BUG=369661
TEST=Existing tests, NavigationTypeClassification_InPageWhilePending.

Review URL: https://codereview.chromium.org/1082083002

Cr-Commit-Position: refs/heads/master@{#325321}

rtc::P2PTransportChannel expects SetRemoteIceCredentials() to be called
to set remove ICE candidate, but LibjingleTransport was never calling
it, and instead it was passing ICE credentials as part of remote
candidates. Fixed LibjingleTransport to call SetRemoteIceCredentials()
explicitly whenever it receives a new candidate.

BUG=476775

Review URL: https://codereview.chromium.org/1070233004

Cr-Commit-Position: refs/heads/master@{#325320}

ajwong is not actively working on the project anymore.

BUG=None

Review URL: https://codereview.chromium.org/1088743004

Cr-Commit-Position: refs/heads/master@{#325319}

Property trees use values computed by PreCalculateMetaInformation,
so it's unfair to count this time against CalcDrawProps in the
draw properties benchmark.

Review URL: https://codereview.chromium.org/1082163003

Cr-Commit-Position: refs/heads/master@{#325318}

The dashboard is currently confused about which direction indicates an
improvement. This CL makes that explicit.

BUG=470272

Review URL: https://codereview.chromium.org/1057823005

Cr-Commit-Position: refs/heads/master@{#325317}

BUG=

Review URL: https://codereview.chromium.org/1089133002

Cr-Commit-Position: refs/heads/master@{#325316}

BUG=477436

Review URL: https://codereview.chromium.org/1090703002

Cr-Commit-Position: refs/heads/master@{#325315}

BUG=429053
TEST=gpu_unittests
R=piman@chromium.org

Review URL: https://codereview.chromium.org/1067733003

Cr-Commit-Position: refs/heads/master@{#325314}

For the full list of changes, see
https://github.com/chromium/web-page-replay/compare/94d7e2a...4e4d540.

WebPageReplay moved the method has_sni() from certutils to platformsettings,
and renamed it to HasSniSupport(). This CL fixes the one location where
has_sni() is used, and also fixes the stubs.

BUG=477130

Review URL: https://codereview.chromium.org/1087093002

Cr-Commit-Position: refs/heads/master@{#325313}

Since we do not know the future behavior of browser CDMs, and have no way to block persistent state or distinctive identifiers, assume that such CDMs always provide such access.

Review URL: https://codereview.chromium.org/1086013002

Cr-Commit-Position: refs/heads/master@{#325312}

The large profiles are typically too new to be used with the reference builds.

BUG=476882

Review URL: https://codereview.chromium.org/1090453002

Cr-Commit-Position: refs/heads/master@{#325311}

Remove obsolete private Pepper interface private/ppb_talk_private.h,
associated Pepper plumbing code, strings, and chrome::PepperTalkHost.

These are no longer needed as support for Pepper GTalk has been
dropped.

BUG=392002
TEST=browser_tests

Review URL: https://codereview.chromium.org/1088183003

Cr-Commit-Position: refs/heads/master@{#325310}

BUG=477154

Review URL: https://codereview.chromium.org/1055933006

Cr-Commit-Position: refs/heads/master@{#325309}

BUG=462581

Review URL: https://codereview.chromium.org/1072533003

Cr-Commit-Position: refs/heads/master@{#325308}