GitLab

This patch pulls in recent UI fixes and changes for memory-infra.

Summary of changes available at:
https://chromium.googlesource.com/external/trace-viewer/+log/eea44df..e2b5374

e2b5374 Handle rows with undefined values in memory dump panes properly
3d85362 Change the main allocator metric from inner_size to outer_size
8b89552 Show generic attributes in memory dump UI
df1e6c7 generic_object_view table
1eff1ad Fix table builder column setter
210d769 Add support for string attributes in memory allocator dumps
86ab68f Adjust CSS to center 'Monitoring' label
2ce2802 Fix memory dump event timestamps
db25ed4 Handle allocators with undefined values in memory dump panes
c39a009 Add precise finding code for slices that obeys nesting rules

Size increase for resources.pak:
3566 bytes (16958824 -> 16962390)

BUG=458295

Review URL: https://codereview.chromium.org/1117643002

Cr-Commit-Position: refs/heads/master@{#327480}

MessageLoopProxy is deprecated.
This basically does a search and replace of:

MessageLoopProxy::current() -> ThreadTaskRunnerHandle::Get().

MessageLoopProxy -> SingleThreadTaskRunner

BUG=391045

Review URL: https://codereview.chromium.org/1117583003

Cr-Commit-Position: refs/heads/master@{#327479}

Reason: the GYP includes in src/third_party/junit/junit.gyp
quickly goes into Android-only terroritory so reusing the target
on non-Android would require a lot more work.

This reverts the DEPS change in commit
0cb25491.
The checklicenses.py change is left in, since it makes sense
to have it even if the code is only checked out for Android.

BUG=webrtc:4499
TBR=jbudorick@chromium.org, phajdan.jr@chromium.org

Review URL: https://codereview.chromium.org/1115643002

Cr-Commit-Position: refs/heads/master@{#327478}

BUG=466848

Review URL: https://codereview.chromium.org/1102353003

Cr-Commit-Position: refs/heads/master@{#327477}

The FrameTreeNode class has a browser-side and globally unique id. we need id with less bits, for example in extension APIs code exporting those ids to JavaScript.

BUG=477770

TBR=sky

Review URL: https://codereview.chromium.org/1093923003

Cr-Commit-Position: refs/heads/master@{#327476}

Review URL: https://codereview.chromium.org/1091353004

Cr-Commit-Position: refs/heads/master@{#327475}

Currently when a window that is not main (e.g. the app list) completes a
fade-out animation and closes, it will trigger a mainMenu update. We
need to ensure the mainMenu says "Chrome" when the last window on the
active space is closed. But if that didn't have main status, it won't
have been influencing the mainMenu to begin with and also won't trigger
a later NSWindowDidBecomeMainNotification.

This regressed in r323157, but the bug was always present. It was
previously avoided because updates were skipped by anticipating whether
AppKit would make a subsequent window main. But that's hard to get right
and conflicts with behaviour used when closing panel windows.

BUG=481803
TEST=On Mac, open a packaged app window with the app launcher. When
launched for the first time, the menubar should say the apps name,
rather than "Chrome".

Review URL: https://codereview.chromium.org/1114803002

Cr-Commit-Position: refs/heads/master@{#327474}

Prefetching the native library by reading its pages speeds up startup on
Android (see the linked bug).

When loading the native library from the APK, there is no file to
associate with the native library. The most reliable way to prefetch the
native library pages is then to look at the pages mapped in the process
address space after the native library has been loaded.

To minimize disruption, the prefetch is done in a separate forked
process. This patch implements the parsing of the address space mappings
and the reading of pages from a forked process. Most of the work is done
before the fork() to avoid potential nasty issues associated with doing
work in a forked process in presence of threads.

BUG=460438

Review URL: https://codereview.chromium.org/1001343002

Cr-Commit-Position: refs/heads/master@{#327473}

Missing return statement in https://codereview.chromium.org/1102093002/.

BUG=479673

Review URL: https://codereview.chromium.org/1114843002

Cr-Commit-Position: refs/heads/master@{#327472}

A MemoryWedge is a single chunk of memory allocated. One passes it a size and the wedge is added to the resident memory used by the current app.

Furthermore, it adds an experimental flag for iOS based on a variation param, to drive the size of the wedge that could be inserted for a given run.

BUG=TBD
R=stuartmorgan,asvitkine

Review URL: https://codereview.chromium.org/1107963002

Cr-Commit-Position: refs/heads/master@{#327471}

Replace MessageLoopProxy usage with ThreadTaskRunnerHandle in icon_loader files of chrome/browser module.

MessageLoopProxy is deprecated.
This basically does a search and replace of:

MessageLoopProxy::current() -> ThreadTaskRunnerHandle::Get().

MessageLoopProxy -> SingleThreadTaskRunner

BUG=391045

Review URL: https://codereview.chromium.org/1110313002

Cr-Commit-Position: refs/heads/master@{#327470}

Revert of [Android] Remove more uses of android_commands from build/android/pylib. (patchset #3 id:40001 of https://codereview.chromium.org/1105323002/)

Reason for revert:
Causes HostDriven_SigninTest.testManagedSignin to fail, and hence prevents rolls into Android downstream repository

BUG=482406

Original issue's description:
> [Android] Remove more uses of android_commands from build/android/pylib.
>
> BUG=267773
>
> Committed: https://crrev.com/166b709fc6c8fefad72520473850ce15ffe45898
> Cr-Commit-Position: refs/heads/master@{#327402}

TBR=perezju@chromium.org,mikecase@chromium.org,rnephew@chromium.org,jbudorick@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=267773

Review URL: https://codereview.chromium.org/1116493003

Cr-Commit-Position: refs/heads/master@{#327469}

rtc::Buffer::data() will soon start defaulting to uint8_t* for its
return type, so we have to start requesting char* explicitly.

Review URL: https://codereview.chromium.org/1097043002

Cr-Commit-Position: refs/heads/master@{#327468}

BUG=

Review URL: https://codereview.chromium.org/1058303004

Cr-Commit-Position: refs/heads/master@{#327467}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/2a2fc95c..d5ebf111

Please follow these instructions for assigning/CC'ing issues:
https://code.google.com/p/v8-wiki/wiki/TriagingIssues

TBR=hablich@chromium.org,machenbach@chromium.org,yangguo@chromium.org,vogelheim@chromium.org

Review URL: https://codereview.chromium.org/1114853002

Cr-Commit-Position: refs/heads/master@{#327466}

Please take look at
https://codereview.chromium.org/1090213002/patch/40001/50001

for details
BUG=

Review URL: https://codereview.chromium.org/1104023002

Cr-Commit-Position: refs/heads/master@{#327465}

BUG=466848

Review URL: https://codereview.chromium.org/1110333002

Cr-Commit-Position: refs/heads/master@{#327464}

This patch modifies a number of tests to have a more realistic threading
set-up by either:

a) adding a TestBrowserThreadBundle
b) constructing any existing TestBrowserThreadBundle earlier or
c) adding a MessageLoop as appropriate.

This makes it possible to later migrate base/ from MessageLoopProxy
over to ThreadTaskRunnerHandle, which requires an active task runner on
the current thread.

BUG=465354, 481632
TBR=tommi@chromium.org

Review URL: https://codereview.chromium.org/1105373002

Cr-Commit-Position: refs/heads/master@{#327463}

Review URL: https://codereview.chromium.org/1110133002

Cr-Commit-Position: refs/heads/master@{#327462}

Call EnsureBrowserStateKeyedServiceFactoriesBuilt() during initialisation of
"chrome_ios_unittests" to ensure the BSKSF are correctly registered against
the DependencyManager before launching the tests.

Call ios::RegisterPathProvider() before running the tests so that they can
use the iOS specific paths by name.

BUG=468694
R=droger@chromium.org

Review URL: https://codereview.chromium.org/1111193002

Cr-Commit-Position: refs/heads/master@{#327461}

BUG=475296

Review URL: https://codereview.chromium.org/1110363002

Cr-Commit-Position: refs/heads/master@{#327460}

BUG=None

Review URL: https://codereview.chromium.org/1114563004

Cr-Commit-Position: refs/heads/master@{#327459}

* Replace ctors of PolicyOAuth2TokenFetcher and Start(), with StartWithSigninContext, and StartWithAuthCode.
* Replace WildcardLoginChecker::Start with WildcardLoginChecker::StartWithSigninContext for consistency.

BUG=470984
TEST=compiles

Review URL: https://codereview.chromium.org/1109133004

Cr-Commit-Position: refs/heads/master@{#327458}

It includes changes:
    ee8ce3d  Update mips64el-linux-android-4.9 built with pre-release binutils-2.25
    a6ebdc1  [MIPS,MIPS64] Add asm/reg.h and mcontext_t struct for MIPS64

Review URL: https://codereview.chromium.org/1111813002

Cr-Commit-Position: refs/heads/master@{#327457}

BUG=412250

Review URL: https://codereview.chromium.org/1105393003

Cr-Commit-Position: refs/heads/master@{#327456}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/901c4c6d..2a2fc95c

Please follow these instructions for assigning/CC'ing issues:
https://code.google.com/p/v8-wiki/wiki/TriagingIssues

TBR=hablich@chromium.org,machenbach@chromium.org,yangguo@chromium.org,vogelheim@chromium.org

Review URL: https://codereview.chromium.org/1116613002

Cr-Commit-Position: refs/heads/master@{#327455}

Build (but don't run) new test targets on the following bots:
  chromium.linux "Linux GN"
  chromium.linux "Linux GN (dbg)"
  tryserver.chromium.linux "linux_chromium_gn_chromeos_dbg"
  tryserver.chromium.linux "linux_chromium_gn_chromeos_rel"
  tryserver.chromium.linux "linux_chromium_gn_dbg"
  tryserver.chromium.linux "linux_chromium_gn_rel"

Add targets to each supported bot:
  html_viewer_unittests (release only, no debug component builds)
  mojo_runner_unittests (release only, no debug component builds)
  mojo_shell_unittests
  mojo_surfaces_lib_unittests
  view_manager_service_unittests (release only, no debug component builds)

These match the Chromium Mojo FYI waterfall unittest targets:
https://code.google.com/p/chromium/codesearch#chromium/build/scripts/slave/recipes/chromium_mojo.py

BUG=478244
TEST=CQ and main waterfall GN bots build new test targets.
R=sky@chromium.org,dpranke@chromium.org

Review URL: https://codereview.chromium.org/1110193004

Cr-Commit-Position: refs/heads/master@{#327454}

Summary of changes available at:
https://chromium.googlesource.com/skia/+log/5b6db07..280a9c8

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

Review URL: https://codereview.chromium.org/1116603002

Cr-Commit-Position: refs/heads/master@{#327453}

Bescause base/strings/string_split.h defines:
typedef std::vector<std::pair<std::string, std::string> > StringPairs;

BUG=412250

Review URL: https://codereview.chromium.org/1109723002

Cr-Commit-Position: refs/heads/master@{#327452}

BUG=384104

Review URL: https://codereview.chromium.org/1088343002

Cr-Commit-Position: refs/heads/master@{#327451}

Similar to NullAudioSink, but I decided not to use FakeAudioWorker
as a generic fake worker since it adds more complexity then we need
for the video renderer sink interface.  I also didn't want to add
pause support or stop callback support to it.

BUG=473424
TEST=new unittest, forthcoming pipeline changes.

Review URL: https://codereview.chromium.org/1116473002

Cr-Commit-Position: refs/heads/master@{#327450}

Had to use WRAPPED_INSTANTIATE_TEST_CASE_P instead to disable this test.

BUG=482121
TBR=yoshiki@chromium.org

Review URL: https://codereview.chromium.org/1117613002

Cr-Commit-Position: refs/heads/master@{#327449}

In Share Dialogue layout xml, the TextView widget is not independent of
the device scale factor i.e. in case of multi-line text for larger font size
preference the text gets truncated. Added ellipsize support to this widget.

BUG=410205

Review URL: https://codereview.chromium.org/1097393005

Cr-Commit-Position: refs/heads/master@{#327448}

BUG=481413

Review URL: https://codereview.chromium.org/1099403008

Cr-Commit-Position: refs/heads/master@{#327447}

This adds histograms that records how often we write to the prefs file. The number of writes is recorded for every 5 minute interval of time spent in Chrome.

To avoid doing any work when Chrome is idle and not recording any writes, we only record metrics at the point when a write occurs. Whenever a write occurs we look to see if the last write occurred more than 5mins ago. If it did, we report the write count for the previous period of time.

The histogram is called:
Settings.JsonDataWriteCount.<pref filename>
e.g.
Settings.JsonDataWriteCount.Local_State

This is in a similar vein to Settings.JsonDataReadSizeKilobytes.*

BUG=476800

Review URL: https://codereview.chromium.org/1083603002

Cr-Commit-Position: refs/heads/master@{#327446}

BUG=171301

Review URL: https://codereview.chromium.org/1111113002

Cr-Commit-Position: refs/heads/master@{#327445}

BUG=466848

Review URL: https://codereview.chromium.org/1061503008

Cr-Commit-Position: refs/heads/master@{#327444}

BUG=466848

Review URL: https://codereview.chromium.org/1106593002

Cr-Commit-Position: refs/heads/master@{#327443}

Cr-Commit-Position: refs/heads/master@{#327442}

This upstreams the bulk of the code necessary to support WebUI in the
iOS web layer. Much of this is forked off of the content/ versions,
which can't be used on iOS.

BUG=464810

Review URL: https://codereview.chromium.org/1110213002

Cr-Commit-Position: refs/heads/master@{#327441}