GitLab

On ChromeOS, kernel notifies Chrome when system gets into low memory
condition. However the low memory calculation misses a parameter
|min_free_kbytes| so the notification comes too late. A kernel fix is
coming to fix the issue.

On Chrome side, TabManager tries to recover memory to normal level. It
mirrors the low memory calculation in kernel to know how many tabs to
kill. So the missing parameter |min_free_kbytes| should be added back.

In the future, try to consolidate the logic in kernel so TabManager
doesn't need to do its own calculation.

BUG=chromium:645512,chrome-os-partner:60498
TEST=cave,peppy,etc.

Review-Url: https://codereview.chromium.org/2618483002
Cr-Commit-Position: refs/heads/master@{#442018}

* Download notifications are grouped together, and Incognito
  is put into its own group.  Summary notifications will come
  in a later patch; Android N doesn't seem to bundle these
  together once they're in their own group, making it hard
  to debug without Android Wear.

* New and existing group name constants are moved into
  NotificationConstants.java.

* Cleans up DownloadNotificationService's constant naming
  schemes to make it clearer what each field does.

BUG=678694,674015

Review-Url: https://codereview.chromium.org/2619433003
Cr-Commit-Position: refs/heads/master@{#442017}

Specifically, this compares making mojo interface calls vs just sending raw IPC::Messages or callbacks.

Review-Url: https://codereview.chromium.org/2608403002
Cr-Commit-Position: refs/heads/master@{#442016}

Crash is reported when doing screen share on hangouts. We found it initiates
at a place where we try to get image for a null icon handle.

This CL fixes the issue.

BUG=676758
TEST=running screen share on hangouts

Review-Url: https://codereview.chromium.org/2617153002
Cr-Commit-Position: refs/heads/master@{#442015}

ASSERT_TRUE stops test execution by retuning, which will not stop the
test if return is called from the block.

BUG=678942

Review-Url: https://codereview.chromium.org/2611283003
Cr-Commit-Position: refs/heads/master@{#442014}

These methods are not implemented by Tab anymore.

BUG=661445

Review-Url: https://codereview.chromium.org/2616843004
Cr-Commit-Position: refs/heads/master@{#442013}

This mode is always enabled now, so we can remove the code for handling
when it is disabled.

BUG=236848
TEST=Tests still pass
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_site_isolation

Review-Url: https://codereview.chromium.org/2617893002
Cr-Commit-Position: refs/heads/master@{#442012}

Previously we were going to use a | delimiter and manually construct
storage keys ourselves. The requires us to write logic to reverse it,
and is inefficient when encoding characters that need to be escaped.
The disadvantages are that we always have 4 bytes (instead of 1) of
overhead, and printing the string is less readable.

BUG=675992

Review-Url: https://codereview.chromium.org/2598113002
Cr-Commit-Position: refs/heads/master@{#442011}

Right now, this code will hide the notification at the end of the
timeout, even we switched to showing a different article in the
notification in the meantime. Fixing that will mean switching to
different tags for different articles.

Notifications should also implicitly be hidden if the articles
themselves expire, since the notifier service will watch for their
removal.

BUG=675561

Review-Url: https://codereview.chromium.org/2616893003
Cr-Commit-Position: refs/heads/master@{#442010}

Move the logic for converting from TitledUrlMatch items to
AutocompleteMatch items out of BookmarkProvider into places where it can
be shared with other AutocompleteProviders.

BUG=630769

Review-Url: https://codereview.chromium.org/2583763003
Cr-Commit-Position: refs/heads/master@{#442009}

CertificateReportingService browser tests wait for the IO thread
after sending a report or resetting the service. This doesn't seem
reliable and the tests occasionally fail. A better approach is to:
- For reports, wait for requests to be destroyed
(CertificateReportingService unit test already does this).
- For service resets, wait for a callback to be called.

This CL does these: It removes WaitForIOThread calls from the browser
test and replaces them with the above. It also refactors the following:
- RequestObserver is shared between browser and unit tests.
- NetworkDelegate is no longer used in the unit tests. RequestObserver is
used instead.
- All tests now expect no URL requests to remain during tear down.
- The code to check expected reports is now shared.

BUG=677560

Review-Url: https://codereview.chromium.org/2605403002
Cr-Commit-Position: refs/heads/master@{#442008}

First step in an effort to refactor content::ChildProcessLauncher, by
unifying
ContentBrowserClient::GetAdditionalMappedFilesForChildProcess() on
Android and Linux by moving the region argument to
content::FileDescriptorInfo.

BUG=675294

Review-Url: https://codereview.chromium.org/2613803003
Cr-Commit-Position: refs/heads/master@{#442007}

This parameter has done nothing for a while.  It became obsolete when
we started using ContextUtils.getAppSharedPreferences().

BUG=678787

Review-Url: https://codereview.chromium.org/2600283002
Cr-Commit-Position: refs/heads/master@{#442006}

Review-Url: https://codereview.chromium.org/2605353002
Cr-Commit-Position: refs/heads/master@{#442005}

We believe that the linked bug is caused by a low memory situation
triggering a bug in MediaCodec. The majority of crashes are
Samsung KK devices and the Lenovo A6000. So this change speculatively
increases the VP8 blacklist for these too.

BUG=600454

Review-Url: https://codereview.chromium.org/2612153004
Cr-Commit-Position: refs/heads/master@{#442004}

This replaces the previous check that was done inside
SyncBackendRegistrar to see whether the corresponding workers for these
types existed.

BUG=663125

Review-Url: https://codereview.chromium.org/2612323003
Cr-Commit-Position: refs/heads/master@{#442003}

Cr-Commit-Position: refs/heads/master@{#442002}

BUG=
TEST=Ran third_party/closure_compiler/run_compiler vr_shell_ui

Review-Url: https://codereview.chromium.org/2617173002
Cr-Commit-Position: refs/heads/master@{#442001}

Add a histogram PhysicalWeb.InitialState.IosChrome to record the startup
state of Bluetooth, Location, and the Physical Web preference.

BUG=630769

Review-Url: https://codereview.chromium.org/2458613004
Cr-Commit-Position: refs/heads/master@{#442000}

Background: Recently in web-platform-tests the format of MANIFEST.json changed, and I filed http://crbug.com/678077 for this. I committed a change http://crrev.com/2610243002 which updated the layout test runner to find web-platform-tests with the new manifest format, but the actual manifest file was too large to upload to Rietveld and was not committed. After that change, no wpt tests are found when running run-webkit-tests imported/wpt.

This change would update MANIFEST.json, although the actual file is still too large to upload to Rietveld, so this CL (in patch 1 at least) appears empty. I believe it would have to be landed directly.

TBR=dpranke

BUG=678077

Review-Url: https://codereview.chromium.org/2615903004 .
Cr-Commit-Position: refs/heads/master@{#441999}

Use TestJavaScriptDialogPresenter instead of deprecated
CRWWebUserInterfaceDelegate.

BUG=661445
TBR=kkhorimoto@chromium.org

Review-Url: https://codereview.chromium.org/2613163002
Cr-Commit-Position: refs/heads/master@{#441998}

BUG=661143
Committed: https://crrev.com/4aa30591e0953da7e0c54ffdb91856e496d16b5f
Cr-Commit-Position: refs/heads/master@{#440418}

Reverted: https://crrev.com/2885403f92c9ab0a8769c2c41e859502da5a7e6f
Cr-Commit-Position: refs/heads/master@{#440479}
Cause of revert fixed in https://codereview.chromium.org/2611053003/.

Review-Url: https://codereview.chromium.org/2592813002
Cr-Commit-Position: refs/heads/master@{#441997}

Paving the way to expose all relevant cookies per 673934.

* Adds filter bar with text match on name, value, and domain
* Converts invisible state buttons into disabled states

BUG=673934

Review-Url: https://codereview.chromium.org/2617043002
Cr-Commit-Position: refs/heads/master@{#441996}

With inert-top-controls turned on by default, we don't change the layout height
when the top controls are normally hidden. That is, if a user hides the URL bar
by scrolling, the page won't resize to use the newly available space. It'll lay
out as though the URL bar was still showing.

This patch makes an exception for cases where Chrome locks the URL bar in the
hidden state. e.g. Fullscreen or WebApp mode. In those cases, since the URL bar
can't be shown at all, there's no reason not to use the full height.

BUG=678649

Review-Url: https://codereview.chromium.org/2614993002
Cr-Commit-Position: refs/heads/master@{#441995}

This CL prefixes the DidFinishLoading and DidReceiveData methods in
BrowserPluginDelegate, so they don't conflict with the
WebAssociatedURLLoaderClient versions which currently start with
lowercase letter. As part of the Blink rewrite to match Chromium style,
those will be colliding in MimeHandlerViewContainer.

BUG=582312

Review-Url: https://codereview.chromium.org/2616123002
Cr-Commit-Position: refs/heads/master@{#441994}

Replace the GLSurface::Format enum with a proper class in preparation to
supporting types with configurable depth_size / samples / stencil_size.
Doing this in the current enum format would lead to a combinatorial
explosion.

The basic idea is that GetFormat can be used to initialize a new
surface with the same format as a pre-existing surface, and a
IsCompatible predicate is used to check if two formats are
equivalent for the purpose of avoiding BAD_MATCH errors. Code
should no longer be comparing formats to SURFACE_DEFAULT or similar.

Also add a new CreateOffscreenGLSurfaceWithFormat function that can
be used to request a specific surface format.

(These changes were suggested during review of https://crrev.com/2461803002 which is now forked into https://crrev.com/2586803003 .)

BUG=655722
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2616723002
Cr-Commit-Position: refs/heads/master@{#441993}

BUG=678983

TBR=rouslan@chromium.org

Review-Url: https://codereview.chromium.org/2617943003
Cr-Commit-Position: refs/heads/master@{#441992}

These fakes will be resued in CRWWebControllerPageDialogOpenPolicyTest
instead of CRWWebUserInterfaceDelegate mock.

BUG=661445
TRR=kkhorimoto@chromium.org

Review-Url: https://codereview.chromium.org/2615093004
Cr-Commit-Position: refs/heads/master@{#441991}

Improves code readability, sharing, and future maintenance. I don't
expect a performance impact but if we do see something material we
could inline.
CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel

Review-Url: https://codereview.chromium.org/2611253003
Cr-Commit-Position: refs/heads/master@{#441990}

The idea is to allow to monitor performance gains while optimizing
parts of the JPEG code path.

BUG=674264

Review-Url: https://codereview.chromium.org/2617843002
Cr-Commit-Position: refs/heads/master@{#441989}

Web forms may not contain a username element, therefore it is not safe
to assume that such an element exists when calling ShowNotSecureWarning.
Instead, use the actual element found by the autofill logic.

BUG=678769

Review-Url: https://codereview.chromium.org/2612893003
Cr-Commit-Position: refs/heads/master@{#441988}

https://pdfium.googlesource.com/pdfium.git/+log/469f6da247ff..8d94b6687f27

$ git log 469f6da24..8d94b6687 --date=short --no-merges --format='%ad %ae %s'
2017-01-06 rbpotter Revert postscript code removal.

Documentation for the AutoRoller is here:
https://skia.googlesource.com/buildbot/+/master/autoroll/README.md

If the roll is causing failures, see:
http://www.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Failures-due-to-DEPS-rolls

TBR=dsinclair@chromium.org

Review-Url: https://codereview.chromium.org/2612303005
Cr-Commit-Position: refs/heads/master@{#441987}

BUG=666792
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_cronet_tester

Review-Url: https://codereview.chromium.org/2617853002
Cr-Commit-Position: refs/heads/master@{#441986}

Fixes compilation failures with various < gcc-5.0  toolchains lacking
std::is_trivially_copyable<T>. Without it, compile errors like

../../base/template_util.h:189:36: error: no type named 'is_trivially_copyable' in namespace 'std'
using is_trivially_copyable = std::is_trivially_copyable<T>;
                              ~~~~~^
../../base/template_util.h:189:57: error: expected ';' after alias declaration
using is_trivially_copyable = std::is_trivially_copyable<T>;
                                                        ^
                                                        ;
will be encountered.

R=danakj
BUG=555754

Review-Url: https://codereview.chromium.org/2612933003
Cr-Commit-Position: refs/heads/master@{#441985}

Currently, all intents received by ChromeTabbedActivity are logged
as MobileReceivedExternalIntent despite there being many codepaths
within Chrome that also trigger intent dispatch.

A code search for Browser.EXTRA_CREATE_NEW_TAB shows a bunch of
callers that trigger intents and we logged them all as external.

This conflates whether we are receiving intents from other applications
or ourself.  The intent of this change (yuck, yuck, yuck) is to
split out intents that we dispatch ourselves and ones that an
actual external app did.

BUG=648386

Review-Url: https://codereview.chromium.org/2604063002
Cr-Commit-Position: refs/heads/master@{#441984}

BUG=678452

Review-Url: https://codereview.chromium.org/2617433005
Cr-Commit-Position: refs/heads/master@{#441983}

BUG=chromium:677972
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.perf:linux_perf_cq;master.tryserver.chromium.perf:mac_retina_perf_cq;master.tryserver.chromium.perf:winx64_10_perf_cq

Review-Url: https://codereview.chromium.org/2615003003
Cr-Commit-Position: refs/heads/master@{#441982}

Ensures the thread safe interface pointer returned by
ThreadSafeAssociatedInterfacePtrProvider can be called before the
interface pointer is actually bound.

BUG=675089
TEST=Run mojo_public_bindings_unittests

Review-Url: https://codereview.chromium.org/2608783003
Cr-Commit-Position: refs/heads/master@{#441981}

The UI is hidden behind a feature flag that enables a menu item (which
uses innovative string compression techniques to avoid bloat due to
i18n) to open the standalone content suggestions UI.

While the activity and menu item are not meant to be shipped in an
enabled state, this will allow iterating on a content suggestions UI
outside of the NTP.

Review-Url: https://codereview.chromium.org/2593523005
Cr-Commit-Position: refs/heads/master@{#441980}

These properties provide the author with the ability to control layout through logical,
rather than physical, direction and dimension mappings.

Intent to Implement and Ship: CSS Logical Properties, Inline/Block size,
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/4qEXWptfVHs. This intent
is a followup for "Intent to Deprecate and Remove: -webkit-{min-,max-}logical-{width,height}",
https://groups.google.com/a/chromium.org/forum/#!topic/blink-dev/VbPI65jBMIc.

When entire css-logical-props are implemented, will review these
-webkit-{min-,max-}logical-{width,height} and see if it is necessary to remove them.

Spec:
https://drafts.csswg.org/css-logical-props-1/#logical-dimension-properties
https://drafts.csswg.org/css-writing-modes-3/#logical-to-physical

BUG=538475

Review-Url: https://codereview.chromium.org/2577363002
Cr-Commit-Position: refs/heads/master@{#441979}