GitLab

The label "changes remaining" is not quite correct.  It's actually set
to either zero or one, depending on whether or not the current sync
cycle should result in more downloads.

If there is an issue with infinite download looping, we should be able
to detect and debug it based on the protocol logs, among other things.
The protocol logs indicate when a GetUpdates response had a non-zero
"changes remaining" value.

BUG=349301

Review URL: https://codereview.chromium.org/256413007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266594 0039d316-1c4b-4281-b951-d872f2087c98

This is a temporary fix to unbreak clank while they remove their dependency on that method.

BUG=367832
TBR=pkasting@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/252913002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266592 0039d316-1c4b-4281-b951-d872f2087c98

BUG=162827
NOTRY=true

Review URL: https://codereview.chromium.org/254563004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266591 0039d316-1c4b-4281-b951-d872f2087c98

Fix for the revert of Blink r172659. The problem was that MediaStreamImpl's constructor was getting a RenderView with an uninitialized routing_id. Also add a dcheck to catch this happening in the future to make debugging easier.

BUG=304341
R=nasko@chromium.org

Review URL: https://codereview.chromium.org/259973002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266588 0039d316-1c4b-4281-b951-d872f2087c98

NOTRY=True

Review URL: https://codereview.chromium.org/254803013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266587 0039d316-1c4b-4281-b951-d872f2087c98

BUG=360664

Review URL: https://codereview.chromium.org/252623002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266586 0039d316-1c4b-4281-b951-d872f2087c98

The patch landed originally in r266470, but was reverted in r266579 due to an
uninitialized variable.  The fix was reviewed in crrev.com/253813003.

This implements a touch dispatch model in which touchmove sending is throttled
while a scroll sequence is active and being consumed. Such touchmove's are
marked with |cancelable = false|, indicating to any potential consumers that the
event cannot be prevented from triggering a platform gesture. Throttling limits
the touchmove sending rate during scrolling to 1 event per 200ms.

BUG=346693,367692
NOTRY=true
TBR=jochen@chromium.org

Review URL: https://codereview.chromium.org/254803012

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266585 0039d316-1c4b-4281-b951-d872f2087c98

Remove the ignore file entry for _M_rep races which previously helped avoid deadlocks in TSan. The recent Clang roll has brought the TSan fix to Chromium.

BUG=356676
TBR=thakis@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/258053002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266584 0039d316-1c4b-4281-b951-d872f2087c98

BUG=366288

Review URL: https://codereview.chromium.org/252393002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266583 0039d316-1c4b-4281-b951-d872f2087c98

Revert of Limit renderer saved frames to avoid running out of fds. (https://codereview.chromium.org/248193003/)

Reason for revert:
Somewhat speculative. SoftwareFrameManagerTest.DoNotEvictVisible started failing reliably on the valgrind bots after this landed. Example: http://build.chromium.org/p/chromium.memory.fyi/builders/Chromium%20OS%20%28valgrind%29%283%29/builds/32121/steps/memory%20test%3A%20content/logs/SoftwareFrameManagerTest.DoNotEvictVisible

http://build.chromium.org/p/chromium.memory.fyi/builders/Chromium%20OS%20%28valgrind%29%283%29/builds/32121/steps/memory%20test%3A%20content/logs/SoftwareFrameManagerTest.DoNotEvictVisible

Original issue's description:
> Limit renderer saved frames to avoid running out of fds.
> 
> Software delegated rendering uses one fd per tile, so with a bunch of tiles that means it can run out of fds. If it seems close to hitting the limit the browser should throw away old frames to avoid this from happening.
> 
> BUG=362603
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=266383

TBR=piman@chromium.org,jbauman@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=362603

Review URL: https://codereview.chromium.org/258093002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266582 0039d316-1c4b-4281-b951-d872f2087c98

valgrind reports uninitialized memory usage:

http://build.chromium.org/p/chromium.memory.fyi/builders/Linux%20Tests%20%28valgrind%29%281%29/builds/33440/steps/memory%20test%3A%20content/logs/AD4CE3561F03B37F

fun:_ZN7content15TouchEventQueue25SendTouchEventImmediatelyERKNS_20EventWithLatencyInfoIN5blink13WebTouchEventEEE
fun:_ZN7content15TouchEventQueue26ForwardNextEventToRendererEv
fun:_ZN7content15TouchEventQueue10QueueEventERKNS_20EventWithLatencyInfoIN5blink13WebTouchEventEEE
fun:_ZN7content19TouchEventQueueTest14SendTouchEventERKN5blink13WebTouchEventE
fun:_ZN7content19TouchEventQueueTest14SendTouchEventEv
fun:_ZN7content19TouchEventQueueTest15PressTouchPointEff
fun:_ZN7content30TouchEventQueueTest_Basic_Test8TestBodyEv

http://build.chromium.org/p/chromium.memory.fyi/builders/Linux%20Tests%20%28valgrind%29%281%29/builds/33440/steps/memory%20test%3A%20content/logs/383BD397F91FBC46

fun:_ZN7content15TouchEventQueue25SendTouchEventImmediatelyERKNS_20EventWithLatencyInfoIN5blink13WebTouchEventEEE
fun:_ZN7content15TouchEventQueue19TouchTimeoutHandler17ConfirmTouchEventENS_18InputEventAckStateE
fun:_ZN7content15TouchEventQueue15ProcessTouchAckENS_18InputEventAckStateERKN2ui11LatencyInfoE
fun:_ZN7content19TouchEventQueueTest17SendTouchEventAckENS_18InputEventAckStateE
fun:_ZN7content42TouchEventQueueTest_TouchTimeoutBasic_Test8TestBodyEv

etc

> Implement async touchmove dispatch during scroll
> 
> This implements a touch dispatch model in which touchmove sending is throttled
> while a scroll sequence is active and being consumed.  Such touchmove's are
> marked with |cancelable = false|, indicating to any potential consumers that the
> event cannot be prevented from triggering a platform gesture. Throttling limits
> the touchmove sending rate during scrolling to 1 event per 200ms.
> 
> BUG=346693
> 
> Review URL: https://codereview.chromium.org/245833002

TBR=jdduke@chromium.org

Review URL: https://codereview.chromium.org/252023005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266579 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266578 0039d316-1c4b-4281-b951-d872f2087c98

http://build.chromium.org/f/chromium/perf/dashboard/ui/changelog_blink.html?url=/trunk&range=172734:172758&mode=html
TBR=
BUG=

Review URL: https://codereview.chromium.org/259823010

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266577 0039d316-1c4b-4281-b951-d872f2087c98

We should not update the local changestamp when something goes wrong.

BUG=255333
TEST=unit_tests

Review URL: https://codereview.chromium.org/256773005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266576 0039d316-1c4b-4281-b951-d872f2087c98

BUG=313454

Review URL: https://codereview.chromium.org/188443003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266575 0039d316-1c4b-4281-b951-d872f2087c98

- Instead of giving generic stats for each event type,
  StatsEventSubsriber will now provide more specific stats. It also
  provide new kinds of stats (packet loss %, latency)
- stats_util to convert output of StatsEventSubscriber to a
  base::DictionaryValue is moved into the subscriber implementation.
- Introduced ReceiverTimeOffsetEstimator to estimate receiver offset
  based on raw events received. It is used by StatsEventSubscriber to
  determine stats such as latency. The estimator itself is implemented
  as a RawEventSubscriber and is shared between the audio and video
  StatsEventSubscriber objects.
- Move stats related structs from logging_defines.{cc,h} into
  stats_event_subscriber.{cc,h}
BUG=353019

Review URL: https://codereview.chromium.org/236123003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266574 0039d316-1c4b-4281-b951-d872f2087c98

BUG=

Review URL: https://codereview.chromium.org/251853003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266573 0039d316-1c4b-4281-b951-d872f2087c98

Test doesn't work with tsan:

http://build.chromium.org/p/chromium.memory.fyi/builders/Linux%20Tests%20%28TSan%20v2%29%281%29/builds/8850/steps/unit_tests/logs/RetainedFilePermissionsObtainedCorrectly

[ RUN      ] AppInfoPermissionsTabTest.RetainedFilePermissionsObtainedCorrectly
../../chrome/browser/ui/views/apps/app_info_dialog/app_info_permissions_tab_unittest.cc:193: Failure
Value of: retained_files[0]
  Actual: 24-byte object \u003C14-66 69-6C 65-5F 33-2E 65-78 74-00 0C-7D 00-00 2B-00 00-00 00-00 00-00>
Expected: base::FilePath("file_1.ext")
Which is: 24-byte object \u003C14-66 69-6C 65-5F 31-2E 65-78 74-00 00-00 00-00 07-7A E9-08 02-00 00-00>
[  FAILED  ] AppInfoPermissionsTabTest.RetainedFilePermissionsObtainedCorrectly (112 ms)

(Possibly because tsan uses libc++ instead of libstdc++?)

> Added collapsible permissions to the app info dialog
> 
> Separated the optional and required permissions in the dialog, and
> added retained files as well. Also made the various types of
> permissions collapsible, and added a message when the app has no
> permissions whatsoever.
> 
> BUG=356985,350746
> 
> Review URL: https://codereview.chromium.org/214163002

TBR=sashab@chromium.org

Review URL: https://codereview.chromium.org/258083002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266572 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266570 0039d316-1c4b-4281-b951-d872f2087c98

When a device-local account is removed, its policy is deleted from disk
immediately. If a session using this account happens to be in progress,
the session is allowed to continue with policy served from an in-memory
cache. If Chrome crashes later in the session, the policy becomes
completely unavailable. Exit the session in that case, rather than
allowing it to continue without policy.

BUG=178290
TEST=Manual in VM

Review URL: https://codereview.chromium.org/252563007

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266568 0039d316-1c4b-4281-b951-d872f2087c98

Remove root_feed.json
Remove other JSON files used in ChangeListProcessorTest

BUG=357038
TEST=unit_tests

Review URL: https://codereview.chromium.org/252563011

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266567 0039d316-1c4b-4281-b951-d872f2087c98

The primary use-case for QueryMouseLocation() is for X11 interactive
tests. The only other use was in WindowEventDispatcher, but that code
is no longer necessary.

BUG=none
R=sky@chromium.org

Review URL: https://codereview.chromium.org/251743004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266565 0039d316-1c4b-4281-b951-d872f2087c98

Opus ARM optimizations has been enabled on Android and ChromeOS, but not on iOS because some of the optimizations don't compile for iOS. In this CL I'm enabling the optimizations that do compile.

BUG=

Review URL: https://codereview.chromium.org/237143003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266564 0039d316-1c4b-4281-b951-d872f2087c98

Move bookmark_model_observer.h and base_bookmark_model_observer.{cc,h}
from chrome/browser/bookmarks to components/bookmarks as they do not
depend on anything in chrome/browser.

BUG=367656
TBR=sky

Review URL: https://codereview.chromium.org/256083003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266563 0039d316-1c4b-4281-b951-d872f2087c98

Previously the name of the enum value was wrong. This CL fixes it:
EntryChangedKind.CREATE -> EntryChangedKind.CREATED.

BUG=364258
TEST=manually

Review URL: https://codereview.chromium.org/257803013

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266562 0039d316-1c4b-4281-b951-d872f2087c98

This reverts commit 945604a5.

BUG=351824
TBR=dbeam@chromium.org, rmsousa@chromium.org
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/257823005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266561 0039d316-1c4b-4281-b951-d872f2087c98

Revert of Revert of Move  service_discovery_client_impl.* into chrome/common. (https://codereview.chromium.org/250793011/)

Reason for revert:
Reland original patch. The revert didn't solve the sizes regression.

Original issue's description:
> Revert of Move  service_discovery_client_impl.* into chrome/common. (https://codereview.chromium.org/256923003/)
> 
> Reason for revert:
> [Sheriff] Speculating that this lead to a sizes regression:
> https://chromeperf.appspot.com/report?masters=Chromium&bots=chromium-rel-linux-64&tests=sizes&rev=266525&checked=core
> 
> Will reland otherwise.
> 
> Original issue's description:
> > Move  service_discovery_client_impl.* into chrome/common.
> > It's need to be used by utility and browser processes.
> > 
> > BUG=349645
> > 
> > Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=266487
> 
> TBR=thestig@chromium.org,vitalybuka@chromium.org
> NOTREECHECKS=true
> NOTRY=true
> BUG=349645
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=266535

TBR=thestig@chromium.org,vitalybuka@chromium.org
NOTREECHECKS=true
NOTRY=true
BUG=349645

Review URL: https://codereview.chromium.org/252023003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266560 0039d316-1c4b-4281-b951-d872f2087c98

Revert of Store default search provider data in dictionary pref. (https://codereview.chromium.org/229763005/)

Reason for revert:
Speculative revert for introducing lots of leaks. See: http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%20LSan%20Tests%20%282%29/builds/2026

Original issue's description:
> Store default search provider data in dictionary pref, and add DefaultSearchManager class to handle the reading and writing of this pref (DefaultSearchManager will eventually all default search related concerns).
> 
> This pref will be used to persist data about user-selected 
> default search providers. In future CLs, support will be added for default search providers set by policy and sync.
> 
> 
> BUG=365762
> 
> Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=266479

BUG=365762
TBR=gab, erikwright, Peter Kasting, Cait Phillips
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/258933009

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266559 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266558 0039d316-1c4b-4281-b951-d872f2087c98

Chromium base revision: 266531 / d751dd54
Old Skia revision: 14349
New Skia revision: 14390
Control CL: https://codereview.chromium.org/250923012

This CL was created by Skia's roll_deps.py script.

Bypassing commit queue trybots:
NOTRY=true

Review URL: https://codereview.chromium.org/251083006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266557 0039d316-1c4b-4281-b951-d872f2087c98

In https://codereview.chromium.org/254783004 the semantics of the scale
parameter for GesturePinchUpdate on Mac is being corrected to be a scale
factor (1.0 no zoom) instead of a weird magnification value (where 0 means
no zoom).  This is important because GesturePinchUpdate events are
coalesced together by multipling (not adding) their scale values.

This CL updates the browser zoom logic to compensate for this change in
scale values.

BUG=289887

Review URL: https://codereview.chromium.org/255843004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266556 0039d316-1c4b-4281-b951-d872f2087c98

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266555 0039d316-1c4b-4281-b951-d872f2087c98

Please reply to the V8 sheriff rossberg@chromium.org in case of problems.

TBR=rossberg@chromium.org

Review URL: https://codereview.chromium.org/258883004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266554 0039d316-1c4b-4281-b951-d872f2087c98

BUG=366585

Review URL: https://codereview.chromium.org/254533003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266553 0039d316-1c4b-4281-b951-d872f2087c98

BUG=366597
R=nasko@chromium.org,jcivelli@chromium.org

Review URL: https://codereview.chromium.org/257463004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266552 0039d316-1c4b-4281-b951-d872f2087c98

Note:
Currently FaviconTabHelper implements both FaviconClient and FaviconDriver.
A follow up CL will create a ContentFaviconDriver class that implements
FaviconDriver. FaviconTabHelper will then own an instance of the
ContentFaviconDriver, instead of inheriting from FaviconDriver.

BUG=362074

Review URL: https://codereview.chromium.org/255843002

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266551 0039d316-1c4b-4281-b951-d872f2087c98

This is a minor bugfix to https://codereview.chromium.org/229063003,
changing the comparison of USB VID to ignore the case.

BUG=288562, 347371

Review URL: https://codereview.chromium.org/255933006

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266550 0039d316-1c4b-4281-b951-d872f2087c98

This is duplicated functionality now, as one can use DevTools touch emulation instead.

BUG=337142

Review URL: https://codereview.chromium.org/258503003

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266549 0039d316-1c4b-4281-b951-d872f2087c98

On some platfroms, mouse events come with a bad timestamp, which leads to
awkward time-dependent gestures. For example, fling animation may be almost
instant.
Using base::TimeTicks::Now() is good enough for emulation purposes, because
emulator spawns new events itself.

BUG=367156

Review URL: https://codereview.chromium.org/250923005

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266548 0039d316-1c4b-4281-b951-d872f2087c98

In order to create a favicon component, we are removing dependencies on chrome/.

TBR=bsalomon1
BUG=359592
NOTRY=true

Review URL: https://codereview.chromium.org/246893004

git-svn-id: svn://svn.chromium.org/chrome/trunk/src@266547 0039d316-1c4b-4281-b951-d872f2087c98