GitLab

Before this change, the testing path invoked RenderWrapper::Stop() by
posting a task, which would execute after PipelineImpl::Stop().

The production path also posts a task, but to a different thread, and
PipelineImpl::Stop() waits for that task to execute before returning.

This change makes the testing path more similar to the production path,
by ensuring that RendererWrapper::Stop() executes before
PipelineImpl::Stop() returns.

The PipelineTeardownTest is also updated so that expectations on the
Demuxer that Stop() the Pipeline are posted rather than directly
called. This better matches the production behavior where they would
be on separate threads and therefore cannot be re-entrant.

Bug: 788508
Change-Id: I894ee6c2febb3b739aaa809c6aa0d9388b76b64a
Reviewed-on: https://chromium-review.googlesource.com/791860
Commit-Queue: Dan Sanders <sandersd@chromium.org>
Reviewed-by: Xiaohan Wang <xhwang@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519862}

Fix the logic that decides whether or not to assign temporary references
in HostFrameSinkManager. This fixes temporary references not being
assigned owners when running with --enable-viz in desktop Chrome. Add a
test to verify temporary references are assigned with a remote
FrameSinkmanager.

Also fix some places in HostFrameSinkManager where |frame_sink_manager_|
and |frame_sink_manager_impl_| were mixed up.

Bug: 787589
Cq-Include-Trybots: master.tryserver.chromium.android:android_optional_gpu_tests_rel
Change-Id: I2b40fa332f8afc8ff7fa7b62482ef174bcbd239a
Reviewed-on: https://chromium-review.googlesource.com/793976Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Fady Samuel <fsamuel@chromium.org>
Commit-Queue: kylechar <kylechar@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519861}

The base::debug::Alias function is normally used to 'fool' the compiler
into retaining an otherwise discardable local variable. In this case
base::debug::Alias is used in a completely different way which initially
seems wrong, but is actually reasonable. A comment will avoid confusion
when auditing uses of Alias.

The initial discussion of using base::debug::Alias is here:
https://codereview.chromium.org/2810833002#msg7

Bug: 756589
Change-Id: Ib00a1444c217680a69178a8bbf6c710179126e86
Reviewed-on: https://chromium-review.googlesource.com/794298Reviewed-by: Brett Wilson <brettw@chromium.org>
Commit-Queue: Bruce Dawson <brucedawson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519860}

This CL updates the open source implementation of the side menu so that
it also shows items from SideMenuItemsProvider. This make it easier for
developer to debug CRD for iOS on Chromium without checking out
ios_internal.

Bug: 786625
Change-Id: I784032e905cfdd78eeef122f302f464b443d2153
Reviewed-on: https://chromium-review.googlesource.com/780151Reviewed-by: Jamie Walch <jamiewalch@chromium.org>
Commit-Queue: Yuwei Huang <yuweih@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519859}

dynamic_annotations is a small library intended to help tsan code
coexist with valgrind.  This CL makes the linux sandbox code use
the RunningOnValgrind() function instead of (multiply) defining its
own version, and does not use the annotations which are being phased
out.

TODO: remove valgrind checks that are no longer needed (perhaps
all of these?)

Bug: 786831,349861
Change-Id: I39f83d5dc839c54c5bc9297e9a92bd3f84ead726
Reviewed-on: https://chromium-review.googlesource.com/778881
Commit-Queue: Mostyn Bramley-Moore <mostynb@vewd.com>
Reviewed-by: Julien Tinnes <jln@chromium.org>
Reviewed-by: danakj <danakj@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519858}

A valid Cast device address must be private. This is checked using
the IPAddress::IsReserved() method, similar to DIAL's device description
service.

The check is performed in several entry points (some are redundant as
extra safety net):
- DnsSdRegistry, when it receives an device advertisement from mDNS
- CastSocketService, before it opens socket
- CastSocketServiceImpl::OpenChannel
- CastChannelOpenFunction (entry point for chrome.cast.channel.open)

Bug: 786109
Change-Id: Iaad91834cd4149fd345b2ada4e2704a0e158ba49
Reviewed-on: https://chromium-review.googlesource.com/792650
Commit-Queue: Derek Cheng <imcheng@chromium.org>
Reviewed-by: mark a. foltz <mfoltz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519857}

For OOP-R, we will eventually have a separate decoder that only handles RasterInterface
(http://crrev.com/c/778242 and https://goo.gl/Qy3aPL), which is a subset of what
a GLES2Decoder handles. This new RasterDecoder will be simpler than a GLES2Decoder and will
eventually evolve into something less GL-like.

This CL adds a new temporary flag to ease development. In following CLs,
--enable-raster-decoder will create a RasterDecoder instead of a GLES2Decoder whenever
a client requests OOP-R (--enable-gpu-rasterization --enable-oop-rasterization from
the command line).

This plumbs through a flag and adds a VLOG where we make the decision to create
a GLES2Decoder.

Bug: 789238
Cq-Include-Trybots: master.tryserver.chromium.android:android_optional_gpu_tests_rel;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
Change-Id: I1c37ecf2372f6770369ee0618fc221ba7f4b9ef9
Reviewed-on: https://chromium-review.googlesource.com/793985Reviewed-by: Tom Sepez <tsepez@chromium.org>
Reviewed-by: Antoine Labour <piman@chromium.org>
Reviewed-by: Victor Miura <vmiura@chromium.org>
Commit-Queue: Jonathan Backer <backer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519856}

This CL adds a fallback to display MostVisited URLs when personalized
zero-query is enabled with ChromeHomePersonalizedOmniboxSuggestions but
the user's default search engine is not Google.

Bug: 781994
Change-Id: I489052f3a33e2f97d86138760f8f2d135515c241
Reviewed-on: https://chromium-review.googlesource.com/777502
Commit-Queue: Gheorghe Comanici <gcomanici@chromium.org>
Reviewed-by: Mark Pearson <mpearson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519855}

https://chromium.googlesource.com/chromium/tools/depot_tools.git/+log/1b30125fbc42..9e51906ffb13

$ git log 1b30125fb..9e51906ff --date=short --no-merges --format='%ad %ae %s'

Created with:
  roll-dep src/third_party/depot_tools


The AutoRoll server is located here: https://depot-tools-chromium-roll.skia.org

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

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.


TBR=phajdan.jr@chromium.org

Change-Id: I422a32985835a92e06ba4c780cbe9a655a3209a7
Reviewed-on: https://chromium-review.googlesource.com/794523Reviewed-by: depot-tools-roller . <depot-tools-roller@chromium.org>
Commit-Queue: depot-tools-roller . <depot-tools-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519854}

This change replaces the usages of ChromeAnimation in the tablet tab
strip. Several properties have been added to facilitate animation and
much of the logic has been simplified. Particularly, onUpdateAnimation
has been removed since most of the logic depending on its param
'jumpToEnd' (which was always false) was unused.

BUG=750381

Change-Id: If5c2346545254917cf690f4e5ddbb76a7dde3239
Reviewed-on: https://chromium-review.googlesource.com/791471Reviewed-by: David Trainor <dtrainor@chromium.org>
Commit-Queue: Matthew Jones <mdjones@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519853}

1. Show scroll (if needed) for the top-level component.
2. Disable scrolls for the individual service worker lists.
3. Fix white background for the "Other origin" list

Bug: 785136
Change-Id: I55e8ff08c72609b34b0aba8b9e5729d86a54c663
Reviewed-on: https://chromium-review.googlesource.com/776115Reviewed-by: Andrey Kosyakov <caseq@chromium.org>
Commit-Queue: Eugene Ostroukhov <eostroukhov@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519852}

https://webrtc.googlesource.com/src.git/+log/b874a352272a..3fb614bc93c3

$ git log b874a3522..3fb614bc9 --date=short --no-merges --format='%ad %ae %s'

Created with:
  roll-dep src/third_party/webrtc
BUG=none


The AutoRoll server is located here: https://webrtc-chromium-roll.skia.org

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

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.


CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_chromium_archive_rel_ng;master.tryserver.chromium.mac:mac_chromium_archive_rel_ng
TBR=webrtc-chromium-sheriffs-robots@google.com

Change-Id: I1a550fb7a4c35c5f61cf6bdd9bacc16456f11128
Reviewed-on: https://chromium-review.googlesource.com/794334Reviewed-by: WebRTC Roll Bot <webrtc-autoroll@chromium.org>
Commit-Queue: WebRTC Roll Bot <webrtc-autoroll@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519851}

process_resources.py and package_resources.py are currently both required to
create the final apk. They currently communicate using zipfiles. This ends up
having 3-4 zips/unzips during the final build of the same apk in order to
perform modifications. This merger allows all the required processing to be
done before apk creation with no need for zipping. It also simplifies the
process_resources.py script to separate the code that runs during library
builds vs apk builds.

Bug: 782316
Change-Id: I958c1d107d593b0932275d3c4452c2c51c499b55
Reviewed-on: https://chromium-review.googlesource.com/791152
Commit-Queue: Mohamed Heikal <mheikal@chromium.org>
Reviewed-by: agrieve <agrieve@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519850}

This should have been caught in https://chromium-review.googlesource.com/c/chromium/src/+/754281

Bug: 
Change-Id: Icca4f209ce474103d6a8231f36c89a36b27384d2
Reviewed-on: https://chromium-review.googlesource.com/794304
Commit-Queue: Shanfeng Zhang <szhangcs@google.com>
Commit-Queue: Roger McFarlane <rogerm@chromium.org>
Reviewed-by: Roger McFarlane <rogerm@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519849}

Add an IllegalStateException when we dismiss the IPH bar twice and add
a "Caused By" exception to track what the original dismiss site was.

This will not fix the bug but should give us more debugging information.

BUG=786916

Change-Id: I66f69a2c3318d5ae7a82c386511ed26a1afb85d9
Reviewed-on: https://chromium-review.googlesource.com/792348Reviewed-by: Ted Choc <tedchoc@chromium.org>
Reviewed-by: Matthew Jones <mdjones@chromium.org>
Commit-Queue: Matthew Jones <mdjones@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519848}

Cleans up flags for enabling ML for tap suppression in Contextual
Search.  Now there are just two flags and they are both normal
Chrome Features. One feature enables the internal Ranker functionality,
but may not have any user-visible effect.  The other flag enables enforcement
of the model -- providing user-visible tap suppression.  Both features
must be enabled for full functionality.  Also there is a param for the
URL of the model to load, and that's associated with the feature to
query Ranker.

BUG=776156

Change-Id: I06492d82c80000ca34aefa195e187f448ddfb513
Reviewed-on: https://chromium-review.googlesource.com/777725Reviewed-by: Ilya Sherman <isherman@chromium.org>
Reviewed-by: Theresa <twellington@chromium.org>
Commit-Queue: Donn Denman <donnd@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519847}

This patch, mirroring the logic in WebUIDataSourceImpl::SetJsonPath,
excludes the configured JSON path (used for localization strings) from
the set of resources that are marked as gzipped.

Bug: 787428
Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Change-Id: I158766444870b93fad44a1117e19f54e8f4b59e7
Reviewed-on: https://chromium-review.googlesource.com/792131
Commit-Queue: Reilly Grant <reillyg@chromium.org>
Reviewed-by: Eugene But <eugenebut@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519846}

Having pointers that are const will make it obvious if they are never
nullptrs, or if they can be sometimes.

Change-Id: I3b817b7cfebdc64372fb203d019eaf1c5ca49d87
Reviewed-on: https://chromium-review.googlesource.com/786824Reviewed-by: Kevin Bailey <krb@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519845}

https://chromium.googlesource.com/catapult.git/+log/65711d3e974c..cde6fc4afe73

$ git log 65711d3e9..cde6fc4af --date=short --no-merges --format='%ad %ae %s'
2017-11-27 hanxi Startup tracing for WebAPKs

Created with:
  roll-dep src/third_party/catapult


The AutoRoll server is located here: https://catapult-roll.skia.org

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

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.


CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel
TBR=sullivan@chromium.org

Change-Id: Id25a969d3d42a6a59cd33793ba1ad8c654b04799
Reviewed-on: https://chromium-review.googlesource.com/793891
Reviewed-by: <catapult-deps-roller@chromium.org>
Commit-Queue: <catapult-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519844}

Back-end can often produce layerTreeDidChange() at a rate that can't
be handled by front-end, so throttle it. A better fix would be to change
the protocol to only send these events when front-end is ready to
process them, but let's have a trivial fix first.

Change-Id: Id24233cf75697363ea1b7995346a735a2055b226
Reviewed-on: https://chromium-review.googlesource.com/786902Reviewed-by: Dmitry Gozman <dgozman@chromium.org>
Commit-Queue: Andrey Kosyakov <caseq@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519843}

All other code paths properly set it. Without this we hit DCHECKs as
the type isn't set.

BUG=786544
TEST=covered by tests

Change-Id: Ia3837afa9641c6fef83ed4441dbcb5ab1d5d9cd4
Reviewed-on: https://chromium-review.googlesource.com/794520Reviewed-by: Elliot Glaysher <erg@chromium.org>
Commit-Queue: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519842}

BUG=696454
TBR=jam@chromium.org

Change-Id: I68bae775dfa3a4a5c74b19659b81cfe976502365
Reviewed-on: https://chromium-review.googlesource.com/791517Reviewed-by: Avi Drissman <avi@chromium.org>
Reviewed-by: Lucas Gadani <lfg@chromium.org>
Reviewed-by: Ilya Sherman <isherman@chromium.org>
Commit-Queue: Avi Drissman <avi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519841}

Use --mus=viz if the mus window-server is the viz-host. If --mus is
not set to viz, then mus is used, but chrome still is the viz host.

For now, always set it to 'viz', so that mus still remains the viz
host. Once some more fixes land, we can turn it off by default.

BUG=786453

Change-Id: If8f6fca0178322311a0a14b718b08f9084bda35c
Reviewed-on: https://chromium-review.googlesource.com/792276
Commit-Queue: Sadrul Chowdhury <sadrul@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519840}

The issue is that ProcessorEntityTracker hits DCHECK when loading entity data
for commit if its specifics doesn't match the one passed to
ProcessorEntityTracker::MakeLocalChange. This can happen if model type code
assembles specifics differently after restart.

I'm removing the DCHECK and adding unittest for this scenario.

BUG=787741
R=pnoland@chromium.org

Change-Id: I1abe5c9e3d89c9f312c8faa0cae889805e2dc4cb
Reviewed-on: https://chromium-review.googlesource.com/792536Reviewed-by: Patrick Noland <pnoland@google.com>
Commit-Queue: Pavel Yatsuk <pavely@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519839}

Bug: 781340
Change-Id: Id9d18f4873ce0d47ed882ff9a6e00140f424b991
Reviewed-on: https://chromium-review.googlesource.com/782709
Commit-Queue: Troy Hildebrandt <thildebr@chromium.org>
Reviewed-by: Jesse Doherty <jwd@chromium.org>
Reviewed-by: Theresa <twellington@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519838}

The test
PasswordSyncableServiceAndroidAutofillTest.TwoWayServerAndLocalMerge
derives a few parameters from a single integer variable in two places.
It seems that in the second place it prepares another variable to
derive from but then uses the same as in the first place. Looks like a
copy-paste typo. In any case, switching to the prepared variable has
strictly more coverage, so this CL changes to the prepared variable.

Bug: 739101
Change-Id: I27bb5895e48c1d1f87f91cac8a9e26fa81390e1d
Reviewed-on: https://chromium-review.googlesource.com/793610Reviewed-by: Jan Wilken Dörrie <jdoerrie@chromium.org>
Commit-Queue: Vaclav Brozek <vabr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519837}

-Reorders order for doesUserQualify
-Adds filtering methods
-Adds tests for filtering methods
-Adds logging for when the last time the random # gen. was rolled
-Adds Chrome Home Force Survey check to doesUserQualify to allow for
  testing of other survey qualifications

Filtering method:
-Gets the max number from the finch config
-Rolls a random number up from 0 (inclusive) to that number (exclusive)
-If the random number = 0, process continues for user
-Else stop


Bug: 778384
Change-Id: Ie7063724614c98121e0e93c1a4b091557635015a
Reviewed-on: https://chromium-review.googlesource.com/741867
Commit-Queue: Daniel Park <danielpark@chromium.org>
Reviewed-by: Ted Choc <tedchoc@chromium.org>
Reviewed-by: Theresa <twellington@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519836}

This regressed in r518459 which added a step to clear out a stale
NSMenu tree once it could never be used again. But it turns out that
-[NSMenuDelegate menuDidClose:] is a bad signal for this. AppKit isn't
actually done with the menu at that point, and doesn't give the signal
we need.

To fix, keep the menu around as before, and improve test coverage.

Bug: 788430
Change-Id: Idbc19cb56e6e52d699d94518b2164d0bd776ce9d
Reviewed-on: https://chromium-review.googlesource.com/790090Reviewed-by: Robert Sesek <rsesek@chromium.org>
Commit-Queue: Trent Apted <tapted@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519835}

With this change, non-WPT layout tests are completely independent of
WPT files and hence won't be affected by imports. The change also makes
the behaviour more consistent as we already have copies of
testharness.js, testdriver.js and idlharness.js, with webidl2.js being
the only left. An apache alias rule is changed accordingly.

See crbug.com/685854 for maintaining these Chromium copies.

Besides, WebIDLParser.js (which is an old version of webidl2.js) is
removed from resources/ to avoid confusion. The only three files using
it are modified to use webidl2.js instead.

Bug: 787829, 685854
Change-Id: I2d9966646cc7db557d87ec403cd0c833ee262279
Reviewed-on: https://chromium-review.googlesource.com/785992
Commit-Queue: Robert Ma <robertma@chromium.org>
Reviewed-by: Quinten Yearsley <qyearsley@chromium.org>
Reviewed-by: Philip Jägenstedt <foolip@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519834}

TBR=agrieve@chromium.org,yfriedman@chromium.org,mef@chromium.org,pasko@chromium.org

Bug: 782848,789177
Cq-Include-Trybots: master.tryserver.chromium.android:android_cronet_tester;master.tryserver.chromium.mac:ios-simulator-cronet
Change-Id: I30cdc80f19ee01619e7ce808423c10ffc752644d
Reviewed-on: https://chromium-review.googlesource.com/794338Reviewed-by: John Budorick <jbudorick@chromium.org>
Reviewed-by: Misha Efimov <mef@chromium.org>
Reviewed-by: agrieve <agrieve@chromium.org>
Commit-Queue: John Budorick <jbudorick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519833}

Introducing a new XML parser service that parses XML into a
base::DictionaryValue. It does so by JSONifying the XML content.

Also factoring out TestServiceManagerListener to its own file, now that
it is used in 2 places (safe_json_parser_browsertest.cc and
safe_xml_parser_browsertest.cc).

Bug: 784667
Change-Id: I1f41f44d49910da43837aac8bca52e24533460ef
Reviewed-on: https://chromium-review.googlesource.com/767890Reviewed-by: Tom Sepez <tsepez@chromium.org>
Reviewed-by: Scott Graham <scottmg@chromium.org>
Reviewed-by: Robert Sesek <rsesek@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
Reviewed-by: Ken Rockot <rockot@chromium.org>
Commit-Queue: Jay Civelli <jcivelli@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519832}

BUG=472742

Cq-Include-Trybots: master.tryserver.chromium.android:android_optional_gpu_tests_rel
Change-Id: Iddaaf37132faf2698c09ae0851c741fb78d97656
Reviewed-on: https://chromium-review.googlesource.com/791010Reviewed-by: danakj <danakj@chromium.org>
Reviewed-by: Florin Malita <fmalita@chromium.org>
Commit-Queue: Herb Derby <herb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519831}

https://skia.googlesource.com/skia.git/+log/0f8e4dbd28d1..592c225b03ca

$ git log 0f8e4dbd2..592c225b0 --date=short --no-merges --format='%ad %ae %s'
2017-11-28 bsheedy Make Skia compatible with Android NDK r16

Created with:
  roll-dep src/third_party/skia


The AutoRoll server is located here: https://autoroll.skia.org

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

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.


CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;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
TBR=allanmac@chromium.org

Change-Id: I575587cd47062e30d6cfe9d39c1edfdb9e18ec38
Reviewed-on: https://chromium-review.googlesource.com/794431Reviewed-by: Skia Deps Roller <skia-deps-roller@chromium.org>
Commit-Queue: Skia Deps Roller <skia-deps-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519830}

When I landed MojoInputMessages by default yesterday in
https://chromium-review.googlesource.com/789590 the
dispatchKeyEvent-focus.js layout test became flaky.

The cause was layout tests don't use a compositor thread and events
could have been placed out of order because events hitting the
WidgetInputHandlerImpl (setFocus would get queued in the
input_event_queue_ whereas events received would get executed right away)

There is no need for the input_event_queue_ to be passed when there is
no compositor_task_queue_ as that causes an extra thread hop that isn't
necessary.

BUG=722928

Change-Id: I6d40ec67072e2ef98f2ebac04e6e8d54c7282ad0
Reviewed-on: https://chromium-review.googlesource.com/793972Reviewed-by: Mustaq Ahmed <mustaq@chromium.org>
Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519829}

Bug: 
Change-Id: I115d8ae2d9694fd98f145cb43181ed361c827427
Reviewed-on: https://chromium-review.googlesource.com/786443Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Commit-Queue: Nate Chapin <japhet@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519828}

TBR=allada@chromium.org

Bug: 785179
Change-Id: I148b8511fa1c7d38fd9d9db8a5caa00d3616bcf1
Reviewed-on: https://chromium-review.googlesource.com/794513Reviewed-by: Scott Little <sclittle@chromium.org>
Commit-Queue: Scott Little <sclittle@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519827}

If the not reachable code handling code gets removed, the rest of the
function works just fine and correctly returns an empty string.

Change-Id: Ia41522564ff242c097463ec804cf868a7b4f7784
Reviewed-on: https://chromium-review.googlesource.com/792298Reviewed-by: David Trainor <dtrainor@chromium.org>
Commit-Queue: Lei Zhang <thestig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519826}

Replaces usages of PersonalDataManager::GetURLRequestContextGetter() with
ApplicationContext::GetSystemURLRequestContext()

Bug: 781942
Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Change-Id: I7e42baf9ce61d722410c1f49ee33a55c5e53485a
Reviewed-on: https://chromium-review.googlesource.com/793850
Commit-Queue: Moe Ahmadi <mahmadi@chromium.org>
Reviewed-by: Mathieu Perreault <mathp@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519825}

The crashes seen in bug 734002 seem to only make sense if
TerminateProcess when used to terminate the current process can return.
This seems impossible. However, if this is happening (especially on
Windows 7 where a disproportionate number of these crashes are
happening) it would be nice to know for sure. Therefore, explicitly
crashing will be informative, one way or another.

Unfortunately the crash rate for this crash is extremely low on canary,
probably due to a population bias away from Windows 7. So, it may be a
while before we get any data.

Bug: 734002
Change-Id: I8c483593de718770b15ff43b699a3e8f2e08bfdc
Reviewed-on: https://chromium-review.googlesource.com/791826
Commit-Queue: Bruce Dawson <brucedawson@chromium.org>
Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519824}

https://chrome-internal.googlesource.com/chrome/src-internal.git/+log/ac0ab16a9d31..d638a44a623a

$ git log ac0ab16a9..d638a44a6 --date=short --no-merges --format='%ad %ae %s'

Created with:
  roll-dep src-internal


The AutoRoll server is located here: https://src-internal-chromium-roll.skia.org

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

If the roll is causing failures, please contact the current sheriff, who should
be CC'd on the roll, and stop the roller if necessary.


TBR=phajdan.jr@chromium.org

Change-Id: I5058eba9b741f305249a84a1642445ed89d412a9
Reviewed-on: https://chromium-review.googlesource.com/794371Reviewed-by: src-internal-roller . <src-internal-roller@chromium.org>
Commit-Queue: src-internal-roller . <src-internal-roller@chromium.org>
Cr-Commit-Position: refs/heads/master@{#519823}