GitLab

Bug: 785275
Change-Id: I30e401025d78ef5e9771c9ee11b3570b8968beb6
Reviewed-on: https://chromium-review.googlesource.com/888624Reviewed-by: Ramin Halavati <rhalavati@chromium.org>
Reviewed-by: Julian Pastarmov <pastarmovj@chromium.org>
Reviewed-by: Sergey Poromov <poromov@chromium.org>
Commit-Queue: Marton Hunyady <hunyadym@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532896}

Turns out pylib is needed to discover all the benchmarks in
benchmark_runner

Bug: 758632
Change-Id: I1b4d1106c8c32ab87b7431fa9331d67777a925a9
Reviewed-on: https://chromium-review.googlesource.com/893208Reviewed-by: Emily Hanley <eyaich@chromium.org>
Commit-Queue: Ashley Enstad <ashleymarie@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532895}

A couple of fuzzers that rely on url/ didn't initialize
ICU. Because url/ depends on ICU for IDN handling, this
results in an assertion failure when ICU's IDN class is
initialized.

Initialize ICU in a test enivornment for the following fuzzers:
  payment_method_manifest_parser_fuzzer
  csv_reader_fuzzer

In addition, print a human readable error message instead of a numeric
error code.

Bug: 754996
Test: The two fuzzers above do not fail the assertion about UIDNA.
Change-Id: I942d339d1ef1d71dd5c9c4e79a774f2c2980555d
Reviewed-on: https://chromium-review.googlesource.com/871350Reviewed-by: Mike West <mkwst@chromium.org>
Reviewed-by: Rouslan Solomakhin <rouslan@chromium.org>
Reviewed-by: Jay Civelli <jcivelli@chromium.org>
Reviewed-by: Vasilii Sukhanov <vasilii@chromium.org>
Reviewed-by: Abhishek Arya <inferno@chromium.org>
Commit-Queue: Abhishek Arya <inferno@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532894}

Some web sites return HTML page instead of PassKit object if request's
User Agent is not a known iOS browser.

Bug: 806759
Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Change-Id: I566fe85eadb889dd715ddf858b07b6cef4a736c5
Reviewed-on: https://chromium-review.googlesource.com/891760Reviewed-by: Sylvain Defresne <sdefresne@chromium.org>
Commit-Queue: Eugene But <eugenebut@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532893}

I believe this is a latent bug. NSURLReuest should reflect the actual
URL to be loaded in the web view, not the virtual URL that is meant to
be displayed to the user. Using virtual URL here breaks
WKBasedNavigationManagerImpl::Restore(), which creates a NavigationItem
whose URL is restore_session.html and whose virtual URL is the to-be-
restored page for display in omnibox.

The virtual URL usage was introduced in
https://codereview.chromium.org/2581193002. The CL description indicates
that virtual URL may not have been the correct value to use here in the
first place.

Bug: 807102
Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Change-Id: I5d40be25ee26e370ffa960b6acb2fb7e74ee91d6
Reviewed-on: https://chromium-review.googlesource.com/892285
Commit-Queue: Danyao Wang <danyao@chromium.org>
Reviewed-by: Eugene But <eugenebut@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532892}

This is allowing us to compute a MEI score during the warm up period
and cut down the warmup period if we know that the score would at worse
already be high enough.

For example, with minimum visits at 10 and MEI threshold at 3, after 3
visits with playback, a website could get MEI privilege instead of
waiting for the 7 following visits.

Bug: 807268
Change-Id: Id5f65f2d97375ed7507b99925459ab11139c2ae4
Reviewed-on: https://chromium-review.googlesource.com/893263Reviewed-by: Tommy Steimel <steimel@chromium.org>
Commit-Queue: Mounir Lamouri (slow) <mlamouri@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532891}

Expect
external/wpt/html/syntax/parsing/html5lib_tests16.html?run_type=uri to
flakily timeout on Mac and Linux (in addition to Win).

TBR=nednguyen@chromium.org,robertma@chromium.org

Bug: 803943
Change-Id: I20e5af250cccf8efec2673ee390a2eae892d4aea
Reviewed-on: https://chromium-review.googlesource.com/893399Reviewed-by: vitaliii <vitaliii@chromium.org>
Reviewed-by: Robert Ma <robertma@chromium.org>
Commit-Queue: vitaliii <vitaliii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532890}

A small cleanup and/or improvement to the way global
data is managed in the crazy linker library:

- Replace ScopedGlobalLock with ScopedLockedGlobals
  which allows reducing the number of Globals::Get()
  call performed at runtime.

- Provide property accessors to the globals that
  don't call Globals::Get() on each call, e.g.

    the static Globals::GetLibraries() is replaced
    by the non-static Globals::libraries() call.

- In crazy_linker_api.cpp, Fix a minor bug where
  the ScopedDelayedCallbackPoster is constructed
  before the global lock is acquired (i.e. a
  thread-race).

  Note that there are still other thread-related
  issues with deferred tasks / callback handling
  but these will be addressed in a future CL.

- Don't use heap allocation for the Globals
  instance storage. This is mostly because there
  were suspicion of invalid pointer addresses
  being passed through callbacks, causing runtime
  crashes (see http://crbug.com/796938 for more
  details). A fixed address in the .bss section
  would be easier to spot in minidump CPU register
  dumps.

BUG=801548
R=agrieve@chromium.org,pasko@chromium.org,rmcilroy@chromium.org

Change-Id: I68cba5360c5b5a3aaa7847172a6e73a87a64d44b
Reviewed-on: https://chromium-review.googlesource.com/893259Reviewed-by: agrieve <agrieve@chromium.org>
Reviewed-by: Egor Pasko <pasko@chromium.org>
Commit-Queue: David Turner <digit@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532889}

This CL reverts d8bd981f, in effect relanding the changes to have
RemoteSuggestionsFetcherImpl and its unittest use IdentityManager. In
the interim I have landed a fix to the bug that necessitated
reversion of those CLs, i.e. the raciness between IdentityManager
getting updated on signin/signout and other SigninManager::Observers
getting updated on those same events
(fix is here:
 https://chromium-review.googlesource.com/c/chromium/src/+/883347).

To test this change, do the following:
- Sign in to Chrome on Android (in a chrome_apk build if building locally)
- Observe that there are personalized suggestions on the NTP
- Sign out
- Observe that there are different, generic suggestions on the NTP

You can also more generally that the suggestions get updated on various
transitions between a signed-out state and a signed-in state.

TBR=gambard@chromium.org

Bug: 798413
Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Change-Id: I37d182718ef0b53a85be1a083f203aa0acdb3d8d
Reviewed-on: https://chromium-review.googlesource.com/888617
Commit-Queue: Colin Blundell <blundell@chromium.org>
Reviewed-by: Jan Krcal <jkrcal@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532888}

Earlier I fixed a crash in updateSelectionState by
returning early. That, however, isn't enough because
onSelectionState also calls isPictureTile, when it
is not safe to do so (because initialize hasn't been
called).

A better way is to move the early return up one
function in the callstack (from updateSelectionState
to onSelectionState), which should catch both of these
issues.

To be sure, I also checked if updateSelectionState could
be called outside of onSelectionState and found two variants:
1) It is called after initialize (when it is safe).
2) It is called after we call isPictureTile (e.g. in setChecked),
in which case we would already be seeing crash stacks for
that (but we're not).

Bug: 760785, 656015
Change-Id: I0f200b16cd34698e103dc4c2b07b3f86c948adad
Reviewed-on: https://chromium-review.googlesource.com/892886
Commit-Queue: Theresa <twellington@chromium.org>
Reviewed-by: Theresa <twellington@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532887}

https://skia.googlesource.com/skia.git/+log/7f861927fcd4..6da98f10951f

$ git log 7f861927f..6da98f109 --date=short --no-merges --format='%ad %ae %s'
2018-01-29 borenet Update DEPS to use googlesource mirrors

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=bungeman@chromium.org

Change-Id: I1f22855cdcd6699dc5752be4a31e6fda0c93b8ed
Reviewed-on: https://chromium-review.googlesource.com/893200Reviewed-by: skia-chromium-autoroll <skia-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Commit-Queue: skia-chromium-autoroll <skia-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#532886}

Adding missing isValid() checks from
https://codereview.chromium.org/2752113005/

Also removing mInitialized which can be misleading.

BUG=806724

Change-Id: I872b3ce27d6efb57dda65b6ae7b84abdf26c250c
Reviewed-on: https://chromium-review.googlesource.com/892106Reviewed-by: Bo <boliu@chromium.org>
Commit-Queue: Changwan Ryu <changwan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532885}

(removed constexpr for In*() methods found in earlier patch sets
as the inlining incurred too big of a code size increase)

Bug: 761570
Change-Id: Iced714ec5541fa67fcafe0c0dbbc56c9353418c1
Reviewed-on: https://chromium-review.googlesource.com/883126Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Yuri Wiitala <miu@chromium.org>
Commit-Queue: Gabriel Charette <gab@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532884}

This is a reland of 38feaec8.
It was reverted because of suspicion of causing red "git commit' failure builds.
However, that was due to a Gerrit bug. Hence, relanding.

Original change's description:
> [CQ.cfg] 50% of chromium_presubmit builds will now run on LUCI.
>
> R=dpranke@chromium.org, jbudorick@chromium.org
>
> No-Try: True
> Bug: 731553
> Change-Id: I12cd97bfee4d1cc7bf81f5c990014a27335a1c50
> Reviewed-on: https://chromium-review.googlesource.com/891666
> Reviewed-by: John Budorick <jbudorick@chromium.org>
> Commit-Queue: Andrii Shyshkalov <tandrii@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#532539}

Bug: 731553
No-Try: True
Change-Id: I89844ce9ae99ae2d2fadb19e6d0db5ab2f759415
Reviewed-on: https://chromium-review.googlesource.com/893538Reviewed-by: John Budorick <jbudorick@chromium.org>
Commit-Queue: Andrii Shyshkalov <tandrii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532883}

This is a reland of 5978ffa4.

Original change's description:
> Reland Group effects support in cc::AnimationPlayer
> 
> Patch https://chromium-review.googlesource.com/c/chromium/src/+/742162
> got reverted because it caused cc_unittest failure on CFI bot. This
> patch fixed the bug.
> 
> Commit message from the original patch:
> 
> The current cc/animations logic assumes a single animation has a single
> keyframe effect and can only affect a single layer. To enable animations
> with multiple keyframe effects, cc::AnimationPlayer need to support
> multiple AnimationTickers each corresponding to one keyframe effect.
> 
> Currently there is a 1:1 relationship between AnimationPlayer and
> AnimationTicker. This patch is to extend it to 1:n. Here is a summary of
> changes:
> - Introduce a sub-class of AnimationPlayer, a.k.a
> SingleTickerAnimationPlayer, to handle the existing logic (single
> effect). SingleTickerAnimationPlayer owns only one AnimationTicker as
> the AnimationPlayer does today.
> - Currently a AnimationTicker is created upon creating AnimationPlayer.
> In this patch, tickers are created separately and added to the player
> afterwards. Tickers that the player owns may belong to different targets
> therefore the player needs to coordinate with AnimationHost regarding
> this situation.
> - Adjust existing unit tests according to the changes above.
> 
> Bug: 767043
> Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel
> Change-Id: If21bad1285c35bbc048fef6b619c8272c0760551
> Reviewed-on: https://chromium-review.googlesource.com/890724
> Reviewed-by: Ian Vollick <vollick@chromium.org>
> Commit-Queue: Yi Gu <yigu@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#532519}

Bug: 767043
Change-Id: I5e6e7d796ff03cf76c3e2b0de3c4df5b506bf12d
Cq-Include-Trybots: master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel
Reviewed-on: https://chromium-review.googlesource.com/893338Reviewed-by: Ian Vollick <vollick@chromium.org>
Commit-Queue: Yi Gu <yigu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532882}

Summary of changes available at:
https://chromium.googlesource.com/v8/v8/+log/9ea6a24f..3c62a1ec

Please follow these instructions for assigning/CC'ing issues:
https://github.com/v8/v8/wiki/Triaging%20issues

Please close rolling in case of a roll revert:
https://v8-roll.appspot.com/
This only works with a Google account.

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;master.tryserver.chromium.android:android_optional_gpu_tests_rel

TBR=hablich@chromium.org,machenbach@chromium.org,kozyatinskiy@chromium.org,sergiyb@chromium.org

Change-Id: I9a3a4bfcbc3ec27290947976788471b768d0c9aa
Reviewed-on: https://chromium-review.googlesource.com/893278Reviewed-by: v8 autoroll <v8-autoroll@chromium.org>
Commit-Queue: v8 autoroll <v8-autoroll@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532881}

Fixed Widget::SetBoundsConstrained & ensured TouchSelectionMenuRunnerViews is clamped to the desktop bounds.

Bug: 791507
Change-Id: I3386fc2a93bee2c10ce98ffa3b2974ab40b9606a
Reviewed-on: https://chromium-review.googlesource.com/888259
Commit-Queue: Allen Bauer <kylixrd@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Sadrul Chowdhury <sadrul@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532880}

This CL was uploaded by git cl split.

R=bauerb@chromium.org

Bug: 177475
Change-Id: Idb94f3a156bf2855422b1f140ef7e2ae5598b60d
Reviewed-on: https://chromium-review.googlesource.com/891923Reviewed-by: Bernhard Bauer <bauerb@chromium.org>
Commit-Queue: Nico Weber <thakis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532879}

A forthcoming CL will add the symbols to the binary, but will also
enforce their ordering, so the orderfile has to contain these symbols
first.
Adding the symbols early is not an issue, as linkers ignore unkown
entries.

Bug: 758566
Change-Id: I6580738f754067ce74acecd5deb796da246bf57d
Reviewed-on: https://chromium-review.googlesource.com/893142Reviewed-by: Matthew Cary <mattcary@chromium.org>
Reviewed-by: Egor Pasko <pasko@chromium.org>
Commit-Queue: Benoit L <lizeb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532878}

Bug: 177475
R=reveman@chromium.org

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: I5466428563e9e1f785a0c977368a6861432e734f
Reviewed-on: https://chromium-review.googlesource.com/889723Reviewed-by: David Reveman <reveman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532877}

AVX2 version of H010ToAR30_Opt is 42% faster on Broadwell.
Add ABGRToAR30 fully optimized for AVX2 and SSSE3.
Add AR30ToABGR, unoptimized,

1000 iterations of 720p:
Was SSSE3 H010ToAR30_Opt (635 ms)
Was SSSE3 I420ToAR30_Opt (635 ms)

Now AVX2  H010ToAR30_Opt (448 ms)
Now AVX2  I420ToAR30_Opt (446 ms)

ffec313 ABGRToAR30 used AVX2 with reversed shuffler
ff8ab9b AR30ToABGR for 10 to 8 bit RGB on Android
ed96b7b AVX2 port of H010ToAR30_AVX2
c95fd57 AVX2 port of I010ToAR30_AVX2
3f43ecc Add H420ToAR30 and a test that does a histogram

Bug: 797347, libyuv:751
Test: try bots passed for local libyuv build
Change-Id: I2ea965b65b6757dc31ab1cdb706dc02c68afeab3
Reviewed-on: https://chromium-review.googlesource.com/892340
Commit-Queue: Miguel Casas <mcasas@chromium.org>
Reviewed-by: Miguel Casas <mcasas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532876}

Bug was closed as WontFix, so removing the corresponding commented
out code.

Bug: 228014
Change-Id: I9d4e389d319523b40b79b21ac94497ff2c0241d6
Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Reviewed-on: https://chromium-review.googlesource.com/893265Reviewed-by: Peter Lee <pkl@chromium.org>
Commit-Queue: Sylvain Defresne <sdefresne@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532875}

Bug: 177475
R=xlai@chromium.org

Change-Id: Iaf7e65f74a55a20591ee49809210ef00548beb2d
Reviewed-on: https://chromium-review.googlesource.com/889784Reviewed-by: Olivia Lai <xlai@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532874}

This is necessary to resolve a use-after-free issue: the CanvasFontCache is promptly freed on heap tear down but not unregistered from TaskObserver.


Change-Id: I0a048bfd8dcae79d57a882ec553eb4aa9877a77a
Reviewed-on: https://chromium-review.googlesource.com/893140Reviewed-by: Kentaro Hara <haraken@chromium.org>
Commit-Queue: Hannes Payer <hpayer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532873}

Fixes a bug where if we received png data and then failed to decode
that, we'd try to decode it as jpeg too unintentionally, cf
https://codereview.chromium.org/2943363003/diff/120001/chrome/renderer/chrome_render_frame_observer.cc

This CL was uploaded by git cl split.

R=thestig@chromium.org

Bug: 177475
Change-Id: I1fbfc4b8b1fa009e10ace632a09619f463305cb6
Reviewed-on: https://chromium-review.googlesource.com/891899
Commit-Queue: Nico Weber <thakis@chromium.org>
Reviewed-by: Lei Zhang <thestig@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532872}

Fixes a bug where device ordinal 5 would fall through to device ordinal 6
after resetting the name, cf
https://codereview.chromium.org/2440043002/diff/120001/device/bluetooth/test/bluetooth_test_bluez.cc

This CL was uploaded by git cl split.

R=ortuno@chromium.org

Bug: 177475
Change-Id: I93fd60b42f634c4b84c464452861f2cf3aa32f37
Reviewed-on: https://chromium-review.googlesource.com/891502
Commit-Queue: Nico Weber <thakis@chromium.org>
Reviewed-by: Giovanni Ortuño Urquidi <ortuno@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532871}

This CL fixes typo password_inputs->explicit_password_inputs. Which in some rare situations
leads to crash.

Bug: 805892
Change-Id: Ibe76806154e7172a4d433d51ec92fb20ba0d900c
Reviewed-on: https://chromium-review.googlesource.com/893298Reviewed-by: Maxim Kolosovskiy <kolos@chromium.org>
Commit-Queue: Maxim Kolosovskiy <kolos@chromium.org>
Commit-Queue: Vadym Doroshenko <dvadym@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532870}

This will allow next_protos_expected_in_ssl_config to test for
ssl_config.alpn_protos being empty.

Bug: 801564
Change-Id: I05ee7803b7ffb5add8ef0a16f0a00c7572967041
Reviewed-on: https://chromium-review.googlesource.com/891445
Commit-Queue: Bence Béky <bnc@chromium.org>
Reviewed-by: Adam Rice <ricea@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532869}

Will help understand why the DCHECK is hit.

Bug: 803056
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: Ibd00659e0ae1024cb863d90cd4e32e3f3f542a7c
Reviewed-on: https://chromium-review.googlesource.com/893318Reviewed-by: Max Morin <maxmorin@chromium.org>
Commit-Queue: Henrik Grunell <grunell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532868}

This is fixing a bug where methods returning all scores would
sometimes have two entries when the incognito profile received new MEI
related events. This is due to some unexpected behaviour from content
settings (in `GetSettingsForOneType`).

Bug: 807269
Change-Id: Id6b8be8ddd2dac6e1b1fec1922896d760e78d31f
Reviewed-on: https://chromium-review.googlesource.com/893260Reviewed-by: Tommy Steimel <steimel@chromium.org>
Commit-Queue: Mounir Lamouri (slow) <mlamouri@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532867}

The account consistency is now injected in the component by the embedder.
Chrome determines the current account consistency using the AccountConsistencyManager.

Note: After this CL, the meaning of the AccountConsistencyMethod values is slightly different:
- kDiceMigration means that the migration is ongoing but not completed yet.
- kDice means that Dice is enabled, either because it is forced or because the migration completed.

Bug: 777774
TBR: noyau, pkasting, treib, zea
Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Change-Id: I82a6464bb2fa76ca219a5fe419cdc26dfe9ffdfa
Reviewed-on: https://chromium-review.googlesource.com/787231
Commit-Queue: David Roger <droger@chromium.org>
Reviewed-by: Marc Treib <treib@chromium.org>
Reviewed-by: Mihai Sardarescu <msarda@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532866}

Restore code to clean up the test instance of the PolicyProvider, so unit tests
that set the PolicyProvider no longer crash when run in succession.

Bug: None
Test: Run unit_tests --gtest_filter=ProfilePolicyConnectorTest*
Change-Id: I3a90fb04f3736c7e1351791b52886d726913aeed
Reviewed-on: https://chromium-review.googlesource.com/893098Reviewed-by: Julian Pastarmov <pastarmovj@chromium.org>
Commit-Queue: Drew Wilson <atwilson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532865}

This CL was uploaded by git cl split.

R=sdefresne@chromium.org

Bug: 177475
Change-Id: Iae7090f66ed567d16e3c3174949a4399b3ca6ad9
Reviewed-on: https://chromium-review.googlesource.com/891879Reviewed-by: Sylvain Defresne <sdefresne@chromium.org>
Commit-Queue: Nico Weber <thakis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532864}

https://webrtc.googlesource.com/src.git/+log/f61b3ba65e3d..49456a5b3327

$ git log f61b3ba65..49456a5b3 --date=short --no-merges --format='%ad %ae %s'

Created with:
  roll-dep src/third_party/webrtc


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;master.tryserver.chromium.win:win-msvc-dbg
TBR=webrtc-chromium-sheriffs-robots@google.com

Change-Id: Ic8f028383bc7ceaa82be7c88a1b72a81f3bc849e
Reviewed-on: https://chromium-review.googlesource.com/893118
Commit-Queue: WebRTC Roll Bot <webrtc-autoroll@chromium.org>
Reviewed-by: WebRTC Roll Bot <webrtc-autoroll@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532863}

This helps with debugging, as the thrown exception will get mangled by
the test runner script.

Bug: 807237
Change-Id: I74677e491c0115521b9b79c051f5213e832545ae
Reviewed-on: https://chromium-review.googlesource.com/892863Reviewed-by: John Budorick <jbudorick@chromium.org>
Commit-Queue: Bernhard Bauer <bauerb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532862}

This CL adds a check in SoundContentSettingObserver to avoid unmuting a
tab if muted by the audio indicator.

Bug: 806482
Change-Id: I433c1fcb7a8e8034a9b100077e2f876e88a7d409
Reviewed-on: https://chromium-review.googlesource.com/891259Reviewed-by: Raymes Khoury <raymes@chromium.org>
Reviewed-by: Yuri Wiitala <miu@chromium.org>
Commit-Queue: Tommy Steimel <steimel@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532861}

Previously, RenderFrameHostImpl::DidCommitProvisionalLoad required that
a non-null |interface_provider_request| was supplied for all cross-
-document navigation commits if a real load had already committed in the
FrameTreeNode.

The check was performed for all commits, including those that the
DidCommitProvisionalLoad method itself later ignored, which led to
false-positive renderer kills in the following scenario with
browser-side navigation and site-per-process enabled:

 1) http://foo.test/main.html is loaded in a main frame.
 2) Either statically in HTML, or dynamically by JS, a frame is added:
      <iframe 'src=http://foo.test/f.html'/>.
 3) The initial byte of the response for `f.html` arrives, causing
    FrameMsg_CommitNavigation to be sent to the same renderer.
 4) Just before processing this message, however, `main.html` navigates
    the iframe to http://bar.test/g.html, which results in a mojom::Frame::
    BeginNavigation call being made to the RenderFrameHost.
 5) Suppose that immediately afterwards, `main.html` enters a busy-loop.
 6) The cross site navigation in the child frame starts, the first response
    byte arrives quickly, and thus the navigation commits quickly.
 6.1) FrameTreeNode::has_committed_real_load is set to true for the child.
 6.2) The same-site RenderFrame in the child FrameTreeNode is swapped out,
      i.e. FrameMsg_SwapOut is sent.
 7) The renderer for site instance `foo.test` exits from the busy loop,
    and starts processing messages in order:
 7.1) The first being processed is FrameMsg_CommitNavigation, so a
      provisional load is created and immediately committed to
      http://foo.test/f.html.
 7.2) Because at the time the same-site child RenderFrame was created,
      there had been no real load committed in the child frame, and because
      the navigation from the initial empty document to the first real
      document was same-origin, the global object is reused and the
      RemoteInterfaceProvider of the RenderFrame is not rebound.
 7.3) The obsoleted load in the same-site child frame commits, calling
      mojom::Frame::DidCommitProvisionalLoad, however, with
      |interface_provider_request| being null.
 8) RenderFrameHostImpl::DidCommitProvisionalLoad sees that a real load was
    already committed in the frame, but |interface_provider_request| is
    missing, so kills the renderer.
 9) RenderFrameHostImpl::DidCommitProvisionalLoad also sees that the frame
    was waiting for a swap-out ACK, so ignores the commit.

Instead, step 8 and step 9 should be reversed: RFHI::DidCommitProvisionalLoad
should first check if the frame already IsWaitingForUnloadACK(), and, if so,
return immediately, without checking |interface_provider_request|.

Bug: 793228
Change-Id: I3e833502ff43d603f1e063098df07a16b5d4b7ec
Reviewed-on: https://chromium-review.googlesource.com/877239
Commit-Queue: Balazs Engedy <engedy@chromium.org>
Reviewed-by: Nasko Oskov <nasko@chromium.org>
Reviewed-by: Łukasz Anforowicz <lukasza@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532860}

The toolbar coordinator shouldn't access implementation details like
the specific view managed by the location bar.
Instead, make it use location bar coordinator, and move the few
protocols that are mostly forwarded by TC to LB straight to LB.

Bug: none
Cq-Include-Trybots: master.tryserver.chromium.mac:ios-simulator-cronet;master.tryserver.chromium.mac:ios-simulator-full-configs
Change-Id: I244d39a065c7f0a03d9821b0c84fd5bddc421208
Reviewed-on: https://chromium-review.googlesource.com/883462
Commit-Queue: Stepan Khapugin <stkhapugin@chromium.org>
Reviewed-by: Gauthier Ambard <gambard@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532859}

This reverts commit 38feaec8.

Reason for revert: commit-git-patch flakily fails on LUCI, e.g. https://ci.chromium.org/p/chromium/builders/luci.chromium.try/chromium_presubmit/21881

Original change's description:
> [CQ.cfg] 50% of chromium_presubmit builds will now run on LUCI.
> 
> R=​dpranke@chromium.org, jbudorick@chromium.org
> 
> No-Try: True
> Bug: 731553
> Change-Id: I12cd97bfee4d1cc7bf81f5c990014a27335a1c50
> Reviewed-on: https://chromium-review.googlesource.com/891666
> Reviewed-by: John Budorick <jbudorick@chromium.org>
> Commit-Queue: Andrii Shyshkalov <tandrii@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#532539}

TBR=dpranke@chromium.org,tandrii@chromium.org,jbudorick@chromium.org

Change-Id: I5ea0844f22c10335ffb0de5a360d76a792442781
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 731553
Reviewed-on: https://chromium-review.googlesource.com/893418Reviewed-by: John Budorick <jbudorick@chromium.org>
Commit-Queue: John Budorick <jbudorick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#532858}

https://skia.googlesource.com/skia.git/+log/2f5b8d81c231..7f861927fcd4

$ git log 2f5b8d81c..7f861927f --date=short --no-merges --format='%ad %ae %s'
2018-01-30 robertphillips Revert "Drop support for unused MSAA extensions"

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=bungeman@chromium.org

Change-Id: I83a934681b38ec6526357ac1769298ea613235ff
Reviewed-on: https://chromium-review.googlesource.com/893199
Commit-Queue: skia-chromium-autoroll <skia-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Reviewed-by: skia-chromium-autoroll <skia-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#532857}