GitLab

Tag a selection of system health stories chosen to have a wide coverage
for catching regressions, but run in a limited amount of time.

The stories chosen caught more than 80% of system health memory mobile
regressions in Q2 2018, and a single repeat of the entire set runs in
under 20 minutes.

Bug: 865959
Change-Id: I397cda35bb71dadf94bb2343c389d00cb8e80ec3
Reviewed-on: https://chromium-review.googlesource.com/1177396Reviewed-by: Timothy Dresser <tdresser@chromium.org>
Reviewed-by: Annie Sullivan <sullivan@chromium.org>
Reviewed-by: Ned Nguyen <nednguyen@google.com>
Commit-Queue: Juan Antonio Navarro Pérez <perezju@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583651}

This reverts commit 2b5dfbb0.

Reason for revert: Strong suspect for performance regression (https://pinpoint-dot-chromeperf.appspot.com/job/14a343e2640000), and disagreements on original CL implementation. Reverting until they can be resolved.

Bug: 872123, 869387

Original change's description:
> Add full saturate-to-infinity behavior to base::TimeDelta
>
> Before this CL TimeDelta had saturation behavior for translating
> from/to other values, but had clamping behavior for its computation.
> That is, invariants like the following did not hold:
>
> TimeDelta::Max() - TimeDelta::FromSeconds(1) == TimeDelta::Max()
>
> This CL fully implements saturation behavior, such that TimeDelta::Max()
> and TimeDelta::Min() are treated as positive and negative infinity
> regardless, with all the expected mathematical consequences.
>
> The one exception is that adding/subtracting a saturated TimeDelta to a
> TimeClass is still undefined. This was left undefined because we are not
> yet making a statement on whether TimeClass should be considered to have
> clamping or saturating behavior for values outsied of range.
>
> See the linked bug, or also the discussion at
> https://groups.google.com/a/chromium.org/forum/#!topic/platform-architecture-dev/S4umX2sOvrk
>
> Bug: 869387
> Change-Id: I2b3f98c69e7fc5030a6e210d75409d667b73c5ac
> Reviewed-on: https://chromium-review.googlesource.com/1155264
> Commit-Queue: Stephen McGruer <smcgruer@chromium.org>
> Reviewed-by: Gabriel Charette <gab@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#579969}

TBR=gab@chromium.org,smcgruer@chromium.org

# Not skipping CQ checks because original CL landed > 1 day ago.

Bug: 869387
Change-Id: I7251e77bd66dc95c9b1aae16fbfc89acd239f19b
Reviewed-on: https://chromium-review.googlesource.com/1175470
Commit-Queue: Stephen McGruer <smcgruer@chromium.org>
Reviewed-by: Stephen McGruer <smcgruer@chromium.org>
Reviewed-by: Gabriel Charette <gab@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583650}

Posted blocking ScriptStreamer tasks can end up in a queue, and have
their Resource complete loading before the task even starts. When this
happens, these tasks will no longer block on network input, but they
will still be posted to the blocking task runner, which does not spin up
new threads if the running task is not blocked. When there are a lot of
small functions posted for streaming, this results in a slightly
paradoxical starvation, where the blocking task runner is processing
lots of no-longer-blocking tasks all in one thread, as it has no reason
to spin up new threads.

To avoid this, if a streaming task isn't started before the resource
finishes loading, we can cancel this posted task, and instead post a new
non-blocking task to the non-blocking thread pool. This frees up the
blocking task runner to process only blocking tasks (and thus allows it
to spin up new threads for them).

Bug: chromium:865098
Bug: chromium:866868
Change-Id: Iceb282582781109ade233366d1009c3be50db3a5
Reviewed-on: https://chromium-review.googlesource.com/1174380
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Reviewed-by: Alexander Timin <altimin@chromium.org>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Gabriel Charette <gab@chromium.org>
Reviewed-by: Sami Kyöstilä <skyostil@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583649}

Current expectations assume a buggy behavior of LegacyNavigationManager,
which doesn't create a pending item on reload (crbug.com/676129). The
test initiates a navigation to the same page to abort the payment
request. This navigation is incorrectly classified as renderer-initiated
by LegacyNavigationManager, which causes the abort reason for payment
request to be incorrectly logged as EVENT_OTHER_ABORTED.

WKBasedNavigationManager fixes this bug, so the test should now check
EVENT_USER_ABORTED.

Bug: 874628
Cq-Include-Trybots: luci.chromium.try:ios-simulator-full-configs;master.tryserver.chromium.mac:ios-simulator-cronet
Change-Id: I74997df64464f196cfcfb34f21b0252d2f2f6e3e
Reviewed-on: https://chromium-review.googlesource.com/1176470
Commit-Queue: Danyao Wang <danyao@chromium.org>
Reviewed-by: Eugene But <eugenebut@chromium.org>
Reviewed-by: Moe Ahmadi <mahmadi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583648}

When building a framework for macOS that has public headers, this flag
is needed for the compiler to find the headers located inside the
framework bundle. This same cflag is set when linking with frameworks
in the build/config/ios/rules.gni template.

E.g. the header may be at `xyz.framework/Headers/x.h` and it is imported
with `#import <xyz/x.h>` in the source.

We need this for using the WebRTC macOS framework from a demo app.

Bug: webrtc:9627
Change-Id: Ib431bacfd85eb4597cb19e60ed575336b76b0598
Reviewed-on: https://chromium-review.googlesource.com/1177400Reviewed-by: Robert Sesek <rsesek@chromium.org>
Commit-Queue: Anders Carlsson <andersc@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583647}

A common pattern on the web is to resize a container to the window's
innerWidth and innerHeight so that it matches the viewport size and
accounts for the URL bar's presence. Unfortunately, these values are do
not report fractional CSS pixels, they're floored. So on high DPI
devices, innerHeight * devicePixelRatio != initial containing block's
size.

This patch allows applying the root scroller on pages following this
pattern by ceil'ing the bounds values before comparing them to the ICB.

Two additional unrelated changes:

- Remove unused parameter in FillsViewport

This is always true since some changes were made recently.

- Early out for non-main frames in ProcessImplicitCandidates

We avoid adding implicit candidates for non-main frames already so this
is a no-op but this early out is convenient for debugging and logging.

Bug: 838683
Change-Id: I740816184a26aa067e135041543170356e90fed5
Reviewed-on: https://chromium-review.googlesource.com/1176550Reviewed-by: Ella Ge <eirage@chromium.org>
Commit-Queue: David Bokan <bokan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583646}

FYI: I am using this opportunity to test the submission flow on
Chrome/Gerrit & co :)

R=ossu@chromium.org

Bug: None
Change-Id: I6d5897a3e64160843d80b423f02c91d68430cb82
Reviewed-on: https://chromium-review.googlesource.com/1177718Reviewed-by: Oskar Sundbom <ossu@chromium.org>
Commit-Queue: Armando Miraglia <armax@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583645}

This reverts commit 902e5475.

Reason for revert: <INSERT REASONING HERE>
Suspect causing layout test failure:
https://ci.chromium.org/buildbot/chromium.webkit/WebKit%20Linux%20Trusty%20ASAN/15204

Original change's description:
> [LayoutNG] Store NGPaintFragment in LayoutObject
> 
> This patch stores a one-way linked list of NGPaintFragment to
> their source LayoutObject. Looking up the HashMap consumes
> non-ignorable time during paint.
> 
> This patch adds IsInLayoutNGInlineFormattingContext flag to
> LayoutObject. This flag represents whether the object is in
> LayoutNG inline formatting context, even if the object itself
> is laid out by legacy.
> 
> With the flag, the storage used for storing InlineBox in
> LayoutText, LayoutInline, and LayoutBox are changed to union
> to store the first NGPaintFramgent.
> 
> Bug: 714962
> Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng;luci.chromium.try:linux_layout_tests_slimming_paint_v2;master.tryserver.blink:linux_trusty_blink_rel
> Change-Id: I7dfd11fafa9b1527b8d77e1fa3a24abf0e7f551e
> Reviewed-on: https://chromium-review.googlesource.com/1170726
> Commit-Queue: Koji Ishii <kojii@chromium.org>
> Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
> Reviewed-by: Emil A Eklund <eae@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#583633}

TBR=eae@chromium.org,kojii@chromium.org,mstensho@chromium.org

Change-Id: I9a870a209f86cfd061df8c82858b58093dff310b
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 714962
Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng;luci.chromium.try:linux_layout_tests_slimming_paint_v2;master.tryserver.blink:linux_trusty_blink_rel
Reviewed-on: https://chromium-review.googlesource.com/1177761Reviewed-by: Xida Chen <xidachen@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583644}

Makes callsites use mojom::blink::FileSystemType instead.

Bug: 787281
Change-Id: If0a3a687fa99b77b3a05ec8c4790f1ee0cbe6dd5
Reviewed-on: https://chromium-review.googlesource.com/1155157Reviewed-by: Kinuko Yasuda <kinuko@chromium.org>
Reviewed-by: Jeremy Roman <jbroman@chromium.org>
Commit-Queue: Jeremy Roman <jbroman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583643}

Mark virtual/new-remote-playback-pipeline/media/controls/modern/doubletap-to-jump-backwards-at-start.html
flaky on Mac

TBR=wjmaclean@chromium.org
NOTRY=true

Bug: 874866
Change-Id: Ie04f74b689e1bea6e7c51e5e30ba4a7ce70863fc
Reviewed-on: https://chromium-review.googlesource.com/1177633Reviewed-by: Xida Chen <xidachen@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583642}

Along the way 2 tests are fixed.

Bug: 831123, 852742
Change-Id: If8466bf72cb81be1edbf5f0b95f7da5d309e6f54
Reviewed-on: https://chromium-review.googlesource.com/1169170
Commit-Queue: Vadym Doroshenko <dvadym@chromium.org>
Reviewed-by: Robert Kaplow (slow) <rkaplow@chromium.org>
Reviewed-by: Vasilii Sukhanov <vasilii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583641}

Omnibox: Prevent ctrl-enter behavior when we believe the ctrl key was pressed in order to perform another action.

For example, if the user presses ctrl-a to select all the text in the omnibox, then presses enter without releasing ctrl, we do not perform ctrl-enter behavior. On the other hand, if the user releases and represses ctrl after performing a ctrl action, we do perform ctrl-enter behavior.

This cl does two changes for the above goal.
1. Previously, ctrl-enter was disabled when the omnibox is focused via ctrl-l but not when the omnibox was already focused and the user presses ctrl-l. Now, we disable ctrl-enter in both cases.
2. Previously, ctrl-enter was disabled if a ctrl action caused either the omnibox selection or text to change; e.g. ctrl-v when the clipboard is non-empty causes a text change, and ctrl-a when the omnibox is non-empty causes a selection change. However, ctrl-enter was not disabled if the ctrl-action did not cause a selction or text change; e.g. ctrl-c, ctrl-v when the clipboard is empty, ctrl-a when the entire omnibox text is already selected, ctrl-left when no omnibox text is selected, etc. Now we disable ctrl-enter if an action was executed, regardless of whether that action caused the omnibox text or selection to change. For example, if text selection is not empty, ctrl-c will disable ctrl-enter, but if selection is empty, ctrl-c will not disable ctrl-enter. Ctrl-left/right/home/end are the 1 notable exception, where they will disable ctrl-enter regardless of whether the cursor was already at either edge.

Bug: 75715

Change-Id: I8b2b4829697bf66855879366e4429494c3e95ab9
Reviewed-on: https://chromium-review.googlesource.com/1173424
Commit-Queue: manuk hovanesian <manukh@chromium.org>
Reviewed-by: Tommy Li <tommycli@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583640}

Monochrome is crashing on startup. Fix this by using the default
proguard instead of proguard 6.0.3 for monochrome. In order to stay
under the DEX method limit use multidex for monochrome release builds.

Bug: 874464, 868770
Change-Id: I9f002c26fb6e2b9b1b5d7471dab890090c446a75
Reviewed-on: https://chromium-review.googlesource.com/1176474
Commit-Queue: Tibor Goldschwendt <tiborg@chromium.org>
Reviewed-by: agrieve <agrieve@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583639}

go/cros-untangle shows that one reason why Chrome OS code depends on
all kinds of Chrome internals, is because it uses g_browser_process
to get hold of various Chroms OS globals, and g_browser_process
keeps a reference to all globals, Chrome and Chrome OS.

This CL removes a few easy-to-remove references to g_browser_process
from chrome/browser/chromeos.

Bug: 446937
Change-Id: I7ed4d1ccede3d0963ab86c80bba6f9327dd0cf61
Reviewed-on: https://chromium-review.googlesource.com/1174448
Commit-Queue: A Olsen <olsen@chromium.org>
Reviewed-by: Maksim Ivanov <emaxx@chromium.org>
Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583638}

These forms should not be filled. This CL implements that nor
PasswordFormManager nor NewPasswordFormManager is created for such forms
so as result they are not filled.

Bug:871191

Change-Id: I509bb6d1ca500ac2bbbea265712001f61bec3c53
Reviewed-on: https://chromium-review.googlesource.com/1174714
Commit-Queue: Vadym Doroshenko <dvadym@chromium.org>
Reviewed-by: Vasilii Sukhanov <vasilii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583637}

Change-Id: Id3e7dfd5a3e60f93ac6653eb7eb32af4a14b4b45
Reviewed-on: https://chromium-review.googlesource.com/1177608
Commit-Queue: Leonard Ge <wangge@google.com>
Commit-Queue: Ned Nguyen <nednguyen@google.com>
Reviewed-by: Ned Nguyen <nednguyen@google.com>
Reviewed-by: Juan Antonio Navarro Pérez <perezju@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583636}

The story should be independently disabled (on separate lines) for each
of Android_Go and Android_One.

TBR=nednguyen@google.com
NOTRY=true

Bug: 871708
Change-Id: Ifa5273b546738bf6e3162785590658671e8fe66f
Reviewed-on: https://chromium-review.googlesource.com/1177713Reviewed-by: Juan Antonio Navarro Pérez <perezju@chromium.org>
Commit-Queue: Juan Antonio Navarro Pérez <perezju@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583635}

TBR=wjmaclean@chromium.org
NOTRY=true

Bug: 874853
Change-Id: I0e28c54f5dfbfa0ae940b48ea5319914e6b49768
Reviewed-on: https://chromium-review.googlesource.com/1177684Reviewed-by: Xida Chen <xidachen@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583634}

This patch stores a one-way linked list of NGPaintFragment to
their source LayoutObject. Looking up the HashMap consumes
non-ignorable time during paint.

This patch adds IsInLayoutNGInlineFormattingContext flag to
LayoutObject. This flag represents whether the object is in
LayoutNG inline formatting context, even if the object itself
is laid out by legacy.

With the flag, the storage used for storing InlineBox in
LayoutText, LayoutInline, and LayoutBox are changed to union
to store the first NGPaintFramgent.

Bug: 714962
Cq-Include-Trybots: luci.chromium.try:linux_layout_tests_layout_ng;luci.chromium.try:linux_layout_tests_slimming_paint_v2;master.tryserver.blink:linux_trusty_blink_rel
Change-Id: I7dfd11fafa9b1527b8d77e1fa3a24abf0e7f551e
Reviewed-on: https://chromium-review.googlesource.com/1170726
Commit-Queue: Koji Ishii <kojii@chromium.org>
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Reviewed-by: Emil A Eklund <eae@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583633}

Add a test case to cover comment 13 in the linked bug.

Bug: 873178
Change-Id: I02fcb38b6737c79b41c32824bcfcc848e4cd79b3
Reviewed-on: https://chromium-review.googlesource.com/1177610Reviewed-by: Paul Jensen <pauljensen@chromium.org>
Commit-Queue: Paul Jensen <pauljensen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583632}

TBR=wjmaclean@chromium.org
NOTRY=true

Bug: 874853
Change-Id: I4d141e987d48df3ee4b4332c8341e5e5c7883301
Reviewed-on: https://chromium-review.googlesource.com/1177683
Commit-Queue: Xida Chen <xidachen@chromium.org>
Reviewed-by: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583631}

R8 doesn't parse the "" version, but is fine with ''.

Bug: 872904
Change-Id: I827798ac9b29031138397dcadde38567e3df63bb
Reviewed-on: https://chromium-review.googlesource.com/1176631Reviewed-by: Sam Maier <smaier@chromium.org>
Commit-Queue: agrieve <agrieve@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583630}

TBR=wjmaclean@chromium.org
NOTRY=true

Bug: 874851
Change-Id: I83cc51e272b418a942f324f3e0a408f7b05961d3
Reviewed-on: https://chromium-review.googlesource.com/1177629Reviewed-by: Xida Chen <xidachen@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583629}

Flakiness dashboard shows that it is flaky on all platform

TBR=wjmaclean@chromium.org
NOTRY=true

Bug: 859169
Change-Id: I5c9193cb1e36962c02b6c8d33739bc1687728ac1
Reviewed-on: https://chromium-review.googlesource.com/1177628Reviewed-by: Xida Chen <xidachen@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583628}

This CL removes the TabGrid implementation of BrowserCommands. This is
done such as the -openNewTab: method of BrowserCommands can be removed
in UI Refresh, merged away in the -openURL: method in
ApplicationCommands.

Bug: 861729
Cq-Include-Trybots: luci.chromium.try:ios-simulator-full-configs;master.tryserver.chromium.mac:ios-simulator-cronet
Change-Id: I001e387a3f4668ff2291c49ef83354888ea81a03
Reviewed-on: https://chromium-review.googlesource.com/1127898Reviewed-by: Mark Cogan <marq@chromium.org>
Commit-Queue: Gauthier Ambard <gambard@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583627}

We mark nodes as needing a layout tree rebuild pass to reconsider the
need for whitespace LayoutText objects if we remove any of its children.
However, we don't need that if all children were removed, if the node no
longer has a LayoutObject, or if the node was already marked for
re-attachment.

Change-Id: I74310936372320599b24f2d5146df855455bce1e
Reviewed-on: https://chromium-review.googlesource.com/1177381
Commit-Queue: Rune Lillesveen <futhark@chromium.org>
Reviewed-by: Anders Ruud <andruud@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583626}

https://webrtc.googlesource.com/src.git/+log/4eb4b05556ff..e9721f2f0830


git log 4eb4b05556ff..e9721f2f0830 --date=short --no-merges --format='%ad %ae %s'
2018-08-16 titovartem@webrtc.org Move SimulatedNetworkConfig on top level.
2018-08-16 danilchap@webrtc.org Remove deprecated fields from RtcpStatistic struct
2018-08-16 nisse@webrtc.org Refactor voe::Channel to not use RtpReceiver.


Created with:
  gclient setdep -r src/third_party/webrtc@e9721f2f0830

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

Change-Id: Ic6e1d945bb71aeb629d95e66a294bf7974b29ed8
Reviewed-on: https://chromium-review.googlesource.com/1177582Reviewed-by: webrtc-chromium-autoroll <webrtc-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Commit-Queue: webrtc-chromium-autoroll <webrtc-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#583625}

Currently, AsyncTask.execute() defaults to the SERIAL_EXECUTOR. This
exector is good for preventing concurrency errors since it guarantees
serial execution, but bad for performance since the entire app shares
this single queue.

We are hoping to get rid of execute() entirely, since it is often
assumed to be a parallel operation.

It looks like this callsite can use the THREAD_POOL_EXECUTOR instead,
since this use doesn't appear to rely on the concurrency guarantees that
SERIAL_EXECUTOR provides.

Bug: 869907
Change-Id: I51f22e371ec1e0d0e5114f4b211c934e214f5784
Reviewed-on: https://chromium-review.googlesource.com/1174900Reviewed-by: Simeon Anfinrud <sanfin@chromium.org>
Commit-Queue: Sam Maier <smaier@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583624}

This CL adds support for I420A VideoFrames in GMBVideoFramePool. These
VFrames are turned into BGRA or RGBA depending on the preferred platform
swizzling (spolier: BGRA is Mac).

Canvas Capture test is changed to premultiplied to follow up the Skia
convention.

Bug: 873345
Change-Id: I20d1f9f59e47a55b67763b2cb7884387f99118a5
Reviewed-on: https://chromium-review.googlesource.com/1174949
Commit-Queue: Miguel Casas <mcasas@chromium.org>
Reviewed-by: Emircan Uysaler <emircan@chromium.org>
Reviewed-by: Dale Curtis <dalecurtis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583623}

FirstChild did not include pseudo elements and some PseudoAware* methods
were already duplicated in the traversal methods.

Change-Id: I002384693edb188eda9ee6388aa191c4727fbc84
Reviewed-on: https://chromium-review.googlesource.com/1177393Reviewed-by: Anders Ruud <andruud@chromium.org>
Commit-Queue: Rune Lillesveen <futhark@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583622}

This mitigation shipped to stable in M61.
https://chromium-review.googlesource.com/518156
https://www.chromestatus.com/feature/5735596811091968

BUG=680970

Change-Id: I827ca161ecc7e3d7b75fc2774cee8e6d0ea04ed3
Reviewed-on: https://chromium-review.googlesource.com/1164967Reviewed-by: Mike West <mkwst@chromium.org>
Commit-Queue: Eric Willigers <ericwilligers@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583621}

TBR=wjmaclean@chromium.org
NOTRY=true

Bug: 874837
Change-Id: I1025816a10ffcc261a9fb21c434bf86b1f9a4dde
Reviewed-on: https://chromium-review.googlesource.com/1177625Reviewed-by: Xida Chen <xidachen@chromium.org>
Commit-Queue: Xida Chen <xidachen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583620}

This reverts commit e2c55011.

Reason for revert: Causes ash_unittests regressions. This is NOT
a speculative revert, as I've bisected to this commit and reproduced
it locally. Please do not reland as is.
Failing build: 
https://ci.chromium.org/p/chromium/builders/luci.chromium.ci/Linux%20Chromium%20OS%20ASan%20LSan%20Tests%20%281%29/28629

Original change's description:
> Reland "cros: Enable touchable app context menus by default."
> 
> This reverts commit 32cd8fcf.
> 
> Reason for revert: Reverting the revert because this change was proven to not have caused the issue.
> 
> Original change's description:
> > Revert "cros: Enable touchable app context menus by default."
> > 
> > This reverts commit b58ccf6d.
> > 
> > Reason for revert: Possibly introduced a use-after-free bug.
> > 
> > This is a speculative fix for a use-after-free bug in two ash_unittests:
> > 
> > 1. OverflowButtonInkDropTest.MouseContextMenu
> > 2. OverflowButtonActiveInkDropTest.MouseContextMenu
> > 
> > [ RUN      ] OverflowButtonInkDropTest.MouseContextMenu
> > =================================================================
> > ==28149==ERROR: AddressSanitizer: heap-use-after-free on address 0x613000031f68 at pc 0x000008c82106 bp 0x7ffea3cd7190 sp 0x7ffea3cd7188
> > READ of size 8 at 0x613000031f68 thread T0
> >     #0 0x8c82105 in begin buildtools/third_party/libc++/trunk/include/vector:1479:30
> >     #1 0x8c82105 in begin<std::__1::vector<aura::WindowObserver *, std::__1::allocator<aura::WindowObserver *> > > buildtools/third_party/libc++/trunk/include/iterator:1670
> >     #2 0x8c82105 in ContainsValue<std::__1::vector<aura::WindowObserver *, std::__1::allocator<aura::WindowObserver *> >, const aura::WindowObserver *, 0> base/stl_util.h:168
> >     #3 0x8c82105 in HasObserver base/observer_list.h:267
> >     #4 0x8c82105 in aura::Window::HasObserver(aura::WindowObserver const*) const ui/aura/window.cc:543
> >     #5 0x8c9d363 in aura::WindowObserver::OnUnobservingWindow(aura::Window*) ui/aura/window_observer.cc:25:15
> >     #6 0x8c799d7 in aura::Window::RemoveObserver(aura::WindowObserver*) ui/aura/window.cc:538:13
> >     #7 0xc2f40e in RemoveAll ui/base/window_tracker_template.h:49:15
> >     #8 0xc2f40e in ui::WindowTrackerTemplate<aura::Window, aura::WindowObserver>::~WindowTrackerTemplate() ui/base/window_tracker_template.h:33
> >     #9 0x9225f3b in ~CompoundEventFilter ui/wm/core/compound_event_filter.cc:49:1
> >     #10 0x9225f3b in wm::CompoundEventFilter::~CompoundEventFilter() ui/wm/core/compound_event_filter.cc:45
> >     #11 0x6b0ae30 in operator() buildtools/third_party/libc++/trunk/include/memory:2321:5
> >     #12 0x6b0ae30 in reset buildtools/third_party/libc++/trunk/include/memory:2634
> >     #13 0x6b0ae30 in ~unique_ptr buildtools/third_party/libc++/trunk/include/memory:2588
> >     #14 0x6b0ae30 in ash::Shell::~Shell() ash/shell.cc:969
> >     #15 0x6b0c81d in ash::Shell::~Shell() ash/shell.cc:737:17
> >     #16 0xc26b344 in ash::AshTestHelper::TearDown() ash/test/ash_test_helper.cc:294:3
> >     #17 0xc26444c in ash::AshTestBase::TearDown() ash/test/ash_test_base.cc:192:21
> >     #18 0x4724c64 in testing::TestInfo::Run() third_party/googletest/src/googletest/src/gtest.cc:2682:11
> >     #19 0x4726036 in testing::TestCase::Run() third_party/googletest/src/googletest/src/gtest.cc:2800:28
> >     #20 0x474b566 in testing::internal::UnitTestImpl::RunAllTests() third_party/googletest/src/googletest/src/gtest.cc:5124:43
> >     #21 0x474a7b5 in testing::UnitTest::Run() third_party/googletest/src/googletest/src/gtest.cc
> >     #22 0x72b431a in RUN_ALL_TESTS third_party/googletest/src/googletest/include/gtest/gtest.h:2331:46
> >     #23 0x72b431a in base::TestSuite::Run() base/test/test_suite.cc:277
> >     #24 0x72b9892 in Run base/callback.h:99:12
> >     #25 0x72b9892 in base::(anonymous namespace)::LaunchUnitTestsInternal(base::OnceCallback<int ()>, unsigned long, int, bool, base::OnceCallback<void ()>) base/test/launcher/unit_test_launcher.cc:224
> >     #26 0x72b9330 in base::LaunchUnitTests(int, char**, base::OnceCallback<int ()>) base/test/launcher/unit_test_launcher.cc:575:10
> >     #27 0x178df94 in main ash/test/ash_unittests.cc:37:10
> >     #28 0x7f78e2b62f44 in __libc_start_main /build/eglibc-ripdx6/eglibc-2.19/csu/libc-start.c:287
> > 
> > 0x613000031f68 is located 296 bytes inside of 328-byte region [0x613000031e40,0x613000031f88)
> > freed by thread T0 here:
> >     #0 0x6b9062 in operator delete(void*) /b/swarming/w/ir/kitchen-workdir/src/third_party/llvm/compiler-rt/lib/asan/asan_new_delete.cc:167:3
> >     #1 0x6a547a0 in ash::RootWindowController::CloseChildWindows() ash/root_window_controller.cc:498:7
> >     #2 0x6b0c5cf in ash::Shell::CloseAllRootWindowChildWindows() ash/shell.cc:1387:19
> >     #3 0x6b06637 in ash::Shell::~Shell() ash/shell.cc:837:3
> >     #4 0x6b0c81d in ash::Shell::~Shell() ash/shell.cc:737:17
> >     #5 0xc26b344 in ash::AshTestHelper::TearDown() ash/test/ash_test_helper.cc:294:3
> >     #6 0xc26444c in ash::AshTestBase::TearDown() ash/test/ash_test_base.cc:192:21
> >     #7 0x4724c64 in testing::TestInfo::Run() third_party/googletest/src/googletest/src/gtest.cc:2682:11
> >     #8 0x4726036 in testing::TestCase::Run() third_party/googletest/src/googletest/src/gtest.cc:2800:28
> >     #9 0x474b566 in testing::internal::UnitTestImpl::RunAllTests() third_party/googletest/src/googletest/src/gtest.cc:5124:43
> >     #10 0x474a7b5 in testing::UnitTest::Run() third_party/googletest/src/googletest/src/gtest.cc
> >     #11 0x72b431a in RUN_ALL_TESTS third_party/googletest/src/googletest/include/gtest/gtest.h:2331:46
> >     #12 0x72b431a in base::TestSuite::Run() base/test/test_suite.cc:277
> >     #13 0x72b9892 in Run base/callback.h:99:12
> >     #14 0x72b9892 in base::(anonymous namespace)::LaunchUnitTestsInternal(base::OnceCallback<int ()>, unsigned long, int, bool, base::OnceCallback<void ()>) base/test/launcher/unit_test_launcher.cc:224
> >     #15 0x72b9330 in base::LaunchUnitTests(int, char**, base::OnceCallback<int ()>) base/test/launcher/unit_test_launcher.cc:575:10
> >     #16 0x178df94 in main ash/test/ash_unittests.cc:37:10
> >     #17 0x7f78e2b62f44 in __libc_start_main /build/eglibc-ripdx6/eglibc-2.19/csu/libc-start.c:287
> > 
> > previously allocated by thread T0 here:
> >     #0 0x6b8422 in operator new(unsigned long) /b/swarming/w/ir/kitchen-workdir/src/third_party/llvm/compiler-rt/lib/asan/asan_new_delete.cc:106:3
> >     #1 0x91cb856 in views::NativeWidgetAura::NativeWidgetAura(views::internal::NativeWidgetDelegate*, bool, aura::Env*) ui/views/widget/native_widget_aura.cc:109:15
> >     #2 0x91d615d in views::internal::NativeWidgetPrivate::CreateNativeWidget(views::Widget::InitParams const&, views::internal::NativeWidgetDelegate*) ui/views/widget/native_widget_aura.cc:1113:14
> >     #3 0x917dabd in CreateNativeWidget ui/views/widget/widget.cc:73:10
> >     #4 0x917dabd in views::Widget::Init(views::Widget::InitParams const&) ui/views/widget/widget.cc:334
> >     #5 0x6af249a in ash::ShelfWidget::ShelfWidget(aura::Window*, ash::Shelf*) ash/shelf/shelf_widget.cc:216:3
> >     #6 0x6a96eaa in ash::Shelf::CreateShelfWidget(aura::Window*) ash/shelf/shelf.cc:73:27
> >     #7 0x6a5847a in ash::RootWindowController::InitLayoutManagers() ash/root_window_controller.cc:699:11
> >     #8 0x6a52166 in ash::RootWindowController::Init(ash::RootWindowController::RootWindowType) ash/root_window_controller.cc:663:3
> >     #9 0x68ef135 in ash::WindowTreeHostManager::InitHosts() ash/display/window_tree_host_manager.cc:238:3
> >     #10 0x6afadce in ash::Shell::Init(ui::ContextFactory*, ui::ContextFactoryPrivate*, std::__1::unique_ptr<base::Value, std::__1::default_delete<base::Value> >, std::__1::unique_ptr<ui::ws2::GpuInterfaceProvider, std::__1::default_delete<ui::ws2::GpuInterfaceProvider> >) ash/shell.cc:1250:30
> >     #11 0x6af7163 in ash::Shell::CreateInstance(ash::ShellInitParams) ash/shell.cc:276:14
> >     #12 0xc26ac31 in ash::AshTestHelper::CreateShell() ash/test/ash_test_helper.cc:390:3
> >     #13 0xc26a2a0 in ash::AshTestHelper::SetUp(bool, bool) ash/test/ash_test_helper.cc:245:3
> >     #14 0xc263f8d in ash::AshTestBase::SetUp() ash/test/ash_test_base.cc:159:21
> >     #15 0x1322b16 in ash::ShelfViewTest::SetUp() ash/shelf/shelf_view_unittest.cc:262:18
> >     #16 0x1326366 in SetUp ash/shelf/shelf_view_unittest.cc:2369:20
> >     #17 0x1326366 in ash::OverflowButtonInkDropTest::SetUp() ash/shelf/shelf_view_unittest.cc:2778
> >     #18 0x4722b42 in testing::Test::Run() third_party/googletest/src/googletest/src/gtest.cc
> >     #19 0x4724c64 in testing::TestInfo::Run() third_party/googletest/src/googletest/src/gtest.cc:2682:11
> >     #20 0x4726036 in testing::TestCase::Run() third_party/googletest/src/googletest/src/gtest.cc:2800:28
> >     #21 0x474b566 in testing::internal::UnitTestImpl::RunAllTests() third_party/googletest/src/googletest/src/gtest.cc:5124:43
> >     #22 0x474a7b5 in testing::UnitTest::Run() third_party/googletest/src/googletest/src/gtest.cc
> >     #23 0x72b431a in RUN_ALL_TESTS third_party/googletest/src/googletest/include/gtest/gtest.h:2331:46
> >     #24 0x72b431a in base::TestSuite::Run() base/test/test_suite.cc:277
> >     #25 0x72b9892 in Run base/callback.h:99:12
> >     #26 0x72b9892 in base::(anonymous namespace)::LaunchUnitTestsInternal(base::OnceCallback<int ()>, unsigned long, int, bool, base::OnceCallback<void ()>) base/test/launcher/unit_test_launcher.cc:224
> >     #27 0x72b9330 in base::LaunchUnitTests(int, char**, base::OnceCallback<int ()>) base/test/launcher/unit_test_launcher.cc:575:10
> >     #28 0x178df94 in main ash/test/ash_unittests.cc:37:10
> >     #29 0x7f78e2b62f44 in __libc_start_main /build/eglibc-ripdx6/eglibc-2.19/csu/libc-start.c:287
> > 
> > SUMMARY: AddressSanitizer: heap-use-after-free buildtools/third_party/libc++/trunk/include/vector:1479:30 in begin
> > Shadow bytes around the buggy address:
> >   0x0c267fffe390: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> >   0x0c267fffe3a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> >   0x0c267fffe3b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 fa
> >   0x0c267fffe3c0: fa fa fa fa fa fa fa fa fd fd fd fd fd fd fd fd
> >   0x0c267fffe3d0: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
> > =>0x0c267fffe3e0: fd fd fd fd fd fd fd fd fd fd fd fd fd[fd]fd fd
> >   0x0c267fffe3f0: fd fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
> >   0x0c267fffe400: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
> >   0x0c267fffe410: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
> >   0x0c267fffe420: fd fd fd fd fd fd fd fd fd fd fd fd fa fa fa fa
> >   0x0c267fffe430: fa fa fa fa fa fa fa fa fd fd fd fd fd fd fd fd
> > Shadow byte legend (one shadow byte represents 8 application bytes):
> >   Addressable:           00
> >   Partially addressable: 01 02 03 04 05 06 07 
> >   Heap left redzone:       fa
> >   Freed heap region:       fd
> >   Stack left redzone:      f1
> >   Stack mid redzone:       f2
> >   Stack right redzone:     f3
> >   Stack after return:      f5
> >   Stack use after scope:   f8
> >   Global redzone:          f9
> >   Global init order:       f6
> >   Poisoned by user:        f7
> >   Container overflow:      fc
> >   Array cookie:            ac
> >   Intra object redzone:    bb
> >   ASan internal:           fe
> >   Left alloca redzone:     ca
> >   Right alloca redzone:    cb
> >   Shadow gap:              cc
> > ==28149==ABORTING
> > 
> > Original change's description:
> > > cros: Enable touchable app context menus by default.
> > > 
> > > This will enable:
> > >  - New context menu UI for:
> > >   - Application icons
> > >   - Shelf context menus
> > >   - Desktop context menus
> > > 
> > > Note:
> > >  - Removed tests whose functionality were already being tested.
> > >    - MouseContextMenu already tests whether the ink drop is being
> > >      shown, so there is no need for
> > >      ShelfButtonShowsInkDropHighlightOnMenuShow.
> > > 
> > > Bug: 871843
> > > Change-Id: I4c92ae52f75b25bc9edd1dd67778c4785d8f4058
> > > Reviewed-on: https://chromium-review.googlesource.com/1170985
> > > Reviewed-by: Scott Violet <sky@chromium.org>
> > > Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
> > > Commit-Queue: Alex Newcomer <newcomer@chromium.org>
> > > Cr-Commit-Position: refs/heads/master@{#582978}
> > 
> > TBR=xiyuan@chromium.org,sky@chromium.org,newcomer@chromium.org
> > 
> > Change-Id: Ib85676861251f41c84f48f956d36c01edb353266
> > No-Presubmit: true
> > No-Tree-Checks: true
> > No-Try: true
> > Bug: 871843
> > Reviewed-on: https://chromium-review.googlesource.com/1174641
> > Reviewed-by: Chris Mumford <cmumford@chromium.org>
> > Commit-Queue: Chris Mumford <cmumford@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#583088}
> 
> TBR=xiyuan@chromium.org,sky@chromium.org,cmumford@chromium.org,newcomer@chromium.org
> 
> Change-Id: I4edeef479d40e1e591ac906aa89ede363668b6a3
> No-Presubmit: true
> No-Tree-Checks: true
> No-Try: true
> Bug: 871843
> Reviewed-on: https://chromium-review.googlesource.com/1174777
> Reviewed-by: Alex Newcomer <newcomer@chromium.org>
> Commit-Queue: Alex Newcomer <newcomer@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#583310}

TBR=xiyuan@chromium.org,sky@chromium.org,cmumford@chromium.org,newcomer@chromium.org

Change-Id: I8398745c9037fb5be605069646b3a655519d023e
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 871843
Reviewed-on: https://chromium-review.googlesource.com/1177482Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583619}

This cl fixes threading model: the proxy object is created on
a gpu main thread, whereas mojo ptr for a WaylandConnection is
set on a IO child thread. ScheduleSwap is also called on GpuMainThread,
but other methods can be called from IO threads.

To make a consitency, always use GpuMainThread's task runner.
This is also needed for a frame callback, which will be passed during
ScheduleSwap calls in a follow up cl.

Bug: 859012, 820047
Change-Id: I6833f3722cb4531da375387a7d443c1835f3c870
Reviewed-on: https://chromium-review.googlesource.com/1170827
Commit-Queue: Maksim Sisov <msisov@igalia.com>
Reviewed-by: Robert Kroeger <rjkroege@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583618}

The zip unpacker NaCl plugin compilation fails too often on ASAN; that
causes persistent time-out flakes. Disable these tests on ASAN only.

Bug: 867738
Change-Id: Ib370b0896afe13a3ebbd33b3d93db2d9f39b39b5
Reviewed-on: https://chromium-review.googlesource.com/1177401
Commit-Queue: Noel Gordon <noel@chromium.org>
Reviewed-by: Sam McNally <sammc@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583617}

It broke find-in-page.

Change-Id: Ic78f5efab61f1b4032fc549cee08e58daf1de7a8
Reviewed-on: https://chromium-review.googlesource.com/1176833
Commit-Queue: agrieve <agrieve@chromium.org>
Reviewed-by: Samuel Huang <huangs@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583616}

Makes the omnibox_util implementation file an Objective-C++ file and
provides better utils (reducing boilerplate at call sites).

Bug: None
Cq-Include-Trybots: luci.chromium.try:ios-simulator-full-configs;master.tryserver.chromium.mac:ios-simulator-cronet
Change-Id: I7f8d42fbe19f9fa65b8697d6d668e861d3b81671
Reviewed-on: https://chromium-review.googlesource.com/1148204
Commit-Queue: Stepan Khapugin <stkhapugin@chromium.org>
Reviewed-by: Justin Cohen <justincohen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583615}

input[type=file]: Move responsibility for PopupOpeningObserver registration from FileInputType to FileChooser.

We saw a DCHECK failure when a user dropped a directory to <input
type=file webkitdiretorty> because FileInputType::FilesChosen()
tried to unregister PopupOpeningObserver though
FileInputType::SetFilesFromDirectory() didn't call
ChromeClient::RegisterPopupOpeningObserver().

This CL moves the registration code from FileInputType to
FileChooser. It's difficult for FileInputType to know exact
timing of showing/hiding FileChooser dialogs because we have a
FileChooser queue in ChromeClientImpl. FileChooser knows it.
Because the FileChooserClient implementation is also a
PopupOpeningObserver implementation, this CL merges them into one
for code simplicity.

Bug: 637639
Change-Id: Ia9f64c212db635b3438431b4c61f06d257300839
Reviewed-on: https://chromium-review.googlesource.com/1177212Reviewed-by: Keishi Hattori <keishi@chromium.org>
Commit-Queue: Kent Tamura <tkent@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583614}

This allows long inline suggestions to not have an ellipsis, instead
they are clipped.

Bug: 439779
Cq-Include-Trybots: luci.chromium.try:ios-simulator-full-configs;master.tryserver.chromium.mac:ios-simulator-cronet
Change-Id: Ia6c47fd901eea2d5fcd04bcfcbc242f3a56bad01
Reviewed-on: https://chromium-review.googlesource.com/1148338Reviewed-by: Justin Cohen <justincohen@chromium.org>
Commit-Queue: Stepan Khapugin <stkhapugin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#583613}

This CL may cause a small binary size increase, roughly proportional
to how long it's been since our last AFDO profile roll. For larger
increases (around or exceeding 100KB), please file a bug against
gbiv@chromium.org. Additional context: https://crbug.com/805539

The AutoRoll server is located here: https://afdo-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=gbiv@chromium.org

Change-Id: I91de1f9e056bac35139c76e10dc48b7969a239c3
Reviewed-on: https://chromium-review.googlesource.com/1177622
Commit-Queue: afdo-chromium-autoroll <afdo-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Reviewed-by: afdo-chromium-autoroll <afdo-chromium-autoroll@skia-buildbots.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#583612}