GitLab

https://chrome-internal.googlesource.com/chrome/src-internal.git/+log/a9c64739fb43..d565d557945e


Created with:
  gclient setdep -r src-internal@d565d557945e

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://skia-autoroll.corp.goog/r/src-internal-chromium-autoroll
Please CC jbudorick@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug

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

CQ_INCLUDE_TRYBOTS=luci.chrome.try:linux-chromeos-chrome
TBR=jbudorick@google.com

Bug: chromium:None
Change-Id: Ibf0aed3641d108989d87d3c7c0829819a495ec75
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1912280Reviewed-by: chromium-internal-autoroll <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com>
Commit-Queue: chromium-internal-autoroll <chromium-internal-autoroll@skia-corp.google.com.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#714569}

As of the 10.15 SDK IMP has changed from:
  id (*IMP)(id receiver, SEL selector, ...);
to:
  void (*IMP)(void);
thus requiring a cast to call it. Rather than having this cast at
every place that invokes an IMP, this change adds a new method to
ScopedObjCClassSwizzler to allow invoking the original method from
the swizzler, which accounts for every invocation of IMP in the tree.

TBR=avi@chromium.org

Bug: 973128
Change-Id: I13eab9a79bab23e9a2c3854fb95e42de61e04d3b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1906334Reviewed-by: Elly Fong-Jones <ellyjones@chromium.org>
Reviewed-by: Mark Mentovai <mark@chromium.org>
Commit-Queue: Elly Fong-Jones <ellyjones@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714568}

Bug: 1011921
Change-Id: I591df9b3e68aced4de83d675df36c0553ec31c9b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1906643
Commit-Queue: Jasper Chapman-Black <jaspercb@chromium.org>
Reviewed-by: Andrew Grieve <agrieve@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714567}

ChromeBrowserMainParts::PreMainMessageLoopRunImpl() may return
early with chrome::RESULT_CODE_NORMAL_EXIT_PROCESS_NOTIFIED in
browser tests. This prevents PreProfileInit() and other stages
from getting called, but PostMainMessageLoopRun() is still
called, so the code needs to be careful about shutting down
singletons initialized in PostMainMessageLoopRun() stages.

Bug: 702403
Change-Id: Icfc7dd93da6f78a3ccaf4e207c720fe2073fc1db
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1893376
Commit-Queue: Steven Bennetts <stevenjb@chromium.org>
Reviewed-by: James Cook <jamescook@chromium.org>
Reviewed-by: Mitsuru Oshima <oshima@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714566}

Generated using:
~~
directories="content chrome components"
regex="DisableForRenderFrameHost(\(([^\(\)]++|(?1))*\))"
label=$(grep -Pzoh $regex -r $directories \
         | tr -d '\000' \
         | grep -Po '\"[^"]+"' \
         | tr -d '"' \
         | sort -u
)

echo "<enum name=\"BackForwardCacheDisabledForRenderFrameHostReason\">"
for i in $label
do
  hash=$(echo -n $i | md5sum | tr -d ' -')
  hash=${hash^^}
  hash=$(echo $hash | head -c 16)
  hash=$(echo "ibase=16; $hash" | bc)
  hash=$(echo "($hash + (2^31)) % (2^32) - 2^31" | bc)
  echo "  <int value=\"$hash\" label=\"$i\"/>"
done
echo "</enum>"
~~

Bug: 1001087
Change-Id: Id2f11f608f7ef8341fb22c3befa56b9d5a28d6e7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1910234
Commit-Queue: Alexander Timin <altimin@chromium.org>
Reviewed-by: Alexander Timin <altimin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714565}

Bug: 991726
Change-Id: I5bdea1502d1b401dcbe79c0405f8745c226b36ec
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1911281
Auto-Submit: Mustafa Emre Acer <meacer@chromium.org>
Reviewed-by: Joe DeBlasio <jdeblasio@chromium.org>
Commit-Queue: Mustafa Emre Acer <meacer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714564}

https://chromium.googlesource.com/chromium/third_party/ffmpeg.git/+log/83304c4e5287..fa04e15e1ad6

$ git log 83304c4e5..fa04e15e1 --date=short --no-merges --format='%ad %ae %s'
2019-11-06 hans Call update.py --package=objdump instead of download_objdump.py

Created with:
  roll-dep src/third_party/ffmpeg

This also points ffmpeg correctly at the master branch rather than sushi branch.

TBR=tmathmeyer

Change-Id: I3ca918bcb38fd40784ed97ed8d14826e5656da6a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1910763Reviewed-by: Dale Curtis <dalecurtis@chromium.org>
Reviewed-by: Ted Meyer <tmathmeyer@chromium.org>
Commit-Queue: Dale Curtis <dalecurtis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714563}

When a new tab is created for loading a non-NTP page, skip waiting
for NTP loading. This decreases the chance of timeout waiting for
NTP loading on slower devices.

Bug: 983170
Change-Id: Ibf73a0664865aba2d10b77d12c4cccfef96ed615
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1911639Reviewed-by: Yue Zhang <yuezhanggg@chromium.org>
Commit-Queue: Wei-Yin Chen (陳威尹) <wychen@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714562}

Bug: None
Change-Id: I17ed1cdbd53033429dc4b045c4225e55f838e775
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1912459
Commit-Queue: Elly Fong-Jones <ellyjones@chromium.org>
Commit-Queue: Alexei Svitkine <asvitkine@chromium.org>
Reviewed-by: Alexei Svitkine <asvitkine@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714561}

https://webrtc.googlesource.com/src.git/+log/051cae5d11be..2bd2d85f65e8

git log 051cae5d11be..2bd2d85f65e8 --date=short --no-merges --format='%ad %ae %s'
2019-11-12 chromium-webrtc-autoroll@webrtc-ci.iam.gserviceaccount.com Roll chromium_revision 38612283..3b0967fe (714324:714463)
2019-11-12 mbonadei@webrtc.org Revert "Reland "Define WEBRTC_ENABLE_SYMBOL_EXPORT if is_component_build=true.""
2019-11-12 mbonadei@webrtc.org Reland "Define WEBRTC_ENABLE_SYMBOL_EXPORT if is_component_build=true."

Created with:
  gclient setdep -r src/third_party/webrtc@2bd2d85f65e8

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/webrtc-chromium-autoroll
Please CC webrtc-chromium-sheriffs-robots@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug

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

TBR=webrtc-chromium-sheriffs-robots@google.com

Bug: chromium:None
Change-Id: I9bcbf7d1022da9060167d96df3700eef37d54049
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1910974Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#714560}

Change-Id: Ie11119d9c018812a17ced065be764547e1dbebef
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1911919
Auto-Submit: Jiajun Ou <garzonou@google.com>
Reviewed-by: Gauthier Ambard <gambard@chromium.org>
Reviewed-by: Eugene But <eugenebut@chromium.org>
Commit-Queue: Eugene But <eugenebut@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714559}

Without it, the manifest merger assumes we're using a library that
targets SDK version 1 and adds some permissions we don't need.

Bug: None
Test: check manifest from CastShellInternal.apk
Change-Id: I98b4597a21e94e5c2d920943df4fd26bf97a556c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1909641Reviewed-by: Luke Halliwell (slow) <halliwell@chromium.org>
Reviewed-by: Simeon Anfinrud <sanfin@chromium.org>
Commit-Queue: Thoren Paulson <thoren@chromium.org>
Auto-Submit: Thoren Paulson <thoren@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714558}

BEFORE: SplitViewController::SnapWindow bails out if called during the
divider snap animation. Callers check for the divider snap animation
before calling SplitViewController::SnapWindow, because correct behavior
is never achieved just by allowing SplitViewController::SnapWindow to
silently fail.

AFTER: SplitViewController::SnapWindow has a DCHECK that fails during
the divider snap animation.

FYI: It behooves me to discuss why snapping a window during the divider
snap animation is problematic in the first place. See [1] for my
commentary about what can happen if the window becomes snapped and the
divider continues to animate. Another idea would be to interrupt the
divider snap animation, but that proposition is not as simple as it may
sound, because if the divider is animating to an edge of the work area,
then split view mode will end.

[1] https://chromium-review.googlesource.com/c/chromium/src/+/1480869/1/ash/wm/splitview/split_view_controller.cc

Bug: None
Change-Id: Ie457e53448ffe440c738f171e4c4b0d0affcf76c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1911704Reviewed-by: Mitsuru Oshima <oshima@chromium.org>
Commit-Queue: Avery Musbach <amusbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714557}

These test cases can be moved to unittests. Additionally, EG coverage
already exists for AlertCoordinator (which is used by DialogPresenter)
from the JavaScriptDialogTestCase tests.

Bug: 754642, 987646
Change-Id: I5e6d9c37f0603bf72bdf854d048381519b920aae
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1907203Reviewed-by: Eugene But <eugenebut@chromium.org>
Commit-Queue: Mike Dougherty <michaeldo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714556}

https://skia.googlesource.com/skia.git/+log/ad21d47cfa8d..7cb52cb5e67c

git log ad21d47cfa8d..7cb52cb5e67c --date=short --no-merges --format='%ad %ae %s'
2019-11-12 senorblanco@chromium.org Dawn: implement dynamic primitive processor texture handling.
2019-11-12 recipe-mega-autoroller@chops-service-accounts.iam.gserviceaccount.com Roll recipe dependencies (trivial).
2019-11-12 skia-autoroll@skia-public.iam.gserviceaccount.com Roll third_party/externals/swiftshader 31edef751a8d..9da287fd0264 (6 commits)
2019-11-12 jvanverth@google.com Detect whether trying to blit a Metal swapchain texture, and fail if so.
2019-11-12 senorblanco@chromium.org Dawn: fix bug in stencil handling.
2019-11-12 nigeltao@google.com Free SkWuffsCodec frame-count decoder earlier
2019-11-12 bungeman@google.com Track and force opsz axis on Mac.
2019-11-12 herb@google.com Make metrics const on SkStrike
2019-11-12 recipe-mega-autoroller@chops-service-accounts.iam.gserviceaccount.com Roll recipe dependencies (trivial).
2019-11-12 skia-autoroll@skia-public.iam.gserviceaccount.com Roll third_party/externals/angle2 012d15196023..e33c1582b4bc (6 commits)
2019-11-12 herb@google.com WS: SkStrike

Created with:
  gclient setdep -r src/third_party/skia@7cb52cb5e67c

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/skia-autoroll
Please CC halcanary@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug

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

CQ_INCLUDE_TRYBOTS=luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux-blink-rel;luci.chromium.try:linux-chromeos-compile-dbg;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel
TBR=halcanary@google.com

Bug: chromium:None,chromium:None
Change-Id: Iefe95942b57a99d293d1bbc2c4e32211660ce9a0
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1911664Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#714555}

Bug: 995052
Change-Id: If782994602e0b1b9d7b723be9417549b0eda125e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1902234Reviewed-by: Alice Boxhall <aboxhall@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Ben Wells <benwells@chromium.org>
Reviewed-by: Steven Holte <holte@chromium.org>
Commit-Queue: Anastasia Helfinstein <anastasi@google.com>
Cr-Commit-Position: refs/heads/master@{#714554}

The analysis and stack trace are in the filed bug.

Bug: 1022630
Change-Id: I58523a59f87e6b0fc9c642133f9c8e70427cea2e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1907268Reviewed-by: Tarun Bansal <tbansal@chromium.org>
Commit-Queue: Callum May <camay@microsoft.com>
Cr-Commit-Position: refs/heads/master@{#714553}

Change-Id: I1aaa91a35248932f5d9e3f4707cd168c3139d547
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1912087Reviewed-by: Sam Maier <smaier@chromium.org>
Commit-Queue: Andrew Grieve <agrieve@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714552}

https://chromium.googlesource.com/chromium/tools/depot_tools.git/+log/79d4f9950680..89863242a428

git log 79d4f9950680..89863242a428 --date=short --no-merges --format='%ad %ae %s'
2019-11-12 yangguo@chromium.org Add fetch configs for devtools-frontend

Created with:
  gclient setdep -r src/third_party/depot_tools@89863242a428

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/depot-tools-chromium-autoroll
Please CC agable@chromium.org on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug

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

TBR=agable@chromium.org

Bug: None
Change-Id: Ide21ab380a0f3eb1031487e15de18a0450e5cc3d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1911703Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#714551}

Translation screenshots project requires Chrome devs to upload UI
screenshots along with string changes in .grd and .grdp files. These
are XML files containing all user visible strings in Chrome.

Translation screenshots has a presubmit checks that extracts the
list of modified strings, looks for image files associated with these
strings and warns the developer to add them if they are missing.

This presubmit parses .grds to extract the list of modified strings but
it skips <part> tags which are used to reference .grdp files. As a
result, the presubmit currently ignores string changes in .grdp files.
.grdp files contain the majority of UI strings in Chrome, so the lack of
presubmit checks for them results in low translation screenshot coverage.

This CL changes how .grdp files are loaded in the presubmit: Previously,
they were written in a temporary directory alongside a fake .grd file
that referenced the loaded .grdp file. The code loaded the fake .grd file
which resulted in loading the strings in the actual .grdp file. However,
test mocks override all os.path methods which in turn breaks temporary
directory creation. As a result, it was not possible to properly test
the .grdp loading code, so we skipped <part> tags as a workaround.

The new loading code writes a temporary copy of the .grdp file and wraps
its contents with proper tags so that it can be loaded as a .grd file
instead. This avoids the need to create a temporary directory and is
fully testable.

The end result is that Translation Screenshots presubmit will now ask
Chrome devs to upload screenshots for changes in .grdp files. This should
significantly increase the coverage of strings with translation screenshots,
leading to better quality in string localizations.

Bug: 924652
Change-Id: Iae7933e8147cefdabc15ef21d2b6daa43d5002bc
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1875554Reviewed-by: anthonyvd <anthonyvd@chromium.org>
Reviewed-by: Jochen Eisinger <jochen@chromium.org>
Commit-Queue: Mustafa Emre Acer <meacer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714550}

All RadioButtonWithDescriptions are hosted under the same layout now.
Logic that injected the additional checkbox option now needs to account
for that change. Also leaving some explinations in there to make the
file a bit more descriptive.

Bug: 1020890
Change-Id: Id46e4d0ef56aee168d55f85882a4a100d872fde0
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1898710
Commit-Queue: Brandon Wylie <wylieb@chromium.org>
Reviewed-by: Theresa  <twellington@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714549}

Currently, a multiline label will in many cases report a minimum size
that can be smaller than the appropriate line-height for the text style,
causing vertical layouts to short the size of the text by a few pixels.

This results in, for example, a visual up-and-down wobble in the
domain label in a hover card as it animates between tabs, due to the
changing hover card size. This CL fixes this issue.

Note that except in vertical flex (and possibly box) layouts minimum
size tends to not be factored in, so this should only affect a handlful
of situations (and likely produce a better visual result).

Change-Id: Icdc6006b8f51588009b0709773f1854ecf5a6a2e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1910807
Commit-Queue: Dana Fried <dfried@chromium.org>
Reviewed-by: Peter Kasting <pkasting@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714548}

- Adds a flag to allow the AEQ UI toggle to be displayed
- UI only displays on devices that support AEQ

BUG=chromium:1021193
TEST=manual

Change-Id: Iaf3a4ef24ed2512cc7096adae40fcdacd7ac3cd5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1894163
Commit-Queue: Zentaro Kavanagh <zentaro@chromium.org>
Reviewed-by: James Cook <jamescook@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714547}

We were counting non-default suggestions with a relevance of 0, which
would later be removed anyways, in our logic to decide whether to
favor or squelch tail suggestions within a round of suggestions.
This CL now ignores them (unless they're the default, of course).

Bug: 1023821
Change-Id: If6a8a49cf6f2f53b487d2a18343af9bc038c33f7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1906963
Commit-Queue: Kevin Bailey <krb@chromium.org>
Reviewed-by: Tommy Li <tommycli@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714546}

BrowserFocusTest.FocusOnReloadCrashedTab is currently flakey on
REL CHROMEOS builds as well. It generally seems off to be
disabling the test on DBG but not REL, so disable it for both.

TBR=sky@chromium.org

Bug: 50025
Change-Id: I94b172f4b472aa361ffc2ba6557ebc12b14691c9
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1910843Reviewed-by: James MacLean <wjmaclean@chromium.org>
Commit-Queue: James MacLean <wjmaclean@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714545}

https://chromium.googlesource.com/breakpad/breakpad.git/+log/5915ea929c43..792e6b2143bf

$ git log 5915ea929..792e6b214 --date=short --no-merges --format='%ad %ae %s'
2019-11-12 sdefresne Remove dependency of uploader.mm on GTMLogger

Created with:
  roll-dep src/third_party/breakpad/breakpad

Bug: 1012115
Change-Id: I33c4f2b7a762be15df97ae7fe1d84ecffd6b317d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1912239
Commit-Queue: Sylvain Defresne <sdefresne@chromium.org>
Commit-Queue: Mark Mentovai <mark@chromium.org>
Auto-Submit: Sylvain Defresne <sdefresne@chromium.org>
Reviewed-by: Mark Mentovai <mark@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714544}

This reverts commit 1d267524.

Reason for revert: Suespect that this broke the build on Windows.

Original change's description:
> Added access operator to color mixers.
> 
> Added the access operator to the color mixer class, updated existing
> code and tests. Changed AddMixer() to return a mixer by reference.
> 
> Bug: 1887666
> Change-Id: I2ae180f77d73e3ce4567d061b5505d920677c450
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1910521
> Commit-Queue: Peter Kasting <pkasting@chromium.org>
> Reviewed-by: Peter Kasting <pkasting@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#714528}

TBR=pkasting@chromium.org,tluk@chromium.org

Change-Id: I7eae3b7e1824b31adaa04c5e998c883da05eea96
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 1887666
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1911709Reviewed-by: James MacLean <wjmaclean@chromium.org>
Commit-Queue: James MacLean <wjmaclean@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714543}

https://swiftshader.googlesource.com/SwiftShader.git/+log/371316798c01..197a5725f2fe

git log 371316798c01..197a5725f2fe --date=short --no-merges --format='%ad %ae %s'
2019-11-12 swiftshader.regress@gmail.com Regres: Update test lists @ 9da287fd

Created with:
  gclient setdep -r src/third_party/swiftshader@197a5725f2fe

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/swiftshader-chromium-autoroll
Please CC swiftshader-team+autoroll@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug

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

CQ_INCLUDE_TRYBOTS=luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_chromium_msan_rel_ng;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel
TBR=swiftshader-team+autoroll@google.com

Bug: None
Change-Id: Ieb054cde46164cdeb1c27291b11db110644139a6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1912159Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#714542}

Goes with spec CL https://github.com/w3c/csswg-drafts/pull/4437 (merged)
Eliminated the use of CSSPseudoElement from WebAnimations to coincide with
recent spec changes. JeyframeEffect targets now use an Element reference
and pseudo-element selector like getComputedStyle and css-animations events.

Change-Id: Ib6957a7b764eac3b4372ad35c1b79b139c356a19
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1894477Reviewed-by: Stephen McGruer <smcgruer@chromium.org>
Reviewed-by: Kevin Ellis <kevers@chromium.org>
Commit-Queue: George Steel <gtsteel@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714541}

Overdraw feedback has been broken for two reasons:
1. When we added explicit fences, we inserted the fence before the
  overdraw colors were drawn
2. When we changed how we do partial swap, we forgot to expand the rect
  where the overdraw feedback is drawn to.

This CL fixes those two issues moving the drawing of the overdraw
feedback at the end of DrawRenderPass.

Test: Run Chrome on nocture with the flag to enable overdraw feedback
Change-Id: I57069f8596d0bdb2bc819c8de5c70cf3901d7199
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1904681
Commit-Queue: Daniele Castagna <dcastagna@chromium.org>
Reviewed-by: kylechar <kylechar@chromium.org>
Reviewed-by: Andres Calderon Jaramillo <andrescj@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714540}

Since caret browsing mode is turned off by default in Chrome and the
detailed report does not show "enable-caret-browsing" among the command
line switches, the null reference read must be from the dereference of
frame.GetSettings() at the start of the method
MoveCommands::UpdateFocusForCaretBrowsing. Note that in earlier
patchsets in:

https://chromium-review.googlesource.com/c/chromium/src/+/1666567

there was a null check for frame.GetSettings():

https://chromium-review.googlesource.com/c/chromium/src/+/1666567/9

These were removed because of reviewer feedback that they were
unnecessary:

https://chromium-review.googlesource.com/c/chromium/src/+/1666567/9/third_party/blink/renderer/core/editing/commands/editor_command.cc#1057

However I see several places under blink\renderer\core\editing where a
null check on return value of Frame::GetSettings is performed. And
Frame::GetSettings has comment:

Settings* GetSettings() const;  // can be null

So the first part of my fix is to move the call to
GetCaretBrowsingEnabled to new method
LocalFrame::IsCaretBrowsingEnabled, which includes a null check on
Frame::GetSettings.

However, this simple initial attempt at a fix that was made before I had
a repro did not answer why an editing command was being executed in
response to an input event when the frame is detached. I finally did
find an easy repro, and used it as the basis for a new webtest.

With a repro in hand. even with the null check debug builds were
hitting a NOTREACHED crash in FrameSelection::DidSetSelectionDeprecated.
And this was a clue to the root cause. The call to SetFocusedNodeIfNeeded
dispatches focus events, and a script handler for these events can
remove the frame. In other words, it is the editing code itself that
is causing the frame to be detached. The fix here is to remove the
NOTREACHED macro and have DidSetSelectionDeprecated return early.
This follows the pattern of other methods in FrameSelection that call
IsAvailable to assure that the frame is still attached.

For trivial changes to local_frame.cc and local_frame.h:
TBR=dcheng@chromium.org

Bug: 1015710
Change-Id: If72e51eadbebbcf8879caccc5bd2c3480d9fbf39
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1869298
Commit-Queue: Bruce Long <brlong@microsoft.com>
Reviewed-by: Yoshifumi Inoue <yosin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714539}

- enable verbose logging.
- try skipping webdriver tests, which seem related to the timeouts.

Also proposing to add per-file owners for testing/scripts/run_wpt_tests.py
to enable quicker poking at these flags in the near term.

Bug: 1023835
Change-Id: I1d7b7b7f08adea6a8f33ff722458faf3bacacbde
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1906897
Commit-Queue: Luke Z <lpz@chromium.org>
Reviewed-by: Dirk Pranke <dpranke@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714538}

Following bot results are included.
253 254 255 261 262

TBR=yosin@chromium.org
NOTRY=true

Bug: 982194
Change-Id: Ib1d1a756c4c0d828d76781a35c5f7e00f3b08f3e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1911339Reviewed-by: Koji Ishii <kojii@chromium.org>
Commit-Queue: Koji Ishii <kojii@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714537}

Idiomatically, the dawn headers are included with angle-bracket style,
using dawn, dawn_native, etc. as the root. This is the include style
used in existing webgpu code in Chromium, eg. webgpu_decoder_impl.cc.
Refactor the DawnContextProvider and SkiaOutputSurfaceDawn code to match
this style.

Bug: 1021566
Change-Id: I3d462b432d763189350a31050ae334dd2bc56b84
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1911600Reviewed-by: Jonathan Backer <backer@chromium.org>
Commit-Queue: Sean Gilhuly <sgilhuly@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714536}

We have some special-code for tall unbreakable lines; if a line in
itself is taller than the fragmentainer, don't break before it like we
normally do when out of space. This piece of code was here for
compatibility reasons (or maybe just to work around another bug), but we
need to avoid this special behavior if the line is the one to break
before in order to honor a widows requirement. Otherwise we'd get
nowhere, stuck in a loop enternally re-laying out, attempting to break
before this line but refusing to do so.

Bug: 1022348
Change-Id: I2b9f702194dc0b7e6cbf8f37dffa95ea5722d045
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1911205Reviewed-by: Koji Ishii <kojii@chromium.org>
Reviewed-by: Ian Kilpatrick <ikilpatrick@chromium.org>
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714535}

https://chromium.googlesource.com/angle/angle.git/+log/b9b1bae4effb..3f647b1bf9a4

git log b9b1bae4effb..3f647b1bf9a4 --date=short --no-merges --format='%ad %ae %s'
2019-11-12 jmadill@chromium.org Vulkan: Improve Bresenham line emulation.

Created with:
  gclient setdep -r src/third_party/angle@3f647b1bf9a4

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/angle-chromium-autoroll
Please CC cnorthrop@google.com on the revert to ensure that a human
is aware of the problem.

To report a problem with the AutoRoller itself, please file a bug:
https://bugs.chromium.org/p/skia/issues/entry?template=Autoroller+Bug

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

CQ_INCLUDE_TRYBOTS=luci.chromium.try:android_optional_gpu_tests_rel;luci.chromium.try:linux_optional_gpu_tests_rel;luci.chromium.try:mac_optional_gpu_tests_rel;luci.chromium.try:win_optional_gpu_tests_rel
TBR=cnorthrop@google.com

Bug: None
Change-Id: Ia807404e27e41ae8b7ec74db796be455a17ca2e4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1910797Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#714534}

Prior to this CL, the month picker would continue to highlight the same month even after
the year is changed. The new behavior is to always highlight the first month when a year
is changed.

This change is validated by adding: date-picker-month-appearance-after-year-change.html.
The expected results for existing tests are also updated.

Bug: 1021331
Change-Id: Ib661e9b3607fd190a7753060a06a108ff53e8be0
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1907487
Commit-Queue: Ionel Popescu <iopopesc@microsoft.com>
Reviewed-by: Kent Tamura <tkent@chromium.org>
Reviewed-by: Mason Freed <masonfreed@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714533}

Bug: 1023397
Change-Id: Iaecffb245ca0da14a8a048966690534eb92f3046
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1912122Reviewed-by: Roman Sorokin [CET] <rsorokin@chromium.org>
Commit-Queue: Denis Kuznetsov [CET] <antrim@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714532}

We'll limit testing on the new builder to devtools-specific tests in a follow up CL.

Bug: 1018204
Change-Id: I45e7442258f4de47999fc1d0b26cd1b31b96e5ca
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1899502
Auto-Submit: Liviu Rau <liviurau@chromium.org>
Commit-Queue: Liviu Rau <liviurau@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714531}

Currently the parsing step of TestExpectations takes ~11 seconds on a
development machine, with the majority of that time in the
_collect_matching_tests() step, building the list of |matching_tests|.

This can be sped up considerably by changing _all_tests from a flat
list of test names to a dictionary tree of path components. Then
matching a directory of tests is a search in the tree + iterating the
subtree below the matching directory node, instead of a search through
the full list with substring matching.

This reduces the time locally for parsing TestExpectations from ~11 to
~3 seconds, saving ~72% of the time on this step.

With logging in web_tests/controllers/manager.py:

+        _log.debug("> build TestExpectations");
+        start = time.time()
         self._expectations = test_expectations.TestExpectations(self._port, test_names)
+        _log.debug("< build TestExpectations %fs", time.time()-start);

Before:
15:28:22.789 167015 < build TestExpectations 11.192820s

After:
11:23:38.824 11280 < build TestExpectations 3.105809s

On the windows try bot:

Before: https://chromium-swarm.appspot.com/task?id=4874d0c111ee7a10
11:48:32.359 11980 Parsing expectations ...
11:49:15.140 11980 Found 3676 tests (total 86570); running 3439, skipping 237.
= 43 seconds

After: https://chromium-swarm.appspot.com/task?id=4874f28bff322210
12:24:46.461 11648 Parsing expectations ...
12:25:22.805 11648 Found 3676 tests (total 86570); running 3439, skipping 237.
= 36 seconds

R=robertma@chromium.org

Bug: 982208
Change-Id: I04f5c3c1b09cf4f1b42ad501d624f83f6f378a97
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1906472Reviewed-by: Robert Ma <robertma@chromium.org>
Commit-Queue: danakj <danakj@chromium.org>
Cr-Commit-Position: refs/heads/master@{#714530}