GitLab

FPS on arm+hires is now ~60fps even with bg blur (w/o notifications)

* Use layer clip during animation
* Remove unnecessary background paint
* Fine tune layer structures: Remove unnecessary texture layer,
  and use smaller layer, or change so that it require less
  repainting.

Refactoring:
* Consolidate the shield/blur/rounded code into TrayBubbleView.
* Introduce layer type to buggle dialog delegate.

Fix:
* Looks like we were painting bg twice which made blur darker.
  It happens once in TrayBubbleView's layer.

Bug: 1033986
Test: no functional change.

Change-Id: I6b31513b04c55c851e468f621b2d18a3183abd56
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1963234Reviewed-by: Tim Song <tengs@chromium.org>
Reviewed-by: Michael Wasserman <msw@chromium.org>
Commit-Queue: Mitsuru Oshima <oshima@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724822}

Add WPT test for invisible in-viewport lazy images and clean up
invisible-image.tentative.html as well.

Bug: 956559
Change-Id: I26ba797fc8fcd3bd82f861c3b1578ab28fb93490
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1962261
Commit-Queue: Rob Buis <rbuis@igalia.com>
Reviewed-by: Dominic Farolino <dom@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724821}

https://chrome-internal.googlesource.com/chrome/src-internal.git/+log/780d83a4fe7e..7281de722b6d


Created with:
  gclient setdep -r src-internal@7281de722b6d

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
Bug: None
Tbr: jbudorick@google.com
Change-Id: Ic9b132c042b90851756a197aa978b08677a7a95f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967824Reviewed-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@{#724820}

Change the Network Service's UpdateCRLSet method to provide
a signal when a CRLSet has been applied. The Network Service
needs to dispatch parsing of the CRLSet onto a worker, as a
CRLSet may be large (up to 750kb)

Bug: 1032713
Change-Id: Ie759c1b47f3751a74be7941bd6d61085e55e1ecf
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1962505Reviewed-by: Maksim Orlovich <morlovich@chromium.org>
Reviewed-by: Tom Sepez <tsepez@chromium.org>
Reviewed-by: Mustafa Emre Acer <meacer@chromium.org>
Reviewed-by: Joshua Pawlicki <waffles@chromium.org>
Commit-Queue: Ryan Sleevi <rsleevi@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724819}

TBR=chasej@chromium.org

Change-Id: Ibf5bca4ed788bdd6d674aff7a125b4294b5f9113
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967737
Commit-Queue: Mason Freed <masonfreed@chromium.org>
Reviewed-by: Mason Freed <masonfreed@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724818}

TBR=kbr

Bug: 1034000
Change-Id: I602ae22cef6226791d3160cbd26a0c3ed8dc46fc
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967915Reviewed-by: Yuly Novikov <ynovikov@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Yuly Novikov <ynovikov@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724817}

* Remove pool.join() call (speeds up by 100ms on my machine)
* Use only 1 multiprocessing job for .info files (down from 3)
  * Saw no slowdown from this.
* Rename .errorprone.stamp -> .errorprone_stamp
  * So that autoninja speed breakdowns consider it a unique extension

Change-Id: I0461527fdc4471aefc2cf0eba986b28096488740
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967560
Commit-Queue: Andrew Grieve <agrieve@chromium.org>
Reviewed-by: Eric Stevenson <estevenson@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724816}

The first CL for tests folder move.

1. Copy the policy tests into chrome/test/enterprise with new folder layout
2. Add a readme.md file as place holder for the net folder
3. Add jxiang as the OWNER of these policy tests

Context: https://docs.google.com/document/d/1cHopqvLxnfhrQDPmrPEbalK8WAyQJyK91lrLqjFk1Os/edit?ts=5deec8b8#
Bug: 35021
Change-Id: I178042b76262bec762ad63013e4f11592c96dd6d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1963198
Auto-Submit: Xiang Ji <jxiang@google.com>
Reviewed-by: Scott Violet <sky@chromium.org>
Reviewed-by: Asanka Herath <asanka@chromium.org>
Reviewed-by: Julian Pastarmov <pastarmovj@chromium.org>
Commit-Queue: Xiang Ji <jxiang@google.com>
Cr-Commit-Position: refs/heads/master@{#724815}

Bug: 1033387
Change-Id: I835a09503aa156ae591b0069cae17846830f5b97
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967929
Auto-Submit: Jeremy Roman <jbroman@chromium.org>
Commit-Queue: Adithya Srinivasan <adithyas@chromium.org>
Reviewed-by: Adithya Srinivasan <adithyas@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724814}

We're working on separating out requests from tabs for different first
parties by keying all requests on NetworkIsolationKeys (NIKs). Internal
requests also need to use non-empty NIKs to avoid inadvertently leaking
information across NIKs, and so we can add checks to make sure all
requests have an associated NIK.

In this case, ProxyLookupRequest is not used often, so err on the side
of caution by using a new transient NIK for each request, which
prevents sharing cached network resources (hostnames, in the case of
proxy lookups).

Bug: 1021661
Change-Id: Ifc7a100e619d81bd89fa7025ace1766c570a6987
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1937859
Commit-Queue: Matt Menke <mmenke@chromium.org>
Reviewed-by: Steven Bennetts <stevenjb@chromium.org>
Reviewed-by: Eric Roman <eroman@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724813}

Some unittests will fail when the feature flags are changed to
"enabled by default". Those tests will be either removed or move to a
difference place in the future.

Original Request in crrev.com/c/1953637

Bug: 932818
Change-Id: Ibaf9d88504d335e824d78bb7680bb22a88e82243
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1966236
Commit-Queue: Siyu An <siyua@chromium.org>
Reviewed-by: David Roger <droger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724812}

Merge-With: eureka-internal/345005

Bug: internal b/142034629
Change-Id: Id83f41d4adb5408fee5a957f2e3001b88f2a8ce4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967871Reviewed-by: Yuchen Liu <yucliu@chromium.org>
Commit-Queue: Kenneth MacKay <kmackay@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724811}

https://pdfium.googlesource.com/pdfium.git/+log/ea5411d97d90..b2cb4336f2e5

git log ea5411d97d90..b2cb4336f2e5 --date=short --first-parent --format='%ad %ae %s'
2019-12-13 thestig@chromium.org Fix some nits in win32 fxge code.
2019-12-13 thestig@chromium.org Initialize more member variables in the header.
2019-12-13 thestig@chromium.org Add ParseSwitchKeyValue() to pdfium_test.cc.

Created with:
  gclient setdep -r src/third_party/pdfium@b2cb4336f2e5

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/pdfium-autoroll
Please CC pdfium-deps-rolls@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

Bug: None
Tbr: pdfium-deps-rolls@chromium.org
Change-Id: Ia9c2bd1f8ceda3a310d558056f09b4e54a503105
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967775Reviewed-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@{#724810}

This CL is
  1) merging the work done by
        EnsureLayoutTextUpdated(...) and
        EnsuresLayoutTextAttributeUpdated(...)
  2) simplifying the truncate text logic
        Only truncate text at the end.

EnsureLayoutTextUpdated(...) is copying text_ into layout_text_ and
it is performing codepoints rewriting. An invariant was assumed:
the amount of codepoints need to stay the same.

EnsuresLayoutTextAttributeUpdated(...) is building the layout break
lists. It is making a copy the text break list.

These two phases can be done at once. That way, there is no more needs
to keep the invariant on amount of codepoints. This is preparing the
code to the text eliding refactoring. Text eliding will break this
assumption.

The truncate logic is only used to avoid slow (huge) cases. The only
limit used in chrome is 10000.

The code for truncate was broken because it was possible to break
graphemes. Unittests are added to cover these cases. Also, of the
text is broken at the beginning or the middle, some invariants
are broken.

Example:
  bool RenderText::IsValidLogicalIndex(size_t index) const {
    return index == 0 || index == text().length() ||
        (index < text().length() &&
         (truncate_length_ == 0 || index < truncate_length_) &&
         IsValidCodePointIndex(text(), index));
  }
The truncated text is assumed to be at the end.

Bug: 1025561
Change-Id: Ifdee81c5a4527175258d0c9335cbe798e6d74775
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1948043
Commit-Queue: Etienne Bergeron <etienneb@chromium.org>
Reviewed-by: Robert Liao <robliao@chromium.org>
Reviewed-by: Peter Kasting <pkasting@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724809}

Still experiencing a lot of infra failures after crrev.com/c/1967566.

Bug: 950413
No-Try: true
Tbr: bpastene@chromium.org,agable@chromium.org
Change-Id: Id539191c5540fd9944ce368572ddb3d9c722f7c8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967878Reviewed-by: John Budorick <jbudorick@chromium.org>
Reviewed-by: Ben Pastene <bpastene@chromium.org>
Commit-Queue: John Budorick <jbudorick@chromium.org>
Auto-Submit: John Budorick <jbudorick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724808}

This reverts commit e401fb7e.

Reason for revert: Build failure on https://ci.chromium.org/p/chrome/builders/ci/android-internal-chromium-tot/45808

Original change's description:
> Replace org.chromium.base.Promise.Function with Function
> 
> Bug: 1034012
> Change-Id: I9dd122eca2169955f2ccf006dc039158be232d14
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1965440
> Commit-Queue: bttk <bttk@chromium.org>
> Reviewed-by: Yaron Friedman <yfriedman@chromium.org>
> Reviewed-by: Ted Choc <tedchoc@chromium.org>
> Reviewed-by: Ender <ender@google.com>
> Cr-Commit-Position: refs/heads/master@{#724788}

TBR=yfriedman@chromium.org,tedchoc@chromium.org,ender@google.com,bttk@chromium.org

Change-Id: I4437405f0c2e31caa5e22d2ad8ead5d346ff7a9d
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 1034012
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1965849Reviewed-by: Shakti Sahu <shaktisahu@chromium.org>
Commit-Queue: Shakti Sahu <shaktisahu@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724807}

Previously, if the caller had sent EOS but not yet received a response,
and the mixer connection was lost, it would get into a state where the
connection was left pending indefinitely (since we would not send EOS on
reconnect, and there was no signal to the caller). Instead, signal EOS
complete to the caller immediately if the connection is lost after EOS
was sent to the mixer.

Change-Id: I50a5adcd35e7c8c200e129076f5d2f274400a856
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967886Reviewed-by: Yuchen Liu <yucliu@chromium.org>
Commit-Queue: Kenneth MacKay <kmackay@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724806}

If RemoteFrameView::AttachToLayout is called during style resolution,
it might not have geometry information available. Rather than
propagating frame rects immediately, we should instead set the flag
to propagate frame rects during post-layout.

BUG=1033746

Change-Id: I962eb41e6ee1b382dc0e46800505eac97e0965de
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1966426Reviewed-by: Philip Rogers <pdr@chromium.org>
Commit-Queue: Stefan Zager <szager@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724805}

- Adds ClipboardItem property raw, and plumbing/support for this property in
  `navigator.clipboard.write` (ClipboardPromise and ClipboardWriter).
- Adds a `RawClipboard` flag, to ensure this feature isn't accidentally
  activated via experimental flags or otherwise.
- Adds tests, including VirtualTestSuite and tentative tests, to verify that
  this works with the flag enabled, but is correctly disabled without the
  flag enabled.


Bug: 897289
Change-Id: I453bb0ad18d9e52a4f2054d0f124fc3e8d5b430d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1927003
Commit-Queue: Darwin Huang <huangdarwin@chromium.org>
Reviewed-by: Matthew Denton <mpdenton@chromium.org>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Victor Costan <pwnall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724804}

This is needed in case metrics callbacks end up calling PostTask, which
isn't possible while the scheduler lock is held.

R=gab@chromium.org
BUG=1031236

Change-Id: I338f7c2ee3bd517937f8557c0f8123b4cb01ca97
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1963375
Commit-Queue: oysteine <oysteine@chromium.org>
Reviewed-by: Gabriel Charette <gab@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724803}

https://swiftshader.googlesource.com/SwiftShader.git/+log/468ee1576da7..fbbfeb700295

git log 468ee1576da7..fbbfeb700295 --date=short --first-parent --format='%ad %ae %s'
2019-12-13 chrisforbes@google.com Add missing source file to android build

Created with:
  gclient setdep -r src/third_party/swiftshader@fbbfeb700295

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
Bug: None
Tbr: swiftshader-team+autoroll@google.com
Change-Id: Icdb02c1bad1cae5e162967ae64047b3f92a3a8f6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1966736
Commit-Queue: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Reviewed-by: chromium-autoroll <chromium-autoroll@skia-public.iam.gserviceaccount.com>
Cr-Commit-Position: refs/heads/master@{#724802}

This patch updates CSSRenderSubtree to be experimental, which makes
it enabled if the experimental web platform features are enabled.

R=chrishtr@chromium.org

Change-Id: I225d849983e16ff46758322dcfc2ab047fd8489d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967876
Auto-Submit: vmpstr <vmpstr@chromium.org>
Commit-Queue: Chris Harrelson <chrishtr@chromium.org>
Reviewed-by: Chris Harrelson <chrishtr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724801}

Change-Id: I67edd64b4042cb518a4206c8aada8ed5a7802a42
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1956007
Commit-Queue: Zhongyi Shi <zhongyi@chromium.org>
Reviewed-by: Steven Holte <holte@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724800}

This CL changes the inactivity logic of the global media controls to
not actually stop the session when inactive, but instead just hide the
session. This fixes an issue where the media keys would not work after
inactivity.

This CL also ensures that sessions inside overlay notifications don't
time out for inactivity.

Bug: 1032179
Change-Id: Ic066c28d8c25738452acc96059abbef2521487a1
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1955335
Commit-Queue: Tommy Steimel <steimel@chromium.org>
Reviewed-by: Mounir Lamouri <mlamouri@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724799}

Change-Id: Ic3d0154e12e4e81063725c51832584bd70f9a31f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1965845
Commit-Queue: Bence Béky <bnc@chromium.org>
Commit-Queue: Ryan Hamilton <rch@chromium.org>
Auto-Submit: Bence Béky <bnc@chromium.org>
Reviewed-by: Ryan Hamilton <rch@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724798}

This moves the shareable parts of Chrome's modal dialog java code, such
as AppModalPresenter and ModalDialogView. It also moves the applicable
tests to the new component.

. Modal dialog specific styles are moved to //components/modaldialog.
. Some generic/widely used styles, dimens and colors are moved from
  //chrome/android/java/res to //components/styles/android/java/res.
. TabModalPresenter is left behind because tabs are a Chrome concept.
. Some test code has non-trivial updates, mainly to cut dependencies
  to ChromeActivity.
. Some of the tests depend on android_chrome_strings, but the tests
  are updated to use testing strings instead, so the question of where
  to put shared strings (a components module?) is punted.
. ModalDialogViewRenderTest.java is left behind due to dependencies.

Bug: 1030785
Change-Id: I3376817f8ec4decc74e59fcdc3feaa5cde6cdfa3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1962642Reviewed-by: Ted Choc <tedchoc@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724797}

InfobarOverlayType is used to describe the types of OverlayRequest
configurations that will be supported for a single InfoBar.  These are
used to show overlays for the banner UI (kBanner), the action sheet
displayed in Messages 2.0 showing detailed options (kDetailSheet), and
the modal UI used to update settings for that infobar (kModal).

InfobarOverlayRequestFactory was updated to take an InfobarOverlayType
in its factory method to allow the creation of OverlayRequests
configured for each of these overlay types for a given InfoBar.
InfobarOverlayRequestFactoryImpl was updated to use templated helpers
to create requests instead of relying on switch statements and boolean
logic to decide which configuration to use for each InfobarType
and InfobarOverlayType pair.

Bug: 1030357
Change-Id: Iaba54c95dfece2381164ccf8a615b8fda28c42d7
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1959936
Commit-Queue: Kurt Horimoto <kkhorimoto@chromium.org>
Reviewed-by: Sergio Collazos <sczs@chromium.org>
Reviewed-by: Rohit Rao <rohitrao@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724796}

R=liaoyuke,sajjadm,jbudorick

Bug: 1034018
Change-Id: Ib6473e949a92926b2678677b59334bef35c7917a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967795
Commit-Queue: Roberto Carrillo <robertocn@chromium.org>
Reviewed-by: John Budorick <jbudorick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724795}

A number of things came together to make the existing tests pass while
the much of the verification was just smoke and mirrors.

- The tests historically call WaitForPendingRead() to verify that the
  Renderer attempts to refill its buffer after its been depleted below
  "full". But the initial Preroll() always left a read "pending"
  because DecoderStream always does one last read past the "full"
  point to queue into its "ready outputs". Most the WaitForPendingRead()
  calls actually just hit this lingering unstatisfied read.

- Fixing Preroll() (really DeliverRemainingAudio()) to actually satisfy
  all pending reads (including DemuxerStream's) revealed a number of
  tests were faking themselves out.

  A typical test pattern was to partially deplete the algorithm buffer
  with a call to ConsumeBufferedData(256). 256 seems like a nice chunk
  since that's what we're sending on the input side see
  DeliverRemainingAudio), BUT these tests use buffer converter, and the
  output frames per buffer was much much larger such that:
    - 6 inputs of 256 frames become 3 outputs of 992, 992, and 1024
      frames
    - the buffer capacity is at 2000, so that third output overfills the
      buffer by over 1000 frames

  Hence consuming 256 isn't nearly enough to cause a pending read on its
  own (the buffer is still quite full). These tests were updated to use
  a new method... ConsumeBufferedDataUntilNotFull(), which will always
  leave the buffer 3/4's full irrespective of how over-filled it may
  start.

- Fixing Preroll() also reveled that many tests call
  WaitForPendingRead() without actually doing anything to trigger a
  pending read. I removed this expectation from those tests.

- Other misclenaious typos and cleanups.

Change-Id: I8eeb715530c0e0ed2814a0f551635e08dbafa8e4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1966020Reviewed-by: Dan Sanders <sandersd@chromium.org>
Commit-Queue: Chrome Cunningham <chcunningham@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724794}

This change adds a call to TraceEvent.setATraceEnabled at a point before
AwBrowserProcess.start() is called so that native startup chromium
traces are captured by atrace. This allows us to turn off chrome tracing
in the system_health.webview_startup benchmark since having both atrace
and chrome tracing on causes incorrect event nesting (see
https://crbug.com/1027261 for more context).

Bug: 1027261
Change-Id: I3435bc2f6a8e4cd340d3d40d70709ecc1ada5c06
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1962632Reviewed-by: Annie Sullivan <sullivan@chromium.org>
Reviewed-by: Changwan Ryu <changwan@chromium.org>
Commit-Queue: Oksana Zhuravlova <oksamyt@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724793}

On permission parsing we no longer collect unsupported optional
permissions into the optional permission list for an extension. This
means we no longer need to give unsupported optional errors for
permissions.request or permissions.remove, as those will be caught by
other errors now.

Bug: 909093
Change-Id: I351c26dafd20d1beb427a97abdae85c70745e3df
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1963273Reviewed-by: Devlin <rdevlin.cronin@chromium.org>
Commit-Queue: Tim Judkins <tjudkins@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724792}

Change-Id: I2398c0c1f95e58423db2ff871aa2e5b70f69699c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1967834
Commit-Queue: Bence Béky <bnc@chromium.org>
Commit-Queue: Ryan Hamilton <rch@chromium.org>
Auto-Submit: Bence Béky <bnc@chromium.org>
Reviewed-by: Ryan Hamilton <rch@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724791}

https://skia.googlesource.com/skia.git/+log/a6069a154d66..1cc767bd0d91

git log a6069a154d66..1cc767bd0d91 --date=short --first-parent --format='%ad %ae %s'
2019-12-13 mtklein@google.com workaround?
2019-12-13 kjlubick@google.com [infra] Update go to 1.13.5
2019-12-13 bungeman@google.com Add more variation support on Mac.
2019-12-13 herb@google.com Rename viewMatrix to drawMatrix, initialMatrix, or currentMatrix

Created with:
  gclient setdep -r src/third_party/skia@1cc767bd0d91

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 djsollen@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
Bug: None
Tbr: djsollen@google.com
Change-Id: Ic0dc7decc9ba05c5632bb8e908a9a7ffe3eed427
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1965719Reviewed-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@{#724790}

NTP process allocation is special:
*) NTP tabs share the same renderer process, because NTP opts into
   process-per-site behavior - see
   ChromeContentBrowserClient::ShouldUseProcessPerSite.
*) NTP tabs might require an "instant" renderer (renderer launched with
   a special cmdline switches - switches that are not provided by the
   spare renderer process).

Because of the above, NTP cannot claim/take the spare renderer process.
Even if the spare is claimed/taken early during navigation to NTP, the
process will have to be thrown away before the navigation commits.  Such
behavior leads to undesirable churn, as pointed out in
https://crbug.com/1029345#c7 and #c6.

Before this CL, the spare renderer process was incorrectly, temporarily
taken when creating a new tab.  The root of the problem was that
GetSiteInstanceForNewTab would create a SiteInstance for an incorrect
URL, because it wouldn't take into account URL rewriting that happens
during main frame navigations via content::BrowserURLHandler.  In this
particular case, a navigation rewrites chrome://newtab/ into
chrome-search://local-ntp/local-ntp.html.
*) SiteInstance creation for a new tab is happening in the following
   callstack:
     content::SiteInstanceImpl::CreateForURL()
     content::SiteInstance::CreateForURL() <- wrong |url| before this CL
     tab_util::GetSiteInstanceForNewTab()
     Navigate()
     chrome::AddTabAt()
     chrome::NewTab()
     ...
*) URL rewriting during an ntp navigation happens in the following
   callstack:
     search::HandleNewTabURLRewrite()
     content::BrowserURLHandlerImpl::RewriteURLIfNecessary()
     content::NavigationControllerImpl::CreateNavigationEntry()
     content::NavigationControllerImpl::CreateNavigationEntryFromLoadParams()
     content::NavigationControllerImpl::NavigateWithoutEntry()
     content::NavigationControllerImpl::LoadURLWithParams()
     (anonymous namespace)::LoadURLInContents()
     Navigate()
     chrome::AddTabAt()
     chrome::NewTab()
     ...

Fixing GetSiteInstanceForNewTab is sufficient to fix the bug and get rid
of the unnecessary churn of the spare renderer process.  This CL also
opportunistically adds process-per-site considerations to
SpareRenderProcessHostManager::MaybeTakeSpareRenderProcessHost to ensure
that other process-per-site cases also avoid needlessly/temporarily
taking/claiming the spare renderer process.

Bug: 1029345
Change-Id: I2b966d50fcec7ed06740fd9171716d792629c5d3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1948488
Commit-Queue: Łukasz Anforowicz <lukasza@chromium.org>
Reviewed-by: Avi Drissman <avi@chromium.org>
Reviewed-by: Peter Kasting <pkasting@chromium.org>
Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724789}

Bug: 1034012
Change-Id: I9dd122eca2169955f2ccf006dc039158be232d14
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1965440
Commit-Queue: bttk <bttk@chromium.org>
Reviewed-by: Yaron Friedman <yfriedman@chromium.org>
Reviewed-by: Ted Choc <tedchoc@chromium.org>
Reviewed-by: Ender <ender@google.com>
Cr-Commit-Position: refs/heads/master@{#724788}

Remove org.chromium.chromecast.base.Function

Bug: 1034012
Change-Id: Ib9cc34f95f3b4d37ac5d00f3ba1229bda3656326
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1966492
Commit-Queue: bttk <bttk@chromium.org>
Reviewed-by: Luke Halliwell (slow) <halliwell@chromium.org>
Reviewed-by: Ender <ender@google.com>
Cr-Commit-Position: refs/heads/master@{#724787}

Remove interface org.chromium.chrome.browser.feed.library.common.functional.Function

Bug: 1034012
Change-Id: I13a0f53753ca7cffe2dece46197e1fc663c4f009
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1966422Reviewed-by: Dan H <harringtond@chromium.org>
Reviewed-by: Ender <ender@google.com>
Commit-Queue: bttk <bttk@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724786}

This is a reland of 1d038c5d

Original change's description:
> Add eligibility check for Quick Answers.
>
> Bug: b/145774037
> Test: Manual
> Change-Id: I81b3d2b7d20cb600db09a1d60bad7ce8c64d9b7b
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1955028
> Commit-Queue: Li Lin <llin@chromium.org>
> Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#724027}

Bug: b/145774037
Change-Id: If9ca0979459e3813c90c1939982bb6ddb404f440
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1963496
Commit-Queue: Li Lin <llin@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724785}

When the DNS-interception-detector is disabled, Chrome no longer sends
detection packets. If the network were to change under this condition
after having learned the redirect domain, the latter would never
change. This CL clears the domain once it detects that the policy was
set to disable the feature.

Bug: 181671
Change-Id: Ie51489272dc3ee92d279e9c3623ba1682ab14330
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1962132
Commit-Queue: Kevin Bailey <krb@chromium.org>
Reviewed-by: Scott Violet <sky@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724784}

WMI can't be used in Chrome as it cause too many issues (hangs,
A/V injecting in the process...), this mean that the new
MemoryPressureMonitor can't work as is and there's no plan to improve
it anytime soon. Remove this code as it's not needed and causes some
crashes.

Fixed: 1032280, 956742, 939092
Change-Id: I70d41346caff8d61dffd3ffc57eb70bdd31043af
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1960077Reviewed-by: Chris Hamilton <chrisha@chromium.org>
Reviewed-by: Brian White <bcwhite@chromium.org>
Commit-Queue: Sébastien Marchand <sebmarchand@chromium.org>
Cr-Commit-Position: refs/heads/master@{#724783}