GitLab

Leaving it there resulted in it being picked up on urelated stacks.

R=ajwong@chromium.org

Bug: 1073500
Change-Id: If8816da156e89122d3c599e1f355e9d6703940b3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161465
Commit-Queue: Albert J. Wong <ajwong@chromium.org>
Reviewed-by: Albert J. Wong <ajwong@chromium.org>
Auto-Submit: Gabriel Charette <gab@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762040}

https://android.googlesource.com/platform/external/perfetto.git/+log/7fc367463824..bb3010037266

git log 7fc367463824..bb3010037266 --date=short --first-parent --format='%ad %ae %s'
2020-04-23 eseckler@google.com Merge "processor: Migrate more TrackEvent arguments to reflection-based parsing"
2020-04-23 eseckler@google.com Merge "processor: Fix 3 clusterfuzz-found issues related to TrackEvent parsing"
2020-04-23 ilkos@google.com Merge "Add ion_stat to config muxer for memory"

Created with:
  gclient setdep -r src/third_party/perfetto@bb3010037266

If this roll has caused a breakage, revert this CL and stop the roller
using the controls here:
https://autoroll.skia.org/r/perfetto-chromium-autoroll
Please CC perfetto-bugs@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:linux-perfetto-rel
Bug: None
Tbr: perfetto-bugs@google.com
Change-Id: Ifc2d084df5f5266d42faa3edb943a4f134cd8e3e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2163448Reviewed-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@{#762039}

fix canvas-getImageData-large-crash.html always passes: it passes regardless
if any exception is thrown. Remove test.

Bug: 1073721

Change-Id: I5abde013194b45e3804ac2c215c5b2c79614c29a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2160346Reviewed-by: Fernando Serboncini <fserb@chromium.org>
Commit-Queue: Yi Xu <yiyix@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762038}

Building the Chrome browser configured to engage the Ozone/Wayland
as its backend by using the gcc compiler collection is failed due
to the 'declaration changes meaning' compile-time error.

This CL fixes the error and gets the sources compiled by both
modern compilers - clang and gcc.

Bug: 1073877
Change-Id: Ib6e480e9b3d1b9dfc3640909b2b3dd1c78cdb7da
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2162865Reviewed-by: Maksim Sisov <msisov@igalia.com>
Reviewed-by: Robert Kroeger <rjkroege@chromium.org>
Commit-Queue: Robert Kroeger <rjkroege@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762037}

Rather than generating a different ${Foo}Jni class for every
target when use_hashed_jni_names=true, also have classes use
org.chromium.base.natives.GEN_JNI.

Then, at the final step, have GEN_JNI forward calls to J.N.$HASH.

Makes a bit more work for R8 (it needs to inline all the GEN_JNI
methods), but move the "hashed functions" logic to live in
one spot at the end of the pipeline.

A byproduct of this is that toggling use_hashed_jni_names no
long requires all libraries to be recompiled (attached bug)

Bug: 1073921
Change-Id: I8df7c751b51d60e878f270c06e2e979dd9dbc821
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2162109Reviewed-by: Michael Thiessen <mthiesse@chromium.org>
Commit-Queue: Andrew Grieve <agrieve@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762036}

Before this change, the browser process would crash if the
SecurityStateClient wasn't set early in the startup path. This change
makes having a SecurityStateClient optional for the embedder and
defaults to the components logic to get security level.

Bug: 1052375
Change-Id: I7336e375908137aed3187debddd3860b7d657e0c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2162985
Commit-Queue: Mugdha Lakhani <nator@chromium.org>
Reviewed-by: Colin Blundell <blundell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762035}

Clean up crypto import, since uses of crypto::ECPrivateKey were removed in change I153949d11b923131b404ee603a6d885aec57706d

Bug: 1012226
Change-Id: Ic698f872e6441894e4330fb784609aba72d25ee1
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161827
Auto-Submit: Naomi Musgrave <nmusgrave@chromium.org>
Commit-Queue: Mikel Astiz <mastiz@chromium.org>
Reviewed-by: Alex Chau <alexchau@chromium.org>
Reviewed-by: Mikel Astiz <mastiz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762034}

Implements the UrlRequestAction FIDL API, which allows embedders to
specify requests to block based on the the requests' host and/or scheme.

* Fix issue where wildcards passed in "hosts_filter" are URL-escaped.
* Add "page_type" edge transition detection to TestNavigationListener.
* Report early navigation errors in NavigationControllerImpl.

Bug: 1054475
Change-Id: If4df4889f980c5a239719d09f039eee32efd1c1d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2070103Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Reviewed-by: Fabrice de Gans-Riberi <fdegans@chromium.org>
Commit-Queue: Kevin Marshall <kmarshall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762033}

This CL ignores settings changes which are <200ms from each other, which
fixes false positives logged with the metric.

Bug: 1073714
Change-Id: I07f81a6243ccad85cd92b34e4b639ba2b6b87442
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2162126Reviewed-by: James Cook <jamescook@chromium.org>
Commit-Queue: Kyle Horimoto <khorimoto@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762032}

Move a brace so that telemetry logic is properly parsed. Previously
multiple telemetry items were only parsed it the battery information is
present.

Bug: none
Change-Id: I0cea4e12c9a8c504bde11917aefd4d0548b77800
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2153694
Commit-Queue: Trent Begin <tbegin@chromium.org>
Reviewed-by: Maksim Ivanov <emaxx@chromium.org>
Reviewed-by: Jesse Schettler <jschettler@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762031}

This changes the default filtering from nearest to bilinear, to match
other canvases.
The filter quality gets passed through from CSS's image-rendering.

Bug: 852089
Change-Id: I704c91940bd495a55a1afe29600dff9cec448b8e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161869Reviewed-by: Austin Eng <enga@chromium.org>
Reviewed-by: Corentin Wallez <cwallez@chromium.org>
Reviewed-by: Kenneth Russell <kbr@chromium.org>
Commit-Queue: Kai Ninomiya <kainino@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762030}

Use LayoutObjectFactory for MathML layout object creation where
possible.

Bug: 1070600
Change-Id: I1f02ca9ab4018798559ad3334fd4b316f31330d8
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2162927
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762029}

Remove the pending_decoded_image_ member of AVIFImageDecoder. It is
apparently intended to avoid calling avifDecoderNthImage() with index 0
twice. But libavif allows us to do that, and the second call will return
immediately.

Move the SetSize() call and size check out of DecodeImage() to the two
call sites.

Bug: 960620
Change-Id: I4c7beaed4ca5aa8040aa7b3cb2facd92e18eb38f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161871Reviewed-by: Leon Scroggins <scroggo@google.com>
Reviewed-by: Dale Curtis <dalecurtis@chromium.org>
Commit-Queue: Wan-Teh Chang <wtc@google.com>
Cr-Commit-Position: refs/heads/master@{#762028}

This reverts commit 8af6bfe2.

Reason for revert: The branched versions have not been updated to run
the non-isolated tests on the builder instead of the tester.

Original change's description:
> Use linux hosts for all mac triggered testers
> 
> We've already shown this to work on Mac10.15 Tests on the FYI
> waterfall, so let's do this for the rest of the Mac triggered
> testers on the main waterfall as well. This will allow us to
> free up some Mac resources to do more important tasks (e.g.
> compile, run tests). It also makes these builders builderless,
> increasing our flexibility and improving our utilization.
> 
> This CL also introduces a new ci.triggered_tester starlark
> function to enable easily specifying builders as "we don't care
> what OS you use", and to enable easily turning all of them into
> "thin testers" if/when we go down that path.
> 
> Bug: 1061582
> Change-Id: Iffd5a4afbadeabc19d9d0e3b28f33151a9474191
> Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161689
> Reviewed-by: Garrett Beaty <gbeaty@chromium.org>
> Commit-Queue: Aaron Gable <agable@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#761718}

TBR=ellyjones@chromium.org,agable@chromium.org,gbeaty@chromium.org

Change-Id: I786079f17db5e32643b11bb43775b7e52a44e4cc
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: 1061582
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2163254Reviewed-by: Garrett Beaty <gbeaty@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762027}

Per:

 "If decoding does not need to be performed for this image (for example
  because it is a vector graphic), resolve promise with undefined."

(https://html.spec.whatwg.org/multipage/embedded-content.html#dom-img-decode)

we can skip the decode request and just resolve the promise up front.
Since the SVG format is context sensitive (and, well, scalable),
decoding them outside of that context will likely just waste resources.

Fixed: 1073361
Change-Id: Iedf3a6636d9525e243ce9a8cf9c3835a1468b34b
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2160915
Commit-Queue: Fredrik Söderquist <fs@opera.com>
Reviewed-by: vmpstr <vmpstr@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762026}

v8::Persistent will leak on destruction by default, unless marked
as copyable. As such, we must call Reset() before destroying any
v8::Persistent with the default, non-copyable, traits.

Then we can remove the JS cleanup code added to the spellcheck tests.

R=avi@chromium.org

Bug: 866140, 1069111
Change-Id: I3da663bda02c8916c1a4ae5d21234c54311a1dab
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161652
Auto-Submit: danakj <danakj@chromium.org>
Reviewed-by: Avi Drissman <avi@chromium.org>
Commit-Queue: danakj <danakj@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762025}

This CL is in preparation for enabling Polymer3 in ChromeOS settings
pages. It fixed issues in os_printing_page/ folder that break
semi-automatic Polymer3 migration. These changes include splitting up
files that contain multiple <dom-module> elements, as this is required
by Polymer3 conversion tools.

Bug: 1045266
Change-Id: Idf9b18aac87b74d34a431ffe5383fd2f845554fc
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2151090Reviewed-by: Kyle Horimoto <khorimoto@chromium.org>
Reviewed-by: Regan Hsu <hsuregan@chromium.org>
Reviewed-by: jimmy gong <jimmyxgong@chromium.org>
Commit-Queue: Nnamdi Theodore Johnson-kanu <tjohnsonkanu@google.com>
Cr-Commit-Position: refs/heads/master@{#762024}

This is for the detailed audio stack tracing. There have been several
audio performance issues reported from the applications that use
Web Audio and MediaSteam together. We need more information on why
the audio rendering thread or AudioWorkletThread is blocked.

Bug: 1072727
Change-Id: I56d33d6bf471938815c7aba3f8b94ad461f9c3b2
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2145557
Commit-Queue: Hongchan Choi <hongchan@chromium.org>
Reviewed-by: Stephen Nusko <nuskos@chromium.org>
Reviewed-by: Sami Kyöstilä <skyostil@chromium.org>
Reviewed-by: Guido Urdaneta <guidou@chromium.org>
Reviewed-by: Kentaro Hara <haraken@chromium.org>
Reviewed-by: Olga Sharonova <olka@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762023}

Encourage to use org.chromium.ui.widget.TextViewWithLeading instead of
android:lineSpacingExtra or android:lineSpacingMultiplier.

Bug: 1069805
Change-Id: I3eea084ac83034d99c27ecdb7b1c69c0877ff800
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2163427Reviewed-by: Theresa  <twellington@chromium.org>
Reviewed-by: Andrew Grieve <agrieve@chromium.org>
Commit-Queue: Lijin Shen <lazzzis@google.com>
Cr-Commit-Position: refs/heads/master@{#762022}

Currently, AdTagging uses the presence of ad scripts on the v8 stack and
the execution context at the time of the fetch as a signal for tagging.
However, CSS stylesheets often defer their resource requests until
needed. E.g., the fonts defined by the stylesheet may not be requested
until a style recalculation occurs. This results in erroneous tagging,
as the stack contains the script which caused the style recalculation
rather than the one that added the stylesheet.

To fix this, we tag stylesheet-requested resources with an is_ad_related
bit. This is set through the CSSParserContext according to the
stylesheet's ResourceRequest. Finally, a was_requested_by_stylesheet bit
is set on these resources, which allows the ad tracker to ignore the v8
stack at request time.

Note that, despite constructing a CSSParserContext, the WebVTT parser
can be ignored as its CSS is prevented from making network requests.

Bug: 1051605
Change-Id: I131586f8c58c4ae0073b1cedae22fa1f92ed408a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2050887
Commit-Queue: Alex Turner <alexmt@chromium.org>
Reviewed-by: Charlie Harrison <csharrison@chromium.org>
Reviewed-by: Nate Chapin <japhet@chromium.org>
Reviewed-by: John Delaney <johnidel@chromium.org>
Reviewed-by: Rune Lillesveen <futhark@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762021}

Make them the same as the ones run on the to-be-retired android-kitkat-arm-rel

Bug: 1032336
Change-Id: I7916451b2a0bb2fb25cdba8147493326292de507
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161660Reviewed-by: Ben Pastene <bpastene@chromium.org>
Commit-Queue: Haiyang Pan <hypan@google.com>
Cr-Commit-Position: refs/heads/master@{#762020}

The DCHECK was checking that the arrays have positive lengths, but the
spec allows zero-length arrays too.  So change the spec to check for
non-negative values.  And also modify getFrequencyResponse to return
quickly if the length is 0.

This applies to both BiquadFilterNodes and IIRFilterNodes.

Bug: 1073557
Change-Id: I8c98cf01021c9ece5731ccf8fdbc1fe8fe97f39c
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2163054Reviewed-by: Hongchan Choi <hongchan@chromium.org>
Commit-Queue: Raymond Toy <rtoy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762019}

Instead of passing WebContentsGetter to ContentBrowserClient, add explicit
callbacks to WebContentsObserver taking RenderFrameHost* or NavigationHandle*
as a parameter.

This allows to attribute cookie/javascript accesses to the correct
frame or navigation request, which is a prerequisite for attributing
content settings to the correct document and making them per-document.

R=falken@chromium.org,engedy@chromium.org
BUG=998171,1061899

Change-Id: Iec55a1db235b26bab44f1c1de76900b73ac6614d
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2151866
Commit-Queue: Alexander Timin <altimin@chromium.org>
Reviewed-by: Balazs Engedy <engedy@chromium.org>
Reviewed-by: Matt Falkenhagen <falken@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762018}

The present CL fixes a bug where if you rip an item off the shelf on one
display, it shows at the right or bottom of the shelf on other displays.

Test: manual
Change-Id: I75813306ec0a11f3ac4ec39ed08ad736e5788fa0
Fixed: 1014721
Bug: 1014721
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161900
Commit-Queue: Avery Musbach <amusbach@chromium.org>
Reviewed-by: Xiyuan Xia <xiyuan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762017}

The drive long welcome banner is removed and the new GM2 looks applied
to drive welcome banner and drive and download warning banners.

Also Dismiss buttons changed from X to text buttons.

Modified cr-button is used for text buttons. cr-button provides most of
the text button styles, specially ripple (inkdrop) effect.

The changed css classes duplicated for existing UI and
files-ng with body:not(.files-ng) and body.files-ng prefixes
respectively.

The icon/logo changed to svg version.

rtl/ltr direction tested. It's supported in this CL by using flex
system and inline start/ends rules while the non-filesng uses dir
attribute and absolution positioning. It helps to reduce the complexity
by using one set of rules for both directions.

For files-ng clean up:
- js: can be caught where isFilesNG() method used
- css: body:not(.files-ng)
- Other: ToDos added in string resource files

Bug: 1002400
Change-Id: Icc04ff6193d164e677d6d39294152ec440c73e48

Patchset 10:
- Fix class of welcome banner in non-filesng mode.
- Update outline rule to comply percentage convention.

Change-Id: Icc04ff6193d164e677d6d39294152ec440c73e48
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2148917
Commit-Queue: Noel Gordon <noel@chromium.org>
Reviewed-by: Noel Gordon <noel@chromium.org>
Reviewed-by: Alex Danilo <adanilo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762016}

Profile Manager and its tests are updated to match the new profile APIs
for multiple OTR profile support.

Bug: 1033903
Change-Id: I01c5c6389c75e9288deb1b03c5a18aa7307bf270
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2144131Reviewed-by: Mihai Sardarescu <msarda@chromium.org>
Commit-Queue: Ramin Halavati <rhalavati@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762015}

When we're hiding the status indicator, we make the Java view GONE and
start animating the cc layer. However, at this point, the cc layer can
have an old texture (or a texture that makes no sense).

With this CL, we invalidate the cc texture when the Java animation
ends and request a render from the LayoutUpdateHost.

Please note that this CL doesn't completely fix the issue. However, it
makes it much less common.

Bug: 1068346
Change-Id: I8e889fec82d5f33c9f5ba30695715a6d6fa67171
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2144629
Commit-Queue: Sinan Sahin <sinansahin@google.com>
Reviewed-by: Theresa  <twellington@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762014}

There are no matching test files for these - deleted or renamed.

Change-Id: I18046adcb151ca7846e7511c39f5721b31218b40
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161418
Auto-Submit: Joshua Bell <jsbell@chromium.org>
Reviewed-by: Stephen McGruer <smcgruer@chromium.org>
Commit-Queue: Joshua Bell <jsbell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762013}

This is a simple string change required by legal.

TBR=sczs@chromium.org

Change-Id: Ia2406a23ca6125d36a4f3675a6d001306c1f9d68
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161786Reviewed-by: edchin <edchin@chromium.org>
Commit-Queue: edchin <edchin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762012}

If Family Link is not installed, open Play store instead.

Bug: 1056458
Change-Id: I31e3356fbe9efba93fd75e9f786d9b84ecc87655
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2135383
Commit-Queue: Xiqi Ruan <xiqiruan@chromium.org>
Reviewed-by: Dominick Ng <dominickn@chromium.org>
Reviewed-by: Aga Wronska <agawronska@chromium.org>
Reviewed-by: James Cook <jamescook@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762011}

Set notification with system priority when create system notification.

Bug: 903545
Change-Id: Ie1b92f2361f7d5564102d1719b95c41753def556
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161771Reviewed-by: Aga Wronska <agawronska@chromium.org>
Reviewed-by: James Cook <jamescook@chromium.org>
Commit-Queue: Xiqi Ruan <xiqiruan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762010}

Invokes the NTPUserDataLogger machinery to log the following histograms
when loading the WebUI NTP:
- NewTabPage.LoadTime
- NewTabPage.LoadTime.MostLikely
- NewTabPage.LoadTime.MostVisited
- NewTabPage.LoadTime.NewTab
- NewTabPage.LoadTime.Startup

Additionally, adds logging for the new histogram
NewTabPage.LoadTime.WebUINTP, which only gets logged when loading the
WebUI NTP while the histograms mentioned above get logged when loading
any NTP.

The local NTP is considered loaded as soon as the most visited tiles
have been displayed. We follow the same logic on the WebUI NTP to make
the load times comparable. Thus, the logging gets triggered as soon as
the most visited tiles have been stamped out.

Fixed: 1072865
Change-Id: I2f908a23dffc911b4338c0611c017bbbc38d0ec1
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2159954Reviewed-by: Ramya Nagarajan <ramyan@chromium.org>
Reviewed-by: Ilya Sherman <isherman@chromium.org>
Reviewed-by: Alex Gough <ajgo@chromium.org>
Commit-Queue: Tibor Goldschwendt <tiborg@chromium.org>
Auto-Submit: Tibor Goldschwendt <tiborg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762009}

Using MAC_ANY caused it to pick up Mac 10.15 machines, which don't
support the GN args currently used by mac-rel. Which causes them to fail
quickly and then get re-used frequently. Until we determine the solution
going forward, changing the os dimension will get the builds back to
green.

# mac-rel is broken, this change will fix it

Bug: 1073694, 1073965
Change-Id: I2620c570c443be89f27ace27cbbd9c3c25cd4010
No-Try: True
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2163151
Commit-Queue: Garrett Beaty <gbeaty@chromium.org>
Auto-Submit: Garrett Beaty <gbeaty@chromium.org>
Reviewed-by: Ben Pastene <bpastene@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762008}

This code was relying on a bug: That SkColors passed to erase were
treated as being in the bitmap's color space, rather than sRGB. (In
every other context, they are interpreted as sRGB). Fixing that bug
in Skia breaks this one unit test.

To keep the current expectations valid, rewrite the code that fills
the bitmap to clearly do what was happening before - draw rectangles
with the test colors, accompanied by the bitmap's color space, to
prevent any color space conversion. This version of the test passes
now, and continues to pass when the Skia bug is fixed.

Bug: skia:8663
Change-Id: I3a7e94847f08472081390b23fcfe0a12d8e5a38f
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2161617Reviewed-by: Mike Klein <mtklein@google.com>
Reviewed-by: Leon Scroggins <scroggo@google.com>
Reviewed-by: mark a. foltz <mfoltz@chromium.org>
Commit-Queue: Brian Osman <brianosman@google.com>
Cr-Commit-Position: refs/heads/master@{#762007}

https://chromium.googlesource.com/chromium/tools/depot_tools.git/+log/fef5cf8e227b..0f15c2729b8c

git log fef5cf8e227b..0f15c2729b8c --date=short --first-parent --format='%ad %ae %s'
2020-04-23 recipe-mega-autoroller@chops-service-accounts.iam.gserviceaccount.com Roll recipe dependencies (trivial).

Created with:
  gclient setdep -r src/third_party/depot_tools@0f15c2729b8c

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

Bug: None
Tbr: agable@chromium.org
Change-Id: Ief6f8a62f3fe28fc16df17dbf1ddcd02ac9d7a12
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2163057Reviewed-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@{#762006}

"Leave Secure Cookies Alone" refers to step 12 in Section 5.4 of RFC
6265bis:
https://tools.ietf.org/html/draft-ietf-httpbis-rfc6265bis-05#section-5.4

If an insecure URL attempts to set a cookie that is "equivalent" to a
preexisting Secure cookie, then that set attempt is rejected.

This CL:
 - Fixes a bug in our calculation of "equivalence" for these purposes,
   which was not correctly computing domain-matching described in
   Section 5.1.3 of the RFC. (See example below.)
 - Adds (perhaps excessively) exhaustive tests for this behavior.
 - Adds comments clarifying the situation.
 - Renames variables to make the direction of the comparison clearer,
   as this is an asymmetric comparison.

Example:

Before this CL: If https://foo.com/ has set a secure host cookie

  cookie=secure; secure

then we allowed an insecure host cookie to be set by http://a.foo.com/:

  cookie=insecure

This cookie should be rejected, according to the spec, because
a.foo.com and foo.com are domain-matching.

Bug: 1069804
Change-Id: I002f9ffd7cf8523a2f4291525db2939306028d7a
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2143865
Commit-Queue: Lily Chen <chlily@chromium.org>
Reviewed-by: Maksim Orlovich <morlovich@chromium.org>
Reviewed-by: Mike West <mkwst@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762005}

Implement safebrowsing shutdown when the browser shuts down
(i.e. when last instance of BrowserImpl is destroyed).

This ensures proper cleanup of the safebrowsing service,
pending requests and the corresponding remote safe browsing
database manager.

In particular in this patch:
- move safe_browsing_service ownership to browser_process
- shutdown safebrowsing on destruction of last BrowserImpl
- add code to count live BrowserImpl instances (corresponds
  to BrowserImpl.Observer on the java side)
- add checks and make sure to stop the safebrowsing
  database manager on IO thread.

BUG=1071533,1015418

Change-Id: I39ad06a489df20454f2970db287b1f11b81ded1e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2132448
Commit-Queue: Tim Volodine <timvolodine@chromium.org>
Reviewed-by: Richard Coles <torne@chromium.org>
Reviewed-by: John Abd-El-Malek <jam@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762004}

IPC messages still might be received while the WebFrameWidgetBase has
been closed but not collected yet. Resetting the receiver prevents
receiving messages in these states.

BUG=1073935

Change-Id: Idef61bae81d142a66efcf5b5ecded7a242383322
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2163248Reviewed-by: Daniel Cheng <dcheng@chromium.org>
Commit-Queue: Dave Tapuska <dtapuska@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762003}

Bug: 997049
Change-Id: I2b6bd94a72110094914e0962c34f8bb893a19ef4
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2154726
Commit-Queue: Matt Menke <mmenke@chromium.org>
Reviewed-by: Eric Orth <ericorth@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762002}

This histogram is still used for the migration of identity manager
to use the Gaia (instead of the email) as the account ID.

Fixed: 1072601
Change-Id: I25fa0b5c093e436a1c3dff799927e1e123ec8dc5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2160968
Commit-Queue: Mihai Sardarescu <msarda@chromium.org>
Reviewed-by: Alex Ilin <alexilin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#762001}