GitLab

Revert of Decrease swarming shards for Linux RandomOrder bots from 18 to 6. (patchset #1 id:1 of https://codereview.chromium.org/2835903002/ )

Reason for revert:
Want to go back to 18 shards to increase likelihood of single shard failure issues.

Original issue's description:
> Decrease swarming shards for Linux RandomOrder bots from 18 to 6.
>
> Review-Url: https://codereview.chromium.org/2835903002
> Cr-Commit-Position: refs/heads/master@{#466599}
> Committed: https://chromium.googlesource.com/chromium/src/+/e510777b49ba34c52aebc3613552cf2f4acbc2dc

TBR=mcgreevy@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.

Review-Url: https://codereview.chromium.org/2842973002
Cr-Commit-Position: refs/heads/master@{#467279}

This patch matches NGInlineLayoutAlgorithm to CSS2 10.8 Line height
calculations: the 'line-height' and 'vertical-align' properties[1]
defining empty inline elements should influence line height.

Also, since the introduction of NGInlineBoxState, we had two places to
compute the union of boxes; NGInlineBoxState.stack_[0] and
NGLineBoxFragmentBuilder. Some code were uniting boxes to wrong one.
This patch unifies them to NGInlineBoxState.

[1] https://drafts.csswg.org/css2/visudet.html#line-height

BUG=636993

Review-Url: https://codereview.chromium.org/2845493002
Cr-Commit-Position: refs/heads/master@{#467278}

Part of project Ribbon

This refactor is one step in the process of refactoring out property
specific logic inside switch cases in
CSSPropertyParser::ParseSingleValue. This is required to implement the
parseSingleValue methods in the property APIs without passing property
ID to the API.

Given that the IsValidAnimationPropertyList function only ever returned
false if the property passed was CSSPropertyTransitionProperty, this
check was moved to the calling methods.

The next step in this process will be to refactor out the need to pass
property to the ConsumeAnimationPropertyList function. At this stage the
call to IsValidAnimationPropertyList will be moved to the
CSSPropertyTransitionProperty case only and the check for this property
will be removed.

The check for this property in
CSSPropertyParser::ConsumeAnimationShorthand will be refactored out at a
later stage in project Ribbon.

BUG=668012

Review-Url: https://codereview.chromium.org/2835893002
Cr-Commit-Position: refs/heads/master@{#467277}

The enum is unnecessary and is only used in tests. I've introduced an internal
namespace, but I'm happy to change if there's a more idiomatic approach.

BUG=712922

Review-Url: https://codereview.chromium.org/2825943002
Cr-Commit-Position: refs/heads/master@{#467276}

BUG=none

Review-Url: https://codereview.chromium.org/2840003002
Cr-Commit-Position: refs/heads/master@{#467275}

This should look better when the user name is too long.

Before: https://drive.google.com/open?id=0ByXziH_JVCGJRDBwcVNuZHN2QkE
After: https://drive.google.com/open?id=0ByXziH_JVCGJZE1xby10SFFKUTA

Related to b/36958290

BUG=661794

Review-Url: https://codereview.chromium.org/2825293002
Cr-Commit-Position: refs/heads/master@{#467274}

This patch marks tests that were fixed in the 'vertical-align' patch[1]
but were missed and not updated in TestExpectations.

[1] https://codereview.chromium.org/2803433002

BUG=636993

Review-Url: https://codereview.chromium.org/2839733005
Cr-Commit-Position: refs/heads/master@{#467273}

The HTMLElementFactory used to take the form element pointer for form
association. That operation happens as a discrete step now for
customized built-in elements. Now that HTMLElementFactory doesn't do
form association any more, HTML and non-HTML element construction can
be merged together the way they are in the specs.

BUG=648828

Review-Url: https://codereview.chromium.org/2842933002
Cr-Commit-Position: refs/heads/master@{#467272}

Linux 4.9 adds /proc/sys/user/max_user_namespaces to limit the number
of concurrent user namespaces.  Sometimes the limit is set to 0 in
which case clone() will return ENOSPC.  This CL avoids PCHECK()ing in
that case.

BUG=715138
R=mdempsky@chromium.org

Review-Url: https://codereview.chromium.org/2842033002
Cr-Commit-Position: refs/heads/master@{#467271}

FileEntry.file is slow operation internally queries file metadata.
However, FileTransferController calls FileEntry.file for all the
selected files each time CHANGE_THROTTLED event is fired.
CHANGE_THROTTLED is fired every time a file is copied
during multi file copy.
In some slow storage devices which do not support parallel requests
such as MTP connected ones, this severely slows down file copy.

BUG=712121
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2843683002
Cr-Commit-Position: refs/heads/master@{#467270}

This is a preparation CL for https://codereview.chromium.org/2778983006.

There are still other tests which will fail when /deep/ (or ::shadow) is disabled.
These tests are being tracked in meta bug 715034.

BUG=715034

Review-Url: https://codereview.chromium.org/2843883002
Cr-Commit-Position: refs/heads/master@{#467269}

BUG=546953

Review-Url: https://codereview.chromium.org/2843773002
Cr-Commit-Position: refs/heads/master@{#467268}

Mainly, some enums were being displayed in their raw form, due to the
Blink Rename. For example, "kSignalingStateStable" instead of "stable".

BUG=chromium:714375

Review-Url: https://codereview.chromium.org/2832263004
Cr-Commit-Position: refs/heads/master@{#467267}

Review-Url: https://codereview.chromium.org/2837423003
Cr-Commit-Position: refs/heads/master@{#467266}

According to latest changes in specification [1,2], when invalid
NFCPushOptions.timeout is provided to nfc.push function, promise
must be rejected with TypeError.

[1] https://w3c.github.io/web-nfc/#dom-nfc-push
[2] https://github.com/w3c/web-nfc/issues/112

BUG=714035

Review-Url: https://codereview.chromium.org/2836813007
Cr-Commit-Position: refs/heads/master@{#467265}

fileManagerPrivate.getFileTasks is slow.
However, every time a file is copied during multi file copy, the method
was called via CHANGE_THROTTLED event listener.
This CL stops creating FileTasks again when file entries for the
previous FileTasks are same as the current ones.
Other possible solution to this problem is to change C++ interface of
fileManagerPrivate.getFileTasks to support incremental update of
FileTasks.

BUG=462989,712121
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:closure_compilation

Review-Url: https://codereview.chromium.org/2833413003
Cr-Commit-Position: refs/heads/master@{#467264}

Resulting vector of statuses must have the same size as the given permissions vector.

Review-Url: https://codereview.chromium.org/2838543003
Cr-Commit-Position: refs/heads/master@{#467263}

Currently generated groups such as StyleSurroundData inherit from
RefCounted, which means we have to explicitly specify a copy
constructor. Inheriting from RefCountedCopyable instead means we can
just use the compiler generated copy constructor.

Diff of generated files:
https://gist.github.com/darrnshn/b3f3492858caedc6359a0507dc173dff/revisions

BUG=628043

Review-Url: https://codereview.chromium.org/2799363002
Cr-Commit-Position: refs/heads/master@{#467262}

Introduces assert_promise_event_order_ that checks the order of promise
resolution and events firing is correct.

Makes assert_event_fires_after_promise and assert_event_fires_before_promise
use this function.

Also removes some obsolete TODOs.

BUG=697698

Review-Url: https://codereview.chromium.org/2825393004
Cr-Commit-Position: refs/heads/master@{#467261}

This CL replace DefineOwnProperty with SetNativeDataProperty in
InstallAttributeInternal<v8::Object>() not to create function
templates or instances on feature installations.

This change enables us to use the function for all data properties
and it improves performance of preparing origin-trial features
and dynamic installs of runtime enabled features.

BUG=617892

Review-Url: https://codereview.chromium.org/2815453005
Cr-Commit-Position: refs/heads/master@{#467260}

Remove redundant WebLocalFrame* parameter from WillReleaseScriptContext
in WebFrameClient.

BUG=361765

Review-Url: https://codereview.chromium.org/2838503002
Cr-Commit-Position: refs/heads/master@{#467259}

According to latest changes in specification [1,2], when argument
of invalid type is provided to nfc.push function, promise must be
rejected with TypeError, instead of SyntaxError.

[1] https://w3c.github.io/web-nfc/#dom-nfc-push
[2] https://github.com/w3c/web-nfc/issues/110

BUG=714034

Review-Url: https://codereview.chromium.org/2836063004
Cr-Commit-Position: refs/heads/master@{#467258}

Decouples IME frame update (cursor/anchor coordinates) from
ContentViewCore, hence redueces the dependency on CVC. Now
RenderWidgetHostViewAndroid uses native ImeAdapter to route
the frame update info directly to Java layer.

BUG=662908

Review-Url: https://codereview.chromium.org/2834343005
Cr-Commit-Position: refs/heads/master@{#467257}

This patch adds another flaky results from 5-6 iterations of running
tests.

I will run a few more iterations and see how it stabilizes.

BUG=591099

Review-Url: https://codereview.chromium.org/2840753003
Cr-Commit-Position: refs/heads/master@{#467256}

Suppress the following on Linux AMD R7 240:
conformance2_textures_canvas_tex_2d_rg8ui_rg_integer_unsigned_byte
conformance2_textures_canvas_tex_2d_rgb8ui_rgb_integer_unsigned_byte
conformance2_textures_canvas_tex_2d_rgba8ui_rgba_integer_unsigned_byte
conformance2_textures_webgl_canvas_tex_2d_rg8ui_rg_integer_unsigned_byte
conformance2_textures_webgl_canvas_tex_2d_rgb8ui_rgb_integer_unsigned_byte
conformance2_textures_webgl_canvas_tex_2d_rgba8ui_rgba_integer_unsigned_byte

BUG=710392
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2841853003
Cr-Commit-Position: refs/heads/master@{#467255}

BUG=582836
NOTRY=true
TBR=suzyh@chromium.org

Review-Url: https://codereview.chromium.org/2838883003
Cr-Commit-Position: refs/heads/master@{#467254}

This makes sure that xvfb actually started correctly and you can connect
to it. We don't use -display argument as we want to make sure the
environment is correct.

BUG=713597

Review-Url: https://codereview.chromium.org/2837773002
Cr-Commit-Position: refs/heads/master@{#467253}

We now call ShowOverlayScrollbars for user and programmatic scrolls, but not for
scroll anchoring scrolls or content area resizes.

In most cases the effects will not be observable until we update the show
triggers in the compositor (an upcoming patch).

BUG=606395
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_layout_tests_slimming_paint_v2

Review-Url: https://codereview.chromium.org/2835403002
Cr-Commit-Position: refs/heads/master@{#467252}

BUG=714166

Review-Url: https://codereview.chromium.org/2842763004
Cr-Commit-Position: refs/heads/master@{#467251}

Revert of Update V8 to version 6.0.66. (patchset #1 id:1 of https://codereview.chromium.org/2837923002/ )

Reason for revert:
Roll caused failed check in IsSane(), see bug for details
BUG=715391

Original issue's description:
> Update V8 to version 6.0.66.
>
> Summary of changes available at:
> https://chromium.googlesource.com/v8/v8/+log/9f6be2e9..ab1af916
>
> Please follow these instructions for assigning/CC'ing issues:
> https://github.com/v8/v8/wiki/Triaging%20issues
>
> Please close rolling in case of a roll revert:
> https://v8-roll.appspot.com/
> This only works with a Google account.
>
> CQ_INCLUDE_TRYBOTS=master.tryserver.blink:linux_trusty_blink_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel;master.tryserver.chromium.android:android_optional_gpu_tests_rel
>
> TBR=hablich@chromium.org,machenbach@chromium.org,kozyatinskiy@chromium.org,vogelheim@chromium.org
>
> Review-Url: https://codereview.chromium.org/2837923002
> Cr-Commit-Position: refs/heads/master@{#466756}
> Committed: https://chromium.googlesource.com/chromium/src/+/ddb812975b721aee41c39a12ac739eab0e8b053f

TBR=hablich@chromium.org,kozyatinskiy@chromium.org,machenbach@chromium.org,vogelheim@chromium.org,v8-autoroll@chromium.org
# Not skipping CQ checks because original CL landed more than 1 days ago.

Review-Url: https://codereview.chromium.org/2842903003
Cr-Commit-Position: refs/heads/master@{#467250}

Keep Pixel_DirectComposition_Video_MP4 disabled on Intel, because it
seems to have some sort of bug.

BUG=704389
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.android:android_optional_gpu_tests_rel;master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2827833002
Cr-Commit-Position: refs/heads/master@{#467249}

This is caused by calling RemoveNotification() while
'Clear All' operation is in progress. A MessageView
could be deleted twice.

BUG=713983

Review-Url: https://codereview.chromium.org/2836023002
Cr-Commit-Position: refs/heads/master@{#467248}

on demand to the client. Make the request hold the stream until the
caller asks for it, and make explicit cancellation unnecessary.

This is in preparation for subsquent cleanups of the session.

Review-Url: https://codereview.chromium.org/2844493002
Cr-Commit-Position: refs/heads/master@{#467247}

Speculative bug fix.

BUG=713671
R=kojii@chromium.org

Review-Url: https://codereview.chromium.org/2840613006
Cr-Commit-Position: refs/heads/master@{#467246}

Replace references to constants in ui/views/layout/layout_constants.h with their
equivalents using ChromeLayoutProvider. Removed some stale includes of layout_constants.h

BUG=691897

Review-Url: https://codereview.chromium.org/2837153003
Cr-Commit-Position: refs/heads/master@{#467245}

Renaming the file since we no longer have the in-memory implementation of
the metadata store. Also removing the cc file for the header and moved
the destructor into the header file.

BUG=708700

Review-Url: https://codereview.chromium.org/2819983003
Cr-Commit-Position: refs/heads/master@{#467244}

Groups inside ComputedStyle, such as StyleSurroundData, are stored as
pointers. When we copy the StyleSurroundData of a ComputedStyle, instead
of copying each member inside StyleSurroundData, we can copy the pointer
itself and do a copy when we write to it (copy-on-write). Similarly,
when we compare two StyleSurroundData pointers, we can first check if
they're pointing to the same object. If so, then we know that the two
StyleSurroundDatas are equal without having to compare each member
inside StyleSurroundData.

This patch introduces two macros: fieldwise_compare and fieldwise_copy,
which generate the code to do this optimisation. These macros take a
a list of fields to compare/copy and generate the cheapest set of code
for that task, taking advantage of group sharing.

This patch removes the perf cost of:
https://codereview.chromium.org/2786883002

by generating the exact same code as what was originally handwritten.

Diff of generated code:
https://gist.github.com/darrnshn/fe97404e680016753030fc788fcf24ff/revisions

BUG=628043

Review-Url: https://codereview.chromium.org/2826633002
Cr-Commit-Position: refs/heads/master@{#467243}

This is a generated CL by a clang refactoring tool at //tools/clang/base_bind_rewriters.
The tool rewrites base::Bind to base::BindOnce where the resulting Callback
is immediately converted to OnceCallback.
E.g.:
  base::PostTask(FROM_HERE, base::Bind(&Foo));
is rewritten to:
  base::PostTask(FROM_HERE, base::BindOnce(&Foo));

In terms of the behavior change after the rewrite, the OnceCallback
implementation moves out the bound argument to the target function, unlike
Callback.

BUG=554299

Review-Url: https://codereview.chromium.org/2826843004
Cr-Commit-Position: refs/heads/master@{#467242}

Changes in this CL:

1. The public interfaces BrowsingDataRemover and BrowsingDataRemoverDelegate
move to content/public/browser. BrowsingDataRemoverImpl moves to
content/browser/browsing_data.

2. Previously, BrowsingDataRemoverImpl was a KeyedService created by
BrowsingDataRemoverFactory. After this CL, it is owned by BrowserContext.
Conversely, ChromeBrowsingRemoverDelegate (which stays in chrome/) becomes
a KeyedService. The former BrowsingDataRemoverFactory is thus adapted into
ChromeBrowsingDataRemoverDelegateFactory.

BrowsingDataRemoverDelegate is now owned by the embedder (rather than
BrowsingDataRemover) and supplied by embedders through
BrowserContext::GetBrowsingDataRemoverDelegate(). Profile (in chrome/)
supplies ChromeBrowsingDataRemoverDelegate. The implementation for other
embedders is left as TODO.

3. BrowsingDataRemoverImplTest moves to content/browser/browsing/data.
browsing_data_remover_test_util that is used by this test as well as various
tests in chrome/ is moved to content/public/test.

4. BrowsingDataRemover can only delete download history if this is feature is
not disabled by administrator. Previously, BrowsingDataRemover queried
PrefService directly. We now pass this information through
BrowsingDataRemoverDelegate. The default value is true,
ChromeBrowsingDataRemoverDelegate queries PrefService.

5. The removal of ContentBrowserClient interfaces ClearCookies(), ClearCache(),
ClearSiteData() and their replacement by BrowsingDataRemover is left as a TODO.

6. All other changes are adding / removing includes, updating BUILD files,
replacing BrowsingDataRemoverFactory with BrowserContext::GetBrowsingDataRemover,
formatting (mostly triggered by prepending content:: to class names), and the
removal of forgotten "#ifdef (EXTENSIONS)" in BrowsingDataRemoverImplTest.

TBR=dbeam@chromium.org
BUG=668114

Review-Url: https://codereview.chromium.org/2827523003
Cr-Commit-Position: refs/heads/master@{#467241}

Add ASSERTs to WebFrameTest to ensure frame parameters are correct in a
few tests, and redesigned part of ContextLifetimeTestWebFrameClient to
support correct WebFrameClient creation for child frames.

This is pre-work to remove the redundant WebLocalFrame parmeters from
didCreateScriptContext and willReleaseScriptContext.

BUG=361765

Review-Url: https://codereview.chromium.org/2837523002
Cr-Commit-Position: refs/heads/master@{#467240}