GitLab

This means that when style sharing we don't always have to push the bloom
filter and parent stack and also avoids pushing the stack for elements that
only have text node children. Similarly this improves the performance of things
like editing that will insert a text node and immediately call recalcStyle to
cause it to attach by avoiding the parent stack since it's not needed.

I had tried to make the StyleResolverParentPusher smarter to push implicitly in
https://codereview.chromium.org/68253003 but it was reverted in
https://codereview.chromium.org/59883011 because it caused a perf regression by
making us push the filter for elements that only had text node children in
recalcStyle. In fixing that I realized that we push inside ::attach() even if
there are only text children and could get a similar perf improvement by avoiding
pushing in those cases.

R=ojan@chromium.org,eseidel@chromium.org

Review URL: https://codereview.chromium.org/102273004

git-svn-id: svn://svn.chromium.org/blink/trunk@170168 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=ager@chromium.org, erik.corry@gmail.com, haraken@chromium.org, oilpan-reviews@chromium.org, tkent@chromium.org, vegorov@chromium.org, zerny@chromium.org
BUG=341815

Review URL: https://codereview.chromium.org/214503002

git-svn-id: svn://svn.chromium.org/blink/trunk@170167 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Implements transform-style as defined in CSS Transforms[1]. Behavior is
identical to -webkit-transform-style.

Note, this property is guarded by the CSSTransformsUnprefixed runtime feature
(enabled only for testing).

[1] http://www.w3.org/TR/css3-transforms/

BUG=154772

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=170013

Review URL: https://codereview.chromium.org/211123002

git-svn-id: svn://svn.chromium.org/blink/trunk@170166 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The browser side is https://codereview.chromium.org/205033002/

BUG=354112

Review URL: https://codereview.chromium.org/205013002

git-svn-id: svn://svn.chromium.org/blink/trunk@170165 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is a minimal change to make [WillBeGarbageCollected] in Event.idl workable.

- This CL adds [WillBeGarbageCollected] to Event.idl and makes minimal changes
to the code base to make it workable.

- This CL is almost just replacing:

  PassRefPtr<XEvent> XEvent::create() { return adoptRef(new XEvent); }

with

  PassRefPtrWillBeRawPtr<XEvent> XEvent::create() { return
adoptRefWillBeRefCountedGarbageCollected(new XEvent); }

BUG=340522
NOTRY=true

Review URL: https://codereview.chromium.org/213363006

git-svn-id: svn://svn.chromium.org/blink/trunk@170164 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Previously, override value was quickly replaced by real data from sensors.
Now, we put sensor data on hold until override is disabled.

This patch also fixes the cleanup/navigation issues with override enabled.

BUG=none

Review URL: https://codereview.chromium.org/209333008

git-svn-id: svn://svn.chromium.org/blink/trunk@170163 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Review URL: https://codereview.chromium.org/211403002

git-svn-id: svn://svn.chromium.org/blink/trunk@170162 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=350806
NOTRY=true

Review URL: https://codereview.chromium.org/212643009

git-svn-id: svn://svn.chromium.org/blink/trunk@170161 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This code is duplicated between setPropertyInternal and setProperty. We
should de-dupe it so we can add specific optimizations to the property
setting code for things like transform and opacity.

Review URL: https://codereview.chromium.org/214333002

git-svn-id: svn://svn.chromium.org/blink/trunk@170160 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Reason for revert:
It seems this caused many memory leaks.

http://build.chromium.org/p/chromium.memory/builders/Linux%20ASan%2BLSan%20Tests%20%281%29/builds/807/steps/browser_tests/logs/stdio


Original issue's description:
> Use v8 Microtask Queue
> 
> V8 now hosts the end-of-microtask work queue. This allows consistent
> scheduling of microtask work regardless of where it is implemented (v8 vs blink).
> 
> R=adamk@chromium.org
> 
> Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=169978

TBR=adamk@chromium.org,haraken@chromium.org,rafaelw@chromium.org
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/214563003

git-svn-id: svn://svn.chromium.org/blink/trunk@170159 bbb929c8-8fbe-4397-9dbb-9b2b20218538

NOTRY=true
TBR=vsevik
BUG=357004

Review URL: https://codereview.chromium.org/213383010

git-svn-id: svn://svn.chromium.org/blink/trunk@170158 bbb929c8-8fbe-4397-9dbb-9b2b20218538

It is reported that, under some condition, an unexpected <div> is inserted
when text is inserted after a paragraph is selected. This was caused by
InsertText::doApply() adjusting the text style according to the current typing
style after the insertion.

This patch clears block properties from the typing style if the selection ends
at the start of a block.

Additionally, this patch fixes another issue in DeleteSelectionCommand, which
may erroneously insert a placeholder <br> if the return value of
VisiblePosition::previous() or next() is a null VisiblePosition.

BUG=349531

Review URL: https://codereview.chromium.org/214313002

git-svn-id: svn://svn.chromium.org/blink/trunk@170157 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=eustas, pfeldman, vsevik
NOTRY=true

Review URL: https://codereview.chromium.org/170273003

git-svn-id: svn://svn.chromium.org/blink/trunk@170156 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=pfeldman@chromium.org, pfeldman
NOTRY=true

Review URL: https://codereview.chromium.org/204333004

git-svn-id: svn://svn.chromium.org/blink/trunk@170155 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This is needed by const-method clients such as "Resource::canDelete() const".

BUG=290193

Review URL: https://codereview.chromium.org/213193002

git-svn-id: svn://svn.chromium.org/blink/trunk@170154 bbb929c8-8fbe-4397-9dbb-9b2b20218538

R=ager@chromium.org, erik.corry@gmail.com, haraken@chromium.org, oilpan-reviews@chromium.org, tkent@chromium.org, vegorov@chromium.org, zerny@chromium.org
BUG=341815

Review URL: https://codereview.chromium.org/212993002

git-svn-id: svn://svn.chromium.org/blink/trunk@170153 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=357097
TBR=tfarina@crhomium.org

Review URL: https://codereview.chromium.org/213283009

git-svn-id: svn://svn.chromium.org/blink/trunk@170152 bbb929c8-8fbe-4397-9dbb-9b2b20218538

- css3/images/cross-fade-invalidation.html [ ImageOnlyFailure Pass ]
- fast/media/lifetime.html [ Crash Pass Timeout ]
- gamepad/gamepad-polling-access.html [ Crash Pass ]
- http/tests/misc/object-embedding-svg-delayed-size-negotiation-2.htm [ Pass Timeout ]
- http/tests/security/contentSecurityPolicy/report-cross-origin-no-cookies.html [ Failure Pass ]
- http/tests/security/contentSecurityPolicy/report-cross-origin-no-cookies.html [ Pass Timeout ]
- inspector/audits/audits-panel-functional.html [ Failure Pass ]
- media/activation-behavior.html [ Pass Failure Timeout ]
- media/autoplay.html [ Failure Pass ]
- virtual/threaded/animations/animation-hit-test.html [ Failure Pass ]
- virtual/threaded/animations/keyframe-timing-functions2.html [ Pass Timeout ]
- virtual/threaded/animations/unanimated-style.html [ Pass Timeout ]
- web-animations-api/timeline-updates-players.html [ Failure Pass ]
- web-animations-api/timeline-updates-players.html [ Failure Pass ]

BUG=355489,357068,357069,357071,357074,357075,357079,357080,357082,357084,357085,357088
TBR=enne@chromium.org

Review URL: https://codereview.chromium.org/214363003

git-svn-id: svn://svn.chromium.org/blink/trunk@170151 bbb929c8-8fbe-4397-9dbb-9b2b20218538

The aux 'makeLayerChildFrameMap' function used in the
ScrollingCoordinator should not be allowed to add empty
key values to the given LayerFrameMap, as this is misuse
of WTF::HashMap which leads to undefined behavior (for
example it was the root cause of
http://code.google.com/p/chromium/issues/detail?id=310975 )

Review URL: https://codereview.chromium.org/212813004

git-svn-id: svn://svn.chromium.org/blink/trunk@170150 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Adds a runtime flag (enabled only for testing) and defines property names
for each property defined in CSS Transforms[1]. Future patches will make
these properties functional.

[1] http://www.w3.org/TR/css3-transforms/

BUG=154772

Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=170005

Review URL: https://codereview.chromium.org/211033002

git-svn-id: svn://svn.chromium.org/blink/trunk@170149 bbb929c8-8fbe-4397-9dbb-9b2b20218538

It is possible for the DocumentLoader to lose it's ArchiveResourceCollection
without it's MHTMLArchive being deallocated as well, as is observed when a
navigation, which may be aborted, is initiated. By not clearing the
ArchiveResourceCollection in such cases, the DocumentLoader is left in a
still valid state.

BUG=348977

Review URL: https://codereview.chromium.org/183763030

git-svn-id: svn://svn.chromium.org/blink/trunk@170148 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=348947
NOTRY=true

Review URL: https://codereview.chromium.org/212103007

git-svn-id: svn://svn.chromium.org/blink/trunk@170147 bbb929c8-8fbe-4397-9dbb-9b2b20218538

With DOMWindow now a garbage collected object, follow up and also turn
it into a heap-residing Supplementable, along with its Supplements.

R=haraken@chromium.org
BUG=340522

Review URL: https://codereview.chromium.org/212933005

git-svn-id: svn://svn.chromium.org/blink/trunk@170146 bbb929c8-8fbe-4397-9dbb-9b2b20218538

r170014 changed multiple GCs to be default, but this test seems to issue GC too much.
This CL modifies the test to call a single GC instead of multiple GCs to avoid taking too much time on the test.
This matches the previous behavior before r170014

BUG=357001

Review URL: https://codereview.chromium.org/214293002

git-svn-id: svn://svn.chromium.org/blink/trunk@170145 bbb929c8-8fbe-4397-9dbb-9b2b20218538

There's currently a comment in core.gypi that reads:
"""
We should probably add CSSUnknownRule.idl to this list,
but it currently causes a compile error.
"""
Now that we've switched to Python (and proper IDL processing),
this no longer causes a compile error, so I think we can switch it on.

Changes:
This adds CSSUnknownRule to the global constructors.

This dates to April 2013:
Introduce bindings.gyp
https://codereview.chromium.org/14110002

Review URL: https://codereview.chromium.org/212653006

git-svn-id: svn://svn.chromium.org/blink/trunk@170144 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Current generated bindings for attribute getters has a
special case for |holder|, as this is sometimes stored as a
local variable.
This simplifies the template flow.

R=haraken
BUG=345503

Review URL: https://codereview.chromium.org/214143003

git-svn-id: svn://svn.chromium.org/blink/trunk@170143 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Enabling the blink_gc_plugin_dump_graph flag will cause the Blink GC plugin for clang to generate files with points-to information for each compilation unit. The output files can be used to statically analyze the complete points-to graph for cycles and other correctness criteria.

R=ager@chromium.org
BUG=

Review URL: https://codereview.chromium.org/207393003

git-svn-id: svn://svn.chromium.org/blink/trunk@170142 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This CL makes the compositing updates for the Polymer calculator sample app
30.7% faster by adding a tree-ordered tree walk before computing compositing
requirements.

Previously, we computed the absolute rects for RenderLayers during a
paint-ordered tree walk in computeCompositingRequirements. However, that's
inefficient for three reasons:

1) Most of the time is spent maintaining the RenderGeometryMap, but the
   RenderGeometryMap is optimized for tree-ordered tree walks. If the
   paint-order walk and the tree-order walk were different, we would thrash the
   geometry map.

2) The dirty bits for the absolute rects need to propagate down the tree-order,
   but when walking in paint-order, we didn't have a good way to propagate the
   bits to the tree-order decendants. Instead, we would set update type to
   "force" for the rest of the siblings in the paint-order walk, which is both
   inefficient and incorrect.

3) Finally, this change lets us move the RenderGeometryMap out of
   computeCompositingRequirements entirely, which means we don't need to
   maintain the map as we visit every RenderLayer in the system. Intead, we
   need to maintain the map only for the RenderLayers that we visit in the
   tree-ordered walk, which is pruned by dirty bits.

Now that we have this tree-ordered phase, we can move a bunch of other
computation of tree-ordered data to this phase. Currently we compute that data
during the paint-ordered walks by walking the tree excessively. However, that
works is reserved for a future CL.

R=ojan@chromium.org

Review URL: https://codereview.chromium.org/213773002

git-svn-id: svn://svn.chromium.org/blink/trunk@170141 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=
R=pfeldman@chromium.org

Review URL: https://codereview.chromium.org/205563007

git-svn-id: svn://svn.chromium.org/blink/trunk@170140 bbb929c8-8fbe-4397-9dbb-9b2b20218538

hostInput() is null checked in several other SliderThumbElement
functions and clearly can return null, so just handle it.

Add GCController.collect() to the test case to make the crash trigger
reliably, which was necessary in order to debug it.

BUG=356148

Review URL: https://codereview.chromium.org/211403003

git-svn-id: svn://svn.chromium.org/blink/trunk@170139 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Becomes:
 * hmac-sign-verify.html
 * sign-verify-badParameters.html

Motivation: I am splitting out per-algorithm tests so the tests are more readable.

Also changing how the error expectations are tested to simplify an upcoming change where the result is no longer just "null".

BUG=245025

Review URL: https://codereview.chromium.org/199663006

git-svn-id: svn://svn.chromium.org/blink/trunk@170138 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Also adds additional tests for extractable=false.

BUG=245025

Review URL: https://codereview.chromium.org/214233002

git-svn-id: svn://svn.chromium.org/blink/trunk@170137 bbb929c8-8fbe-4397-9dbb-9b2b20218538

With the previous defaults, tests could timeout before the LayoutTests
timeout, which would be reported as a failure. Making the timeout very
long should fix tests that are flaky with the 2 second timeout and will
make the SlowTests mechanism work for testharness.js tests, should the
need arise.

BUG=355489

Review URL: https://codereview.chromium.org/213283002

git-svn-id: svn://svn.chromium.org/blink/trunk@170136 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This patch fixes a compositor chicken and egg bug in
LocalFrame::nodeImage where stale compositor state was read through
setBaseBackgroundColor. An updateLayoutAndStyleForPainting call has
been moved before setBaseBackgroundColor in nodeImage.

Additionally, two unnecessary protector RefPtrs have been removed
because the callers already ref'ed the objects. The Node RefPtr is
ref'ed by Clipboard, the nodeImage() caller. The LocalFrame RefPtr
is ref'ed by DragController::doSystemDrag.

As a cleanup, LocalFrame::nodeImage has been changed to
take a reference since the node cannot be null.

BUG=354373

Review URL: https://codereview.chromium.org/214053003

git-svn-id: svn://svn.chromium.org/blink/trunk@170135 bbb929c8-8fbe-4397-9dbb-9b2b20218538

This set of tests all attempt to do a "repaint" test but aren't about
painting at all.  If the display() call from repaint.js is removed,
these tests all continue to pass, indicating that their output pixels
are not affected by the WebTestProxy::displayRepaintMask that covers
part of the screen in grey to indicate what has changed.

Because of this, it also means that they are not a good candidate for a
true text-based repaint rect, as they are either spit out no repaints or
full-screen repaints, depending on when paint rect tracking begins.

My conclusion is that these tests are really trying to test some
incremental layout- or style-related change, and so merely forcing a
style/layout recalc, tickling some property, and outputting the final
test pixels is a sufficient test.

BUG=345027

Review URL: https://codereview.chromium.org/208583004

git-svn-id: svn://svn.chromium.org/blink/trunk@170134 bbb929c8-8fbe-4397-9dbb-9b2b20218538

Don't use dangling reference in ReplaceSelectionCommand::removeRedundantStylesAndKeepStyleSpanInline

This patch changes to avoid dangling reference in ReplaceSelectionCommand::removeRedundantStylesAndKeepStyleSpanInline().
In this function, we hold pointer to inlineStyle member in ElemendData from
|node|. The |node| doesn't change in loop body unless it doesn't have a
self-conflicting inline style, e.g. <b style="font-weight: normal">. In this
case, we should update |inlineStyle| variable to new value from ElementData of
newly created element.

Note: Attached test case is failed only on ASAN build by heap-use-after-free.

BUG=356517
TEST=LayoutTests/editing/execCommand/remove-redundant-styles-and-keep-style-span-inline-crashed.html

Review URL: https://codereview.chromium.org/212323004

git-svn-id: svn://svn.chromium.org/blink/trunk@170133 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=170119

TBR=fmalita@chromium.org

Review URL: https://codereview.chromium.org/212653007

git-svn-id: svn://svn.chromium.org/blink/trunk@170132 bbb929c8-8fbe-4397-9dbb-9b2b20218538

If an element has at least one shadow tree, the ::shadow pseudo-element matches the shadow roots themselves.

c.f. Shadow DOM styling spec:
http://drafts.csswg.org/css-scoping/#shadow-pseudoelement

The new spec doesn't support /shadow/ combinator. Instead, it supports ::shadow.

BUG=355954
TEST=fast/dom/shadow/content-pseudo-element-with-hat.html,fast/dom/shadow/querySelector-for-multiple-shadowroots.html,fast/dom/shadow/querySelector-for-useragent-shadowroot.html,fast/dom/shadow/querySelector-with-shadow-all-and-shadow-deep.html,fast/dom/shadow/style-with-hat.html

Review URL: https://codereview.chromium.org/210713002

git-svn-id: svn://svn.chromium.org/blink/trunk@170131 bbb929c8-8fbe-4397-9dbb-9b2b20218538

http://src.chromium.org/viewvc/blink?view=revision&revision=170115

BUG=345452
TBR=jchaffraix@chromium.org

Review URL: https://codereview.chromium.org/214243002

git-svn-id: svn://svn.chromium.org/blink/trunk@170130 bbb929c8-8fbe-4397-9dbb-9b2b20218538

BUG=347001,347002,347003,347004
TBR=enne@chromium.org

Review URL: https://codereview.chromium.org/212663009

git-svn-id: svn://svn.chromium.org/blink/trunk@170129 bbb929c8-8fbe-4397-9dbb-9b2b20218538