WebKit-https.git
4 years agoDevice motion and orientation should only be visible from the main frame's security...
dino@apple.com [Tue, 13 Oct 2015 21:46:10 +0000 (21:46 +0000)]
Device motion and orientation should only be visible from the main frame's security origin
https://bugs.webkit.org/show_bug.cgi?id=150072
<rdar://problem/23082036>

Reviewed by Brent Fulgham.

.:

Add a manual test for cross-origin device orientation events, while
we're waiting on the mock client to be supported everywhere.

* ManualTests/deviceorientation-child-frame.html: Added.
* ManualTests/deviceorientation-main-frame-only.html: Added.

Source/WebCore:

There are reports that gyroscope and accelerometer information can
be used to detect keyboard entry. One initial step to reduce the
risk is to forbid device motion and orientation events from
being fired in frames that are a different security origin from the main page.

Manual test: deviceorientation-main-frame-only.html

* page/DOMWindow.cpp:
(WebCore::DOMWindow::isSameSecurityOriginAsMainFrame): New helper function.
(WebCore::DOMWindow::addEventListener): Check if we are the main frame, or the
same security origin as the main frame. If not, don't add the event
listeners.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@191008 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoASSERT(m_motionManager) on emgn.com page
dino@apple.com [Tue, 13 Oct 2015 21:45:43 +0000 (21:45 +0000)]
ASSERT(m_motionManager) on emgn.com page
https://bugs.webkit.org/show_bug.cgi?id=150070
<rdar://problem/18383732>

Reviewed by Simon Fraser.

In the WebCoreMotionManager init method, we call
into another setup method on the main thread.
However, if a listener was attached before that
ran, we'd ASSERT. This wasn't actually causing a bug
because the main thread initialization would
check the listeners once it got a chance to run.

The fix is to only check status when we've
actually initialized.

* platform/ios/WebCoreMotionManager.h: New m_initialized boolean.
* platform/ios/WebCoreMotionManager.mm: Check m_initialized before doing real work.
(-[WebCoreMotionManager init]):
(-[WebCoreMotionManager addMotionClient:]):
(-[WebCoreMotionManager removeMotionClient:]):
(-[WebCoreMotionManager addOrientationClient:]):
(-[WebCoreMotionManager removeOrientationClient:]):
(-[WebCoreMotionManager initializeOnMainThread]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@191007 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoA lot of the http/tests/cache/disk-cache/ tests are flaky on mac-wk2, marking them...
commit-queue@webkit.org [Tue, 13 Oct 2015 21:28:21 +0000 (21:28 +0000)]
A lot of the http/tests/cache/disk-cache/ tests are flaky on mac-wk2, marking them as such.
https://bugs.webkit.org/show_bug.cgi?id=149087

Patch by Ryan Haddad <ryanhaddad@apple.com> on 2015-10-13
Reviewed by Alexey Proskuryakov.

* platform/mac-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@191006 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoMac Debug and 32-bit queues should be separate
ap@apple.com [Tue, 13 Oct 2015 20:30:41 +0000 (20:30 +0000)]
Mac Debug and 32-bit queues should be separate
https://bugs.webkit.org/show_bug.cgi?id=150092

Reviewed by Lucas Forschler.

* QueueStatusServer/config/queues.py:
* QueueStatusServer/model/queues.py:
(Queue._capitalize_after_dash):
(Queue._caplitalize_after_dash): Deleted.
* Scripts/webkitpy/common/config/ews.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@191005 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION(r189567): Web Inspector: Clipped filter icons when debugger sidebar is...
nvasilyev@apple.com [Tue, 13 Oct 2015 20:23:10 +0000 (20:23 +0000)]
REGRESSION(r189567): Web Inspector: Clipped filter icons when debugger sidebar is narrow
https://bugs.webkit.org/show_bug.cgi?id=150023

r189567 changed flexbox items' default min-width from 0 to auto.
Explicitly set it to 0 in CSS.

Reviewed by Brian Burg.

* UserInterface/Views/FilterBar.css:
(.filter-bar > input[type="search"]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@191004 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoCodeBlock write barriers should be precise
ggaren@apple.com [Tue, 13 Oct 2015 20:08:54 +0000 (20:08 +0000)]
CodeBlock write barriers should be precise
https://bugs.webkit.org/show_bug.cgi?id=150042

Reviewed by Saam Barati.

CodeBlock performs lots of unnecessary write barriers. This wastes
performance and makes the code a bit harder to follow, and it might mask
important bugs. Now is a good time to unmask important bugs.

* bytecode/CodeBlock.h:
(JSC::CodeBlockSet::mark): Don't write barrier all CodeBlocks on the
stack. Only CodeBlocks that do value profiling need write barriers, and
they do those themselves.

In steady state, when most of our CodeBlocks are old and FTL-compiled,
and we're doing eden GC's, we should almost never visit a CodeBlock.

* dfg/DFGOSRExitCompilerCommon.cpp:
(JSC::DFG::osrWriteBarrier):
(JSC::DFG::adjustAndJumpToTarget): Don't write barrier all inlined
CodeBlocks on exit. That's not necessary. Instead, write barrier the
CodeBlock(s) we will exit to, along with the one we will write a value
profile to.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@191003 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAvoid useless copies in range-loops that are using 'auto'
cdumez@apple.com [Tue, 13 Oct 2015 19:15:55 +0000 (19:15 +0000)]
Avoid useless copies in range-loops that are using 'auto'
https://bugs.webkit.org/show_bug.cgi?id=150091

Reviewed by Sam Weinig.

Avoid useless copies in range-loops that are using 'auto'. Also use
'auto*' instead of 'auto' when range values are pointers for clarity.
Source/bmalloc:

* bmalloc/Deallocator.cpp:
(bmalloc::Deallocator::processObjectLog):

Source/WebKit/mac:

* WebView/WebFrame.mm:
(-[WebFrame getDictationResultRanges:andMetadatas:]):

Source/WebKit2:

* UIProcess/VisitedLinkStore.cpp:
(WebKit::VisitedLinkStore::pendingVisitedLinksTimerFired):
(WebKit::VisitedLinkStore::resizeTable):
* UIProcess/WebProcessProxy.cpp:
(WebKit::WebProcessProxy::releaseRemainingIconsForPageURLs):
* UIProcess/WebsiteData/WebsiteDataStore.cpp:
(WebKit::WebsiteDataStore::fetchData):
(WebKit::WebsiteDataStore::removeData):
(WebKit::WebsiteDataStore::plugins):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@191002 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: Don't shadow global Object constructor in WebInspector.Object class...
utatane.tea@gmail.com [Tue, 13 Oct 2015 19:03:46 +0000 (19:03 +0000)]
Web Inspector: Don't shadow global Object constructor in WebInspector.Object class definition
https://bugs.webkit.org/show_bug.cgi?id=150093

Reviewed by Joseph Pecoraro.

Currently, class expression does not define any variables. So even in `class Object { }`,
we were able to touch the global Object constructor (like `Object.keys`).
However, that is a bug. After https://bugs.webkit.org/show_bug.cgi?id=150089 fixes that bug,
the global `Object` is shadowed by the user-defined `class Object` name.

To solve this issue, we changed the `class Object` to `class WebInspectorObject`.
And we expose this class as a `WebInspector.Object`.

* UserInterface/Base/Object.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@191001 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoMove Image::drawPattern for CG into GraphicsContext
simon.fraser@apple.com [Tue, 13 Oct 2015 19:03:37 +0000 (19:03 +0000)]
Move Image::drawPattern for CG into GraphicsContext
https://bugs.webkit.org/show_bug.cgi?id=150077

Reviewed by Myles C. Maxfield.

In order to consolidate code that calls into Core Graphics inside
GraphicsContext, move the body of Image::drawPattern() into
GraphicsContextCG.cpp, and do the same for Cairo.

* platform/graphics/GraphicsContext.h:
* platform/graphics/cairo/GraphicsContextCairo.cpp:
(WebCore::GraphicsContext::drawPattern):
* platform/graphics/cairo/ImageCairo.cpp:
(WebCore::Image::drawPattern):
* platform/graphics/cg/GraphicsContextCG.cpp:
(WebCore::drawPatternCallback):
(WebCore::patternReleaseCallback):
(WebCore::GraphicsContext::drawPattern):
* platform/graphics/cg/ImageCG.cpp:
(WebCore::Image::drawPattern):
(WebCore::drawPatternCallback): Deleted.
(WebCore::patternReleaseCallback): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@191000 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnprefix font-kerning
mmaxfield@apple.com [Tue, 13 Oct 2015 18:49:58 +0000 (18:49 +0000)]
Unprefix font-kerning
https://bugs.webkit.org/show_bug.cgi?id=150080

Reviewed by Sam Weinig.

This is the last property in CSS3 Fonts which is prefixed.

Source/WebCore:

Test: fast/text/font-kerning.html

* css/CSSPropertyNames.in:

LayoutTests:

* fast/text/font-kerning-expected.html:
* fast/text/font-kerning.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190999 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd debug settings for using giant tiles (4096x4096)
commit-queue@webkit.org [Tue, 13 Oct 2015 18:46:53 +0000 (18:46 +0000)]
Add debug settings for using giant tiles (4096x4096)
https://bugs.webkit.org/show_bug.cgi?id=149977
<rdar://problem/23017093>

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2015-10-13
Reviewed by Tim Horton.

Source/WebCore:

Instead of creating the GraphicsLayer with a fixed size 512x512, we need
to read the useGiantTiles setting. If its value is true, we set the layer
tileSize to 4096x4096.

* page/Settings.in:
Define the name of the setting and its default value.

* platform/graphics/GraphicsLayerClient.h:
(WebCore::GraphicsLayerClient::tileSize):
Define tileSize() in the base class GraphicsLayerClient. This is going to
be overridden RenderLayerBacking.

* platform/graphics/TiledBacking.h:
(WebCore::defaultTileSize):
Define the default tileSize which will depend on the useGiantTiles
setting.

* platform/graphics/ca/GraphicsLayerCA.cpp:
(WebCore::GraphicsLayerCA::platformCALayerTileSize):
Implement the virtual function GraphicsLayerCA::platformCALayerTileSize().
It passes the call to client GraphicsLayerClient which can be RenderLayerBacking
in our case.

* platform/graphics/ca/GraphicsLayerCA.h:
Override base class function PlatformCALayerClient::PlatformCALayerClient().

* platform/graphics/ca/PlatformCALayerClient.h:
(WebCore::PlatformCALayerClient::platformCALayerTileSize):
Define platformCALayerTileSize() in the base class PlatformCALayerClient.
This is going to be overridden GraphicsLayerCA.

* platform/graphics/ca/TileController.cpp:
(WebCore::TileController::TileController):
No need for the member m_tileSize.

(WebCore::TileController::computeTileCoverageRect):
Use the function tileSize() instead of using the static values.

(WebCore::TileController::tileSize):
The tileSize will be retrieved from the owning graphics layer.

* platform/graphics/ca/TileController.h:
No need for the member m_tileSize. The tileSize will be retrieved from the owning graphics layer.

* rendering/RenderLayerBacking.cpp:
(WebCore::RenderLayerBacking::setTiledBackingHasMargins):
Use the function tileSize() instead of using the static values.

(WebCore::RenderLayerBacking::tileSize):
Override base class function GraphicsLayerClient::tileSize().

* rendering/RenderLayerBacking.h:
Define the concrete method RenderLayerBacking::tilSize().

Source/WebKit2:

* Shared/WebPreferencesDefinitions.h:
* UIProcess/API/C/WKPreferences.cpp:
(WKPreferencesSetUseGiantTiles):
(WKPreferencesGetUseGiantTiles):
* UIProcess/API/C/WKPreferencesRefPrivate.h:
* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updatePreferences):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190998 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoTry to fix ENABLE(DETAILS_ELEMENT) with SHADOW_DOM disabled.
antti@apple.com [Tue, 13 Oct 2015 18:34:22 +0000 (18:34 +0000)]
Try to fix ENABLE(DETAILS_ELEMENT) with SHADOW_DOM disabled.

* dom/Element.cpp:
(WebCore::Element::attributeChanged):
* dom/Node.cpp:
(WebCore::traverseStyleParent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190995 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agohttp/tests/multipart/multipart-replace-non-html-content.php shouldn't exercise
ap@apple.com [Tue, 13 Oct 2015 18:05:17 +0000 (18:05 +0000)]
http/tests/multipart/multipart-replace-non-html-content.php shouldn't exercise
edge cases of multipart parsing
https://bugs.webkit.org/show_bug.cgi?id=149978
rdar://problem/22981062

Reviewed by Anders Carlsson.

This test is not about HTTP edge cases, but about handling of multipart content in WebKit.

* http/tests/multipart/multipart-replace-non-html-content.php:
* platform/mac/http/tests/multipart/multipart-replace-non-html-content-expected.txt: Removed.
* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190994 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFix license and copyrights of WebCore js binding builtin files
youenn.fablet@crf.canon.fr [Tue, 13 Oct 2015 17:44:18 +0000 (17:44 +0000)]
Fix license and copyrights of WebCore js binding builtin files
https://bugs.webkit.org/show_bug.cgi?id=150027

Reviewed by Darin Adler.

Fixing copyright on all three files.
Moving to BSD-like license as they should have been in the first place.
Ordering lexicographically the builtin names in WebCoreBuiltinNames.h.

No change in behaviour.

* bindings/js/WebCoreBuiltinNames.h:
* bindings/js/WebCoreJSBuiltins.cpp:
* bindings/js/WebCoreJSBuiltins.h:
* bindings/js/WebCoreJSBuiltinInternals.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190993 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[iOS] Avoid crash due to invalid screen bounds
bfulgham@apple.com [Tue, 13 Oct 2015 16:54:16 +0000 (16:54 +0000)]
[iOS] Avoid crash due to invalid screen bounds
https://bugs.webkit.org/show_bug.cgi?id=150048
<rdar://problem/22112664>

Reviewed by Jer Noble.

Check for an invalid bounds, and reset it to a known state before
attempting to use it.

* WebProcess/ios/WebVideoFullscreenManager.mm:
(WebKit::WebVideoFullscreenManager::setVideoLayerFrameFenced):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190992 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION: ASSERT (impl->isAtomic()) @ facebook.com
utatane.tea@gmail.com [Tue, 13 Oct 2015 16:03:22 +0000 (16:03 +0000)]
REGRESSION: ASSERT (impl->isAtomic()) @ facebook.com
https://bugs.webkit.org/show_bug.cgi?id=149965

Reviewed by Geoffrey Garen.

Edge filtering for CheckIdent ensures that a given value is either Symbol or StringIdent.
However, this filtering is not applied to CheckIdent when propagating a constant value in
the constant folding phase. As a result, it is not guaranteeed that a constant value
propagated in constant folding is Symbol or StringIdent.

* dfg/DFGConstantFoldingPhase.cpp:
(JSC::DFG::ConstantFoldingPhase::foldConstants):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190991 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[iOS] Update anonymous table results for iOS port.
zalan@apple.com [Tue, 13 Oct 2015 15:42:50 +0000 (15:42 +0000)]
[iOS] Update anonymous table results for iOS port.

Unreviewed gardening.

* platform/ios-simulator/css2.1/tables/table-anonymous-objects-015-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-016-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-023-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-024-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-035-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-036-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-037-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-038-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-045-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-046-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-047-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-048-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-049-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-050-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-055-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-056-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-091-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-092-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-099-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-100-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-105-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-106-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-107-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-108-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-109-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-110-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-111-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-112-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-113-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-114-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-115-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-116-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-121-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-122-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-123-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-124-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-139-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-140-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-149-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-150-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-155-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-156-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-159-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-160-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-165-expected.txt: Added.
* platform/ios-simulator/css2.1/tables/table-anonymous-objects-166-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190990 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Fix build for ENABLE_TOUCH_EVENTS=OFF
svillar@igalia.com [Tue, 13 Oct 2015 14:42:23 +0000 (14:42 +0000)]
[GTK] Fix build for ENABLE_TOUCH_EVENTS=OFF
https://bugs.webkit.org/show_bug.cgi?id=150085

Reviewed by Carlos Garcia Campos.

Source/WebKit2:

* Shared/gtk/NativeWebTouchEventGtk.cpp:
* Shared/gtk/WebEventFactory.cpp:
* Shared/gtk/WebEventFactory.h:
* UIProcess/API/gtk/PageClientImpl.cpp:
* UIProcess/API/gtk/PageClientImpl.h:
* UIProcess/API/gtk/WebKitWebViewBase.cpp:
(webkit_web_view_base_class_init):

Tools:

* WebKitTestRunner/gtk/EventSenderProxyGtk.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190987 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoImplement iterator for traversing composed DOM
antti@apple.com [Tue, 13 Oct 2015 13:12:25 +0000 (13:12 +0000)]
Implement iterator for traversing composed DOM
https://bugs.webkit.org/show_bug.cgi?id=149997

Reviewed by Ryosuke Niwa.

Source/WebCore:

ComposedTreeIterator traverses the DOM in composed tree order. This means it enters
shadow trees and follows slots created by Shadow DOM API correctly.

auto children = composedTreeChildren(containerNode);
for (auto& composedChild : children)
    ...

auto descendants = composedTreeDescendants(containerNode);
for (auto& composedDescendant : descendants)
    ...

* WebCore.xcodeproj/project.pbxproj:
* dom/ComposedTreeIterator.cpp: Added.
(WebCore::ComposedTreeIterator::initializeShadowStack):
(WebCore::ComposedTreeIterator::traverseNextInShadowTree):
(WebCore::ComposedTreeIterator::traverseNextSiblingSlot):
(WebCore::ComposedTreeIterator::traversePreviousSiblingSlot):
(WebCore::ComposedTreeIterator::traverseParentInShadowTree):
* dom/ComposedTreeIterator.h: Added.
(WebCore::ComposedTreeIterator::operator*):
(WebCore::ComposedTreeIterator::operator->):
(WebCore::ComposedTreeIterator::operator==):
(WebCore::ComposedTreeIterator::operator!=):
(WebCore::ComposedTreeIterator::ShadowContext::ShadowContext):
(WebCore::ComposedTreeIterator::ComposedTreeIterator):
(WebCore::ComposedTreeIterator::traverseNext):
(WebCore::ComposedTreeIterator::traverseNextSibling):
(WebCore::ComposedTreeIterator::traversePreviousSibling):
(WebCore::ComposedTreeIterator::traverseParent):
(WebCore::ComposedTreeChildAdapter::Iterator::Iterator):
(WebCore::ComposedTreeChildAdapter::Iterator::operator++):
(WebCore::ComposedTreeChildAdapter::Iterator::operator--):
(WebCore::ComposedTreeChildAdapter::ComposedTreeChildAdapter):
(WebCore::ComposedTreeChildAdapter::begin):
(WebCore::ComposedTreeChildAdapter::end):
(WebCore::ComposedTreeChildAdapter::at):
(WebCore::composedTreeChildren):
* dom/NodeRenderingTraversal.cpp:
(WebCore::NodeRenderingTraversal::parentSlow):
(WebCore::NodeRenderingTraversal::nextInScope):
(WebCore::NodeRenderingTraversal::firstChildSlow): Deleted.
(WebCore::NodeRenderingTraversal::nextSiblingSlow): Deleted.
(WebCore::NodeRenderingTraversal::previousSiblingSlow): Deleted.
* dom/NodeRenderingTraversal.h:
(WebCore::NodeRenderingTraversal::parent):
(WebCore::NodeRenderingTraversal::firstChild): Deleted.
(WebCore::NodeRenderingTraversal::nextSibling): Deleted.
(WebCore::NodeRenderingTraversal::previousSibling): Deleted.
* style/RenderTreePosition.cpp:
(WebCore::RenderTreePosition::computeNextSibling):

    Restore the full assert.

(WebCore::RenderTreePosition::invalidateNextSibling):
(WebCore::RenderTreePosition::previousSiblingRenderer):
(WebCore::RenderTreePosition::nextSiblingRenderer):

    Make these member functions.
    Use the iterator. This is fixes some bugs and allows enabling a test case.

* style/RenderTreePosition.h:
* style/StyleResolveTree.cpp:
(WebCore::Style::textRendererIsNeeded):

LayoutTests:

* TestExpectations:

    Re-enable fast/html/details-replace-text.html which is fixed by this change.

* fast/forms/select-listbox-focus-displaynone-expected.txt:
* fast/repaint/text-in-relative-positioned-inline-expected.txt:
* fullscreen/full-screen-fixed-pos-parent-expected.txt:
* platform/mac-mavericks/fast/html/details-open2-expected.txt:
* platform/mac/fast/html/details-add-child-2-expected.txt:
* platform/mac/fast/html/details-open2-expected.txt:

    Non-visual whitespace changes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190983 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Use GUniquePtr for GtkIconInfo
changseok.oh@collabora.com [Tue, 13 Oct 2015 11:30:48 +0000 (11:30 +0000)]
[GTK] Use GUniquePtr for GtkIconInfo
https://bugs.webkit.org/show_bug.cgi?id=150082

Reviewed by Carlos Garcia Campos.

GtkIconInfo cab be wrapped by a smart pointer, GUniqueptr.

No new test required since no functionality changed.

* rendering/RenderThemeGtk.cpp:
(WebCore::getStockSymbolicIconForWidgetType):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190970 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, add myself to the committers list.
slomo@coaxion.net [Tue, 13 Oct 2015 10:53:47 +0000 (10:53 +0000)]
Unreviewed, add myself to the committers list.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190959 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAvoid useless copying of Key::HashType in loops in NetworkCache::Storage::synchronize()
cdumez@apple.com [Tue, 13 Oct 2015 08:53:21 +0000 (08:53 +0000)]
Avoid useless copying of Key::HashType in loops in NetworkCache::Storage::synchronize()
https://bugs.webkit.org/show_bug.cgi?id=150061

Reviewed by Carlos Garcia Campos.

Avoid useless copying of Key::HashType in loops in NetworkCache::Storage::synchronize().
Key::HashType is currently a SHA1::Digest, which is a std::array<uint8_t, 20>.

* NetworkProcess/cache/NetworkCacheStorage.cpp:
(WebKit::NetworkCache::Storage::synchronize):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190930 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, register symbol structure to fix Debug build
utatane.tea@gmail.com [Tue, 13 Oct 2015 06:30:02 +0000 (06:30 +0000)]
Unreviewed, register symbol structure to fix Debug build
https://bugs.webkit.org/show_bug.cgi?id=149622

Since InferredTypes for String or Symbol claim that they don't have any structure,
`registerInferredType` does not register the structure for Symbol.
We take the similar way to String to fix this issue; Registering Symbol structure
explicitly in DFGStructureRegisterationPhase. Because,

1. InferredType::structure is only allowed for ObjectWithStructure / ObjectWithStructureOrOther.
   It looks clear to me that only ObjectWithStructure has structure.
2. Symbol is similar primitive value to String. So handling its structure in similar way to String is nice.

* dfg/DFGStructureRegistrationPhase.cpp:
(JSC::DFG::StructureRegistrationPhase::run):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190927 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, fix debug build warning.
ljaehun.lim@samsung.com [Tue, 13 Oct 2015 06:00:00 +0000 (06:00 +0000)]
Unreviewed, fix debug build warning.

%llu needs 'long long unsigned int'.

Type casting unit64_t to long long unsigned int.

* Modules/indexeddb/client/IDBOpenDBRequestImpl.cpp:
(WebCore::IDBClient::IDBOpenDBRequest::onUpgradeNeeded):
* Modules/indexeddb/server/UniqueIDBDatabase.cpp:
(WebCore::IDBServer::UniqueIDBDatabase::notifyConnectionsOfVersionChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190926 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUpdate bot watcher's dashboard to reflect that EWS is on iOS 9 now
ap@apple.com [Tue, 13 Oct 2015 05:13:05 +0000 (05:13 +0000)]
Update bot watcher's dashboard to reflect that EWS is on iOS 9 now
https://bugs.webkit.org/show_bug.cgi?id=149983

Reviewed by Dan Bates.

* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BubbleQueueServer.js:
(BubbleQueueServer):
* BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/Dashboard.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190925 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFix webkitpy tests after r190922.
ap@apple.com [Tue, 13 Oct 2015 05:08:52 +0000 (05:08 +0000)]
Fix webkitpy tests after r190922.

* Scripts/webkitpy/common/config/ews.json:
* Scripts/webkitpy/tool/commands/earlywarningsystem.py:
* Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190924 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoIterator loops over key twice after delete
utatane.tea@gmail.com [Tue, 13 Oct 2015 03:24:02 +0000 (03:24 +0000)]
Iterator loops over key twice after delete
https://bugs.webkit.org/show_bug.cgi?id=149811

Reviewed by Geoffrey Garen.

When an object is the dictionary mode, JSPropertyNameEnumerator collects property names through generic property name enumeration `getPropertyNames`.
The result vector contains indexed property names. But in this case, `publicLength()` may not be 0.
So without disabling indexed names enumeration phase explicitly, JSPropertyNameEnumerator produces indexed property names twice.
One in indexed name enumeration phase, and another in generic property name enumeration phase.
This patch disables indexed names enumeration by setting `indexedLength` to 0 when collecting names through generic property name enumeration.

* runtime/JSPropertyNameEnumerator.h:
(JSC::propertyNameEnumerator):
* tests/stress/property-name-enumerator-should-not-look-into-indexed-values-when-it-is-a-dictionary.js: Added.
(shouldBe):
(col2.of.Reflect.enumerate):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190923 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd a Mac debug EWS queue
ap@apple.com [Tue, 13 Oct 2015 02:49:33 +0000 (02:49 +0000)]
Add a Mac debug EWS queue
https://bugs.webkit.org/show_bug.cgi?id=150069

Reviewed by Geoffrey Garen.

* QueueStatusServer/config/queues.py:
* Scripts/webkitpy/common/config/ews.json:
* Scripts/webkitpy/tool/commands/earlywarningsystem.py:
(AbstractEarlyWarningSystem):
(AbstractEarlyWarningSystem.archive_last_test_results):
(AbstractEarlyWarningSystem.build_style):
(AbstractEarlyWarningSystem.refetch_patch):
(AbstractEarlyWarningSystem.load_ews_classes):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190922 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd canvas line dash test
jonlee@apple.com [Tue, 13 Oct 2015 02:12:11 +0000 (02:12 +0000)]
Add canvas line dash test
https://bugs.webkit.org/show_bug.cgi?id=150078
<rdar://problem/23082347>

Reviewed by Dean Jackson.

* Animometer/runner/resources/tests.js: Add line dash test.
* Animometer/tests/simple/resources/simple-canvas-paths.js:
(CanvasLineDashStage): Create a new stage with a default dash pattern and stroke style.
Maintain a step which will animate the dash.
(CanvasLineDashStage.prototype.animate): Render the line path with the dash. Increment
the dash offset to animate.
(CanvasPathBenchmark.prototype.createStage): Add the test.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190921 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd canvas line path tests
jonlee@apple.com [Tue, 13 Oct 2015 02:11:44 +0000 (02:11 +0000)]
Add canvas line path tests
https://bugs.webkit.org/show_bug.cgi?id=150076
<rdar://problem/23082285>

Reviewed by Dean Jackson.

* Animometer/runner/resources/tests.js: Add new line path tests, with different
line joins.
* Animometer/tests/simple/resources/simple-canvas-paths.js:
(CanvasLinePathStage): The stage is the same as the SimpleCanvasPathStrokeStage
but sets the lineJoin on the context.
(CanvasPathBenchmark.prototype.createStage): Add the tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190920 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd missing paint objects for arc and arcTo fills
jonlee@apple.com [Tue, 13 Oct 2015 02:11:06 +0000 (02:11 +0000)]
Add missing paint objects for arc and arcTo fills
https://bugs.webkit.org/show_bug.cgi?id=150075
<rdar://problem/23082233>

Reviewed by Dean Jackson.

* Animometer/tests/simple/resources/simple-canvas-paths.js: Add missing paint objects
for arcTo and arc segment fills.
(CanvasArcToSegmentFill):
(CanvasArcToSegmentFill.prototype.draw):
(CanvasArcSegmentFill):
(CanvasArcSegmentFill.prototype.draw):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190919 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd canvas line segment tests
jonlee@apple.com [Tue, 13 Oct 2015 02:10:42 +0000 (02:10 +0000)]
Add canvas line segment tests
https://bugs.webkit.org/show_bug.cgi?id=150073
<rdar://problem/23082138>

Reviewed by Dean Jackson.

* Animometer/runner/resources/tests.js: Add new line segment tests, with different
line caps.
* Animometer/tests/simple/resources/simple-canvas-paths.js:
(CanvasLineSegment): Add new line segment paint object.
(CanvasLineSegmentStage): Create a new stage that sets the lineCap.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190918 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd canvas path fill tests
jonlee@apple.com [Tue, 13 Oct 2015 02:10:08 +0000 (02:10 +0000)]
Add canvas path fill tests
https://bugs.webkit.org/show_bug.cgi?id=150071
<rdar://problem/23082001>

Reviewed by Dean Jackson.

* Animometer/runner/resources/tests.js: Add new pathTypes for path fills.
* Animometer/tests/simple/resources/simple-canvas-paths.js:
(CanvasLinePoint): Add basic point for a line, and call lineTo.
(SimpleCanvasPathFillStage): Add a new stage similar to SimpleCanvasPathStrokeStage.
(CanvasPathBenchmark.prototype.createStage): Add the tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190917 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoIntroduce Symbol type for property type inference
utatane.tea@gmail.com [Tue, 13 Oct 2015 02:09:34 +0000 (02:09 +0000)]
Introduce Symbol type for property type inference
https://bugs.webkit.org/show_bug.cgi?id=149622

Reviewed by Geoffrey Garen.

This patch introduces Symbol type into property type inference.
One of the use cases of ES6 Symbol is enum value. In this case,
we may hold different symbols as the same property of the same structure.
Current property type inference does not support Symbol type, so in the
above case, the property will be inferred as Top type.

* bytecode/PutByIdFlags.h:
* dfg/DFGAbstractValue.cpp:
(JSC::DFG::AbstractValue::set):
* dfg/DFGInferredTypeCheck.cpp:
(JSC::DFG::insertInferredTypeCheck):
* ftl/FTLLowerDFGToLLVM.cpp:
(JSC::FTL::DFG::LowerDFGToLLVM::checkInferredType):
* jit/AssemblyHelpers.cpp:
(JSC::AssemblyHelpers::branchIfNotType):
* llint/LLIntData.cpp:
(JSC::LLInt::Data::performAssertions):
* llint/LowLevelInterpreter.asm:
* llint/LowLevelInterpreter32_64.asm:
* llint/LowLevelInterpreter64.asm:
* runtime/InferredType.cpp:
(JSC::InferredType::kindForFlags):
(JSC::InferredType::Descriptor::forValue):
(JSC::InferredType::Descriptor::putByIdFlags):
(JSC::InferredType::Descriptor::merge):
(WTF::printInternal):
* runtime/InferredType.h:
* tests/stress/prop-type-symbol-then-object.js: Added.
(foo):
(bar):
(toString):
* tests/stress/prop-type-symbol-then-string.js: Added.
(foo):
(bar):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190916 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Win] Update anonymous table results for Windows port.
zalan@apple.com [Tue, 13 Oct 2015 01:54:52 +0000 (01:54 +0000)]
[Win] Update anonymous table results for Windows port.

Unreviewed gardening.

* platform/win/css2.1/tables/table-anonymous-objects-015-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-016-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-023-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-024-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-035-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-036-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-037-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-038-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-045-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-046-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-047-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-048-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-049-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-050-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-055-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-056-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-091-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-092-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-099-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-100-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-105-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-106-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-107-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-108-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-109-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-110-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-111-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-112-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-113-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-114-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-115-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-116-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-121-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-122-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-123-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-124-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-139-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-140-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-149-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-150-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-155-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-156-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-159-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-160-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-165-expected.txt: Added.
* platform/win/css2.1/tables/table-anonymous-objects-166-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190915 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoSpeculative Cairo build fixes after r190910.
simon.fraser@apple.com [Tue, 13 Oct 2015 01:39:30 +0000 (01:39 +0000)]
Speculative Cairo build fixes after r190910.

* platform/graphics/cairo/ImageBufferCairo.cpp:
(WebCore::ImageBuffer::drawPattern):
* platform/graphics/cairo/ImageCairo.cpp:
(WebCore::Image::drawPattern):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190914 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd canvas path tests
jonlee@apple.com [Tue, 13 Oct 2015 01:18:03 +0000 (01:18 +0000)]
Add canvas path tests
https://bugs.webkit.org/show_bug.cgi?id=150067
rdar://problem/23081463

Reviewed by Dean Jackson.

* Animometer/runner/resources/tests.js: Add a quadratic and bezier path test.

* Animometer/tests/simple/resources/simple-canvas.js:
(SimpleCanvasStage.prototype.tune): This kind of test joins all of the segments
into one long path, and tries to render that one path. Random points make it
difficult to tell what is going on, so add a parameter to the constructor to
confine the area where the random coordinates can land. The more complicated the
case is, the larger an area the path will cover. Add an artificial minimum so
that the first 200 points aren't confined to a space that is too small.

* Animometer/tests/simple/resources/simple-canvas-paths.js:
(SimpleCanvasPathStrokeStage): Add a new kind of stage that inherits from
SimpleCanvasStage. Each time the frame animates a random line width and stroke
color are chosen. The path setup is done outside of each paint object.
(CanvasQuadraticPoint): This point just calls quadraticCurveTo.
(CanvasPathBenchmark.prototype.createStage): Add the tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190913 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd basic canvas tests
jonlee@apple.com [Tue, 13 Oct 2015 01:11:45 +0000 (01:11 +0000)]
Add basic canvas tests
https://bugs.webkit.org/show_bug.cgi?id=150066
rdar://problem/23081143

Reviewed by Dean Jackson.

This adds a new test suite that will cover all of the path-based canvas calls.
The patch will be divided up to cover tests with similar techniques.

The simplest version uses a SimpleCanvasStage.

* Animometer/runner/resources/tests.js: Add tests for quadratic, bezier, arcTo,
arc, and rect segments. Also include arcTo, arc, and rect fills.
* Animometer/tests/resources/stage.js:
(Stage.prototype.randomBool): Added for counterclockwise property for arc segments.
(Stage.prototype.randomInt): Fix how values are rounded, used by randomBool. It should
round instead of flooring everything.
* Animometer/tests/simple/resources/simple-canvas.js: Added. Defines common classes
used by all simple canvas tests. The page reads best bottom to top.
(SimpleCanvasStage): Basic stage. Pass a canvasObject which will be used to create new
objects as needed in tune().
(SimpleCanvasStage.prototype.tune): Modeled on other tests. Adds and removed objects
as specified by the provided |count|.
(SimpleCanvasStage.prototype.animate): Iterate over all the objects and ask them to draw.
There is no "animating" of the objects; they will just paint statically on the canvas.
(SimpleCanvasAnimator): Basic animator clears the canvas prior to painting.
(SimpleCanvasBenchmark): Hard-code the feedback loop parameters instead of including
them in the query parameters to the test URLs.
(SimpleCanvasBenchmark.prototype.createAnimator):
* Animometer/tests/simple/simple-canvas-paths.html: Added.

* Animometer/tests/simple/resources/simple-canvas-paths.js: Added. There is no "animating"
of these objects--they just paint statically on the canvas.
(CanvasQuadraticSegment): Paint a quadratic segment stroke.
(CanvasBezierSegment): Paint a bezier segment stroke.
(CanvasArcToSegment): Paint an arcTo stroke.
(CanvasArcSegment): Paint an arc stroke.
(CanvasRect): Paint a rect.
(CanvasRectFill): Paint a filled rect.

(CanvasPathBenchmark):
(CanvasPathBenchmark.prototype.createStage): Look for the pathType and create the
stage using the right paint object.
(window.benchmarkClient.create):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190912 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFix iOS and Efl builds.
simon.fraser@apple.com [Tue, 13 Oct 2015 00:55:06 +0000 (00:55 +0000)]
Fix iOS and Efl builds.

* platform/graphics/NamedImageGeneratedImage.cpp:
(WebCore::NamedImageGeneratedImage::drawPattern):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190911 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove Image::spaceSize() and ImageBuffer::spaceSize()
simon.fraser@apple.com [Tue, 13 Oct 2015 00:52:13 +0000 (00:52 +0000)]
Remove Image::spaceSize() and ImageBuffer::spaceSize()
https://bugs.webkit.org/show_bug.cgi?id=150064

Reviewed by Tim Horton.

Image spacing when tiled should not be a property of the image; but a description
of how it's drawn, like tile size. So remove spacing from Image and ImageBuffer,
and pass it in as an argument.

* platform/graphics/BitmapImage.cpp:
(WebCore::BitmapImage::drawPattern):
* platform/graphics/BitmapImage.h:
* platform/graphics/CrossfadeGeneratedImage.cpp:
(WebCore::CrossfadeGeneratedImage::drawPattern):
* platform/graphics/CrossfadeGeneratedImage.h:
* platform/graphics/GeneratedImage.h:
* platform/graphics/GradientImage.cpp:
(WebCore::GradientImage::drawPattern):
* platform/graphics/GradientImage.h:
* platform/graphics/GraphicsContext.cpp:
(WebCore::GraphicsContext::drawTiledImage):
* platform/graphics/GraphicsContext.h:
* platform/graphics/Image.cpp:
(WebCore::Image::drawTiled):
* platform/graphics/Image.h:
(WebCore::Image::spaceSize): Deleted.
(WebCore::Image::setSpaceSize): Deleted.
* platform/graphics/ImageBuffer.h:
(WebCore::ImageBuffer::spaceSize): Deleted.
(WebCore::ImageBuffer::setSpaceSize): Deleted.
* platform/graphics/NamedImageGeneratedImage.cpp:
(WebCore::NamedImageGeneratedImage::drawPattern):
* platform/graphics/NamedImageGeneratedImage.h:
* platform/graphics/cg/ImageBufferCG.cpp:
(WebCore::ImageBuffer::copyImage):
(WebCore::ImageBuffer::drawPattern):
* platform/graphics/cg/ImageCG.cpp:
(WebCore::Image::drawPattern):
* rendering/RenderBoxModelObject.cpp:
(WebCore::RenderBoxModelObject::paintFillLayerExtended):
* svg/graphics/SVGImage.cpp:
(WebCore::SVGImage::drawPatternForContainer):
* svg/graphics/SVGImage.h:
* svg/graphics/SVGImageForContainer.cpp:
(WebCore::SVGImageForContainer::drawPattern):
* svg/graphics/SVGImageForContainer.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190910 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] OSX linker doesn't understand --whole-archive
commit-queue@webkit.org [Tue, 13 Oct 2015 00:37:48 +0000 (00:37 +0000)]
[GTK] OSX linker doesn't understand --whole-archive
https://bugs.webkit.org/show_bug.cgi?id=144557

Patch by Philip Chimento <philip.chimento@gmail.com> on 2015-10-12
Reviewed by Martin Robinson.

.:

* Source/cmake/OptionsGTK.cmake: Turn the macro
ADD_WHOLE_ARCHIVE_TO_LIBRARIES into a no-op on Darwin systems,
because XCode's linker doesn't have the --whole-archive option.

Source/WebKit2:

* PlatformGTK.cmake: Link with extra libraries on Darwin, since
we don't have the --whole-archive linker option.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190909 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRefactor tune() to pass in an integer-based count
jonlee@apple.com [Tue, 13 Oct 2015 00:11:16 +0000 (00:11 +0000)]
Refactor tune() to pass in an integer-based count
https://bugs.webkit.org/show_bug.cgi?id=150060
<rdar://problem/23079425>

Reviewed by Dean Jackson.

All of the tune functions do an initial step of flooring the
# of painting objects to add or remove. Factor that out since
the complexity of all of these tests is always expressed in terms
of countable items.

* Animometer/tests/resources/main.js:
(Benchmark.prototype.update): Update the tuneValue.

* Animometer/tests/bouncing-particles/resources/bouncing-particles.js:
(BouncingParticlesStage.prototype.tune): Remove the count adjustment.
* Animometer/tests/examples/resources/canvas-electrons.js:
(CanvasElectronsStage.prototype.tune): Ditto.
* Animometer/tests/examples/resources/canvas-stars.js:
(CanvasStarsStage.prototype.tune): Ditto.
* Animometer/tests/text/resources/layering-text.js:
(LayeringTextStage.prototype.tune): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190908 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoStore test-interval in localStorage
jonlee@apple.com [Tue, 13 Oct 2015 00:03:31 +0000 (00:03 +0000)]
Store test-interval in localStorage
https://bugs.webkit.org/show_bug.cgi?id=150055
<rdar://problem/23078879>

Reviewed by Dean Jackson.

* Animometer/runner/resources/animometer.js:
(populateSettings): Keep track of the specified test interval in localStorage
to make it easy to repeat a test suite.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190907 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoHave TransformState::mappedSecondaryQuad() return an Optional<FloatQuad>.
akling@apple.com [Mon, 12 Oct 2015 23:39:48 +0000 (23:39 +0000)]
Have TransformState::mappedSecondaryQuad() return an Optional<FloatQuad>.
<https://webkit.org/b/150057>

Reviewed by Simon Fraser.

Using Optional instead of std::unique_ptr allows us to signal the absence of a
secondary quad without incurring a heap allocation in cases where one is present.

This dodges ~132000 malloc/free pairs on loading thelocal.se

* platform/graphics/ca/GraphicsLayerCA.cpp:
(WebCore::GraphicsLayerCA::computeVisibleAndCoverageRect):
* platform/graphics/transforms/TransformState.cpp:
(WebCore::TransformState::mappedSecondaryQuad):
* platform/graphics/transforms/TransformState.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190906 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoWeb Inspector: Rebaseline Inspector generator tests and make better use of RWIProtoco...
commit-queue@webkit.org [Mon, 12 Oct 2015 23:37:51 +0000 (23:37 +0000)]
Web Inspector: Rebaseline Inspector generator tests and make better use of RWIProtocol constant
https://bugs.webkit.org/show_bug.cgi?id=150044

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2015-10-12
Reviewed by Brian Burg.

* inspector/scripts/codegen/generate_objc_configuration_header.py:
(ObjCConfigurationHeaderGenerator.generate_output):
(ObjCConfigurationHeaderGenerator._generate_configuration_interface_for_domains):
* inspector/scripts/codegen/generate_objc_configuration_implementation.py:
(ObjCBackendDispatcherImplementationGenerator._generate_configuration_implementation_for_domains):
* inspector/scripts/codegen/generate_objc_header.py:
(ObjCHeaderGenerator.generate_output):
* inspector/scripts/codegen/generate_objc_internal_header.py:
(ObjCInternalHeaderGenerator.generate_output):
* inspector/scripts/tests/expected/commands-with-async-attribute.json-result:
* inspector/scripts/tests/expected/commands-with-optional-call-return-parameters.json-result:
* inspector/scripts/tests/expected/domains-with-varying-command-sizes.json-result:
* inspector/scripts/tests/expected/enum-values.json-result:
* inspector/scripts/tests/expected/events-with-optional-parameters.json-result:
* inspector/scripts/tests/expected/generate-domains-with-feature-guards.json-result:
* inspector/scripts/tests/expected/same-type-id-different-domain.json-result:
* inspector/scripts/tests/expected/shadowed-optional-type-setters.json-result:
* inspector/scripts/tests/expected/type-declaration-aliased-primitive-type.json-result:
* inspector/scripts/tests/expected/type-declaration-array-type.json-result:
* inspector/scripts/tests/expected/type-declaration-enum-type.json-result:
* inspector/scripts/tests/expected/type-declaration-object-type.json-result:
* inspector/scripts/tests/expected/type-requiring-runtime-casts.json-result:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190905 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed build fix
mmaxfield@apple.com [Mon, 12 Oct 2015 23:26:36 +0000 (23:26 +0000)]
Unreviewed build fix

* runtime/JSObject.cpp:
(JSC::JSObject::reallocateAndShrinkButterfly):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190904 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove "../tests/" from the URLs of all tests
jonlee@apple.com [Mon, 12 Oct 2015 23:25:41 +0000 (23:25 +0000)]
Remove "../tests/" from the URLs of all tests
https://bugs.webkit.org/show_bug.cgi?id=150054
<rdar://problem/23078784>

Reviewed by Dean Jackson.

Presumably all tests will be running from the tests/ directory,
so we don't have to specify it in all of the test URLs.

* Animometer/runner/resources/benchmark-runner.js:
(BenchmarkRunnerState.prototype.prepareCurrentTest): Prepend
"../tests/" to all tests.
* Animometer/runner/resources/tests.js: Remove "../tests/" from
all URLs.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190903 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRefactor test suites to a separate class.
jonlee@apple.com [Mon, 12 Oct 2015 23:22:23 +0000 (23:22 +0000)]
Refactor test suites to a separate class.
https://bugs.webkit.org/show_bug.cgi?id=150053
<rdar://problem/23078645>

Reviewed by Dean Jackson.

Create a Suite class to refactor out prepare() and run().
Generate the checkboxes representing the suites using Suites
instead of maintaining a separate list. Also, save the
selections out to localStorage.

* Animometer/runner/animometer.html: Remove the explicitly listed
suites. These will be generated from Suites instead.
* Animometer/runner/resources/animometer.js:
(populateSettings): Iterate through Suites, and create the
label and checkbox. Attach the Suite object to the checkbox so
when the benchmark is started, we get direct access. Initialize
the checkmark based on its value in localStorage. Set this to
run when DOMContentLoaded is dispatched.
(startBenchmark): Grab all of the checkboxes, inspect their
values, add it to enabledSuites if selected. Remember whether
the suite was enabled in localStorage, so that it's easy to do
repeated runs.
* Animometer/runner/resources/tests.js:
(Suite): Create a new Suite class. Refactor out prepare() and
run(), since all of them have the same implementation. Populate
Suites with Suite instances instead of generic objects.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190902 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUpdate graph styles for legibility.
jonlee@apple.com [Mon, 12 Oct 2015 23:20:24 +0000 (23:20 +0000)]
Update graph styles for legibility.
https://bugs.webkit.org/show_bug.cgi?id=150052
<rdar://problem/23078503>

Reviewed by Dean Jackson.

* Animometer/runner/resources/animometer.css: Update colors and
stroke thicknesses to make the graphs easier to read.
(.smaple-time): Correct to .sample-time
* Animometer/runner/resources/graph.js:
(graph): Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190901 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoPerf dashboard tools shouldn't require server credentials in multiple configuration...
rniwa@webkit.org [Mon, 12 Oct 2015 23:05:43 +0000 (23:05 +0000)]
Perf dashboard tools shouldn't require server credentials in multiple configuration files
https://bugs.webkit.org/show_bug.cgi?id=149994

Reviewed by Chris Dumez.

Made detect-changes.js and pull-svn.py pull username and passwords from the server config JSON to reduce
the number of JSON files that need to include credentials.

Also made each script reload the server config after sleep to allow dynamic credential updates.

In addition, change the server config JSON's format to include scheme, host, and port numbers separately
instead of a url since detect-changes.js needs each value separately.

This reduces the number of JSONs with credentials to two for our internal dashboard.

* tools/detect-changes.js:
(main): Added a property argument parsing. Now takes --server-config-json, --change-detection-config-json,
and --seconds-to-sleep like other scripts.
(parseArgument): Added.
(fetchManifestAndAnalyzeData): Reload the server config JSON.
(loadServerConfig): Added. Set settings.perfserver and settings.slave from the server config JSON. Also
add settings.perfserver.host to match the old format.
(configurationsForTesting): Fixed a bug that we throw an exception when a dashboard contains an empty cell.

* tools/pull-os-versions.py:
(main): Use load_server_config after each sleep.

* tools/pull-svn.py:
(main): Use load_server_config after each sleep.
(fetch_commits_and_submit): Use the perf dashboard's auth as subversion credential when useServerAuth is set.

* tools/sync-with-buildbot.py:
(main): Use load_server_config after each sleep.

* tools/util.py:
(load_server_config): Extracted from python scripts. Computes server's url from scheme, host, and port number
to match the old format python scripts except.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190900 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFollowup to:
beidson@apple.com [Mon, 12 Oct 2015 22:51:59 +0000 (22:51 +0000)]
Followup to:
Modern IDB: Start version change transaction for connections to new database.
https://bugs.webkit.org/show_bug.cgi?id=150033

Reviewed by NOBODY (Fixing existing test flakiness)

No new tests (Covered by existing tests)

* Modules/indexeddb/server/UniqueIDBDatabase.cpp:
(WebCore::IDBServer::UniqueIDBDatabase::handleOpenDatabaseOperations): Early return if a version change transaction
  is already in progress.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190899 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoBuild fix after r190817. Now that pull-os-versions store fake timestamps, we need...
rniwa@webkit.org [Mon, 12 Oct 2015 22:43:49 +0000 (22:43 +0000)]
Build fix after r190817.  Now that pull-os-versions store fake timestamps, we need to bypass timestamp
checks for OS versions when bots try to report new results. Otherwise, we fail to process the reports
with a MismatchingCommitTime error.

* public/include/report-processor.php:
(ReportProcessor::resolve_build_id):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190898 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, revert an unintended commit.
fpizlo@apple.com [Mon, 12 Oct 2015 22:42:21 +0000 (22:42 +0000)]
Unreviewed, revert an unintended commit.

* JetStream/Octane2/crypto.js:
(montRevert):
(montReduce):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190897 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoGC should have a Baker barrier for concurrent copying
fpizlo@apple.com [Mon, 12 Oct 2015 22:41:01 +0000 (22:41 +0000)]
GC should have a Baker barrier for concurrent copying
https://bugs.webkit.org/show_bug.cgi?id=149852

Reviewed by Geoffrey Garen.

This adds a Baker-style read barrier [1] to copied space accesses. This barrier incurs some
overhead (0%-2% depending on benchmark suite), but what it buys is the ability to make the GC copy
phase concurrent.

The barrier relies on copied space pointers having two "space bits" in the low pointer bits. The
space bits indicate whether the backing store is being copied right now or not, and if it is being
copied, what stage of copying it's in. Two barrier variants are supported:

Read only barrier: if you load a backing store and immediately load from it without doing anything
else, you can just mask off the bits. In the worst case, you'll get the old backing store while
some copying thread is already allocating and populating the new version of the backing store. But
in that case, forwarding to the new backing store will not enable you to load a more up-to-date
value from the backing store. So, just masking the bits is enough. The read-only barrier is only
used in ICs where we know that we are only reading, and opportunistically within the DFG and FTL
thanks to the CopyBarrierOptimizationPhase. We never explicitly emit a read-only barrier in those
compilers; instead the phase will turn a GetButterfly into GetButterflyReadOnly if it proves that a
bunch of requirements are met.

Normal barrier: if the space bits are non-zero, call a slow path. The slow path will either do
nothing (if the copy phase hasn't started yet), or it will copy the backing store and update the
pointer (if the copy phase hasn't gotten around to copying this particular backing store), or it
will wait for the copying thread to finish (if some thread is copying this backing store right
now), or it will do nothing (if by the time we called into the slow path the backing store was
already copied). This is just like Baker's CAR/CDR barrier, but with a lock thrown in to handle
concurrent execution.

This is a 1% slow-down on SunSpider, a 1.5% slow-down on Octane, a 1.5% slow-down on Kraken, and a
0% slow-down on AsmBench. Note that the Octane slow-down is excluding the SplayLatency benchmark.
That benchmark will eventually speed up a lot once we finish doing all of this stuff. Probably, the
JetStream splay-latency will see an even larger speed-up, since our version of the latency tests do
a better job of punishing bad worst-case behavior.

[1] http://dspace.mit.edu/bitstream/handle/1721.1/41976/AI_WP_139.pdf, look for the CAR and CDR
procedures on page 9.

* CMakeLists.txt:
* JavaScriptCore.vcxproj/JavaScriptCore.vcxproj:
* JavaScriptCore.xcodeproj/project.pbxproj:
* bytecode/PolymorphicAccess.cpp:
(JSC::AccessCase::generate):
* dfg/DFGAbstractInterpreterInlines.h:
(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):
* dfg/DFGArgumentsEliminationPhase.cpp:
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
* dfg/DFGCopyBarrierOptimizationPhase.cpp: Added.
(JSC::DFG::performCopyBarrierOptimization):
* dfg/DFGCopyBarrierOptimizationPhase.h: Added.
* dfg/DFGDoesGC.cpp:
(JSC::DFG::doesGC):
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
* dfg/DFGHeapLocation.cpp:
(WTF::printInternal):
* dfg/DFGHeapLocation.h:
* dfg/DFGLICMPhase.cpp:
(JSC::DFG::LICMPhase::run):
* dfg/DFGNodeType.h:
* dfg/DFGOperations.cpp:
* dfg/DFGOperations.h:
* dfg/DFGPlan.cpp:
(JSC::DFG::Plan::compileInThreadImpl):
* dfg/DFGPredictionPropagationPhase.cpp:
(JSC::DFG::PredictionPropagationPhase::propagate):
* dfg/DFGSafeToExecute.h:
(JSC::DFG::safeToExecute):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileGetIndexedPropertyStorage):
(JSC::DFG::SpeculativeJIT::compileGetTypedArrayByteOffset):
(JSC::DFG::SpeculativeJIT::compileReallocatePropertyStorage):
(JSC::DFG::SpeculativeJIT::compileGetButterfly):
(JSC::DFG::SpeculativeJIT::temporaryRegisterForPutByVal):
* dfg/DFGSpeculativeJIT.h:
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGTypeCheckHoistingPhase.cpp:
(JSC::DFG::TypeCheckHoistingPhase::identifyRedundantStructureChecks):
(JSC::DFG::TypeCheckHoistingPhase::identifyRedundantArrayChecks):
* ftl/FTLCapabilities.cpp:
(JSC::FTL::canCompile):
* ftl/FTLLowerDFGToLLVM.cpp:
(JSC::FTL::DFG::LowerDFGToLLVM::compileNode):
(JSC::FTL::DFG::LowerDFGToLLVM::compileGetButterfly):
(JSC::FTL::DFG::LowerDFGToLLVM::compileGetButterflyReadOnly):
(JSC::FTL::DFG::LowerDFGToLLVM::compileConstantStoragePointer):
(JSC::FTL::DFG::LowerDFGToLLVM::compileGetIndexedPropertyStorage):
(JSC::FTL::DFG::LowerDFGToLLVM::compileCheckArray):
(JSC::FTL::DFG::LowerDFGToLLVM::compileGetTypedArrayByteOffset):
(JSC::FTL::DFG::LowerDFGToLLVM::compileMultiGetByOffset):
(JSC::FTL::DFG::LowerDFGToLLVM::compileMultiPutByOffset):
(JSC::FTL::DFG::LowerDFGToLLVM::compileGetDirectPname):
(JSC::FTL::DFG::LowerDFGToLLVM::storageForTransition):
(JSC::FTL::DFG::LowerDFGToLLVM::getById):
(JSC::FTL::DFG::LowerDFGToLLVM::loadButterflyWithBarrier):
(JSC::FTL::DFG::LowerDFGToLLVM::loadVectorWithBarrier):
(JSC::FTL::DFG::LowerDFGToLLVM::copyBarrier):
(JSC::FTL::DFG::LowerDFGToLLVM::loadButterflyReadOnly):
(JSC::FTL::DFG::LowerDFGToLLVM::loadVectorReadOnly):
(JSC::FTL::DFG::LowerDFGToLLVM::removeSpaceBits):
(JSC::FTL::DFG::LowerDFGToLLVM::baseIndex):
* ftl/FTLOperations.cpp:
(JSC::FTL::operationNewObjectWithButterfly):
(JSC::FTL::operationPopulateObjectInOSR):
* ftl/FTLOutput.h:
(JSC::FTL::Output::testNonZero32):
(JSC::FTL::Output::testIsZero64):
(JSC::FTL::Output::testNonZero64):
(JSC::FTL::Output::testIsZeroPtr):
(JSC::FTL::Output::testNonZeroPtr):
(JSC::FTL::Output::select):
(JSC::FTL::Output::extractValue):
* heap/CopyBarrier.h: Copied from Source/JavaScriptCore/heap/CopyWriteBarrier.h.
(JSC::CopyBarrierBase::CopyBarrierBase):
(JSC::CopyBarrierBase::operator!):
(JSC::CopyBarrierBase::operator bool):
(JSC::CopyBarrierBase::getWithoutBarrier):
(JSC::CopyBarrierBase::get):
(JSC::CopyBarrierBase::copyState):
(JSC::CopyBarrierBase::setCopyState):
(JSC::CopyBarrierBase::clear):
(JSC::CopyBarrierBase::set):
(JSC::CopyBarrierBase::setWithoutBarrier):
(JSC::CopyBarrierBase::weakCASWithoutBarrier):
(JSC::CopyBarrier::CopyBarrier):
(JSC::CopyBarrier::getWithoutBarrier):
(JSC::CopyBarrier::get):
(JSC::CopyBarrier::set):
(JSC::CopyBarrier::setWithoutBarrier):
(JSC::CopyBarrier::weakCASWithoutBarrier):
(JSC::CopyWriteBarrier::CopyWriteBarrier): Deleted.
(JSC::CopyWriteBarrier::operator!): Deleted.
(JSC::CopyWriteBarrier::operator bool): Deleted.
(JSC::CopyWriteBarrier::get): Deleted.
(JSC::CopyWriteBarrier::operator*): Deleted.
(JSC::CopyWriteBarrier::operator->): Deleted.
(JSC::CopyWriteBarrier::set): Deleted.
(JSC::CopyWriteBarrier::setWithoutWriteBarrier): Deleted.
(JSC::CopyWriteBarrier::clear): Deleted.
* heap/CopyVisitorInlines.h:
(JSC::CopyVisitor::checkIfShouldCopy):
* heap/CopyWriteBarrier.h: Removed.
* heap/Heap.cpp:
(JSC::Heap::addToRememberedSet):
(JSC::Heap::copyBarrier):
(JSC::Heap::collectAndSweep):
* heap/Heap.h:
(JSC::Heap::writeBarrierBuffer):
* heap/HeapInlines.h:
* jit/AssemblyHelpers.h:
(JSC::AssemblyHelpers::branchStructure):
(JSC::AssemblyHelpers::branchIfNotToSpace):
(JSC::AssemblyHelpers::removeSpaceBits):
(JSC::AssemblyHelpers::addressForByteOffset):
* jit/JIT.cpp:
(JSC::JIT::privateCompileMainPass):
(JSC::JIT::privateCompileSlowCases):
* jit/JITOpcodes.cpp:
(JSC::JIT::emitSlow_op_has_indexed_property):
(JSC::JIT::emit_op_get_direct_pname):
(JSC::JIT::emitSlow_op_get_direct_pname):
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_get_direct_pname):
(JSC::JIT::emitSlow_op_get_direct_pname):
* jit/JITPropertyAccess.cpp:
(JSC::JIT::emitDoubleLoad):
(JSC::JIT::emitContiguousLoad):
(JSC::JIT::emitArrayStorageLoad):
(JSC::JIT::emitSlow_op_get_by_val):
(JSC::JIT::emitGenericContiguousPutByVal):
(JSC::JIT::emitArrayStoragePutByVal):
(JSC::JIT::emitSlow_op_put_by_val):
(JSC::JIT::emit_op_get_from_scope):
(JSC::JIT::emitSlow_op_get_from_scope):
(JSC::JIT::emit_op_put_to_scope):
(JSC::JIT::emitSlow_op_put_to_scope):
(JSC::JIT::emitIntTypedArrayGetByVal):
(JSC::JIT::emitFloatTypedArrayGetByVal):
(JSC::JIT::emitIntTypedArrayPutByVal):
(JSC::JIT::emitFloatTypedArrayPutByVal):
* llint/LowLevelInterpreter.asm:
* llint/LowLevelInterpreter64.asm:
* runtime/DirectArguments.cpp:
(JSC::DirectArguments::visitChildren):
(JSC::DirectArguments::copyBackingStore):
(JSC::DirectArguments::overrideThings):
(JSC::DirectArguments::overrideThingsIfNecessary):
(JSC::DirectArguments::overrideArgument):
(JSC::DirectArguments::copyToArguments):
* runtime/DirectArguments.h:
(JSC::DirectArguments::canAccessIndexQuickly):
(JSC::DirectArguments::canAccessArgumentIndexQuicklyInDFG):
* runtime/JSArray.cpp:
(JSC::JSArray::setLength):
(JSC::JSArray::pop):
(JSC::JSArray::push):
(JSC::JSArray::fastSlice):
(JSC::JSArray::fastConcatWith):
(JSC::JSArray::shiftCountWithArrayStorage):
(JSC::JSArray::shiftCountWithAnyIndexingType):
(JSC::JSArray::unshiftCountWithAnyIndexingType):
(JSC::JSArray::fillArgList):
(JSC::JSArray::copyToArguments):
* runtime/JSArrayBufferView.cpp:
(JSC::JSArrayBufferView::ConstructionContext::ConstructionContext):
(JSC::JSArrayBufferView::JSArrayBufferView):
(JSC::JSArrayBufferView::finishCreation):
(JSC::JSArrayBufferView::finalize):
* runtime/JSArrayBufferView.h:
(JSC::JSArrayBufferView::vector):
(JSC::JSArrayBufferView::length):
* runtime/JSArrayBufferViewInlines.h:
(JSC::JSArrayBufferView::neuter):
(JSC::JSArrayBufferView::byteOffset):
* runtime/JSGenericTypedArrayView.h:
(JSC::JSGenericTypedArrayView::typedVector):
* runtime/JSGenericTypedArrayViewInlines.h:
(JSC::JSGenericTypedArrayView<Adaptor>::visitChildren):
(JSC::JSGenericTypedArrayView<Adaptor>::copyBackingStore):
(JSC::JSGenericTypedArrayView<Adaptor>::slowDownAndWasteMemory):
* runtime/JSMap.h:
(JSC::JSMap::JSMap):
* runtime/JSObject.cpp:
(JSC::JSObject::copyButterfly):
(JSC::JSObject::visitChildren):
(JSC::JSObject::copyBackingStore):
(JSC::JSObject::getOwnPropertySlotByIndex):
(JSC::JSObject::putByIndex):
(JSC::JSObject::enterDictionaryIndexingMode):
(JSC::JSObject::createInitialIndexedStorage):
(JSC::JSObject::createArrayStorage):
(JSC::JSObject::convertUndecidedToInt32):
(JSC::JSObject::convertUndecidedToDouble):
(JSC::JSObject::convertUndecidedToContiguous):
(JSC::JSObject::constructConvertedArrayStorageWithoutCopyingElements):
(JSC::JSObject::convertUndecidedToArrayStorage):
(JSC::JSObject::convertInt32ToDouble):
(JSC::JSObject::convertInt32ToContiguous):
(JSC::JSObject::convertInt32ToArrayStorage):
(JSC::JSObject::convertDoubleToContiguous):
(JSC::JSObject::convertDoubleToArrayStorage):
(JSC::JSObject::convertContiguousToArrayStorage):
(JSC::JSObject::setIndexQuicklyToUndecided):
(JSC::JSObject::ensureArrayStorageExistsAndEnterDictionaryIndexingMode):
(JSC::JSObject::deletePropertyByIndex):
(JSC::JSObject::getOwnPropertyNames):
(JSC::JSObject::putIndexedDescriptor):
(JSC::JSObject::defineOwnIndexedProperty):
(JSC::JSObject::putByIndexBeyondVectorLengthWithoutAttributes):
(JSC::JSObject::putDirectIndexBeyondVectorLength):
(JSC::JSObject::getNewVectorLength):
(JSC::JSObject::ensureLengthSlow):
(JSC::JSObject::reallocateAndShrinkButterfly):
(JSC::JSObject::growOutOfLineStorage):
(JSC::JSObject::getOwnPropertyDescriptor):
(JSC::JSObject::getEnumerableLength):
* runtime/JSObject.h:
(JSC::JSObject::getArrayLength):
(JSC::JSObject::getVectorLength):
(JSC::JSObject::canGetIndexQuickly):
(JSC::JSObject::getIndexQuickly):
(JSC::JSObject::tryGetIndexQuickly):
(JSC::JSObject::canSetIndexQuickly):
(JSC::JSObject::canSetIndexQuicklyForPutDirect):
(JSC::JSObject::setIndexQuickly):
(JSC::JSObject::initializeIndex):
(JSC::JSObject::hasSparseMap):
(JSC::JSObject::inSparseIndexingMode):
(JSC::JSObject::inlineStorage):
(JSC::JSObject::butterfly):
(JSC::JSObject::outOfLineStorage):
(JSC::JSObject::locationForOffset):
(JSC::JSObject::ensureInt32):
(JSC::JSObject::ensureDouble):
(JSC::JSObject::ensureContiguous):
(JSC::JSObject::ensureArrayStorage):
(JSC::JSObject::arrayStorage):
(JSC::JSObject::arrayStorageOrNull):
(JSC::JSObject::ensureLength):
(JSC::JSObject::putDirectWithoutTransition):
* runtime/JSSet.h:
(JSC::JSSet::JSSet):
* runtime/MapData.h:
(JSC::JSIterator>::MapDataImpl):
(JSC::JSIterator>::IteratorData::next):
(JSC::JSIterator>::IteratorData::refreshCursor):
* runtime/MapDataInlines.h:
(JSC::JSIterator>::clear):
(JSC::JSIterator>::find):
(JSC::JSIterator>::add):
(JSC::JSIterator>::remove):
(JSC::JSIterator>::replaceAndPackBackingStore):
(JSC::JSIterator>::replaceBackingStore):
(JSC::JSIterator>::ensureSpaceForAppend):
(JSC::JSIterator>::visitChildren):
(JSC::JSIterator>::copyBackingStore):
* runtime/Options.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190896 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Font Features] Tiny cleanup regarding FontCascade::typesettingFeatures()
mmaxfield@apple.com [Mon, 12 Oct 2015 22:40:36 +0000 (22:40 +0000)]
[Font Features] Tiny cleanup regarding FontCascade::typesettingFeatures()
https://bugs.webkit.org/show_bug.cgi?id=150051

Reviewed by Simon Fraser.

There are no typesetting features which aren't kerning nor ligatures.

No new tests because there is no behavior difference.

* platform/graphics/FontCascade.cpp:
(WebCore::FontCascade::codePath):
* platform/graphics/WidthIterator.h:
(WebCore::WidthIterator::supportsTypesettingFeatures): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190895 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd a CGContextStateSaver and use it
simon.fraser@apple.com [Mon, 12 Oct 2015 22:36:03 +0000 (22:36 +0000)]
Add a CGContextStateSaver and use it
https://bugs.webkit.org/show_bug.cgi?id=150049

Reviewed by Tim Horton.

Add a stack-based graphics state save/restore class for CGContext,
like the one we have for GraphicsContext, and use it in GraphicsContextCG.

* platform/graphics/cg/GraphicsContextCG.cpp:
(WebCore::GraphicsContext::drawNativeImage):
(WebCore::GraphicsContext::drawLine):
(WebCore::GraphicsContext::drawJoinedLines):
(WebCore::GraphicsContext::fillPath):
(WebCore::GraphicsContext::strokePath):
(WebCore::GraphicsContext::fillRect):
(WebCore::GraphicsContext::platformFillRoundedRect):
(WebCore::GraphicsContext::fillRectWithRoundedHole):
(WebCore::GraphicsContext::strokeRect):
* platform/graphics/cg/GraphicsContextCG.h:
(WebCore::CGContextStateSaver::CGContextStateSaver):
(WebCore::CGContextStateSaver::~CGContextStateSaver):
(WebCore::CGContextStateSaver::save):
(WebCore::CGContextStateSaver::restore):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190894 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agodisplay: table-cell; bug when resizing window
zalan@apple.com [Mon, 12 Oct 2015 22:33:03 +0000 (22:33 +0000)]
display: table-cell; bug when resizing window
https://bugs.webkit.org/show_bug.cgi?id=138167

Reviewed by David Hyatt.

Clean up anonymous table wrappers all the way up to RenderTable.
This patch ensures that we don't keep the generated RenderTable/RenderSection/RenderCaption/RenderRow
objects around in the tree anymore when the last child is destroyed.

Import W3C CSS2.1 anonymous table tests.

Source/WebCore:

* rendering/RenderObject.cpp:
(WebCore::RenderObject::destroyAndCleanupAnonymousWrappers):

LayoutTests:

* css2.1/tables/table-anonymous-objects-015.xht: Added.
* css2.1/tables/table-anonymous-objects-016.xht: Added.
* css2.1/tables/table-anonymous-objects-023.xht: Added.
* css2.1/tables/table-anonymous-objects-024.xht: Added.
* css2.1/tables/table-anonymous-objects-035.xht: Added.
* css2.1/tables/table-anonymous-objects-036.xht: Added.
* css2.1/tables/table-anonymous-objects-037.xht: Added.
* css2.1/tables/table-anonymous-objects-038.xht: Added.
* css2.1/tables/table-anonymous-objects-045.xht: Added.
* css2.1/tables/table-anonymous-objects-046.xht: Added.
* css2.1/tables/table-anonymous-objects-047.xht: Added.
* css2.1/tables/table-anonymous-objects-048.xht: Added.
* css2.1/tables/table-anonymous-objects-049.xht: Added.
* css2.1/tables/table-anonymous-objects-050.xht: Added.
* css2.1/tables/table-anonymous-objects-055.xht: Added.
* css2.1/tables/table-anonymous-objects-056.xht: Added.
* css2.1/tables/table-anonymous-objects-091.xht: Added.
* css2.1/tables/table-anonymous-objects-092.xht: Added.
* css2.1/tables/table-anonymous-objects-099.xht: Added.
* css2.1/tables/table-anonymous-objects-100.xht: Added.
* css2.1/tables/table-anonymous-objects-105.xht: Added.
* css2.1/tables/table-anonymous-objects-106.xht: Added.
* css2.1/tables/table-anonymous-objects-107.xht: Added.
* css2.1/tables/table-anonymous-objects-108.xht: Added.
* css2.1/tables/table-anonymous-objects-109.xht: Added.
* css2.1/tables/table-anonymous-objects-110.xht: Added.
* css2.1/tables/table-anonymous-objects-111.xht: Added.
* css2.1/tables/table-anonymous-objects-112.xht: Added.
* css2.1/tables/table-anonymous-objects-113.xht: Added.
* css2.1/tables/table-anonymous-objects-114.xht: Added.
* css2.1/tables/table-anonymous-objects-115.xht: Added.
* css2.1/tables/table-anonymous-objects-116.xht: Added.
* css2.1/tables/table-anonymous-objects-121.xht: Added.
* css2.1/tables/table-anonymous-objects-122.xht: Added.
* css2.1/tables/table-anonymous-objects-123.xht: Added.
* css2.1/tables/table-anonymous-objects-124.xht: Added.
* css2.1/tables/table-anonymous-objects-139.xht: Added.
* css2.1/tables/table-anonymous-objects-140.xht: Added.
* css2.1/tables/table-anonymous-objects-149.xht: Added.
* css2.1/tables/table-anonymous-objects-150.xht: Added.
* css2.1/tables/table-anonymous-objects-155.xht: Added.
* css2.1/tables/table-anonymous-objects-156.xht: Added.
* css2.1/tables/table-anonymous-objects-159.xht: Added.
* css2.1/tables/table-anonymous-objects-160.xht: Added.
* css2.1/tables/table-anonymous-objects-165.xht: Added.
* css2.1/tables/table-anonymous-objects-166.xht: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-015-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-016-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-023-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-024-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-035-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-036-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-037-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-038-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-045-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-046-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-047-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-048-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-049-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-050-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-055-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-056-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-091-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-092-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-099-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-100-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-105-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-106-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-107-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-108-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-109-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-110-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-111-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-112-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-113-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-114-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-115-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-116-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-121-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-122-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-123-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-124-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-139-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-140-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-149-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-150-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-155-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-156-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-159-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-160-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-165-expected.txt: Added.
* platform/mac/css2.1/tables/table-anonymous-objects-166-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190893 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, more Mac test gardening after r190629.
jhoneycutt@apple.com [Mon, 12 Oct 2015 22:15:49 +0000 (22:15 +0000)]
Unreviewed, more Mac test gardening after r190629.

* platform/mac-wk1/TestExpectations:
Add a new expected time out. Group expected time outs for imported
Blink tests using requestAnimationFrame.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190892 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION(r182192): Ligatures do not interact correctly with SHY in some fonts
mmaxfield@apple.com [Mon, 12 Oct 2015 22:08:27 +0000 (22:08 +0000)]
REGRESSION(r182192): Ligatures do not interact correctly with SHY in some fonts
https://bugs.webkit.org/show_bug.cgi?id=150006

Reviewed by Simon Fraser.

Source/WebCore:

When performing font transforms and we encounter kCGFontIndexInvalid, we filter it out of the
GlyphBuffer. However, this filter was only interacting with part of the GlyphBuffer instead
of the whole thing. This causes glyph IDs from one font to be rendered with other fonts,
thereby showing garbage glyphs.

However, now that <rdar://problem/20230073> is fixed, we don't need to perform this filter in
the first place.

Test: fast/text/undefined-glyph-with-ligature.html

* platform/graphics/GlyphBuffer.h:
(WebCore::GlyphBuffer::copyItem):
(WebCore::GlyphBuffer::swap):
* platform/graphics/WidthIterator.cpp:
(WebCore::WidthIterator::applyFontTransforms):

LayoutTests:

* fast/text/resources/tiny-ligature-font.svg: Added.
* fast/text/undefined-glyph-with-ligature-expected.html: Added.
* fast/text/undefined-glyph-with-ligature.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190891 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[SVG] Handle endEvent for svg animations
commit-queue@webkit.org [Mon, 12 Oct 2015 21:48:32 +0000 (21:48 +0000)]
[SVG] Handle endEvent for svg animations
https://bugs.webkit.org/show_bug.cgi?id=121587

Patch by Antoine Quint <graouts@apple.com> on 2015-10-12
Reviewed by Dean Jackson.

Source/WebCore:

Add support for the "endEvent" SVG event triggered when an animation completes, as
specified in http://www.w3.org/TR/SMIL3/smil-timing.html#q135. This event doesn't
bubble and can't be canceled. Added test coverage for the event through the DOM
Events API as well as the declarative SMIL Animation syntax.

Adapted from a Chromium patch by pavan.e@samsung.com
https://chromium.googlesource.com/chromium/blink/+/4d415ca0268231aa80e3552fe21bf3480a6978f8

Tests: svg/animations/end-event-declarative-expected.svg
       svg/animations/end-event-declarative.svg
       svg/animations/end-event-script-expected.svg
       svg/animations/end-event-script.svg

* svg/animation/SMILTimeContainer.cpp:
(WebCore::SMILTimeContainer::updateAnimations):
* svg/animation/SVGSMILElement.cpp:
(WebCore::smilEndEventSender):
(WebCore::SVGSMILElement::~SVGSMILElement):
(WebCore::SVGSMILElement::progress):
(WebCore::SVGSMILElement::dispatchPendingEvent):
* svg/animation/SVGSMILElement.h:
(WebCore::SVGSMILElement::hasConditionsConnected):

LayoutTests:

Tests for the "endEvent" event for SVG animations.

* svg/animations/end-event-declarative-expected.svg: Added.
* svg/animations/end-event-declarative.svg: Added.
* svg/animations/end-event-script-expected.svg: Added.
* svg/animations/end-event-script.svg: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190890 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUpdate JSC features.json
sbarati@apple.com [Mon, 12 Oct 2015 21:31:28 +0000 (21:31 +0000)]
Update JSC features.json
https://bugs.webkit.org/show_bug.cgi?id=150043

Reviewed by Mark Lam.

There were a lot of things implemented that weren't in
the list. We should be better about updating the list
as we land patches for new ES6 features.

* features.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190889 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoCleanup Heap.h and some related headers
joepeck@webkit.org [Mon, 12 Oct 2015 21:29:21 +0000 (21:29 +0000)]
Cleanup Heap.h and some related headers
https://bugs.webkit.org/show_bug.cgi?id=149981

Reviewed by Geoffrey Garen.

* heap/Heap.h:
- Some functions did not need export.
- threadDupStrings never had an implementation.

* heap/ConservativeRoots.cpp:
* heap/ConservativeRoots.h:
* heap/Heap.cpp:
* heap/ListableHandler.h:
* heap/WeakReferenceHarvester.h:
* jit/Repatch.cpp:
* runtime/JSONObject.h:
* runtime/VM.h:
- Stale forward declarations / includes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190888 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Curl] Increase limit of parallel network requests.
peavo@outlook.com [Mon, 12 Oct 2015 21:22:01 +0000 (21:22 +0000)]
[Curl] Increase limit of parallel network requests.
https://bugs.webkit.org/show_bug.cgi?id=150035

Reviewed by Alex Christensen.

If the limit is too low, other network requests will often be blocked until
active requests finishes. This can affect performance in a negative way.

* platform/network/curl/ResourceHandleManager.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190887 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agobmalloc: Don't ASSERT that all syscalls succeed
ggaren@apple.com [Mon, 12 Oct 2015 21:04:08 +0000 (21:04 +0000)]
bmalloc: Don't ASSERT that all syscalls succeed
https://bugs.webkit.org/show_bug.cgi?id=150047
<rdar://problem/22649531>

Reviewed by Mark Lam.

madvise can fail due to VM behaviors outside of our control:
copy-on-write, fork, mprotect, and other stuff.

Older darwin kernels sometimes return this error value, and new kernels
might again in future.

We haven't gained much from this ASSERT so far, so let's remove it.

Perhaps in future we can come up with a scheme that makes madvise
never fail, or that responds to failure.

* bmalloc/Syscall.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190886 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoEach *ById inline cache in the FTL must have its own CallSiteIndex
sbarati@apple.com [Mon, 12 Oct 2015 20:37:05 +0000 (20:37 +0000)]
Each *ById inline cache in the FTL must have its own CallSiteIndex
https://bugs.webkit.org/show_bug.cgi?id=150039

Reviewed by Geoffrey Garen and Filip Pizlo.

When lowering to LLVM, we create a patchpoint intrinsic for each
*ById in DFG IR. LLVM may choose to duplicate these patchpoints.
Therefore, we want each resulting inline cache to have a unique
CallSiteIndex because each inline cache will have its own set of
used registers. This change is necessary when we implement try/catch
in the FTL because an inline cache will ask for the set of used
registers it will need to restore in the event of an exception
being thrown. It asks for this set of registers by giving JITCode
a CallSiteIndex. Because each corresponding inline cache that results
from a duplicated patchpoint may all ask this for this set of registers,
we must assign each inline cache a unique CallSiteIndex.

* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::newExceptionHandlingCallSiteIndex):
* dfg/DFGCommonData.cpp:
(JSC::DFG::CommonData::addCodeOrigin):
(JSC::DFG::CommonData::addUniqueCallSiteIndex):
(JSC::DFG::CommonData::addCodeOriginUnconditionally): Deleted.
* dfg/DFGCommonData.h:
* ftl/FTLCompile.cpp:
(JSC::FTL::mmAllocateDataSection):
* ftl/FTLInlineCacheDescriptor.h:
(JSC::FTL::InlineCacheDescriptor::InlineCacheDescriptor):
(JSC::FTL::InlineCacheDescriptor::stackmapID):
(JSC::FTL::InlineCacheDescriptor::codeOrigin):
(JSC::FTL::InlineCacheDescriptor::uid):
(JSC::FTL::GetByIdDescriptor::GetByIdDescriptor):
(JSC::FTL::PutByIdDescriptor::PutByIdDescriptor):
(JSC::FTL::CheckInDescriptor::CheckInDescriptor):
(JSC::FTL::LazySlowPathDescriptor::LazySlowPathDescriptor):
(JSC::FTL::InlineCacheDescriptor::callSiteIndex): Deleted.
* ftl/FTLLowerDFGToLLVM.cpp:
(JSC::FTL::DFG::LowerDFGToLLVM::compilePutById):
(JSC::FTL::DFG::LowerDFGToLLVM::compileIn):
(JSC::FTL::DFG::LowerDFGToLLVM::getById):
(JSC::FTL::DFG::LowerDFGToLLVM::lazySlowPath):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190885 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoModern IDB: Start version change transaction for connections to new database.
beidson@apple.com [Mon, 12 Oct 2015 20:24:32 +0000 (20:24 +0000)]
Modern IDB: Start version change transaction for connections to new database.
https://bugs.webkit.org/show_bug.cgi?id=150033

Reviewed by Alex Christensen.

Source/WebCore:

No new tests (Covered by changes to existing tests).

* CMakeLists.txt:
* Modules/indexeddb/IDBDatabase.h:

* Modules/indexeddb/client/IDBAnyImpl.cpp:
(WebCore::IDBClient::IDBAny::IDBAny):
(WebCore::IDBClient::IDBAny::~IDBAny):
(WebCore::IDBClient::IDBAny::idbDatabase):
(WebCore::IDBClient::IDBAny::domStringList):
(WebCore::IDBClient::IDBAny::idbCursor):
(WebCore::IDBClient::IDBAny::idbCursorWithValue):
(WebCore::IDBClient::IDBAny::idbFactory):
(WebCore::IDBClient::IDBAny::idbIndex):
(WebCore::IDBClient::IDBAny::idbObjectStore):
(WebCore::IDBClient::IDBAny::idbTransaction):
(WebCore::IDBClient::IDBAny::scriptValue):
(WebCore::IDBClient::IDBAny::integer):
(WebCore::IDBClient::IDBAny::string):
(WebCore::IDBClient::IDBAny::keyPath):
* Modules/indexeddb/client/IDBAnyImpl.h:
(WebCore::IDBClient::IDBAny::create):

* Modules/indexeddb/client/IDBConnectionToServer.cpp:
(WebCore::IDBClient::IDBConnectionToServer::fireVersionChangeEvent):
(WebCore::IDBClient::IDBConnectionToServer::registerDatabaseConnection):
(WebCore::IDBClient::IDBConnectionToServer::unregisterDatabaseConnection):
* Modules/indexeddb/client/IDBConnectionToServer.h:

* Modules/indexeddb/client/IDBDatabaseImpl.cpp: Added.
(WebCore::IDBClient::IDBDatabase::create):
(WebCore::IDBClient::IDBDatabase::IDBDatabase):
(WebCore::IDBClient::IDBDatabase::~IDBDatabase):
(WebCore::IDBClient::IDBDatabase::name):
(WebCore::IDBClient::IDBDatabase::version):
(WebCore::IDBClient::IDBDatabase::objectStoreNames):
(WebCore::IDBClient::IDBDatabase::createObjectStore):
(WebCore::IDBClient::IDBDatabase::transaction):
(WebCore::IDBClient::IDBDatabase::deleteObjectStore):
(WebCore::IDBClient::IDBDatabase::close):
(WebCore::IDBClient::IDBDatabase::activeDOMObjectName):
(WebCore::IDBClient::IDBDatabase::canSuspendForPageCache):
(WebCore::IDBClient::IDBDatabase::startVersionChangeTransaction):
* Modules/indexeddb/client/IDBDatabaseImpl.h: Added.
(WebCore::IDBClient::IDBDatabase::info):

* Modules/indexeddb/client/IDBOpenDBRequestImpl.cpp:
(WebCore::IDBClient::IDBOpenDBRequest::onError):
(WebCore::IDBClient::IDBOpenDBRequest::onSuccess):
(WebCore::IDBClient::IDBOpenDBRequest::onUpgradeNeeded):
(WebCore::IDBClient::IDBOpenDBRequest::requestCompleted):
* Modules/indexeddb/client/IDBOpenDBRequestImpl.h:

* Modules/indexeddb/client/IDBRequestImpl.cpp:
(WebCore::IDBClient::IDBRequest::IDBRequest):
(WebCore::IDBClient::IDBRequest::result):
(WebCore::IDBClient::IDBRequest::source):
(WebCore::IDBClient::IDBRequest::transaction):
* Modules/indexeddb/client/IDBRequestImpl.h:
(WebCore::IDBClient::IDBRequest::connection):

* Modules/indexeddb/client/IDBTransactionImpl.cpp:
(WebCore::IDBClient::IDBTransaction::create):
(WebCore::IDBClient::IDBTransaction::IDBTransaction):
(WebCore::IDBClient::IDBTransaction::~IDBTransaction):
(WebCore::IDBClient::IDBTransaction::mode):
(WebCore::IDBClient::IDBTransaction::db):
(WebCore::IDBClient::IDBTransaction::error):
(WebCore::IDBClient::IDBTransaction::objectStore):
(WebCore::IDBClient::IDBTransaction::abort):
(WebCore::IDBClient::IDBTransaction::activeDOMObjectName):
(WebCore::IDBClient::IDBTransaction::canSuspendForPageCache):
* Modules/indexeddb/client/IDBTransactionImpl.h:
(WebCore::IDBClient::IDBTransaction::info):

* Modules/indexeddb/client/IDBVersionChangeEventImpl.cpp:
(WebCore::IDBClient::IDBVersionChangeEvent::IDBVersionChangeEvent):
(WebCore::IDBClient::IDBVersionChangeEvent::eventInterface):
* Modules/indexeddb/client/IDBVersionChangeEventImpl.h:
(WebCore::IDBClient::IDBVersionChangeEvent::create):

* Modules/indexeddb/legacy/LegacyDatabase.h:

* Modules/indexeddb/server/IDBConnectionToClient.cpp:
(WebCore::IDBServer::IDBConnectionToClient::fireVersionChangeEvent):
* Modules/indexeddb/server/IDBConnectionToClient.h:
* Modules/indexeddb/server/IDBConnectionToClientDelegate.h:

* Modules/indexeddb/server/IDBServer.cpp:
(WebCore::IDBServer::IDBServer::registerDatabaseConnection):
(WebCore::IDBServer::IDBServer::unregisterDatabaseConnection):
(WebCore::IDBServer::IDBServer::deleteDatabase):
* Modules/indexeddb/server/IDBServer.h:

* Modules/indexeddb/server/UniqueIDBDatabase.cpp:
(WebCore::IDBServer::UniqueIDBDatabase::info):
(WebCore::IDBServer::UniqueIDBDatabase::handleOpenDatabaseOperations):
(WebCore::IDBServer::UniqueIDBDatabase::hasAnyOpenConnections):
(WebCore::IDBServer::UniqueIDBDatabase::startVersionChangeTransaction):
(WebCore::IDBServer::UniqueIDBDatabase::notifyConnectionsOfVersionChange):
(WebCore::IDBServer::UniqueIDBDatabase::addOpenDatabaseConnection):
* Modules/indexeddb/server/UniqueIDBDatabase.h:

* Modules/indexeddb/server/UniqueIDBDatabaseConnection.cpp:
(WebCore::IDBServer::nextDatabaseConnectionIdentifier):
(WebCore::IDBServer::UniqueIDBDatabaseConnection::create):
(WebCore::IDBServer::UniqueIDBDatabaseConnection::UniqueIDBDatabaseConnection):
(WebCore::IDBServer::UniqueIDBDatabaseConnection::fireVersionChangeEvent):
(WebCore::IDBServer::UniqueIDBDatabaseConnection::createVersionChangeTransaction):

* Modules/indexeddb/server/UniqueIDBDatabaseConnection.h:
(WebCore::IDBServer::UniqueIDBDatabaseConnection::identifier):
(WebCore::IDBServer::UniqueIDBDatabaseConnection::database):
(WebCore::IDBServer::UniqueIDBDatabaseConnection::connectionToClient):
(WebCore::IDBServer::UniqueIDBDatabaseConnection::closePending):

* Modules/indexeddb/server/UniqueIDBDatabaseTransaction.cpp:
(WebCore::IDBServer::UniqueIDBDatabaseTransaction::create):
(WebCore::IDBServer::UniqueIDBDatabaseTransaction::UniqueIDBDatabaseTransaction):
* Modules/indexeddb/server/UniqueIDBDatabaseTransaction.h:
(WebCore::IDBServer::UniqueIDBDatabaseTransaction::databaseConnection):
(WebCore::IDBServer::UniqueIDBDatabaseTransaction::info):

* Modules/indexeddb/shared/IDBDatabaseInfo.cpp:
(WebCore::IDBDatabaseInfo::IDBDatabaseInfo):
* Modules/indexeddb/shared/IDBDatabaseInfo.h:
(WebCore::IDBDatabaseInfo::name):
(WebCore::IDBDatabaseInfo::version):
(WebCore::IDBDatabaseInfo::IDBDatabaseInfo): Deleted.

* Modules/indexeddb/shared/IDBError.h:

* Modules/indexeddb/shared/IDBRequestData.cpp:
(WebCore::IDBRequestData::IDBRequestData):
(WebCore::IDBRequestData::requestedVersion):
* Modules/indexeddb/shared/IDBRequestData.h:

* Modules/indexeddb/shared/IDBResourceIdentifier.cpp:
(WebCore::nextClientResourceNumber):
(WebCore::nextServerResourceNumber):
(WebCore::IDBResourceIdentifier::IDBResourceIdentifier):
(WebCore::IDBResourceIdentifier::emptyValue):
(WebCore::IDBResourceIdentifier::deletedValue):
(WebCore::nextResourceNumber): Deleted.
* Modules/indexeddb/shared/IDBResourceIdentifier.h:

* Modules/indexeddb/shared/IDBResultData.cpp:
(WebCore::IDBResultData::IDBResultData):
(WebCore::IDBResultData::error):
(WebCore::IDBResultData::openDatabaseSuccess):
(WebCore::IDBResultData::openDatabaseUpgradeNeeded):
(WebCore::IDBResultData::databaseInfo):
(WebCore::IDBResultData::transactionInfo):
* Modules/indexeddb/shared/IDBResultData.h:
(WebCore::IDBResultData::type):
(WebCore::IDBResultData::databaseConnectionIdentifier):

* Modules/indexeddb/shared/IDBTransactionInfo.cpp:
(WebCore::IDBTransactionInfo::IDBTransactionInfo):
(WebCore::IDBTransactionInfo::versionChange):
(WebCore::IDBTransactionInfo::isolatedCopy):
* Modules/indexeddb/shared/IDBTransactionInfo.h:
(WebCore::IDBTransactionInfo::identifier):
(WebCore::IDBTransactionInfo::mode):
(WebCore::IDBTransactionInfo::newVersion):

* Modules/indexeddb/shared/InProcessIDBServer.cpp:
(WebCore::InProcessIDBServer::fireVersionChangeEvent):
* Modules/indexeddb/shared/InProcessIDBServer.h:

* WebCore.xcodeproj/project.pbxproj:

* platform/CrossThreadCopier.cpp:
(WebCore::IDBTransactionInfo>::copy):
* platform/CrossThreadCopier.h:

LayoutTests:

* storage/indexeddb/modern/opendatabase-request-event-expected.txt:
* storage/indexeddb/modern/opendatabase-request-event.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190884 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION(r184895): border-image should always slice the SVG image to nine pieces...
commit-queue@webkit.org [Mon, 12 Oct 2015 19:51:11 +0000 (19:51 +0000)]
REGRESSION(r184895): border-image should always slice the SVG image to nine pieces when drawing it in the container element
https://bugs.webkit.org/show_bug.cgi?id=149901
<rdar://problem/21995596>

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2015-10-12
Reviewed by Darin Adler.

Source/WebCore:

The nine-pieces algorithm should be applied to the border-image regardless
whether the image has an intrinsic size or not. It is not guaranteed to have
a meaningful border-image in all the cases of non-intrinsic size images. But
it should work as expected in most of the cases.

* rendering/RenderBoxModelObject.cpp:
* rendering/RenderBoxModelObject.h:
(WebCore::RenderBoxModelObject::calculateImageIntrinsicDimensions):
Revert the changes which were added to return whether the image has
intrinsic size or not.

(WebCore::RenderBoxModelObject::calculateFillTileSize):
(WebCore::RenderBoxModelObject::paintNinePieceImage):
Size of the image is now the return value of calculateImageIntrinsicDimensions().

* rendering/RenderListMarker.cpp:
(WebCore::RenderListMarker::updateContent):
* rendering/shapes/ShapeOutsideInfo.cpp:
(WebCore::ShapeOutsideInfo::createShapeForImage):
Size of the image is now the return value of calculateImageIntrinsicDimensions().

* rendering/style/NinePieceImage.cpp:
* rendering/style/NinePieceImage.h:
(WebCore::NinePieceImage::paint):
Delete the logic for the non-intrinsic case. Both intrinsic and non-intrinsic
cases will be treated the same.

(WebCore::NinePieceImage::computeNineRects):
(WebCore::NinePieceImage::computeSideTileScale):
(WebCore::NinePieceImage::computeMiddleTileScale):
(WebCore::NinePieceImage::computeTileScales):
(WebCore::NinePieceImage::computeIntrinsicRects): Deleted.
(WebCore::NinePieceImage::computeIntrinsicSideTileScale): Deleted.
(WebCore::NinePieceImage::computeIntrinsicMiddleTileScale): Deleted.
(WebCore::NinePieceImage::computeIntrinsicTileScales): Deleted.
Remove *Intrinsic* from the name of the functions.

(WebCore::NinePieceImage::computeNonIntrinsicRects): Deleted.
(WebCore::NinePieceImage::computeNonIntrinsicTileScales): Deleted.
Delete the *NonIntrinsic* functions.

LayoutTests:

* fast/borders/border-image-fill-no-intrinsic-size-expected.html:
* fast/borders/border-image-fill-no-intrinsic-size.html:
* fast/borders/resources/svg-border-100x100-relative.svg: Added.
* fast/borders/resources/svg-border-100x100-viewbox.svg: Added.
* fast/borders/resources/svg-border-140x140-intrinsic.svg:
New test cases for relative lengths border images were added. The relative
lengths can be specified by using percentage lengths or adding viewBox to
the root element.

One of the non-intrinsic image-borders must be removed from this test.
The SVG image in the expected file has to be scaled by (100 / 140), which
can be done. But run-webkit-tests found a very slight difference between
the actual and the result files.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190883 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago"A + B" with strings shouldn't copy if A or B is empty.
akling@apple.com [Mon, 12 Oct 2015 19:44:52 +0000 (19:44 +0000)]
"A + B" with strings shouldn't copy if A or B is empty.
<https://webkit.org/b/150034>

Reviewed by Anders Carlsson.

Source/JavaScriptCore:

* runtime/JSStringBuilder.h:
(JSC::jsMakeNontrivialString):
* runtime/Lookup.cpp:
(JSC::reifyStaticAccessor):
* runtime/ObjectPrototype.cpp:
(JSC::objectProtoFuncToString):

Source/WTF:

Add a fast path to WTF's operator+ magic for concatenation of two strings where
one of them is empty. In that case, try to avoid allocation altogether by returning
the non-empty string.

Spotted this while analyzing memory peaks during page load; it turns out we were
duplicating whole text resources (JS, CSS) at the end of decoding, below
TextResourceDecoder::decodeAndFlush(). That function effectively does:

    return decode() + flush();

Very often, flush() returns an empty string, so due to the naive operator+,
we'd allocate a new StringImpl of length (decode().length() + flush().length())
and copy the return value from decode() into it. So silly!

Had to make the tryMakeString() machinery use String as a return type instead of
RefPtr<StringImpl> to make all the right overloads gel. StringTypeAdapter templates
are now required to provide a toString() function.

* wtf/text/StringConcatenate.h:
(WTF::StringTypeAdapter<char>::toString):
(WTF::StringTypeAdapter<UChar>::toString):
(WTF::StringTypeAdapter<Vector<char>>::toString):
(WTF::StringTypeAdapter<String>::toString):
(WTF::tryMakeString):
(WTF::makeString):
* wtf/text/StringOperators.h:
(WTF::StringAppend::operator String):
* wtf/text/StringView.h:
(WTF::StringTypeAdapter<StringView>::toString):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190882 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoVisitedValueCount GC Counter misses parallel SlotVisitors
commit-queue@webkit.org [Mon, 12 Oct 2015 19:36:03 +0000 (19:36 +0000)]
VisitedValueCount GC Counter misses parallel SlotVisitors
https://bugs.webkit.org/show_bug.cgi?id=149980

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2015-10-12
Reviewed by Geoffrey Garen.

* heap/Heap.cpp:
(JSC::Heap::updateObjectCounts):
Include threaded slot visitor's object counts in the debugging value.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190881 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Win] Unreviewed gardening
bfulgham@apple.com [Mon, 12 Oct 2015 19:18:49 +0000 (19:18 +0000)]
[Win] Unreviewed gardening

Unskip a bunch of passing tests.

* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190880 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoClip-path transitions sometimes trigger endless animation timers
simon.fraser@apple.com [Mon, 12 Oct 2015 19:15:28 +0000 (19:15 +0000)]
Clip-path transitions sometimes trigger endless animation timers
https://bugs.webkit.org/show_bug.cgi?id=150018

Reviewed by Tim Horton.

Source/WebCore:

Transitioning -webkit-clip-path could trigger endless animation
timers, because when CompositeAnimation::updateTransitions() calls
isTargetPropertyEqual(), a false negative answer triggers canceling the
current transition and starting a new one over and over.

This happened because StyleRareNonInheritedData simply tested pointer
equality for m_clipPath and m_shapeOutside. Both of these need to do deep
equality testing, requiring the implementation of operator== in BasicShapes
classes.

In addition, the PropertyWrappers in CSSPropertyAnimation need equals()
implementations that also do more than pointer equality testing.

Tests: transitions/clip-path-transitions.html
       transitions/shape-outside-transitions.html

* page/animation/CSSPropertyAnimation.cpp:
(WebCore::PropertyWrapperClipPath::equals):
(WebCore::PropertyWrapperShape::equals):
* rendering/ClipPathOperation.h:
* rendering/style/BasicShapes.cpp:
(WebCore::BasicShapeCircle::operator==):
(WebCore::BasicShapeEllipse::operator==):
(WebCore::BasicShapePolygon::operator==):
(WebCore::BasicShapeInset::operator==):
* rendering/style/BasicShapes.h:
(WebCore::BasicShapeCenterCoordinate::operator==):
(WebCore::BasicShapeRadius::operator==):
* rendering/style/ShapeValue.cpp:
(WebCore::pointersOrValuesEqual):
(WebCore::ShapeValue::operator==):
* rendering/style/ShapeValue.h:
(WebCore::ShapeValue::operator!=):
(WebCore::ShapeValue::operator==): Deleted.
(WebCore::ShapeValue::ShapeValue): Deleted.
* rendering/style/StyleRareNonInheritedData.cpp:
(WebCore::StyleRareNonInheritedData::operator==):
(WebCore::StyleRareNonInheritedData::clipPathOperationsEquivalent):
(WebCore::StyleRareNonInheritedData::shapeOutsideDataEquivalent):
* rendering/style/StyleRareNonInheritedData.h:

LayoutTests:

New tests for transitions of clip-path and shape-outside.

* transitions/clip-path-transitions-expected.txt: Added.
* transitions/clip-path-transitions.html: Added.
* transitions/resources/transition-test-helpers.js:
(parseClipPath):
(checkExpectedValue):
* transitions/shape-outside-transitions-expected.txt: Added.
* transitions/shape-outside-transitions.html: Added.
* transitions/svg-transitions-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190879 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, fix non-FTL build for real.
fpizlo@apple.com [Mon, 12 Oct 2015 19:15:17 +0000 (19:15 +0000)]
Unreviewed, fix non-FTL build for real.

* ftl/FTLLazySlowPath.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190878 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoMarking storage/domstorage/events/basic-body-attribute.html as flaky
commit-queue@webkit.org [Mon, 12 Oct 2015 19:03:17 +0000 (19:03 +0000)]
Marking storage/domstorage/events/basic-body-attribute.html as flaky
https://bugs.webkit.org/show_bug.cgi?id=148435

Patch by Ryan Haddad <ryanhaddad@apple.com> on 2015-10-12
Reviewed by Alexey Proskuryakov.

* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190877 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoTest font-variant-* and font-feature-settings on Yosemite and Mavericks
mmaxfield@apple.com [Mon, 12 Oct 2015 19:00:47 +0000 (19:00 +0000)]
Test font-variant-* and font-feature-settings on Yosemite and Mavericks
https://bugs.webkit.org/show_bug.cgi?id=149778

Reviewed by Simon Fraser.

Source/WebCore:

We can simply call the function which enables features on Yosemite and Mavericks.

* platform/graphics/cocoa/FontCacheCoreText.cpp:
(WebCore::platformFontLookupWithFamily):
(WebCore::fontWithFamily):

LayoutTests:

* platform/mac-wk1/TestExpectations:
* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190876 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, fix style in the hopes that it fixes Windows.
fpizlo@apple.com [Mon, 12 Oct 2015 18:45:00 +0000 (18:45 +0000)]
Unreviewed, fix style in the hopes that it fixes Windows.

* wtf/ParallelHelperPool.cpp:
(WTF::ParallelHelperClient::~ParallelHelperClient):
(WTF::ParallelHelperClient::setTask):
(WTF::ParallelHelperClient::doSomeHelping):
(WTF::ParallelHelperClient::runTaskInParallel):
(WTF::ParallelHelperClient::finish):
(WTF::ParallelHelperClient::claimTask):
(WTF::ParallelHelperClient::runTask):
(WTF::ParallelHelperPool::doSomeHelping):
(WTF::ParallelHelperPool::helperThreadBody):
* wtf/ParallelHelperPool.h:
(WTF::ParallelHelperClient::setFunction):
(WTF::ParallelHelperClient::runFunctionInParallel):
(WTF::ParallelHelperClient::pool):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190875 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Win] Unreviewed test gardening after r190800.
bfulgham@apple.com [Mon, 12 Oct 2015 18:43:00 +0000 (18:43 +0000)]
[Win] Unreviewed test gardening after r190800.

Skip mac-specific 'fast/forms/hidpi-textfield-background-bleeding.html' test
on Windows.

* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190874 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoREGRESSION(r190735, not fixed by r190741) failures.
bfulgham@apple.com [Mon, 12 Oct 2015 18:39:06 +0000 (18:39 +0000)]
REGRESSION(r190735, not fixed by r190741) failures.
https://bugs.webkit.org/show_bug.cgi?id=150040

* platform/win/TestExpectations: Skip several test failures introduced by JSC changes.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190873 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Win] Unreviewed test gardening after r190667.
bfulgham@apple.com [Mon, 12 Oct 2015 18:11:42 +0000 (18:11 +0000)]
[Win] Unreviewed test gardening after r190667.

Mark fast/css/will-change/will-change-create-stacking-context-inline.html as failing, because
Windows does not support backdrop-filter.

* platform/win/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190868 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, clarify a comment. The example code had a bug.
fpizlo@apple.com [Mon, 12 Oct 2015 18:11:26 +0000 (18:11 +0000)]
Unreviewed, clarify a comment. The example code had a bug.

* ftl/FTLLowerDFGToLLVM.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190867 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, fix no-FTL build.
fpizlo@apple.com [Mon, 12 Oct 2015 18:08:50 +0000 (18:08 +0000)]
Unreviewed, fix no-FTL build.

* ftl/FTLLazySlowPath.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190865 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agowebkit-gtk 2.3.3 fails to build on OS X - Conflicting type "Fixed"
mcatanzaro@igalia.com [Mon, 12 Oct 2015 17:59:45 +0000 (17:59 +0000)]
webkit-gtk 2.3.3 fails to build on OS X - Conflicting type "Fixed"
https://bugs.webkit.org/show_bug.cgi?id=126433

Patch by Philip Chimento <philip.chimento@gmail.com> on 2015-10-12
Reviewed by Philippe Normand

Don't include CoreFoundation.h when building the GTK port.

* Source/JavaScriptCore/API/WebKitAvailability.h: Add !defined(BUILDING_GTK__) to defined(__APPLE__).

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190862 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[Win] Unreviewed test gardening after r190840.
bfulgham@apple.com [Mon, 12 Oct 2015 17:59:05 +0000 (17:59 +0000)]
[Win] Unreviewed test gardening after r190840.

Rebaseline tests to match new output (same changes as Mac).

* platform/win/fast/css-generated-content/details-summary-before-after-expected.txt:
* platform/win/fast/html/details-add-child-2-expected.txt:
* platform/win/fast/html/details-open2-expected.txt:
* platform/win/fast/html/details-remove-summary-1-and-click-expected.txt:
* platform/win/fast/html/details-remove-summary-4-and-click-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190861 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoFTL should generate code to call slow paths lazily
fpizlo@apple.com [Mon, 12 Oct 2015 17:56:26 +0000 (17:56 +0000)]
FTL should generate code to call slow paths lazily
https://bugs.webkit.org/show_bug.cgi?id=149936

Reviewed by Saam Barati.

Source/JavaScriptCore:

We often have complex slow paths in FTL-generated code. Those slow paths may never run. Even
if they do run, they don't need stellar performance. So, it doesn't make sense to have LLVM
worry about compiling such slow path code.

This patch enables us to use our own MacroAssembler for compiling the slow path inside FTL
code. It does this by using a crazy lambda thingy (see FTLLowerDFGToLLVM.cpp's lazySlowPath()
and its documentation). The result is quite natural to use.

Even for straight slow path calls via something like vmCall(), the lazySlowPath offers the
benefit that the call marshalling and the exception checking are not expressed using LLVM IR
and do not require LLVM to think about it. It also has the benefit that we never generate the
code if it never runs. That's great, since function calls usually involve ~10 instructions
total (move arguments to argument registers, make the call, check exception, etc.).

This patch adds the lazy slow path abstraction and uses it for some slow paths in the FTL.
The code we generate with lazy slow paths is worse than the code that LLVM would have
generated. Therefore, a lazy slow path only makes sense when we have strong evidence that
the slow path will execute infrequently relative to the fast path. This completely precludes
the use of lazy slow paths for out-of-line Nodes that unconditionally call a C++ function.
It also precludes their use for the GetByVal out-of-bounds handler, since when we generate
a GetByVal with an out-of-bounds handler it means that we only know that the out-of-bounds
case executed at least once. So, for all we know, it may actually be the common case. So,
this patch just deployed the lazy slow path for GC slow paths and masquerades-as-undefined
slow paths. It makes sense for GC slow paths because those have a statistical guarantee of
slow path frequency - probably bounded at less than 1/10. It makes sense for masquerades-as-
undefined because we can say quite confidently that this is an uncommon scenario on the
modern Web.

Something that's always been challenging about abstractions involving the MacroAssembler is
that linking is a separate phase, and there is no way for someone who is just given access to
the MacroAssembler& to emit code that requires linking, since linking happens once we have
emitted all code and we are creating the LinkBuffer. Moreover, the FTL requires that the
final parts of linking happen on the main thread. This patch ran into this issue, and solved
it comprehensively, by introducing MacroAssembler::addLinkTask(). This takes a lambda and
runs it at the bitter end of linking - when performFinalization() is called. This ensure that
the task added by addLinkTask() runs on the main thread. This patch doesn't replace all of
the previously existing idioms for dealing with this issue; we can do that later.

This shows small speed-ups on a bunch of things. No big win on any benchmark aggregate. But
mainly this is done for https://bugs.webkit.org/show_bug.cgi?id=149852, where we found that
outlining the slow path in this way was a significant speed boost.

* CMakeLists.txt:
* JavaScriptCore.vcxproj/JavaScriptCore.vcxproj:
* JavaScriptCore.xcodeproj/project.pbxproj:
* assembler/AbstractMacroAssembler.h:
(JSC::AbstractMacroAssembler::replaceWithAddressComputation):
(JSC::AbstractMacroAssembler::addLinkTask):
(JSC::AbstractMacroAssembler::AbstractMacroAssembler):
* assembler/LinkBuffer.cpp:
(JSC::LinkBuffer::linkCode):
(JSC::LinkBuffer::allocate):
(JSC::LinkBuffer::performFinalization):
* assembler/LinkBuffer.h:
(JSC::LinkBuffer::wasAlreadyDisassembled):
(JSC::LinkBuffer::didAlreadyDisassemble):
(JSC::LinkBuffer::vm):
(JSC::LinkBuffer::executableOffsetFor):
* bytecode/CodeOrigin.h:
(JSC::CodeOrigin::CodeOrigin):
(JSC::CodeOrigin::isSet):
(JSC::CodeOrigin::operator bool):
(JSC::CodeOrigin::isHashTableDeletedValue):
(JSC::CodeOrigin::operator!): Deleted.
* ftl/FTLCompile.cpp:
(JSC::FTL::mmAllocateDataSection):
* ftl/FTLInlineCacheDescriptor.h:
(JSC::FTL::InlineCacheDescriptor::InlineCacheDescriptor):
(JSC::FTL::CheckInDescriptor::CheckInDescriptor):
(JSC::FTL::LazySlowPathDescriptor::LazySlowPathDescriptor):
* ftl/FTLJITCode.h:
* ftl/FTLJITFinalizer.cpp:
(JSC::FTL::JITFinalizer::finalizeFunction):
* ftl/FTLJITFinalizer.h:
* ftl/FTLLazySlowPath.cpp: Added.
(JSC::FTL::LazySlowPath::LazySlowPath):
(JSC::FTL::LazySlowPath::~LazySlowPath):
(JSC::FTL::LazySlowPath::generate):
* ftl/FTLLazySlowPath.h: Added.
(JSC::FTL::LazySlowPath::createGenerator):
(JSC::FTL::LazySlowPath::patchpoint):
(JSC::FTL::LazySlowPath::usedRegisters):
(JSC::FTL::LazySlowPath::callSiteIndex):
(JSC::FTL::LazySlowPath::stub):
* ftl/FTLLazySlowPathCall.h: Added.
(JSC::FTL::createLazyCallGenerator):
* ftl/FTLLowerDFGToLLVM.cpp:
(JSC::FTL::DFG::LowerDFGToLLVM::compileCreateActivation):
(JSC::FTL::DFG::LowerDFGToLLVM::compileNewFunction):
(JSC::FTL::DFG::LowerDFGToLLVM::compileCreateDirectArguments):
(JSC::FTL::DFG::LowerDFGToLLVM::compileNewArrayWithSize):
(JSC::FTL::DFG::LowerDFGToLLVM::compileMakeRope):
(JSC::FTL::DFG::LowerDFGToLLVM::compileNotifyWrite):
(JSC::FTL::DFG::LowerDFGToLLVM::compileIsObjectOrNull):
(JSC::FTL::DFG::LowerDFGToLLVM::compileIsFunction):
(JSC::FTL::DFG::LowerDFGToLLVM::compileIn):
(JSC::FTL::DFG::LowerDFGToLLVM::compileMaterializeNewObject):
(JSC::FTL::DFG::LowerDFGToLLVM::compileMaterializeCreateActivation):
(JSC::FTL::DFG::LowerDFGToLLVM::compileCheckWatchdogTimer):
(JSC::FTL::DFG::LowerDFGToLLVM::allocatePropertyStorageWithSizeImpl):
(JSC::FTL::DFG::LowerDFGToLLVM::allocateObject):
(JSC::FTL::DFG::LowerDFGToLLVM::allocateJSArray):
(JSC::FTL::DFG::LowerDFGToLLVM::buildTypeOf):
(JSC::FTL::DFG::LowerDFGToLLVM::sensibleDoubleToInt32):
(JSC::FTL::DFG::LowerDFGToLLVM::lazySlowPath):
(JSC::FTL::DFG::LowerDFGToLLVM::speculate):
(JSC::FTL::DFG::LowerDFGToLLVM::emitStoreBarrier):
* ftl/FTLOperations.cpp:
(JSC::FTL::operationMaterializeObjectInOSR):
(JSC::FTL::compileFTLLazySlowPath):
* ftl/FTLOperations.h:
* ftl/FTLSlowPathCall.cpp:
(JSC::FTL::SlowPathCallContext::SlowPathCallContext):
(JSC::FTL::SlowPathCallContext::~SlowPathCallContext):
(JSC::FTL::SlowPathCallContext::keyWithTarget):
(JSC::FTL::SlowPathCallContext::makeCall):
(JSC::FTL::callSiteIndexForCodeOrigin):
(JSC::FTL::storeCodeOrigin): Deleted.
(JSC::FTL::callOperation): Deleted.
* ftl/FTLSlowPathCall.h:
(JSC::FTL::callOperation):
* ftl/FTLState.h:
* ftl/FTLThunks.cpp:
(JSC::FTL::genericGenerationThunkGenerator):
(JSC::FTL::osrExitGenerationThunkGenerator):
(JSC::FTL::lazySlowPathGenerationThunkGenerator):
(JSC::FTL::registerClobberCheck):
* ftl/FTLThunks.h:
* interpreter/CallFrame.h:
(JSC::CallSiteIndex::CallSiteIndex):
(JSC::CallSiteIndex::operator bool):
(JSC::CallSiteIndex::bits):
* jit/CCallHelpers.h:
(JSC::CCallHelpers::setupArgument):
(JSC::CCallHelpers::setupArgumentsWithExecState):
* jit/JITOperations.cpp:

Source/WTF:

Enables SharedTask to handle any function type, not just void().

It's probably better to use SharedTask instead of std::function in performance-sensitive
code. std::function uses the system malloc and has copy semantics. SharedTask uses FastMalloc
and has aliasing semantics. So, you can just trust that it will have sensible performance
characteristics.

* wtf/ParallelHelperPool.cpp:
(WTF::ParallelHelperClient::~ParallelHelperClient):
(WTF::ParallelHelperClient::setTask):
(WTF::ParallelHelperClient::doSomeHelping):
(WTF::ParallelHelperClient::runTaskInParallel):
(WTF::ParallelHelperClient::finish):
(WTF::ParallelHelperClient::claimTask):
(WTF::ParallelHelperClient::runTask):
(WTF::ParallelHelperPool::doSomeHelping):
(WTF::ParallelHelperPool::helperThreadBody):
* wtf/ParallelHelperPool.h:
(WTF::ParallelHelperClient::setFunction):
(WTF::ParallelHelperClient::runFunctionInParallel):
(WTF::ParallelHelperClient::pool):
* wtf/SharedTask.h:
(WTF::createSharedTask):
(WTF::SharedTask::SharedTask): Deleted.
(WTF::SharedTask::~SharedTask): Deleted.
(WTF::SharedTaskFunctor::SharedTaskFunctor): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190860 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoDon't allow plug-ins to override image types for <embed> elements
andersca@apple.com [Mon, 12 Oct 2015 17:25:37 +0000 (17:25 +0000)]
Don't allow plug-ins to override image types for <embed> elements
https://bugs.webkit.org/show_bug.cgi?id=149979

Reviewed by Tim Horton.

Source/WebCore:

Stop allowing plug-ins to take over image types for <embed> elements. We already do this
for <object> elements, but had to make <embed> elements exempt because of webkit.org/b/49016.
The QuickTime plug-in hasn't supported image types since Lion, so there's no point in keeping this code around.

* html/HTMLAppletElement.cpp:
(WebCore::HTMLAppletElement::HTMLAppletElement):
* html/HTMLEmbedElement.cpp:
(WebCore::HTMLEmbedElement::HTMLEmbedElement):
* html/HTMLObjectElement.cpp:
(WebCore::HTMLObjectElement::HTMLObjectElement):
(WebCore::HTMLObjectElement::parametersForPlugin):
* html/HTMLPlugInImageElement.cpp:
(WebCore::HTMLPlugInImageElement::HTMLPlugInImageElement):
(WebCore::HTMLPlugInImageElement::isImageType):
(WebCore::HTMLPlugInImageElement::wouldLoadAsNetscapePlugin):
* html/HTMLPlugInImageElement.h:
(WebCore::HTMLPlugInImageElement::shouldPreferPlugInsForImages): Deleted.
* loader/EmptyClients.h:
* loader/FrameLoaderClient.h:
* loader/SubframeLoader.cpp:
(WebCore::SubframeLoader::resourceWillUsePlugin):
(WebCore::SubframeLoader::requestObject):
(WebCore::SubframeLoader::shouldUsePlugin):
* loader/SubframeLoader.h:

Source/WebKit/mac:

* WebCoreSupport/WebFrameLoaderClient.h:
* WebCoreSupport/WebFrameLoaderClient.mm:
(WebFrameLoaderClient::objectContentType):

Source/WebKit/win:

* WebCoreSupport/WebFrameLoaderClient.cpp:
(WebFrameLoaderClient::objectContentType):
* WebCoreSupport/WebFrameLoaderClient.h:

Source/WebKit2:

* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::objectContentType):
* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:

LayoutTests:

Remove test.

* platform/efl/TestExpectations:
* platform/gtk/TestExpectations:
* plugins/embed-prefers-plugins-for-images-expected.txt: Removed.
* plugins/embed-prefers-plugins-for-images.html: Removed.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190859 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoMarking http/tests/notifications/events.html as flaky again as patch did not resolve...
commit-queue@webkit.org [Mon, 12 Oct 2015 17:23:06 +0000 (17:23 +0000)]
Marking http/tests/notifications/events.html as flaky again as patch did not resolve crashes
https://bugs.webkit.org/show_bug.cgi?id=149218

Patch by Ryan Haddad <ryanhaddad@apple.com> on 2015-10-12
Reviewed by Brian Burg.

* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190858 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agowebkit-gtk-2.3.4 fails to link JavaScriptCore, missing symbols add_history and readline
commit-queue@webkit.org [Mon, 12 Oct 2015 17:03:48 +0000 (17:03 +0000)]
webkit-gtk-2.3.4 fails to link JavaScriptCore, missing symbols add_history and readline
https://bugs.webkit.org/show_bug.cgi?id=127059

Patch by Philip Chimento <philip.chimento@gmail.com> on 2015-10-12
Reviewed by Philippe Normand.

* shell/CMakeLists.txt: Link JSC with -ledit on Mac OSX.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190857 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Use --version-script only on Linux
commit-queue@webkit.org [Mon, 12 Oct 2015 17:01:05 +0000 (17:01 +0000)]
[GTK] Use --version-script only on Linux
https://bugs.webkit.org/show_bug.cgi?id=144555

Patch by Philip Chimento <philip.chimento@gmail.com> on 2015-10-12
Reviewed by Philippe Normand.

* Source/cmake/OptionsGTK.cmake: Don't add --version-script
option on Darwin (whose linker doesn't support it.)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190856 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Update jhbuild's version of harfbuzz
mario@webkit.org [Mon, 12 Oct 2015 15:30:32 +0000 (15:30 +0000)]
[GTK] Update jhbuild's version of harfbuzz
https://bugs.webkit.org/show_bug.cgi?id=149749

Reviewed by Martin Robinson.

Update harfbuzz to 0.9.35, following the lead of Debian Stable.

* gtk/jhbuild.modules: Updated details for harfbuzz.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190855 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[GTK] Media tests crashing with gst-plugins-bad 1.4.4 and cpp 5.2
mario@webkit.org [Mon, 12 Oct 2015 12:37:05 +0000 (12:37 +0000)]
[GTK] Media tests crashing with gst-plugins-bad 1.4.4 and cpp 5.2
https://bugs.webkit.org/show_bug.cgi?id=150026

Reviewed by Philippe Normand.

Backport upstream patch from gst-plugins-bad so that the right version
of FAAD2 can be found with newer versions of cpp, so that gst-plugins-bad
gets compiled properly, avoiding crashes on runtime while running tests.

* gtk/patches/gst-plugins-bad-fix-faad2-version-check.patch: Added.
* gtk/jhbuild.modules: Referenced the newly added patch for gst-plugins-bad.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190854 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, fixing debug builds with Clang on Linux
zandobersek@gmail.com [Mon, 12 Oct 2015 11:31:26 +0000 (11:31 +0000)]
Unreviewed, fixing debug builds with Clang on Linux
by including the stdio.h header where it's required
under a debug configuration.

* page/scrolling/ScrollingStateTree.cpp:
(WebCore::ScrollingStateTree::ScrollingStateTree):
* rendering/SimpleLineLayoutFunctions.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190853 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed, followup to r190643.
zandobersek@gmail.com [Mon, 12 Oct 2015 11:24:40 +0000 (11:24 +0000)]
Unreviewed, followup to r190643.

Inline the std::function<> constructor wrappings around lambdas
into a single line, instead of spanning it across four lines.

* platform/graphics/gstreamer/InbandTextTrackPrivateGStreamer.cpp:
(WebCore::InbandTextTrackPrivateGStreamer::handleSample):
(WebCore::InbandTextTrackPrivateGStreamer::streamChanged):
(WebCore::InbandTextTrackPrivateGStreamer::notifyTrackOfSample):
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:
(WebCore::MediaPlayerPrivateGStreamer::videoChanged):
(WebCore::MediaPlayerPrivateGStreamer::videoCapsChanged):
(WebCore::MediaPlayerPrivateGStreamer::notifyPlayerOfVideo):
(WebCore::MediaPlayerPrivateGStreamer::audioChanged):
(WebCore::MediaPlayerPrivateGStreamer::notifyPlayerOfAudio):
(WebCore::MediaPlayerPrivateGStreamer::textChanged):
(WebCore::MediaPlayerPrivateGStreamer::notifyPlayerOfText):
* platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:
(WebCore::MediaPlayerPrivateGStreamerBase::volumeChanged):
(WebCore::MediaPlayerPrivateGStreamerBase::networkState):
(WebCore::MediaPlayerPrivateGStreamerBase::muteChanged):
* platform/graphics/gstreamer/TrackPrivateBaseGStreamer.cpp:
(WebCore::TrackPrivateBaseGStreamer::activeChanged):
(WebCore::TrackPrivateBaseGStreamer::tagsChanged):
(WebCore::TrackPrivateBaseGStreamer::notifyTrackOfActiveChanged):
* platform/graphics/gstreamer/VideoSinkGStreamer.cpp:
(webkitVideoSinkRender):
* platform/graphics/gstreamer/WebKitWebSourceGStreamer.cpp:
(webKitWebSrcChangeState):
(webKitWebSrcNeedDataCb):
(webKitWebSrcEnoughDataMainCb):
(webKitWebSrcEnoughDataCb):
(webKitWebSrcSeekMainCb):
(webKitWebSrcSeekDataCb):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190852 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years ago[EFL] Bump Version/X in user agent
gyuyoung.kim@webkit.org [Mon, 12 Oct 2015 07:52:53 +0000 (07:52 +0000)]
[EFL] Bump Version/X in user agent
https://bugs.webkit.org/show_bug.cgi?id=150013

Reviewed by Csaba Osztrogon√°c.

EFL port needs to ensure maximum compatibility with existing web sites.
Thus version number needs to be bumped from 5.0 to 8.0 as GTK port.

* UIProcess/efl/WebPageProxyEfl.cpp:
(WebKit::WebPageProxy::standardUserAgent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190851 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoAdd missing builtins files to the Xcode projects
bburg@apple.com [Mon, 12 Oct 2015 04:41:36 +0000 (04:41 +0000)]
Add missing builtins files to the Xcode projects
https://bugs.webkit.org/show_bug.cgi?id=150015

Reviewed by Yusuke Suzuki.

* WebCore.xcodeproj/project.pbxproj:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190850 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoCleanup and simplification of SVG path-related classes
simon.fraser@apple.com [Mon, 12 Oct 2015 03:34:08 +0000 (03:34 +0000)]
Cleanup and simplification of SVG path-related classes
https://bugs.webkit.org/show_bug.cgi?id=150011

Reviewed by Zalan Bujtas.

Many SVG path-related subclasses were stateful, but only because code in
SVGPathUtilities kept global copies around for no reason. A microbenchmark
showed that there was no benefit to keeping global singletons of SVGPathBuilder,
SVGPathSegListBuilder, SVGPathByteStreamBuilder, SVGPathStringBuilder,
SVGPathTraversalStateBuilder, SVGPathParser and SVGPathBlender.

Making these classes not be re-usable makes the code much simpler, allowing
their SVGPathSources, SVGPathConsumers, SVGPathByteStream etc. to be stored
by reference, and eliminating the cleanup() function which created annoying
ordering issues.

Code that uses SVGPathParser and SVGPathBlender is further simplified by having
these classes expose only static functions, hiding any internal statefulness.

* svg/SVGPathBlender.cpp: Remove the m_progress member variable and instead
pass progress to the various blend functions, as we do for other blend functions.
Expose two only static functions. Pointers to references.
(WebCore::SVGPathBlender::addAnimatedPath):
(WebCore::SVGPathBlender::blendAnimatedPath):
(WebCore::SVGPathBlender::SVGPathBlender):
(WebCore::SVGPathBlender::blendAnimatedDimensonalFloat):
(WebCore::SVGPathBlender::blendAnimatedFloatPoint):
(WebCore::SVGPathBlender::blendMoveToSegment):
(WebCore::SVGPathBlender::blendLineToSegment):
(WebCore::SVGPathBlender::blendLineToHorizontalSegment):
(WebCore::SVGPathBlender::blendLineToVerticalSegment):
(WebCore::SVGPathBlender::blendCurveToCubicSegment):
(WebCore::SVGPathBlender::blendCurveToCubicSmoothSegment):
(WebCore::SVGPathBlender::blendCurveToQuadraticSegment):
(WebCore::SVGPathBlender::blendCurveToQuadraticSmoothSegment):
(WebCore::SVGPathBlender::blendArcToSegment):
(WebCore::SVGPathBlender::cleanup): Deleted.
* svg/SVGPathBlender.h: Make the constructor take a ref to the destination Path,
which is stored by reference.
* svg/SVGPathBuilder.cpp:
(WebCore::SVGPathBuilder::SVGPathBuilder):
(WebCore::SVGPathBuilder::moveTo):
(WebCore::SVGPathBuilder::lineTo):
(WebCore::SVGPathBuilder::curveToCubic):
(WebCore::SVGPathBuilder::closePath):
* svg/SVGPathBuilder.h:
(WebCore::SVGPathBuilder::setCurrentPath): Deleted.
* svg/SVGPathByteStreamBuilder.cpp: References, assertions removed.
(WebCore::SVGPathByteStreamBuilder::SVGPathByteStreamBuilder):
* svg/SVGPathByteStreamBuilder.h:
(WebCore::SVGPathByteStreamBuilder::writeType):
(WebCore::SVGPathByteStreamBuilder::setCurrentByteStream): Deleted.
* svg/SVGPathConsumer.h:
* svg/SVGPathElement.cpp:
* svg/SVGPathParser.cpp: Expose some static helper functions for parsing
to byte streams and strings. References.
(WebCore::SVGPathParser::parse):
(WebCore::SVGPathParser::parseToByteStream):
(WebCore::SVGPathParser::parseToString):
(WebCore::SVGPathParser::SVGPathParser):
(WebCore::SVGPathParser::parseClosePathSegment):
(WebCore::SVGPathParser::parseMoveToSegment):
(WebCore::SVGPathParser::parseLineToSegment):
(WebCore::SVGPathParser::parseLineToHorizontalSegment):
(WebCore::SVGPathParser::parseLineToVerticalSegment):
(WebCore::SVGPathParser::parseCurveToCubicSegment):
(WebCore::SVGPathParser::parseCurveToCubicSmoothSegment):
(WebCore::SVGPathParser::parseCurveToQuadraticSegment):
(WebCore::SVGPathParser::parseCurveToQuadraticSmoothSegment):
(WebCore::SVGPathParser::parseArcToSegment):
(WebCore::SVGPathParser::parsePathData):
(WebCore::SVGPathParser::decomposeArcToCubic):
(WebCore::SVGPathParser::parsePathDataFromSource): Deleted.
(WebCore::SVGPathParser::cleanup): Deleted.
* svg/SVGPathParser.h:
(WebCore::SVGPathParser::setCurrentConsumer): Deleted.
(WebCore::SVGPathParser::setCurrentSource): Deleted.
* svg/SVGPathSegListBuilder.cpp:
(WebCore::SVGPathSegListBuilder::SVGPathSegListBuilder):
(WebCore::SVGPathSegListBuilder::moveTo):
(WebCore::SVGPathSegListBuilder::lineTo):
(WebCore::SVGPathSegListBuilder::lineToHorizontal):
(WebCore::SVGPathSegListBuilder::lineToVertical):
(WebCore::SVGPathSegListBuilder::curveToCubic):
(WebCore::SVGPathSegListBuilder::curveToCubicSmooth):
(WebCore::SVGPathSegListBuilder::curveToQuadratic):
(WebCore::SVGPathSegListBuilder::curveToQuadraticSmooth):
(WebCore::SVGPathSegListBuilder::arcTo):
(WebCore::SVGPathSegListBuilder::closePath):
* svg/SVGPathSegListBuilder.h:
(WebCore::SVGPathSegListBuilder::setCurrentSVGPathElement): Deleted.
(WebCore::SVGPathSegListBuilder::setCurrentSVGPathSegList): Deleted.
(WebCore::SVGPathSegListBuilder::setCurrentSVGPathSegRole): Deleted.
* svg/SVGPathStringBuilder.cpp:
(WebCore::SVGPathStringBuilder::cleanup): Deleted.
* svg/SVGPathStringBuilder.h:
* svg/SVGPathTraversalStateBuilder.cpp:
(WebCore::SVGPathTraversalStateBuilder::SVGPathTraversalStateBuilder):
(WebCore::SVGPathTraversalStateBuilder::moveTo):
(WebCore::SVGPathTraversalStateBuilder::lineTo):
(WebCore::SVGPathTraversalStateBuilder::curveToCubic):
(WebCore::SVGPathTraversalStateBuilder::closePath):
(WebCore::SVGPathTraversalStateBuilder::continueConsuming):
(WebCore::SVGPathTraversalStateBuilder::totalLength):
(WebCore::SVGPathTraversalStateBuilder::currentPoint):
(WebCore::SVGPathTraversalStateBuilder::setDesiredLength): Deleted.
* svg/SVGPathTraversalStateBuilder.h:
(WebCore::SVGPathTraversalStateBuilder::pathSegmentIndex):
(WebCore::SVGPathTraversalStateBuilder::setCurrentTraversalState): Deleted.
* svg/SVGPathUtilities.cpp: Remove globals accessors, making things on the stack
instead. Use SVGPathParser helper functions where possible.
(WebCore::buildPathFromString):
(WebCore::buildSVGPathByteStreamFromSVGPathSegList):
(WebCore::appendSVGPathByteStreamFromSVGPathSeg):
(WebCore::buildPathFromByteStream):
(WebCore::buildSVGPathSegListFromByteStream):
(WebCore::buildStringFromByteStream):
(WebCore::buildStringFromSVGPathSegList):
(WebCore::buildSVGPathByteStreamFromString):
(WebCore::buildAnimatedSVGPathByteStream):
(WebCore::addToSVGPathByteStream):
(WebCore::getSVGPathSegAtLengthFromSVGPathByteStream):
(WebCore::getTotalLengthOfSVGPathByteStream):
(WebCore::getPointAtLengthOfSVGPathByteStream):
(WebCore::buildStringFromPath):
(WebCore::globalSVGPathBuilder): Deleted.
(WebCore::globalSVGPathSegListBuilder): Deleted.
(WebCore::globalSVGPathByteStreamBuilder): Deleted.
(WebCore::globalSVGPathStringBuilder): Deleted.
(WebCore::globalSVGPathTraversalStateBuilder): Deleted.
(WebCore::globalSVGPathParser): Deleted.
(WebCore::globalSVGPathBlender): Deleted.
* svg/SVGPathUtilities.h:
* svg/SVGToOTFFontConversion.cpp: CFFBuilder no longer inherits from SVGPathBuilder.
It did nothing with the Path, re-implemented all the functions, and only made use of
the m_current member var, so just make it inherit from SVGPathConsumer, and have
its own m_current.
(WebCore::SVGToOTFFontConverter::transcodeGlyphPaths):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190849 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoTest font-variant-* and font-feature-settings with mocked preinstalled fonts
mmaxfield@apple.com [Sun, 11 Oct 2015 20:23:26 +0000 (20:23 +0000)]
Test font-variant-* and font-feature-settings with mocked preinstalled fonts
https://bugs.webkit.org/show_bug.cgi?id=149777

Reviewed by Simon Fraser.

Tools:

Include FontWithFeatures in DumpRenderTree and WebKitTestRunner. Add it
to the list of tests to preinstall at startup time.

Also, update the generation of the 'name' table in the FontWithFeatures font.

* DumpRenderTree/mac/DumpRenderTree.mm:
(allowedFontFamilySet):
(activateTestingFonts):
(activateFontsIOS):
* FontWithFeatures/FontWithFeatures/FontCreator.cpp:
(Generator::generate):
(Generator::appendCFFTable):
(Generator::appendNameSubtable):
(Generator::appendNAMETable):
* FontWithFeatures/FontWithFeatures/main.cpp:
(main):
* WebKitTestRunner/InjectedBundle/cocoa/ActivateFontsCocoa.mm:
(WTR::allowedFontFamilySet):
(WTR::activateFonts):
* WebKitTestRunner/mac/TestControllerMac.mm:
(WTR::allowedFontFamilySet):

LayoutTests:

See per-file comments.

* css3/font-feature-settings-preinstalled-fonts.html: Removed. Superseded by
css3/font-feature-settings-rendering.html.
* css3/font-feature-settings-rendering-2-expected.html: Removed.
* css3/font-feature-settings-rendering-2.html: Removed. Renamed to
css3/font-feature-settings-rendering.html
* css3/font-feature-settings-rendering-expected.html: Added.
* css3/font-feature-settings-rendering.html:
* css3/font-variant-all-expected.html:
* css3/font-variant-all.html:
* css3/resources/FontWithFeatures.otf: Update for new 'name' table.
* css3/resources/FontWithFeatures.ttf: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190848 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoES6 classes: When a class extends B, super() invokes B.prototype.constructor() instea...
utatane.tea@gmail.com [Sun, 11 Oct 2015 18:46:52 +0000 (18:46 +0000)]
ES6 classes: When a class extends B, super() invokes B.prototype.constructor() instead of B()
https://bugs.webkit.org/show_bug.cgi?id=149001

Reviewed by Saam Barati.

Source/JavaScriptCore:

This patch matches the `super()` call in the constructor to the latest spec.
Before this patch, when calling `super()`, it loads `callee.[[HomeObject]].__proto__.constructor`
as a super constructor. But after this patch, it loads `callee.__proto__` as a super constructor.
This behavior corresponds to the section 12.3.5.2[1].

[1]: http://www.ecma-international.org/ecma-262/6.0/#sec-getsuperconstructor

* bytecompiler/NodesCodegen.cpp:
(JSC::SuperNode::emitBytecode):
* tests/stress/super-call-does-not-look-up-constructor.js: Added.
(shouldBe):
(B):
(C):
(B.prototype):

LayoutTests:

An error message becomes changed.

* js/class-syntax-call-expected.txt:
* js/class-syntax-extends-expected.txt:
* js/class-syntax-super-expected.txt:
* js/script-tests/class-syntax-call.js:
* js/script-tests/class-syntax-extends.js:
* js/script-tests/class-syntax-super.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190847 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUnreviewed. Add Jinyoung Hur as a committer.
hur.ims@navercorp.com [Sun, 11 Oct 2015 16:09:49 +0000 (16:09 +0000)]
Unreviewed. Add Jinyoung Hur as a committer.

* Scripts/webkitpy/common/config/contributors.json:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190846 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoRemove InsertionPoint and ContentDistributor
antti@apple.com [Sun, 11 Oct 2015 07:18:58 +0000 (07:18 +0000)]
Remove InsertionPoint and ContentDistributor
https://bugs.webkit.org/show_bug.cgi?id=150004

Rubber-stamped by Sam Weinig.

Now that <details> is on top of the modern Shadow DOM remove the last vestiges of the V0 Shadow DOM API.

* CMakeLists.txt:
* WebCore.vcxproj/WebCore.vcxproj:
* WebCore.xcodeproj/project.pbxproj:
* css/SelectorChecker.cpp:
* css/StyleResolver.cpp:
* dom/ContainerNode.cpp:
* dom/Element.cpp:
(WebCore::shouldUseNodeRenderingTraversalSlowPath):
(WebCore::Element::resetNeedsNodeRenderingTraversalSlowPath):
* dom/Node.cpp:
(WebCore::Node::parentOrShadowHostElement):
(WebCore::Node::insertedInto):
(WebCore::Node::insertionParentForBinding): Deleted.
* dom/Node.h:
(WebCore::Node::isCharacterDataNode):
(WebCore::Node::isFrameOwnerElement):
(WebCore::Node::isPluginElement):
(WebCore::Node::isImageControlsRootElement):
(WebCore::Node::isImageControlsButtonElement):
(WebCore::Node::isDocumentFragment):
(WebCore::Node::isShadowRoot):
(WebCore::Node::isInsertionPointNode): Deleted.
(WebCore::Node::isInsertionPoint): Deleted.
* dom/NodeRenderingTraversal.cpp:
(WebCore::NodeRenderingTraversal::traverseParent):
(WebCore::NodeRenderingTraversal::traverseFirstChild):
(WebCore::NodeRenderingTraversal::traverseLastChild):
(WebCore::NodeRenderingTraversal::traverseNextSibling):
(WebCore::NodeRenderingTraversal::traversePreviousSibling):
(WebCore::NodeRenderingTraversal::parentSlow):
(WebCore::NodeRenderingTraversal::nextInScope):
(WebCore::NodeRenderingTraversal::previousInScope):
(WebCore::NodeRenderingTraversal::parentInScope):
(WebCore::NodeRenderingTraversal::lastChildInScope):
(WebCore::NodeRenderingTraversal::nodeCanBeDistributed): Deleted.
(WebCore::NodeRenderingTraversal::findFirstSiblingEnteringInsertionPoints): Deleted.
(WebCore::NodeRenderingTraversal::findFirstEnteringInsertionPoints): Deleted.
(WebCore::NodeRenderingTraversal::findFirstFromDistributedNode): Deleted.
(WebCore::NodeRenderingTraversal::findLastSiblingEnteringInsertionPoints): Deleted.
(WebCore::NodeRenderingTraversal::findLastEnteringInsertionPoints): Deleted.
(WebCore::NodeRenderingTraversal::findLastFromDistributedNode): Deleted.
* dom/ShadowRoot.h:
(WebCore::ShadowRoot::distributor): Deleted.
* html/HTMLInputElement.cpp:
* html/HTMLSlotElement.cpp:
* html/HTMLSlotElement.h:
* html/shadow/ContentDistributor.cpp: Removed.
* html/shadow/ContentDistributor.h: Removed.
* html/shadow/InsertionPoint.cpp: Removed.
* html/shadow/InsertionPoint.h: Removed.
* page/FocusController.cpp:
(WebCore::FocusController::previousFocusableElement):
* testing/Internals.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190845 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoUse references and more const in SVGPathUtilities
simon.fraser@apple.com [Sun, 11 Oct 2015 06:16:18 +0000 (06:16 +0000)]
Use references and more const in SVGPathUtilities
https://bugs.webkit.org/show_bug.cgi?id=150007

Reviewed by Zalan Bujtas.

SVGPathUtilities had lots of pointers whose non-nullness was asserted,
little const, and was very trigger-happy with stack allocations. Clean
that all up.

* svg/SVGAnimatedPath.cpp:
(WebCore::SVGAnimatedPathAnimator::constructFromString):
(WebCore::SVGAnimatedPathAnimator::resetAnimValToBaseVal):
(WebCore::SVGAnimatedPathAnimator::addAnimatedTypes):
(WebCore::SVGAnimatedPathAnimator::calculateAnimatedValue):
* svg/SVGPathByteStream.h:
(WebCore::SVGPathByteStream::begin):
(WebCore::SVGPathByteStream::end):
(WebCore::SVGPathByteStream::append):
* svg/SVGPathByteStreamSource.cpp:
(WebCore::SVGPathByteStreamSource::SVGPathByteStreamSource):
* svg/SVGPathByteStreamSource.h:
* svg/SVGPathElement.cpp:
(WebCore::SVGPathElement::getTotalLength):
(WebCore::SVGPathElement::getPointAtLength):
(WebCore::SVGPathElement::getPathSegAtLength):
(WebCore::SVGPathElement::parseAttribute):
(WebCore::SVGPathElement::svgAttributeChanged):
(WebCore::SVGPathElement::pathByteStream):
(WebCore::SVGPathElement::lookupOrCreateDWrapper):
(WebCore::SVGPathElement::pathSegListChanged):
(WebCore::SVGPathElement::SVGPathElement): Deleted.
* svg/SVGPathElement.h:
* svg/SVGPathUtilities.cpp:
(WebCore::globalSVGPathBuilder):
(WebCore::globalSVGPathSegListBuilder):
(WebCore::globalSVGPathByteStreamBuilder):
(WebCore::globalSVGPathStringBuilder):
(WebCore::globalSVGPathTraversalStateBuilder):
(WebCore::globalSVGPathParser):
(WebCore::globalSVGPathBlender):
(WebCore::buildPathFromString):
(WebCore::buildSVGPathByteStreamFromSVGPathSegList):
(WebCore::appendSVGPathByteStreamFromSVGPathSeg):
(WebCore::buildPathFromByteStream):
(WebCore::buildSVGPathSegListFromByteStream):
(WebCore::buildStringFromByteStream):
(WebCore::buildStringFromSVGPathSegList):
(WebCore::buildSVGPathByteStreamFromString):
(WebCore::buildAnimatedSVGPathByteStream):
(WebCore::addToSVGPathByteStream):
(WebCore::getSVGPathSegAtLengthFromSVGPathByteStream):
(WebCore::getTotalLengthOfSVGPathByteStream):
(WebCore::getPointAtLengthOfSVGPathByteStream):
(WebCore::buildStringFromPath):
* svg/SVGPathUtilities.h:
* svg/properties/SVGAnimatedPathSegListPropertyTearOff.h:
(WebCore::SVGAnimatedPathSegListPropertyTearOff::animValDidChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190844 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4 years agoReduce pointless malloc traffic in CodeBlock construction.
akling@apple.com [Sat, 10 Oct 2015 21:14:56 +0000 (21:14 +0000)]
Reduce pointless malloc traffic in CodeBlock construction.
<https://webkit.org/b/149999>

Reviewed by Antti Koivisto.

Create the RefCountedArray<Instruction> for CodeBlock's m_instructions directly
instead of first creating a Vector<Instruction> and then creating a RefCountedArray
from that. None of the Vector functionality is needed here anyway.

* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::finishCreation):
(JSC::CodeBlock::insertBasicBlockBoundariesForControlFlowProfiler):
* bytecode/CodeBlock.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@190843 268f45cc-cd09-0410-ab3c-d52691b4dbfc