WebKit-https.git
3 years agoGardening: change to use old header guard to appease Win EWS.
mark.lam@apple.com [Tue, 6 Sep 2016 21:47:13 +0000 (21:47 +0000)]
Gardening: change to use old header guard to appease Win EWS.

Not reviewed.

* runtime/AuxiliaryBarrier.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205509 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r205494.
ryanhaddad@apple.com [Tue, 6 Sep 2016 21:43:08 +0000 (21:43 +0000)]
Unreviewed, rolling out r205494.
https://bugs.webkit.org/show_bug.cgi?id=161646

This change broke the Windows build (Requested by ryanhaddad
on #webkit).

Reverted changeset:

"Typed arrays should use MarkedSpace instead of CopiedSpace"
https://bugs.webkit.org/show_bug.cgi?id=161100
http://trac.webkit.org/changeset/205494

Patch by Commit Queue <commit-queue@webkit.org> on 2016-09-06

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205508 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r205504.
ryanhaddad@apple.com [Tue, 6 Sep 2016 21:35:24 +0000 (21:35 +0000)]
Unreviewed, rolling out r205504.
https://bugs.webkit.org/show_bug.cgi?id=161645

Broke the iOS device build (Requested by ryanhaddad on
#webkit).

Reverted changeset:

"Make JSMap and JSSet faster"
https://bugs.webkit.org/show_bug.cgi?id=160989
http://trac.webkit.org/changeset/205504

Patch by Commit Queue <commit-queue@webkit.org> on 2016-09-06

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205507 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove unused overlap checking functions from FrameView.
zalan@apple.com [Tue, 6 Sep 2016 21:27:15 +0000 (21:27 +0000)]
Remove unused overlap checking functions from FrameView.
https://bugs.webkit.org/show_bug.cgi?id=161635

Reviewed by Simon Fraser.

No change in functionality.

* page/FrameView.cpp:
(WebCore::FrameView::isOverlappedIncludingAncestors): Deleted.
* page/FrameView.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205506 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAlign element.scroll() / scrollTo() / scrollBy() with the CSSOM specification
simon.fraser@apple.com [Tue, 6 Sep 2016 21:18:06 +0000 (21:18 +0000)]
Align element.scroll() / scrollTo() / scrollBy() with the CSSOM specification
https://bugs.webkit.org/show_bug.cgi?id=161610

Reviewed by Darin Adler, Chris Dumez.

Source/WebCore:

Implement Element.scroll(), scrollBy() and scrollTo() with x,y and ScrollToOptions
on Element, according to <https://drafts.csswg.org/cssom-view/#element-scrolling-members>

WebKit's behavior of treating scrolls on the body element in both quirks and strict mode
is preserved.

Tests: fast/dom/Element/scrolling-funtions-on-body-quirks.html
       fast/dom/Element/scrolling-funtions-on-body.html
       fast/dom/Element/scrolling-funtions-on-element.html

* dom/Element.cpp:
(WebCore::Element::scrollBy):
(WebCore::normalizeNonFiniteValue):
(WebCore::Element::scrollTo):
* dom/Element.h:
* dom/Element.idl:
* html/HTMLBodyElement.cpp:
(WebCore::HTMLBodyElement::scrollTo):
* html/HTMLBodyElement.h:

LayoutTests:

* fast/dom/Element/scrolling-funtions-on-body-expected.txt: Added.
* fast/dom/Element/scrolling-funtions-on-body-quirks-expected.txt: Added.
* fast/dom/Element/scrolling-funtions-on-body-quirks.html: Added.
* fast/dom/Element/scrolling-funtions-on-body.html: Added.
* fast/dom/Element/scrolling-funtions-on-element-expected.txt: Added.
* fast/dom/Element/scrolling-funtions-on-element.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205505 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMake JSMap and JSSet faster
sbarati@apple.com [Tue, 6 Sep 2016 21:13:25 +0000 (21:13 +0000)]
Make JSMap and JSSet faster
https://bugs.webkit.org/show_bug.cgi?id=160989

Reviewed by Filip Pizlo.

JSTests:

* microbenchmarks/dense-set.js: Added.
(bench):
* microbenchmarks/large-map-iteration-with-additions.js: Added.
(bar):
(foo):
* microbenchmarks/large-map-iteration-with-mutation.js: Added.
(bar):
(foo):
* microbenchmarks/large-map-iteration.js: Added.
(bar):
(foo):
* microbenchmarks/map-get-get-cse.js: Added.
(bar):
(foo):
* microbenchmarks/map-has-get-cse-opportunity.js: Added.
(bar):
(foo):
* microbenchmarks/sparse-set.js: Added.
(bench):
* stress/map-cse-correctness.js: Added.
(assert):
(testHas):
(testGet):
(foo):
* stress/map-iteration.js: Added.
(assert):
(test1):
(test2):
(test3):
(test4):
(test5):
(test6):
(test7):
(test8):
(test9):
(test10):
(test11):
(test12):
(test13):
(test14):
(test15):
(test16):
(test17):
(test18):

Source/JavaScriptCore:

This patch revamps how we implement Map and Set. It uses
a new hash map implementation. The hash map uses linear
probing and it uses Wang's 64 bit hash function for JSValues
that aren't strings. Strings use StringImpl's hash function.
The reason I wanted to roll our own HashTable is twofold:
I didn't want to inline WTF::HashMap's implementation into our
JIT, since that seems error prone and unmaintainable. Also, I wanted
a different structure for hash map buckets where buckets also exist in
a linked list.

The reason for making buckets part of a linked list is that iteration
is now simple. Iteration works by just traversing a linked list.
This design also allows for a simple implementation when doing iteration
while the hash table is mutating. Whenever we remove a bucket from
the hash table, it is removed from the list, meaning items in the
list don't point to it. However, the removed bucket will still point
to things that are either in the list, or have also been removed.
e.g, from a removed bucket, you can always follow pointers until you
either find an item in the list, or you find the tail of the list.
This is a really nice property because it means that a Map or Set
does not need to reason about the all the iterators that point
into its list. Also, whenever we add items to the Map or Set, we
hijack the tail as the new item, and make the new item point to a newly
created tail. This means that any iterator that pointed to the "tail" now
points to non-tail items. This makes the implementation of adding things
to the Map/Set while iterating easy.

I also made Map.prototype.get, Map.prototype.has, and Set.prototype.has
into intrinsics in the DFG. The IR can now reason about hash map
operations and can even do CSE over Wang's hash function, hash map
bucket lookups, hash map bucket loads, and testing if a key is in
the hash table. This makes code patterns for Map like so, super fast
in the FTL, since we will only be doing a single hash and hash bucket lookup:

```
function getKeyIfPresent(map, key) {
    if (map.has(key))
        return map.get(key);
}
```

This patch is roughly an 8% speedup on ES6SampleBench.

* CMakeLists.txt:
* JavaScriptCore.xcodeproj/project.pbxproj:
* bytecode/SpeculatedType.cpp:
(JSC::speculationFromClassInfo):
* bytecode/SpeculatedType.h:
* dfg/DFGAbstractInterpreterInlines.h:
(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):
(JSC::DFG::AbstractInterpreter<AbstractStateType>::execute):
* dfg/DFGByteCodeParser.cpp:
(JSC::DFG::ByteCodeParser::handleIntrinsicCall):
* dfg/DFGClobberize.h:
(JSC::DFG::clobberize):
* dfg/DFGDoesGC.cpp:
(JSC::DFG::doesGC):
* dfg/DFGEdge.h:
(JSC::DFG::Edge::shift):
(JSC::DFG::Edge::makeWord):
* dfg/DFGFixupPhase.cpp:
(JSC::DFG::FixupPhase::fixupNode):
* dfg/DFGHeapLocation.cpp:
(WTF::printInternal):
* dfg/DFGHeapLocation.h:
* dfg/DFGNode.h:
(JSC::DFG::Node::hasHeapPrediction):
* dfg/DFGNodeType.h:
* dfg/DFGOperations.cpp:
* dfg/DFGOperations.h:
* dfg/DFGPredictionPropagationPhase.cpp:
* dfg/DFGSafeToExecute.h:
(JSC::DFG::SafeToExecuteEdge::operator()):
(JSC::DFG::safeToExecute):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::speculateMapObject):
(JSC::DFG::SpeculativeJIT::speculateSetObject):
(JSC::DFG::SpeculativeJIT::speculate):
* dfg/DFGSpeculativeJIT.h:
(JSC::DFG::SpeculativeJIT::callOperation):
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
* dfg/DFGUseKind.cpp:
(WTF::printInternal):
* dfg/DFGUseKind.h:
(JSC::DFG::typeFilterFor):
(JSC::DFG::isCell):
* ftl/FTLAbstractHeapRepository.h:
* ftl/FTLCapabilities.cpp:
(JSC::FTL::canCompile):
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileNode):
(JSC::FTL::DFG::LowerDFGToB3::compileMapHash):
(JSC::FTL::DFG::LowerDFGToB3::compileGetMapBucket):
(JSC::FTL::DFG::LowerDFGToB3::compileLoadFromJSMapBucket):
(JSC::FTL::DFG::LowerDFGToB3::compileIsNonEmptyMapBucket):
(JSC::FTL::DFG::LowerDFGToB3::lowMapObject):
(JSC::FTL::DFG::LowerDFGToB3::lowSetObject):
(JSC::FTL::DFG::LowerDFGToB3::lowMapBucket):
(JSC::FTL::DFG::LowerDFGToB3::speculate):
(JSC::FTL::DFG::LowerDFGToB3::speculateMapObject):
(JSC::FTL::DFG::LowerDFGToB3::speculateSetObject):
(JSC::FTL::DFG::LowerDFGToB3::setMapBucket):
(JSC::FTL::DFG::LowerDFGToB3::lowRegExpObject): Deleted.
(JSC::FTL::DFG::LowerDFGToB3::lowStorage): Deleted.
(JSC::FTL::DFG::LowerDFGToB3::speculateRegExpObject): Deleted.
(JSC::FTL::DFG::LowerDFGToB3::setStorage): Deleted.
* jit/AssemblyHelpers.cpp:
(JSC::AssemblyHelpers::wangsInt64Hash):
* jit/AssemblyHelpers.h:
(JSC::AssemblyHelpers::emitAllocateDestructibleObject): Deleted.
* jit/JITOperations.h:
* parser/ModuleAnalyzer.cpp:
(JSC::ModuleAnalyzer::ModuleAnalyzer):
* runtime/HashMapImpl.cpp: Added.
(JSC::HashMapBucket<Data>::visitChildren):
(JSC::HashMapImpl<HashMapBucket>::visitChildren):
(JSC::HashMapImpl<HashMapBucket>::copyBackingStore):
* runtime/HashMapImpl.h: Added.
(JSC::HashMapBucket::selectStructure):
(JSC::HashMapBucket::createStructure):
(JSC::HashMapBucket::create):
(JSC::HashMapBucket::HashMapBucket):
(JSC::HashMapBucket::setNext):
(JSC::HashMapBucket::setPrev):
(JSC::HashMapBucket::setKey):
(JSC::HashMapBucket::setValue):
(JSC::HashMapBucket::key):
(JSC::HashMapBucket::value):
(JSC::HashMapBucket::next):
(JSC::HashMapBucket::prev):
(JSC::HashMapBucket::deleted):
(JSC::HashMapBucket::setDeleted):
(JSC::HashMapBucket::offsetOfKey):
(JSC::HashMapBucket::offsetOfValue):
(JSC::HashMapBuffer::allocationSize):
(JSC::HashMapBuffer::buffer):
(JSC::HashMapBuffer::create):
(JSC::areKeysEqual):
(JSC::normalizeMapKey):
(JSC::jsMapHash):
(JSC::HashMapImpl::selectStructure):
(JSC::HashMapImpl::createStructure):
(JSC::HashMapImpl::create):
(JSC::HashMapImpl::HashMapImpl):
(JSC::HashMapImpl::buffer):
(JSC::HashMapImpl::finishCreation):
(JSC::HashMapImpl::emptyValue):
(JSC::HashMapImpl::isEmpty):
(JSC::HashMapImpl::deletedValue):
(JSC::HashMapImpl::isDeleted):
(JSC::HashMapImpl::findBucket):
(JSC::HashMapImpl::get):
(JSC::HashMapImpl::has):
(JSC::HashMapImpl::add):
(JSC::HashMapImpl::remove):
(JSC::HashMapImpl::size):
(JSC::HashMapImpl::clear):
(JSC::HashMapImpl::bufferSizeInBytes):
(JSC::HashMapImpl::offsetOfBuffer):
(JSC::HashMapImpl::offsetOfCapacity):
(JSC::HashMapImpl::head):
(JSC::HashMapImpl::tail):
(JSC::HashMapImpl::approximateSize):
(JSC::HashMapImpl::findBucketAlreadyHashedAndNormalized):
(JSC::HashMapImpl::rehash):
(JSC::HashMapImpl::makeAndSetNewBuffer):
* runtime/Intrinsic.h:
* runtime/JSCJSValue.h:
* runtime/JSCJSValueInlines.h:
(JSC::sameValue):
* runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::init):
* runtime/JSMap.cpp:
(JSC::JSMap::destroy): Deleted.
(JSC::JSMap::estimatedSize): Deleted.
(JSC::JSMap::visitChildren): Deleted.
(JSC::JSMap::copyBackingStore): Deleted.
(JSC::JSMap::has): Deleted.
(JSC::JSMap::size): Deleted.
(JSC::JSMap::get): Deleted.
(JSC::JSMap::set): Deleted.
(JSC::JSMap::clear): Deleted.
(JSC::JSMap::remove): Deleted.
* runtime/JSMap.h:
(JSC::JSMap::createStructure):
(JSC::JSMap::create):
(JSC::JSMap::get):
(JSC::JSMap::set):
(JSC::JSMap::JSMap):
(JSC::JSMap::Entry::key): Deleted.
(JSC::JSMap::Entry::value): Deleted.
(JSC::JSMap::Entry::visitChildren): Deleted.
(JSC::JSMap::Entry::setKey): Deleted.
(JSC::JSMap::Entry::setKeyWithoutWriteBarrier): Deleted.
(JSC::JSMap::Entry::setValue): Deleted.
(JSC::JSMap::Entry::clear): Deleted.
* runtime/JSMapIterator.cpp:
(JSC::JSMapIterator::finishCreation):
(JSC::JSMapIterator::visitChildren):
(JSC::JSMapIterator::clone):
* runtime/JSMapIterator.h:
(JSC::JSMapIterator::advanceIter):
(JSC::JSMapIterator::next):
(JSC::JSMapIterator::nextKeyValue):
(JSC::JSMapIterator::JSMapIterator):
(JSC::JSMapIterator::setIterator):
(JSC::JSMapIterator::finish): Deleted.
(JSC::JSMapIterator::iteratorData): Deleted.
* runtime/JSModuleLoader.cpp:
(JSC::JSModuleLoader::finishCreation):
* runtime/JSModuleLoader.h:
(JSC::JSModuleLoader::create):
* runtime/JSModuleRecord.cpp:
(JSC::JSModuleRecord::finishCreation):
* runtime/JSModuleRecord.h:
(JSC::JSModuleRecord::create):
* runtime/JSSet.cpp:
(JSC::JSSet::destroy): Deleted.
(JSC::JSSet::estimatedSize): Deleted.
(JSC::JSSet::visitChildren): Deleted.
(JSC::JSSet::copyBackingStore): Deleted.
(JSC::JSSet::has): Deleted.
(JSC::JSSet::size): Deleted.
(JSC::JSSet::add): Deleted.
(JSC::JSSet::clear): Deleted.
(JSC::JSSet::remove): Deleted.
* runtime/JSSet.h:
(JSC::JSSet::createStructure):
(JSC::JSSet::create):
(JSC::JSSet::add):
(JSC::JSSet::JSSet):
(JSC::JSSet::Entry::key): Deleted.
(JSC::JSSet::Entry::value): Deleted.
(JSC::JSSet::Entry::visitChildren): Deleted.
(JSC::JSSet::Entry::setKey): Deleted.
(JSC::JSSet::Entry::setKeyWithoutWriteBarrier): Deleted.
(JSC::JSSet::Entry::setValue): Deleted.
(JSC::JSSet::Entry::clear): Deleted.
* runtime/JSSetIterator.cpp:
(JSC::JSSetIterator::finishCreation):
(JSC::JSSetIterator::visitChildren):
(JSC::JSSetIterator::clone):
* runtime/JSSetIterator.h:
(JSC::JSSetIterator::advanceIter):
(JSC::JSSetIterator::next):
(JSC::JSSetIterator::JSSetIterator):
(JSC::JSSetIterator::setIterator):
(JSC::JSSetIterator::finish): Deleted.
(JSC::JSSetIterator::iteratorData): Deleted.
* runtime/JSType.h:
* runtime/MapBase.cpp: Added.
(JSC::MapBase<HashMapBucketType>::visitChildren):
(JSC::MapBase<HashMapBucketType>::estimatedSize):
* runtime/MapBase.h: Added.
(JSC::MapBase::size):
(JSC::MapBase::has):
(JSC::MapBase::clear):
(JSC::MapBase::remove):
(JSC::MapBase::findBucket):
(JSC::MapBase::offsetOfHashMapImpl):
(JSC::MapBase::impl):
(JSC::MapBase::finishCreation):
(JSC::MapBase::MapBase):
* runtime/MapConstructor.cpp:
(JSC::constructMap):
* runtime/MapIteratorPrototype.cpp:
(JSC::MapIteratorPrototypeFuncNext):
* runtime/MapPrototype.cpp:
(JSC::MapPrototype::finishCreation):
(JSC::getMap):
(JSC::privateFuncIsMap):
(JSC::privateFuncMapIteratorNext):
* runtime/PropertyDescriptor.cpp:
(JSC::sameValue): Deleted.
* runtime/PropertyDescriptor.h:
* runtime/SetConstructor.cpp:
(JSC::constructSet):
* runtime/SetIteratorPrototype.cpp:
(JSC::SetIteratorPrototypeFuncNext):
* runtime/SetPrototype.cpp:
(JSC::SetPrototype::finishCreation):
(JSC::getSet):
(JSC::privateFuncSetIteratorNext):
* runtime/VM.cpp:
(JSC::VM::VM):
* runtime/VM.h:

Source/WebCore:

* ForwardingHeaders/runtime/HashMapImpl.h: Added.
* ForwardingHeaders/runtime/MapBase.h: Added.
* bindings/js/SerializedScriptValue.cpp:
(WebCore::CloneSerializer::serialize):
(WebCore::CloneDeserializer::deserialize):

Source/WTF:

I made s_flagCount public since JSC's JITs now use this field.

* wtf/text/StringImpl.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205504 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSkip some asserting Fetch API tests.
ryanhaddad@apple.com [Tue, 6 Sep 2016 20:57:38 +0000 (20:57 +0000)]
Skip some asserting Fetch API tests.
https://bugs.webkit.org/show_bug.cgi?id=161632

Unreviewed test gardening.

* TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205503 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Xcode] WebKitLegacy.framework is re-signed on incremental build even if nothing...
mitz@apple.com [Tue, 6 Sep 2016 20:50:26 +0000 (20:50 +0000)]
[Xcode] WebKitLegacy.framework is re-signed on incremental build even if nothing’s changed
https://bugs.webkit.org/show_bug.cgi?id=161642

Reviewed by Tim Horton.

* WebKit.xcodeproj/project.pbxproj:  Removed the output file listed under the
Symlink WebKitPluginHost script build phase, because it doesn’t produce that file on iOS.
With no outputs listed, the script will still run every time, but the fact that it has run
won’t trigger re-signing the framework.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205502 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRevert r205491 because it did not resolve the test failures, update bug number in...
ryanhaddad@apple.com [Tue, 6 Sep 2016 20:44:00 +0000 (20:44 +0000)]
Revert r205491 because it did not resolve the test failures, update bug number in TestExpectations.
https://bugs.webkit.org/show_bug.cgi?id=161366

Unreviewed test gardening.

* platform/mac-wk2/TestExpectations:
* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205501 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago<rdar://problem/8064563> Ugly NSLog in DOMInternal.mm
mitz@apple.com [Tue, 6 Sep 2016 20:39:37 +0000 (20:39 +0000)]
<rdar://problem/8064563> Ugly NSLog in DOMInternal.mm

Rubber-stamped by Anders Carlsson.

* DOM/DOMInternal.mm:
(-[WebScriptObject _initializeScriptDOMNodeImp]): Removed useless call to NSLog.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205500 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Cannot undo in breakpoint editor
commit-queue@webkit.org [Tue, 6 Sep 2016 19:56:37 +0000 (19:56 +0000)]
Web Inspector: Cannot undo in breakpoint editor
https://bugs.webkit.org/show_bug.cgi?id=152858

Patch by Devin Rousso <dcrousso+webkit@gmail.com> on 2016-09-06
Reviewed by Brian Burg.

* UserInterface/Controllers/BreakpointPopoverController.js:
(WebInspector.BreakpointPopoverController.prototype._conditionCodeMirrorBeforeChange):
Since the breakpoint editor only allows a single line of JavaScript, it attempts to remove
all "\n" characters.  During an "undo" action, this is not possible.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205499 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUse TemporaryChange<> to maintain m_userIsInteracting
simon.fraser@apple.com [Tue, 6 Sep 2016 18:55:12 +0000 (18:55 +0000)]
Use TemporaryChange<> to maintain m_userIsInteracting
https://bugs.webkit.org/show_bug.cgi?id=161497

Reviewed by Tim Horton.

Replace some error-prone manual set/clear of m_userIsInteracting with
TemporaryChange<>.

No behavior change.

* WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::mouseEvent):
(WebKit::WebPage::keyEvent):
(WebKit::WebPage::dispatchTouchEvent):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205498 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Win] Enable strict runtime stack buffer checks.
pvollan@apple.com [Tue, 6 Sep 2016 18:55:02 +0000 (18:55 +0000)]
[Win] Enable strict runtime stack buffer checks.
https://bugs.webkit.org/show_bug.cgi?id=161629

Reviewed by Brent Fulgham.

* wtf/Platform.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205497 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove SwipeTransitionStyle; only Overlap has ever been used
timothy_horton@apple.com [Tue, 6 Sep 2016 18:38:15 +0000 (18:38 +0000)]
Remove SwipeTransitionStyle; only Overlap has ever been used
https://bugs.webkit.org/show_bug.cgi?id=161617

Reviewed by Alex Christensen.

* UIProcess/mac/ViewGestureController.h:
* UIProcess/mac/ViewGestureControllerMac.mm:
(WebKit::ViewGestureController::beginSwipeGesture):
(WebKit::ViewGestureController::handleSwipeGesture):
Remove SwipeTransitionStyle::Push, and SwipeTransitionStyle itself entirely,
because we've never used it and it is only supported by the Mac
implementation of ViewGestureController.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205496 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Win] Null pointer crash under WebView::scrollOffset.
pvollan@apple.com [Tue, 6 Sep 2016 18:21:26 +0000 (18:21 +0000)]
[Win] Null pointer crash under WebView::scrollOffset.
https://bugs.webkit.org/show_bug.cgi?id=161620

Reviewed by Darin Adler.

Added null pointer checks.

* WebView.cpp:
(WebView::scrollOffset):
(WebView::scrollBy):
(WebView::visibleContentRect):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205495 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoTyped arrays should use MarkedSpace instead of CopiedSpace
fpizlo@apple.com [Tue, 6 Sep 2016 18:18:10 +0000 (18:18 +0000)]
Typed arrays should use MarkedSpace instead of CopiedSpace
https://bugs.webkit.org/show_bug.cgi?id=161100

Reviewed by Geoffrey Garen.

This moves typed array backing stores out of CopiedSpace and into Auxiliary MarkedSpace.

This is a purely mechanical change since Auxiliary MarkedSpace already knows how to do
everything that typed arrays want.

* dfg/DFGOperations.cpp:
(JSC::DFG::newTypedArrayWithSize):
* dfg/DFGOperations.h:
(JSC::DFG::operationNewTypedArrayWithSizeForType):
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::compileNewTypedArray):
* dfg/DFGSpeculativeJIT.h:
(JSC::DFG::SpeculativeJIT::callOperation):
(JSC::DFG::SpeculativeJIT::emitAllocateBasicStorage): Deleted.
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileNewTypedArray):
(JSC::FTL::DFG::LowerDFGToB3::initializeArrayElements):
(JSC::FTL::DFG::LowerDFGToB3::splatWords):
(JSC::FTL::DFG::LowerDFGToB3::allocateBasicStorageAndGetEnd): Deleted.
(JSC::FTL::DFG::LowerDFGToB3::allocateBasicStorage): Deleted.
* heap/CopyToken.h:
* heap/SlotVisitor.cpp:
(JSC::SlotVisitor::markAuxiliary):
* jit/JITOperations.h:
* runtime/JSArrayBufferView.cpp:
(JSC::JSArrayBufferView::ConstructionContext::ConstructionContext):
(JSC::JSArrayBufferView::JSArrayBufferView):
* runtime/JSArrayBufferView.h:
* runtime/JSGenericTypedArrayView.h:
* runtime/JSGenericTypedArrayViewInlines.h:
(JSC::JSGenericTypedArrayView<Adaptor>::createWithFastVector):
(JSC::JSGenericTypedArrayView<Adaptor>::visitChildren):
(JSC::JSGenericTypedArrayView<Adaptor>::slowDownAndWasteMemory):
(JSC::JSGenericTypedArrayView<Adaptor>::copyBackingStore): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205494 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoImplement relative file urls and begin implementing character encoding in URLParser
achristensen@apple.com [Tue, 6 Sep 2016 18:16:07 +0000 (18:16 +0000)]
Implement relative file urls and begin implementing character encoding in URLParser
https://bugs.webkit.org/show_bug.cgi?id=161618

Reviewed by Tim Horton.

Source/WebCore:

Covered by new API tests.
Also, this is a significant step towards passing the URL web platform tests when using the URLParser,
which is still off by default.

* platform/URLParser.cpp:
(WebCore::isInSimpleEncodeSet):
(WebCore::isInDefaultEncodeSet):
(WebCore::isInUserInfoEncodeSet):
(WebCore::isInvalidDomainCharacter):
(WebCore::shouldCopyFileURL):
(WebCore::percentEncode):
(WebCore::utf8PercentEncode):
(WebCore::encodeQuery):
(WebCore::isDefaultPort):
(WebCore::isPercentEncodedDot):
(WebCore::URLParser::parse):
(WebCore::percentDecode):
(WebCore::domainToASCII):
(WebCore::hasInvalidDomainCharacter):
(WebCore::URLParser::parsePort):
(WebCore::URLParser::parseHost):
(WebCore::isTabOrNewline): Deleted.
* platform/URLParser.h:

Tools:

* TestWebKitAPI/Tests/WebCore/URLParser.cpp:
(TestWebKitAPI::TEST_F):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205493 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFix the Apple-internal build following <https://trac.webkit.org/changeset/205488>
dbates@webkit.org [Tue, 6 Sep 2016 18:02:28 +0000 (18:02 +0000)]
Fix the Apple-internal build following <https://trac.webkit.org/changeset/205488>
(https://bugs.webkit.org/show_bug.cgi?id=161090)

Use the default linkage for QuartzCore constants to match the linkage used for such constants
in the Apple-internal SDK.

* platform/spi/cocoa/QuartzCoreSPI.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205492 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSkipping editing/inserting/smart-link-when-caret-is-moved-before-URL.html.
ryanhaddad@apple.com [Tue, 6 Sep 2016 17:43:30 +0000 (17:43 +0000)]
Skipping editing/inserting/smart-link-when-caret-is-moved-before-URL.html.
https://bugs.webkit.org/show_bug.cgi?id=161366

Unreviewed test gardening.

* platform/mac-wk2/TestExpectations:
* platform/mac/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205491 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMarking imported/w3c/web-platform-tests/html/browsers/browsing-the-web/scroll-to...
ryanhaddad@apple.com [Tue, 6 Sep 2016 17:43:27 +0000 (17:43 +0000)]
Marking imported/w3c/web-platform-tests/html/browsers/browsing-the-web/scroll-to-fragid/scroll-to-id-top.html as flaky
https://bugs.webkit.org/show_bug.cgi?id=161631

Unreviewed test gardening.

* platform/ios-simulator-wk2/TestExpectations:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205490 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago<table> inside <div align="right"> with large content inside = no scrollbar
dbates@webkit.org [Tue, 6 Sep 2016 17:20:58 +0000 (17:20 +0000)]
<table> inside <div align="right"> with large content inside = no scrollbar
https://bugs.webkit.org/show_bug.cgi?id=3352

Reviewed by Brent Fulgham.

Source/WebCore:

Respect "right" text-alignment only if the width of the child does not overflow the
width of its containing block. This makes our margin computation more closely conform
to section "Block-level, non-replaced elements in normal flow" of the CSS 2.1
spec., <http://www.w3.org/TR/CSS21/>, and to the behavior observed in Firefox, Chrome,
IE version 9 or later and Microsoft Edge.

Test: fast/block/align-overflow-child.html

* rendering/RenderBox.cpp:
(WebCore::RenderBox::computeInlineDirectionMargins):

LayoutTests:

Add a new test that is derived from the test LayoutTests/fast/block/align-bigger-child.html included
in <https://src.chromium.org/viewvc/blink?revision=187157&view=revision>.

Also update expected results of existing tests.

* fast/block/align-overflow-child-expected.txt: Added.
* fast/block/align-overflow-child.html: Added.
* platform/efl/tables/mozilla_expected_failures/marvin/table_overflow_td_align_right-expected.txt:
* platform/gtk/tables/mozilla_expected_failures/marvin/table_overflow_td_align_right-expected.txt:
* platform/ios-simulator/tables/mozilla_expected_failures/marvin/table_overflow_td_align_right-expected.txt:
* platform/mac/tables/mozilla_expected_failures/marvin/table_overflow_td_align_right-expected.txt:
* platform/win/tables/mozilla_expected_failures/marvin/table_overflow_td_align_right-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205489 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove EXTERN_C from WTF
dbates@webkit.org [Tue, 6 Sep 2016 17:17:11 +0000 (17:17 +0000)]
Remove EXTERN_C from WTF
https://bugs.webkit.org/show_bug.cgi?id=161090

Reviewed by Brent Fulgham.

Guard external C declarations in WTF_EXTERN_C_BEGIN, WTF_EXTERN_C_END.

Source/WebCore:

* platform/spi/cf/CFLocaleSPI.h:
* platform/spi/cf/CFNetworkConnectionCacheSPI.h:
* platform/spi/cf/CFNetworkSPI.h: Additionally, remove the workaround for <rdar://problem/18337182>
as this bug was fixed in El Capitan.
* platform/spi/cf/CFUtilitiesSPI.h:
* platform/spi/cocoa/CoreTextSPI.h:
* platform/spi/cocoa/DataDetectorsCoreSPI.h:
* platform/spi/cocoa/IOPMLibSPI.h:
* platform/spi/cocoa/MachVMSPI.h:
* platform/spi/cocoa/QuartzCoreSPI.h:
* platform/spi/ios/MobileGestaltSPI.h:
* platform/spi/ios/QuickLookSPI.h:
* platform/spi/win/CoreTextSPIWin.cpp:
* platform/spi/win/CoreTextSPIWin.h:

Source/WebKit2:

* Platform/spi/ios/ManagedConfigurationSPI.h:

Source/WTF:

* wtf/Compiler.h: Remove macro definition EXTERN_C.
* wtf/spi/cf/CFBundleSPI.h:
* wtf/spi/cocoa/NSMapTableSPI.h:
* wtf/spi/cocoa/SecuritySPI.h:
* wtf/spi/darwin/CommonCryptoSPI.h:
* wtf/spi/darwin/SandboxSPI.h:
* wtf/spi/darwin/XPCSPI.h:
* wtf/spi/darwin/dyldSPI.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205488 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r205407.
ryanhaddad@apple.com [Tue, 6 Sep 2016 16:55:13 +0000 (16:55 +0000)]
Unreviewed, rolling out r205407.

Not the correct way to implement this functionality

Reverted changeset:

"Consult with the FrameLoaderClient about whether or not
content extensions should be enabled when loading this URL."
https://bugs.webkit.org/show_bug.cgi?id=161441
http://trac.webkit.org/changeset/205407

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205487 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r205480.
commit-queue@webkit.org [Tue, 6 Sep 2016 16:45:38 +0000 (16:45 +0000)]
Unreviewed, rolling out r205480.
https://bugs.webkit.org/show_bug.cgi?id=161630

tests-options.json is not known from knockout filesystem
(Requested by youenn on #webkit).

Reverted changeset:

"Using tests-options.json to compute whether tests are slow"
https://bugs.webkit.org/show_bug.cgi?id=161626
http://trac.webkit.org/changeset/205480

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205486 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK][Wayland] evince-browser-plugin prevents viewing PDFs
carlosgc@webkit.org [Tue, 6 Sep 2016 16:28:03 +0000 (16:28 +0000)]
[GTK][Wayland] evince-browser-plugin prevents viewing PDFs
https://bugs.webkit.org/show_bug.cgi?id=158697

Reviewed by Michael Catanzaro.

Use a different cache file for plugins depending on the current platform display. Plugins can claim to work on
X11 but not on Wayland, for example, if they need XEmebed to work. That's the case of the evince browser plugin.

* UIProcess/Plugins/gtk/PluginInfoCache.cpp:
(WebKit::cacheFilenameForCurrentDisplay):
(WebKit::PluginInfoCache::PluginInfoCache):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205485 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Threaded Compositor] Update timer should have lower priority than tasks scheduled...
carlosgc@webkit.org [Tue, 6 Sep 2016 16:26:05 +0000 (16:26 +0000)]
[Threaded Compositor] Update timer should have lower priority than tasks scheduled in compositing thread
https://bugs.webkit.org/show_bug.cgi?id=161625

Reviewed by Sergio Villar Senin.

This way, if a resize is scheduled, for example, and there are pending updates, the resize will happen first,
and then the updates will happen at the right size.

* Shared/CoordinatedGraphics/threadedcompositor/CompositingRunLoop.cpp:
(WebKit::CompositingRunLoop::CompositingRunLoop): Use G_PRIORITY_HIGH_IDLE as update timer priority. The actual
priority doesn't really matter as long as it's lower than G_PRIORITY_DEFAULT, because there are only two
possible sources in the compositing run loop, one for the tasks scheduled, and another one for the update timer.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205484 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago"Redundant argument in sprintf" warning spam from prepare-ChangeLog
mcatanzaro@igalia.com [Tue, 6 Sep 2016 16:24:02 +0000 (16:24 +0000)]
"Redundant argument in sprintf" warning spam from prepare-ChangeLog
https://bugs.webkit.org/show_bug.cgi?id=161606

Reviewed by Darin Adler.

Avoid extra arguments in call to sprintf, hopefully without introducing new warnings this
time.

* Scripts/prepare-ChangeLog:
(statusDescription):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205483 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRebaseline http/tests/cookies/third-party-cookie-relaxing.html for Sierra.
ryanhaddad@apple.com [Tue, 6 Sep 2016 16:15:30 +0000 (16:15 +0000)]
Rebaseline http/tests/cookies/third-party-cookie-relaxing.html for Sierra.

Unreviewed test gardening.

* platform/mac/http/tests/cookies/third-party-cookie-relaxing-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205482 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agohttp/tests/security/contentSecurityPolicy/worker-csp-blocks-xhr-redirect-cross-origin...
commit-queue@webkit.org [Tue, 6 Sep 2016 16:03:39 +0000 (16:03 +0000)]
http/tests/security/contentSecurityPolicy/worker-csp-blocks-xhr-redirect-cross-origin.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=161627

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-06
Reviewed by Darin Adler.

Source/WebCore:

Covered by changed expectation.

* loader/DocumentThreadableLoader.cpp:
(WebCore::DocumentThreadableLoader::redirectReceived): Clearing the resource in lieu of clearing the request. This will stop the resource load.
* loader/SubresourceLoader.cpp:
(WebCore::SubresourceLoader::didReceiveResponse): Adding an ASSERT to ensure that no valid redirect responses is mistakenly processed here.

LayoutTests:

* TestExpectations: Removing flakiness expectation.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205481 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUsing tests-options.json to compute whether tests are slow
commit-queue@webkit.org [Tue, 6 Sep 2016 16:01:54 +0000 (16:01 +0000)]
Using tests-options.json to compute whether tests are slow
https://bugs.webkit.org/show_bug.cgi?id=161626

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-06
Reviewed by Darin Adler.

Tools:

* Scripts/webkitpy/layout_tests/controllers/manager.py:
(Manager.__init__): Loading tests-options.json
(Manager._test_is_slow): Computing timeout value based on tests-options.json.

LayoutTests:

* TestExpectations: Removing slow WPT expectations now that they are handled through tests-options.json.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205480 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoASSERTION FAILED: !currBox->needsLayout() in WebCore::RenderBlock::checkPositionedObj...
zalan@apple.com [Tue, 6 Sep 2016 16:00:11 +0000 (16:00 +0000)]
ASSERTION FAILED: !currBox->needsLayout() in WebCore::RenderBlock::checkPositionedObjectsNeedLayout
https://bugs.webkit.org/show_bug.cgi?id=120291
<rdar://problem/27683456>

Reviewed by David Hyatt.

Source/WebCore:

This patch ensures that we always set the 'positioned child needs layout' bit on the containing block
when a new positioned descendant gets inserted. It fixes cases for simplified layout when
we ended up not laying out the descendant element.

Test: fast/block/positioning/assert-when-positioned-descendant-is-not-getting-laid-out.html

* rendering/RenderBlock.cpp:
(WebCore::RenderBlock::insertPositionedObject):

LayoutTests:

* fast/block/positioning/assert-when-positioned-descendant-is-not-getting-laid-out-expected.txt: Added.
* fast/block/positioning/assert-when-positioned-descendant-is-not-getting-laid-out.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205479 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r205461.
commit-queue@webkit.org [Tue, 6 Sep 2016 14:32:02 +0000 (14:32 +0000)]
Unreviewed, rolling out r205461.
https://bugs.webkit.org/show_bug.cgi?id=161628

Introduced "Use of uninitialized value $original in sprintf"
warning (Requested by mcatanzaro on #webkit).

Reverted changeset:

""Redundant argument in sprintf" warning spam from prepare-
ChangeLog"
https://bugs.webkit.org/show_bug.cgi?id=161606
http://trac.webkit.org/changeset/205461

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205478 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSilence GCC warning spam introduced in r205462
mcatanzaro@igalia.com [Tue, 6 Sep 2016 14:29:43 +0000 (14:29 +0000)]
Silence GCC warning spam introduced in r205462

Rubber-stamped by Filip Pizlo.

* bytecode/Opcode.h:
(JSC::padOpcodeName):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205477 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoW3C test importer should be able to identify slow test
commit-queue@webkit.org [Tue, 6 Sep 2016 12:30:51 +0000 (12:30 +0000)]
W3C test importer should be able to identify slow test
https://bugs.webkit.org/show_bug.cgi?id=161601

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-06
Reviewed by Darin Adler.

Tools:

Test parser is now identifying slow tests based on "meta element name=timeout content=long" elements.
Test importer is updating test-options.json file according that information.

In case of clean import, the test-options.json is cleaned from 'slow' options for all W3C tests.

* Scripts/webkitpy/w3c/test_importer.py:
(TestImporter.__init__):
(TestImporter.find_importable_tests):
(TestImporter.import_tests):
(TestImporter.update_test_options):
(TestImporter):
(TestImporter.remove_slow_from_w3c_test_options):
* Scripts/webkitpy/w3c/test_parser.py:
(TestParser.analyze_test):
(TestParser.is_slow_test):
(TestParser.potential_ref_filename):

LayoutTests:

* tests-options.json: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205476 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] glImageTargetTexture2D() is a GL function, rename accordingly
commit-queue@webkit.org [Tue, 6 Sep 2016 11:28:32 +0000 (11:28 +0000)]
[GTK] glImageTargetTexture2D() is a GL function, rename accordingly
https://bugs.webkit.org/show_bug.cgi?id=161622

Patch by Emanuele Aina <emanuele.aina@collabora.com> on 2016-09-06
Reviewed by Carlos Garcia Campos.

glImageTargetTexture2D() is a GL function, not EGL, using the 'egl'
prefix is misleading.

* UIProcess/gtk/WaylandCompositor.cpp:
(WebKit::WaylandCompositor::Surface::prepareTextureForPainting):
(WebKit::WaylandCompositor::initializeEGL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205475 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK][EFL] Make *_PROCESS_CMD_PREFIX conditional on DEVELOPER_MODE
commit-queue@webkit.org [Tue, 6 Sep 2016 11:19:52 +0000 (11:19 +0000)]
[GTK][EFL] Make *_PROCESS_CMD_PREFIX conditional on DEVELOPER_MODE
https://bugs.webkit.org/show_bug.cgi?id=161621

Patch by Emanuele Aina <emanuele.aina@collabora.com> on 2016-09-06
Reviewed by Carlos Garcia Campos.

Instead of enabling support for the *_PROCESS_CMD_PREFIX set of
environment valiables (eg. WEB_PROCESS_CMD_PREFIX) on debug builds
only, enable it when DEVELOPER_MODE is enabled as it already guards
similar features (eg. WEBKIT_EXEC_PATH).

* UIProcess/ChildProcessProxy.cpp:
(WebKit::ChildProcessProxy::getLaunchOptions):
* UIProcess/Launcher/ProcessLauncher.h:
* UIProcess/Launcher/efl/ProcessLauncherEfl.cpp:
(WebKit::ProcessLauncher::launchProcess):
* UIProcess/Launcher/gtk/ProcessLauncherGtk.cpp:
(WebKit::ProcessLauncher::launchProcess):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205474 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoCachedResourceLoader is not taking into account fetch options to use or not cached...
commit-queue@webkit.org [Tue, 6 Sep 2016 11:06:52 +0000 (11:06 +0000)]
CachedResourceLoader is not taking into account fetch options to use or not cached resources
https://bugs.webkit.org/show_bug.cgi?id=161389

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-06
Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Updated as new console log messages appear now that cors checks are done at SubresourceLoader level.

* web-platform-tests/XMLHttpRequest/security-consideration.sub-expected.txt:
* web-platform-tests/fetch/api/cors/cors-basic-expected.txt:
* web-platform-tests/fetch/api/cors/cors-basic-worker-expected.txt:
* web-platform-tests/fetch/api/cors/cors-basic.js: Fixing a typo in the test making the test always passing, since the fetch promise was not taken into account.
* web-platform-tests/fetch/api/cors/cors-multiple-origins-expected.txt:
* web-platform-tests/fetch/api/cors/cors-multiple-origins-worker-expected.txt:
* web-platform-tests/fetch/api/cors/cors-origin-expected.txt:
* web-platform-tests/fetch/api/cors/cors-origin-worker-expected.txt:
* web-platform-tests/fetch/api/cors/cors-origin.js:
(corsOrigin): Fixing a typo in the test making the tests always passing, since the fetch promise was not taken into account.
* web-platform-tests/fetch/api/cors/cors-redirect-credentials-expected.txt:
* web-platform-tests/fetch/api/cors/cors-redirect-credentials-worker-expected.txt:

Source/WebCore:

Tests: http/tests/fetch/fetching-same-resource-with-diffferent-options.html
       http/tests/security/cross-origin-cached-resource-parallel.html
       http/tests/security/cross-origin-cached-resource.html
       http/tests/security/load-image-after-redirection-2.html
       http/tests/security/shape-outside-and-cached-resources.html

Adding CORS checks for the response in case of CORS fetch mode, in SubresourceLoader.
Removing the CORS checks in Image and DocumentThreadableLoader.

The direction of this patch is to make CachedResource origin-specific/fetch mode specific.

This will remove the need for CachedResource clients to do CORS checks when receiving the notifyFinished call.
This will also make the computation of whether a resource is clean or not much easier since the CachedResource knowd its origin and its response tainting.

Removing the CORS checks at ImageLoader creates the risk of using some cached resources loaded from previously no-cors mode without doing the actual CORS check.
Note that the risk was already there in case of a resource loaded through redirections.
Reusing a cached resource for a load with different options also leads to bad computation of the resource tainting.

As a first step, improvements are done but only for CachedImage resources.

This patch limits the direct reuse of cached resources as follow:
- If the request and existing resources have different origins.
- If the fetch mode is different between request and existing resource.

In those cases, a new CachedResource is created with the correct options and origin.
The data and response of the CachedResource found in the cache are copied efficiently in the new CachedResource, if the matching CachedResource finished loading (CachedImage specific).

If the matching CachedResource is still loading, we trigger a reload (with caching=false to not disturb the being loaded resource).
This should be made more efficient at some point, especially if the matching CachedResource already has its response set.

This triggers a change of behavior: previously, the CORS checks were done by the ImageLoader when the resource was finished loading.
The CORS checks were controlled by the crossOrigin attribute, which may be set or unset between the load start and the load end.

Now the crossOrigin attribute is checked at load start. If it is set, the CORS checks will happen even if the attribute is unset before the end of the load.
This is more consistent as the actual request was built with CORS enabled.

* loader/CrossOriginPreflightChecker.cpp:
(WebCore::CrossOriginPreflightChecker::startPreflight): Setting correctly the preflight options as per fetch spec.
* loader/DocumentThreadableLoader.cpp:
(WebCore::DocumentThreadableLoader::didReceiveResponse): Removing CORS check.
(WebCore::DocumentThreadableLoader::loadRequest): Adding CORS check in sync mode.
* loader/ImageLoader.cpp:
(WebCore::ImageLoader::updateFromElement):
(WebCore::ImageLoader::notifyFinished):
* loader/SubresourceLoader.cpp:
(WebCore::SubresourceLoader::didReceiveResponse): Adding CORS checks to the response
(WebCore::SubresourceLoader::checkResponseCrossOriginAccessControl): Helper routine to do CORS checks
* loader/SubresourceLoader.h:
* loader/cache/CachedImage.cpp:
(WebCore::CachedImage::cloneData): Responsible to set image content from another CachedImage.
* loader/cache/CachedImage.h:
* loader/cache/CachedResource.cpp:
(WebCore::CachedResource::computeOrigin): Helper routine to set the origin and whether the resource is cross-origin or not.
(WebCore::CachedResource::load): Using computeOrigin.
(WebCore::CachedResource::loadFrom): Loading from a CachedResource from the same type and which finished loading.
* loader/cache/CachedResource.h:
(WebCore::CachedResource::cloneData):
* loader/cache/CachedResourceLoader.cpp:
(WebCore::CachedResourceLoader::updateCachedResourceWithCurrentRequest): Helper routine responsible to adapt the CachedResource
that can be reused to the origin and options of a new request.
(WebCore::CachedResourceLoader::requestResource): Calling updateCachedResourceWithCurrentRequest before actually returning the resource.
(WebCore::CachedResourceLoader::determineRevalidationPolicy): Space clean-up.
* loader/cache/CachedResourceLoader.h:
* loader/cache/CachedResourceRequest.h:
(WebCore::CachedResourceRequest::setCachingPolicy):
* style/StylePendingResources.cpp:
(WebCore::Style::loadPendingImage): Allowing data URLs for ShapeOutside data.

LayoutTests:

Added specific expectations for fetch cors-origin* tests for mac-wk2 and ios-simulator-wk2 as these tests use
HTTPS, and the connection is refused.

* TestExpectations: Marking http/tests/security/contentSecurityPolicy/worker-csp-blocks-xhr-redirect-cross-origin.html as flaky.
* http/tests/eventsource/eventsource-cors-basic-expected.txt:
* http/tests/eventsource/eventsource-cors-with-credentials-expected.txt:
* http/tests/fetch/fetching-same-resource-with-diffferent-options-expected.txt: Added.
* http/tests/fetch/fetching-same-resource-with-diffferent-options.html: Added.
* http/tests/loading/cross-origin-XHR-willLoadRequest-expected.txt:
* http/tests/resources/download-json-with-delay.php:
* http/tests/resources/redirect.php:
* http/tests/security/cross-origin-cached-resource-expected.txt: Added.
* http/tests/security/cross-origin-cached-resource-parallel-expected.txt: Added.
* http/tests/security/cross-origin-cached-resource-parallel.html: Added.
* http/tests/security/cross-origin-cached-resource.html: Added.
* http/tests/security/img-with-failed-cors-check-fails-to-load-expected.txt:
* http/tests/security/load-image-after-redirection-2-expected.txt: Added.
* http/tests/security/load-image-after-redirection-2.html: Added.
* http/tests/security/resources/abe-allow-star.php:
* http/tests/security/resources/allow-if-origin.php: Added.
* http/tests/security/resources/cross-origin-cached-resource-iframe.html: Added.
* http/tests/security/resources/rgbalpha.png: Added.
* http/tests/security/shape-outside-and-cached-resources-expected.html: Added.
* http/tests/security/shape-outside-and-cached-resources.html: Added.
* http/tests/security/video-poster-cross-origin-crash-expected.txt:
* http/tests/security/video-poster-cross-origin-crash2-expected.txt:
* http/tests/xmlhttprequest/access-control-and-redirects-async-same-origin-expected.txt:
* http/tests/xmlhttprequest/access-control-repeated-failed-preflight-crash-expected.txt:
* http/tests/xmlhttprequest/cross-origin-no-authorization-expected.txt:
* http/tests/xmlhttprequest/cross-origin-no-credential-prompt-expected.txt:
* http/tests/xmlhttprequest/cross-site-denied-response-expected.txt:
* http/tests/xmlhttprequest/onerror-event-expected.txt:
* http/tests/xmlhttprequest/origin-whitelisting-https-expected.txt:
* http/tests/xmlhttprequest/origin-whitelisting-ip-addresses-with-subdomains-expected.txt:
* http/tests/xmlhttprequest/post-blob-content-type-async-expected.txt:
* http/tests/xmlhttprequest/redirect-cross-origin-2-expected.txt:
* http/tests/xmlhttprequest/redirect-cross-origin-expected.txt:
* http/tests/xmlhttprequest/simple-cross-origin-denied-events-expected.txt:
* http/tests/xmlhttprequest/simple-cross-origin-progress-events-expected.txt:
* http/tests/xmlhttprequest/xmlhttprequest-unsafe-redirect-expected.txt:
* platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-expected.txt:
* platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-worker-expected.txt:
* platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin-expected.txt: Added.
* platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin-worker-expected.txt: Added.
* platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-expected.txt:
* platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-worker-expected.txt:
* platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin-expected.txt: Added.
* platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin-worker-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205473 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Win] Crash logs have incorrect symbols.
pvollan@apple.com [Tue, 6 Sep 2016 09:48:18 +0000 (09:48 +0000)]
[Win] Crash logs have incorrect symbols.
https://bugs.webkit.org/show_bug.cgi?id=161603

Reviewed by Darin Adler.

Add build path to _NT_SYMBOL_PATH environment variable.

* Scripts/webkitpy/port/win.py:
(WinPort.setup_crash_log_saving):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205472 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed GTK gardening
philn@webkit.org [Tue, 6 Sep 2016 07:09:01 +0000 (07:09 +0000)]
Unreviewed GTK gardening

* platform/gtk/TestExpectations: Mark 2 mediastream tests as crashing

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205471 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFix ENABLE(GAMEPAD) build errors on non-Cocoa platforms
zandobersek@gmail.com [Tue, 6 Sep 2016 05:57:16 +0000 (05:57 +0000)]
Fix ENABLE(GAMEPAD) build errors on non-Cocoa platforms
https://bugs.webkit.org/show_bug.cgi?id=161596

Reviewed by Darin Adler.

This patch fixes build errors popping up when enabling Gamepad
feature on non-Cocoa platforms.

* CMakeLists.txt: Add the Source/WebCore/platform/gamepad/ directory to
WebKit2_INCLUDE_DIRECTORIES. Add GamepadData.cpp, UIGamepad.cpp files to the build.
* UIProcess/Gamepad/UIGamepadProvider.cpp: Don't include Cocoa-specific
HIDGamepadProvider.h file. Update methods for the UIGamepadProvider class.
(WebKit::UIGamepadProvider::platformWebPageProxyForGamepadInput):
(WebKit::UIGamepadProvider::platformStartMonitoringGamepads): Deleted.
(WebKit::UIGamepadProvider::platformStopMonitoringGamepads): Deleted.
(WebKit::UIGamepadProvider::platformGamepads): Deleted.
(WebKit::UIGamepadProvider::platformWebProcessProxyForGamepadInput): Deleted.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205470 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove some dead code in ViewGestureControllerMac
timothy_horton@apple.com [Tue, 6 Sep 2016 04:21:23 +0000 (04:21 +0000)]
Remove some dead code in ViewGestureControllerMac
https://bugs.webkit.org/show_bug.cgi?id=161616

Reviewed by Darin Adler.

* UIProcess/mac/ViewGestureControllerMac.mm:
(WebKit::ViewGestureController::endSwipeGesture):
All other LEGACY_SWIPE_SHADOW_STYLE code has been removed, except for this.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205469 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMore bindings improvements, particularly things not needed for JavaScript bindings
darin@apple.com [Tue, 6 Sep 2016 03:33:50 +0000 (03:33 +0000)]
More bindings improvements, particularly things not needed for JavaScript bindings
https://bugs.webkit.org/show_bug.cgi?id=161572

Reviewed by Sam Weinig.

Source/WebCore:

* WebCore.xcodeproj/project.pbxproj: Export a couple of files needed for legacy bindings.

* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::addCrossOriginWindowPropertyNames): Mark array const.
* bindings/js/JSLocationCustom.cpp:
(WebCore::addCrossOriginLocationPropertyNames): Ditto.

* dom/DOMImplementation.cpp:
(WebCore::addString): Deleted.
(WebCore::isSupportedSVG10Feature): Deleted.
(WebCore::isSupportedSVG11Feature): Deleted.
(WebCore::DOMImplementation::hasFeature): Deleted.

* dom/DOMImplementation.h: Changed hasFeature to take no arguments and to always
return true, as specified in the DOM spec.

* dom/DOMImplementation.idl: Removed the arguments to hasFeature.

* dom/Document.cpp:
(WebCore::Document::defaultCharsetForLegacyBindings): Renamed from
defaultCharsetForBindings, because this function is used only by the
non-JavaScript bindings.
* dom/Document.h: Updated for the above.

* dom/DocumentType.h: Removed entitiesForBindings,
notationsForBindings, and internalSubsetForBindings, all of which were
empty functions. The empty implementations are now in the legacy bindings.

* dom/Element.cpp:
(WebCore::Element::dispatchSimulatedClickForBindings): Deleted.
This code is now at the single call site, in HTMLElement.
* dom/Element.h: Updated for the above change.

* dom/Node.cpp:
(WebCore::Node::isSupportedForBindings): Deleted. This is only used in the
legacy bindings, and since it only returns a value other than "true" for
SVG features, it is now in the SVGTests class.
* dom/Node.h: Updated for the above change.

* html/HTMLElement.cpp:
(WebCore::HTMLElement::click): Call simulateClick with the appropriate
arguments rather than calling dispatchSimulatedClickForBindings.

* svg/SVGTests.cpp:
(WebCore::supportedSVGFeatures): Added. Moved here from DOMImplementation,
but also refactored.
(WebCore::SVGTests::isValid): Changed to use the function above, rather than
calling DOMImplementation::hasFeature. This change fixes a bug where feature names
that do not look like valid SVG feature names, names such as "xxx", were returning
true claiming that the feature was supported. This behavior was helpful in the
general DOMImplementation function, but harmful here. This bug was causing test
failures in some of our SVG tests.
(WebCore::SVGTests::hasFeatureForLegacyBindings): Added. Uses the set above to
implement the legacy behavior of hasFeature. It's better to have this in parallel
with the SVGTests::isValid function rather than having either depend on the other.

* svg/SVGTests.h: Added exported SVGTests::isValid function.

Source/WebKit/mac:

* DOM/DOMDOMImplementation.mm:
(-[DOMImplementation hasFeature:version:]): Call SVGTests::hasFeatureForLegacyBindings
instead of DOMImplementation::hasFeature. Also removed JSMainThreadNullState
since this does not do anything that can invoke JavaScript.

* DOM/DOMDocument.mm:
(-[DOMDocument defaultCharset]): Call defaultCharsetForLegacyBindings. Also removed
JSMainThreadNullState.

* DOM/DOMDocumentType.mm:
(-[DOMDocumentType entities]): Changed to simply return nil; same behavior as before,
but with no dependency on WebCore. Also removed JSMainThreadNullState.
(-[DOMDocumentType notations]): Ditto.
(-[DOMDocumentType internalSubset]): Ditto, except empty string rather than nil.

* DOM/DOMNode.mm:
(-[DOMNode isSupported:version:]): Call SVGTests::hasFeatureForLegacyBindings
instead of Node::isSupportedForBindings. Also removed JSMainThreadNullState.

Source/WebKit2:

* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDOMImplementation.cpp:
(webkit_dom_dom_implementation_has_feature): Call SVGTests::hasFeatureForLegacyBindings
instead of DOMImplementation::hasFeature. Also removed JSMainThreadNullState
since this does not do anything that can invoke JavaScript.

* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDeprecated.cpp:
(webkit_dom_document_get_default_charset): Call defaultCharsetForLegacyBindings.
Also removed JSMainThreadNullState.

* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMDocumentType.cpp:
(webkit_dom_document_type_get_entities): Changed to simply return nullptr; same behavior
as before, but with no dependency on WebCore. Also removed JSMainThreadNullState.
(webkit_dom_document_type_get_notations): Ditto.
(webkit_dom_document_type_get_internal_subset): Ditto.

* WebProcess/InjectedBundle/API/gtk/DOM/WebKitDOMNode.cpp:
(webkit_dom_node_is_supported): Call SVGTests::hasFeatureForLegacyBindings
instead of Node::isSupportedForBindings. Also removed JSMainThreadNullState.

LayoutTests:

* TestExpectations: Updated to expect success instead of failure for the
imported/mozilla/svg tests named conditions-01.svg, conditions-06.svg,
dynamic-conditions-01.svg, switch-01.svg, and
svg-integration/dynamic-conditions-outer-svg-03.xhtml. These are now all passing,
likely because of a progression due to correctly handling bogus feature strings.
Also, a comment in bug 139115 claims that requiredFeatures it going to be removed
from the SVG specifications and from other implementations. Updated to expect
possible failure for dynamic-conditions-08.svg and dynamic-conditions-12.svg. Before
these were reliably passing because of the incorrect handling of bogus feature strings.
Now, they are sometimes failing because the tests include required code that runs on
a timer, but no code to guarantee the test engine will wait for the timer to fire.

* platform/efl/TestExpectations: Removed now-redundant expectations that these
tests will pass; now expected on all platforms.

* platform/ios-simulator/svg/W3C-SVG-1.1/struct-dom-03-b-expected.txt:
* platform/ios-simulator/svg/W3C-SVG-1.1/struct-dom-04-b-expected.txt:
* platform/ios-simulator/svg/W3C-SVG-1.1/struct-dom-05-b-expected.txt:
* platform/mac/svg/W3C-SVG-1.1/struct-dom-03-b-expected.png:
* platform/mac/svg/W3C-SVG-1.1/struct-dom-03-b-expected.txt:
* platform/mac/svg/W3C-SVG-1.1/struct-dom-04-b-expected.png:
* platform/mac/svg/W3C-SVG-1.1/struct-dom-04-b-expected.txt:
* platform/mac/svg/W3C-SVG-1.1/struct-dom-05-b-expected.png:
* platform/mac/svg/W3C-SVG-1.1/struct-dom-05-b-expected.txt:
* svg/custom/svg-features-expected.txt:
Updated results for all these tests now that DOMImplementation.hasFeature always
returns true. The tests aren't super-valuable any more, but I don't see a good
reason to locally modify the imported tests from the W3C-SVG-1.1 suite, and
while the svg-features test isn't great, keeping it does provide a little test
coverage for the "always return true" behavior.

* platform/ios-simulator/svg/W3C-SVG-1.1-SE/types-dom-06-f-expected.txt:
* platform/mac/svg/W3C-SVG-1.1-SE/types-dom-06-f-expected.txt:
* platform/win/svg/W3C-SVG-1.1-SE/types-dom-06-f-expected.txt:
* svg/W3C-SVG-1.1-SE/types-dom-06-f-expected.txt:
Updated for progression due to correctly handling bogus feature strings.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205468 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, update stale variable name
mcatanzaro@igalia.com [Tue, 6 Sep 2016 03:16:22 +0000 (03:16 +0000)]
Unreviewed, update stale variable name

* UIProcess/API/gtk/WebKitWebViewBase.cpp:
(webkitWebViewBaseCreateWebPage):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205467 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoHeap::isMarked() should use concurrent lazy flipping
fpizlo@apple.com [Tue, 6 Sep 2016 02:39:11 +0000 (02:39 +0000)]
Heap::isMarked() should use concurrent lazy flipping
https://bugs.webkit.org/show_bug.cgi?id=161613

Reviewed by Michael Catanzaro.

I found out about this race condition via
https://bugs.webkit.org/show_bug.cgi?id=160125#c233.

The problem is that we use isMarked, and maybe even isLive, inside the concurrent mark
phase. So, they need to lazy-flip in a non-racy way.

* heap/HeapInlines.h:
(JSC::Heap::isLive):
(JSC::Heap::isMarked):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205466 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, reset generator test results after the butterflies.
fpizlo@apple.com [Tue, 6 Sep 2016 02:16:03 +0000 (02:16 +0000)]
Unreviewed, reset generator test results after the butterflies.

* Scripts/tests/builtins/expected/JavaScriptCore-Builtin.Promise-Combined.js-result:
* Scripts/tests/builtins/expected/JavaScriptCore-Builtin.Promise-Separate.js-result:
* Scripts/tests/builtins/expected/JavaScriptCore-Builtin.prototype-Combined.js-result:
* Scripts/tests/builtins/expected/JavaScriptCore-Builtin.prototype-Separate.js-result:
* Scripts/tests/builtins/expected/JavaScriptCore-BuiltinConstructor-Combined.js-result:
* Scripts/tests/builtins/expected/JavaScriptCore-BuiltinConstructor-Separate.js-result:
* Scripts/tests/builtins/expected/JavaScriptCore-InternalClashingNames-Combined.js-result:
* Scripts/tests/builtins/expected/WebCore-AnotherGuardedInternalBuiltin-Separate.js-result:
* Scripts/tests/builtins/expected/WebCore-ArbitraryConditionalGuard-Separate.js-result:
* Scripts/tests/builtins/expected/WebCore-GuardedBuiltin-Separate.js-result:
* Scripts/tests/builtins/expected/WebCore-GuardedInternalBuiltin-Separate.js-result:
* Scripts/tests/builtins/expected/WebCore-UnguardedBuiltin-Separate.js-result:
* Scripts/tests/builtins/expected/WebCore-xmlCasingTest-Separate.js-result:
* Scripts/tests/builtins/expected/WebCoreJSBuiltins.h-result:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205465 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r205450.
commit-queue@webkit.org [Tue, 6 Sep 2016 02:08:52 +0000 (02:08 +0000)]
Unreviewed, rolling out r205450.
https://bugs.webkit.org/show_bug.cgi?id=161614

Made tests flaky, causing false positives on EWS (Requested by
ap on #webkit).

Reverted changeset:

"CachedResourceLoader is not taking into account fetch options
to use or not cached resources"
https://bugs.webkit.org/show_bug.cgi?id=161389
http://trac.webkit.org/changeset/205450

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205464 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, fix cloop build.
fpizlo@apple.com [Tue, 6 Sep 2016 01:11:19 +0000 (01:11 +0000)]
Unreviewed, fix cloop build.

* bytecode/SuperSampler.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205463 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoButterflies should be allocated in Auxiliary MarkedSpace instead of CopiedSpace and...
fpizlo@apple.com [Tue, 6 Sep 2016 01:02:22 +0000 (01:02 +0000)]
Butterflies should be allocated in Auxiliary MarkedSpace instead of CopiedSpace and we should rewrite as much of the GC as needed to make this not a regression
https://bugs.webkit.org/show_bug.cgi?id=160125

Reviewed by Geoffrey Garen and Keith Miller.
JSTests:

Most of the things I did properly covered by existing tests, but I found some simple cases of
unshifting that had sketchy coverage.

* stress/array-storage-array-unshift.js: Added.
* stress/contiguous-array-unshift.js: Added.
* stress/double-array-unshift.js: Added.
* stress/int32-array-unshift.js: Added.

Source/bmalloc:

I needed to tryMemalign, so I added such a thing.

* bmalloc/Allocator.cpp:
(bmalloc::Allocator::allocate):
(bmalloc::Allocator::tryAllocate):
(bmalloc::Allocator::allocateImpl):
* bmalloc/Allocator.h:
* bmalloc/Cache.h:
(bmalloc::Cache::tryAllocate):
* bmalloc/bmalloc.h:
(bmalloc::api::tryMemalign):

Source/JavaScriptCore:

In order to make the GC concurrent (bug 149432), we would either need to enable concurrent
copying or we would need to not copy. Concurrent copying carries a 1-2% throughput overhead
from the barriers alone. Considering that MarkedSpace does a decent job of avoiding
fragmentation, it's unlikely that it's worth paying 1-2% throughput for copying. So, we want
to get rid of copied space. This change moves copied space's biggest client over to marked
space.

Moving butterflies to marked space means having them use the new Auxiliary HeapCell
allocation path. This is a fairly mechanical change, but it caused performance regressions
everywhere, so this change also fixes MarkedSpace's performance issues.

At a high level the mechanical changes are:

- We use AuxiliaryBarrier instead of CopyBarrier.

- We use tryAllocateAuxiliary instead of tryAllocateStorage. I got rid of the silly
  CheckedBoolean stuff, since it's so much more trouble than it's worth.

- The JITs have to emit inlined marked space allocations instead of inline copy space
  allocations.

- Everyone has to get used to zeroing their butterflies after allocation instead of relying
  on them being pre-zeroed by the GC. Copied space would zero things for you, while marked
  space doesn't.

That's about 1/3 of this change. But this led to performance problems, which I fixed with
optimizations that amounted to a major MarkedSpace rewrite:

- MarkedSpace always causes internal fragmentation for array allocations because the vector
  length we choose when we resize usually leads to a cell size that doesn't correspond to any
  size class. I got around this by making array allocations usually round up vectorLength to
  the maximum allowed by the size class that we would have allocated in. Also,
  ensureLengthSlow() and friends first make sure that the requested length can't just be
  fulfilled with the current allocation size. This safeguard means that not every array
  allocation has to do size class queries. For example, the fast path of new Array(length)
  never does any size class queries, under the assumption that (1) the speed gained from
  avoiding an ensureLengthSlow() call, which then just changes the vectorLength by doing the
  size class query, is too small to offset the speed lost by doing the query on every
  allocation and (2) new Array(length) is a pretty good hint that resizing is not very
  likely.

- Size classes in MarkedSpace were way too precise, which led to external fragmentation. This
  changes MarkedSpace size classes to use a linear progression for very small sizes followed
  by a geometric progression that naturally transitions to a hyperbolic progression. We want
  hyperbolic sizes when we get close to blockSize: for example the largest size we want is
  payloadSize / 2 rounded down, to ensure we get exactly two cells with minimal slop. The
  next size down should be payloadSize / 3 rounded down, and so on. After the last precise
  size (80 bytes), we proceed using a geometric progression, but round up each size to
  minimize slop at the end of the block. This naturally causes the geometric progression to
  turn hyperbolic for large sizes. The size class configuration happens at VM start-up, so
  it can be controlled with runtime options. I found that a base of 1.4 works pretty well.

- Large allocations caused massive internal fragmentation, since the smallest large
  allocation had to use exactly blockSize, and the largest small allocation used
  blockSize / 2. The next size up - the first large allocation size to require two blocks -
  also had 50% internal fragmentation. This is because we required large allocations to be
  blockSize aligned, so that MarkedBlock::blockFor() would work. I decided to rewrite all of
  that. Cells no longer have to be owned by a MarkedBlock. They can now alternatively be
  owned by a LargeAllocation. These two things are abstracted as CellContainer. You know that
  a cell is owned by a LargeAllocation if the MarkedBlock::atomSize / 2 bit is set.
  Basically, large allocations are deliberately misaligned by 8 bytes. This actually works
  out great since (1) typed arrays won't use large allocations anyway since they have their
  own malloc fallback and (2) large array butterflies already have a 8 byte header, which
  means that the 8 byte base misalignment aligns the large array payload on a 16 byte
  boundary. I took extreme care to make sure that the isLargeAllocation bit checks are as
  rare as possible; for example, ExecState::vm() skips the check because we know that callees
  must be small allocations. It's also possible to use template tricks to do one check for
  cell container kind, and then invoke a function specialized for MarkedBlock or a function
  specialized for LargeAllocation. LargeAllocation includes stubs for all MarkedBlock methods
  that get used from functions that are template-specialized like this. That's mostly to
  speed up the GC marking code. Most other code can use CellContainer API or HeapCell API
  directly. That's another thing: HeapCell, the common base of JSCell and auxiliary
  allocations, is now smart enough to do a lot of things for you, like HeapCell::vm(),
  HeapCell::heap(), HeapCell::isLargeAllocation(), and HeapCell::cellContainer(). The size
  cutoff for large allocations is runtime-configurable, so long as you don't choose something
  so small that callees end up large. I found that 400 bytes is roughly optimal. This means
  that the MarkedBlock size classes end up being:

  16, 32, 48, 64, 80, 112, 160, 224, 320

  The next size class would have been 432, but that's above the 400 byte cutoff. All of this
  is configurable with --sizeClassProgression and --largeAllocationCutoff. You can see what
  size classes you end up with by doing --dumpSizeClasses=true.

- Copied space uses 64KB blocks, while marked space used to use 16KB blocks. Allocating a lot
  of stuff in 16KB blocks was slower than allocating it in 64KB blocks because the GC had a
  lot of per-block overhead. I removed this overhead: It's now 2x faster to scan all
  MarkedBlocks because the list that contains the interesting meta-data is allocated on the
  side, for better locality during a sequential walk. It's no longer necessary to scan
  MarkedBlocks to find WeakSets, since the sets of WeakSets for eden scan and full scan are
  maintained on-the-fly. It's no longer necessary to scan all MarkedBlocks to clear mark
  bits because we now use versioned mark bits: to clear then, just increment the 64-bit
  heap version. It's no longer necessary to scan retired MarkedBlocks while allocating
  because marking retires them on-the-fly. It's no longer necessary to sort all blocks in
  the IncrementalSweeper's snapshot because blocks now know if they are in the snapshot. Put
  together, these optimizations allowed me to reduce block size to 16KB without losing much
  performance. There is some small perf loss on JetStream/splay, but not enough to hurt
  JetStream overall. I tried reducing block sizes further, to 4KB, since that is a
  progression on membuster. That's not possible yet, since there is still enough per-block
  overhead yet that such a reduction hurts JetStream too much. I filed a bug about improving
  this further: https://bugs.webkit.org/show_bug.cgi?id=161581.

- Even after all of that, copying butterflies was still faster because it allowed us to skip
  sweeping dead space. A good GC allocates over dead bytes without explicitly freeing them,
  so the GC pause is O(size of live), not O(size of live + dead). O(dead) is usually much
  larger than O(live), especially in an eden collection. Copying satisfies this premise while
  mark+sweep does not. So, I invented a new kind of allocator: bump'n'pop. Previously, our
  MarkedSpace allocator was a freelist pop. That's simple and easy to inline but requires
  that we walk the block to build a free list. This means walking dead space. The new
  allocator allows totally free MarkedBlocks to simply set up a bump-pointer arena instead.
  The allocator is a hybrid of bump-pointer and freelist pop. It tries bump first. The bump
  pointer always bumps by cellSize, so the result of filling a block with bumping looks as if
  we had used freelist popping to fill it. Additionally, each MarkedBlock now has a bit to
  quickly tell if the block is entirely free. This makes sweeping O(1) whenever a MarkedBlock
  is completely empty, which is the common case because of the generational hypothesis: the
  number of objects that survive an eden collection is a tiny fraction of the number of
  objects that had been allocated, and this fraction is so small that there are typically
  fewer than one survivors per MarkedBlock. This change was enough to make this change a net
  win over tip-of-tree.

- FTL now shares the same allocation fast paths as everything else, which is great, because
  bump'n'pop has gnarly control flow. We don't really want B3 to have to think about that
  control flow, since it won't be able to improve the machine code we write ourselves. GC
  fast paths are best written in assembly. So, I've empowered B3 to have even better support
  for Patchpoint terminals. It's now totally fine for a Patchpoint terminal to be non-Void.
  So, the new FTL allocation fast paths are just Patchpoint terminals that call through to
  AssemblyHelpers::emitAllocate(). B3 still reasons about things like constant-folding the
  size class calculation and constant-hoisting the allocator. Also, I gave the FTL the
  ability to constant-fold some allocator logic (in case we first assume that we're doing a
  variable-length allocation but then realize that the length is known). I think it makes
  sense to have constant folding rules in FTL::Output, or whatever the B3 IR builder is,
  since this makes lowering easier (you can constant fold during lowering more easily) and it
  reduces the amount of malloc traffic. In the future, we could teach B3 how to better
  constant-fold this code. That would require allowing loads to be constant-folded, which is
  doable but hella tricky.

- It used to be that if a logical object allocation required two physical allocations (first
  the butterfly and then the cell), then the JIT would emit the code in such a way that a
  failure in the second fast path would cause us to forget the successful first physical
  allocation. This was pointlessly wasteful. It turns out that it's very cheap to devote a
  register to storing either the butterfly or null, because the butterfly register is anyway
  going to be free inside the first allocation. The only overhead here is zeroing the
  butterfly register. With that in place, we can just pass the butterfly-or-null to the slow
  path, which can then either allocate a butterfly or not. So now we never waste a successful
  allocation. This patch implements such a solution both in DFG (where it's easy to do this
  since we control registers already) and in FTL (where it's annoying, because mutable
  "butterfly-or-null" variables are hard to say in SSA; also I realized that we had code
  duplicated the JSArray allocation utility, so I deduplicated it). This came up because in
  one version of this patch, this wastage would resonate with some Kraken benchmark: the
  benchmark would always allocate N small things followed by one bigger thing. The problem
  was I accidentally adjusted the various fixed overheads in MarkedBlock in such a way that
  the JSObject size class, which both the small and big thing shared for their cell, could
  hold exactly N cells per MarkedBlock. Then the benchmark would always call slow path when
  it allocated the big thing. So, it would end up having to allocate the big thing's large
  butterfly twice, every single time! Ouch!

- It used to be that we zeroed CopiedBlocks using memset, and so array allocations enjoyed
  amortization of the cost of zeroing. This doesn't work anymore - it's now up to the client
  of the allocator to initialize the object to whatever state they need. It used to be that
  we would just use a dumb loop. I initially changed this so that we would end up in memset
  for large allocations, but this didn't actually help performance that much. I got a much
  better result by playing with different memsets written in assembly. First I wrote one
  using non-temporal stores. That was a small speed-up over memset. Then I tried the classic
  "rep stos" approach, and holy cow that version was fast. It's a ~20% speed-up on array
  allocation microbenchmarks. So, this patch adds code paths to do "rep stos" on x86_64, or
  memset, or use a loop, as appropriate, for both "contiguous" arrays (holes are zero) and
  double arrays (holes are PNaN). Note that the JIT always emits either a loop or a flat slab
  of stores (if the size is known), but those paths in the JIT won't trigger for
  NewArrayWithSize() if the size is large, since that takes us to the
  operationNewArrayWithSize() slow path, which calls into JSArray::create(). That's why the
  optimizations here are all in JSArray::create() - that's the hot place for large arrays
  that need to be filled with holes.

All of this put together gives us neutral perf on JetStream,  membuster, and PLT3, a ~1%
regression on Speedometer, and up to a 4% regression Kraken. The Kraken regression is
because Kraken was allocating exactly 1024 element arrays at a rate of 400MB/sec. This is a
best-case scenario for bump allocation. I think that we should fix bmalloc to make up the
difference, but take the hit for now because it's a crazy corner case. By comparison, the
alternative approach of using a copy barrier would have cost us 1-2%. That's the real
apples-to-apples comparison if your premise is that we should have a concurrent GC. After we
finish removing copied space, we will be barrier-ready for concurrent GC: we already have a
marking barrier and we simply won't need a copying barrier. This change gets us there for
the purposes of our benchmarks, since the remaining clients of copied space are not very
important. On the other hand, if we keep copying, then getting barrier-ready would mean
adding back the copy barrier, which costs more perf.

We might get bigger speed-ups once we remove CopiedSpace altogether. That requires moving
typed arrays and a few other weird things over to Aux MarkedSpace.

This also includes some header sanitization. The introduction of AuxiliaryBarrier, HeapCell,
and CellContainer meant that I had to include those files from everywhere. Fortunately,
just including JSCInlines.h (instead of manually including the files that includes) is
usually enough. So, I made most of JSC's cpp files include JSCInlines.h, which is something
that we were already basically doing. In places where JSCInlines.h would be too much, I just
included HeapInlines.h. This got weird, because we previously included HeapInlines.h from
JSObject.h. That's bad because it led to some circular dependencies, so I fixed it - but that
meant having to manually include HeapInlines.h from the places that previously got it
implicitly via JSObject.h. But that led to more problems for some reason: I started getting
build errors because non-JSC files were having trouble including Opcode.h. That's just silly,
since Opcode.h is meant to be an internal JSC header. So, I made it an internal header and
made it impossible to include it from outside JSC. This was a lot of work, but it was
necessary to get the patch to build on all ports. It's also a net win. There were many places
in WebCore that were transitively including a *ton* of JSC headers just because of the
JSObject.h->HeapInlines.h edge and a bunch of dependency edges that arose from some public
(for WebCore) JSC headers needing Interpreter.h or Opcode.h for bad reasons.

* API/JSManagedValue.mm:
(-[JSManagedValue initWithValue:]):
* API/JSTypedArray.cpp:
* API/ObjCCallbackFunction.mm:
* API/tests/testapi.mm:
(testObjectiveCAPI):
(testWeakValue): Deleted.
* CMakeLists.txt:
* JavaScriptCore.xcodeproj/project.pbxproj:
* Scripts/builtins/builtins_generate_combined_implementation.py:
(BuiltinsCombinedImplementationGenerator.generate_secondary_header_includes):
* Scripts/builtins/builtins_generate_internals_wrapper_implementation.py:
(BuiltinsInternalsWrapperImplementationGenerator.generate_secondary_header_includes):
* Scripts/builtins/builtins_generate_separate_implementation.py:
(BuiltinsSeparateImplementationGenerator.generate_secondary_header_includes):
* assembler/AbstractMacroAssembler.h:
(JSC::AbstractMacroAssembler::JumpList::link):
(JSC::AbstractMacroAssembler::JumpList::linkTo):
* assembler/MacroAssembler.h:
* assembler/MacroAssemblerARM64.h:
(JSC::MacroAssemblerARM64::add32):
* assembler/MacroAssemblerCodeRef.cpp: Added.
(JSC::MacroAssemblerCodePtr::createLLIntCodePtr):
(JSC::MacroAssemblerCodePtr::dumpWithName):
(JSC::MacroAssemblerCodePtr::dump):
(JSC::MacroAssemblerCodeRef::createLLIntCodeRef):
(JSC::MacroAssemblerCodeRef::dump):
* assembler/MacroAssemblerCodeRef.h:
(JSC::MacroAssemblerCodePtr::createLLIntCodePtr): Deleted.
(JSC::MacroAssemblerCodePtr::dumpWithName): Deleted.
(JSC::MacroAssemblerCodePtr::dump): Deleted.
(JSC::MacroAssemblerCodeRef::createLLIntCodeRef): Deleted.
(JSC::MacroAssemblerCodeRef::dump): Deleted.
* b3/B3BasicBlock.cpp:
(JSC::B3::BasicBlock::appendBoolConstant):
* b3/B3BasicBlock.h:
* b3/B3DuplicateTails.cpp:
* b3/B3StackmapGenerationParams.h:
* b3/testb3.cpp:
(JSC::B3::testPatchpointTerminalReturnValue):
(JSC::B3::run):
* bindings/ScriptValue.cpp:
* bytecode/AdaptiveInferredPropertyValueWatchpointBase.cpp:
* bytecode/BytecodeBasicBlock.cpp:
* bytecode/BytecodeLivenessAnalysis.cpp:
* bytecode/BytecodeUseDef.h:
* bytecode/CallLinkInfo.cpp:
(JSC::CallLinkInfo::callTypeFor):
* bytecode/CallLinkInfo.h:
(JSC::CallLinkInfo::callTypeFor): Deleted.
* bytecode/CallLinkStatus.cpp:
* bytecode/CodeBlock.cpp:
(JSC::CodeBlock::finishCreation):
(JSC::CodeBlock::clearLLIntGetByIdCache):
(JSC::CodeBlock::predictedMachineCodeSize):
* bytecode/CodeBlock.h:
(JSC::CodeBlock::jitCodeMap): Deleted.
(JSC::clearLLIntGetByIdCache): Deleted.
* bytecode/ExecutionCounter.h:
* bytecode/Instruction.h:
* bytecode/LLIntPrototypeLoadAdaptiveStructureWatchpoint.cpp:
(JSC::LLIntPrototypeLoadAdaptiveStructureWatchpoint::fireInternal):
* bytecode/ObjectAllocationProfile.h:
(JSC::ObjectAllocationProfile::isNull):
(JSC::ObjectAllocationProfile::initialize):
* bytecode/Opcode.h:
(JSC::padOpcodeName):
* bytecode/PolymorphicAccess.cpp:
(JSC::AccessCase::generateImpl):
(JSC::PolymorphicAccess::regenerate):
* bytecode/PolymorphicAccess.h:
* bytecode/PreciseJumpTargets.cpp:
* bytecode/StructureStubInfo.cpp:
* bytecode/StructureStubInfo.h:
* bytecode/UnlinkedCodeBlock.cpp:
(JSC::UnlinkedCodeBlock::vm): Deleted.
* bytecode/UnlinkedCodeBlock.h:
* bytecode/UnlinkedInstructionStream.cpp:
* bytecode/UnlinkedInstructionStream.h:
* dfg/DFGOperations.cpp:
* dfg/DFGSpeculativeJIT.cpp:
(JSC::DFG::SpeculativeJIT::emitAllocateRawObject):
(JSC::DFG::SpeculativeJIT::compileMakeRope):
(JSC::DFG::SpeculativeJIT::compileAllocatePropertyStorage):
(JSC::DFG::SpeculativeJIT::compileReallocatePropertyStorage):
* dfg/DFGSpeculativeJIT.h:
(JSC::DFG::SpeculativeJIT::emitAllocateJSCell):
(JSC::DFG::SpeculativeJIT::emitAllocateJSObject):
* dfg/DFGSpeculativeJIT32_64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
(JSC::DFG::SpeculativeJIT::compileAllocateNewArrayWithSize):
* dfg/DFGSpeculativeJIT64.cpp:
(JSC::DFG::SpeculativeJIT::compile):
(JSC::DFG::SpeculativeJIT::compileAllocateNewArrayWithSize):
* dfg/DFGStrengthReductionPhase.cpp:
(JSC::DFG::StrengthReductionPhase::handleNode):
* ftl/FTLAbstractHeapRepository.h:
* ftl/FTLCompile.cpp:
* ftl/FTLJITFinalizer.cpp:
* ftl/FTLLowerDFGToB3.cpp:
(JSC::FTL::DFG::LowerDFGToB3::compileCreateDirectArguments):
(JSC::FTL::DFG::LowerDFGToB3::compileCreateRest):
(JSC::FTL::DFG::LowerDFGToB3::allocateArrayWithSize):
(JSC::FTL::DFG::LowerDFGToB3::compileNewArrayWithSize):
(JSC::FTL::DFG::LowerDFGToB3::compileMakeRope):
(JSC::FTL::DFG::LowerDFGToB3::compileMaterializeNewObject):
(JSC::FTL::DFG::LowerDFGToB3::initializeArrayElements):
(JSC::FTL::DFG::LowerDFGToB3::allocatePropertyStorageWithSizeImpl):
(JSC::FTL::DFG::LowerDFGToB3::allocateHeapCell):
(JSC::FTL::DFG::LowerDFGToB3::allocateCell):
(JSC::FTL::DFG::LowerDFGToB3::allocateObject):
(JSC::FTL::DFG::LowerDFGToB3::allocatorForSize):
(JSC::FTL::DFG::LowerDFGToB3::allocateVariableSizedObject):
(JSC::FTL::DFG::LowerDFGToB3::allocateJSArray):
(JSC::FTL::DFG::LowerDFGToB3::compileAllocateArrayWithSize): Deleted.
* ftl/FTLOutput.cpp:
(JSC::FTL::Output::constBool):
(JSC::FTL::Output::add):
(JSC::FTL::Output::shl):
(JSC::FTL::Output::aShr):
(JSC::FTL::Output::lShr):
(JSC::FTL::Output::zeroExt):
(JSC::FTL::Output::equal):
(JSC::FTL::Output::notEqual):
(JSC::FTL::Output::above):
(JSC::FTL::Output::aboveOrEqual):
(JSC::FTL::Output::below):
(JSC::FTL::Output::belowOrEqual):
(JSC::FTL::Output::greaterThan):
(JSC::FTL::Output::greaterThanOrEqual):
(JSC::FTL::Output::lessThan):
(JSC::FTL::Output::lessThanOrEqual):
(JSC::FTL::Output::select):
(JSC::FTL::Output::appendSuccessor):
(JSC::FTL::Output::addIncomingToPhi):
* ftl/FTLOutput.h:
* ftl/FTLValueFromBlock.h:
(JSC::FTL::ValueFromBlock::operator bool):
(JSC::FTL::ValueFromBlock::ValueFromBlock): Deleted.
* ftl/FTLWeightedTarget.h:
(JSC::FTL::WeightedTarget::frequentedBlock):
* heap/CellContainer.h: Added.
(JSC::CellContainer::CellContainer):
(JSC::CellContainer::operator bool):
(JSC::CellContainer::isMarkedBlock):
(JSC::CellContainer::isLargeAllocation):
(JSC::CellContainer::markedBlock):
(JSC::CellContainer::largeAllocation):
* heap/CellContainerInlines.h: Added.
(JSC::CellContainer::isMarked):
(JSC::CellContainer::isMarkedOrNewlyAllocated):
(JSC::CellContainer::noteMarked):
(JSC::CellContainer::cellSize):
(JSC::CellContainer::weakSet):
(JSC::CellContainer::flipIfNecessary):
* heap/ConservativeRoots.cpp:
(JSC::ConservativeRoots::ConservativeRoots):
(JSC::ConservativeRoots::~ConservativeRoots):
(JSC::ConservativeRoots::grow):
(JSC::ConservativeRoots::genericAddPointer):
(JSC::ConservativeRoots::genericAddSpan):
* heap/ConservativeRoots.h:
(JSC::ConservativeRoots::roots):
* heap/CopyToken.h:
* heap/FreeList.cpp: Added.
(JSC::FreeList::dump):
* heap/FreeList.h: Added.
(JSC::FreeList::FreeList):
(JSC::FreeList::list):
(JSC::FreeList::bump):
(JSC::FreeList::operator==):
(JSC::FreeList::operator!=):
(JSC::FreeList::operator bool):
(JSC::FreeList::allocationWillFail):
(JSC::FreeList::allocationWillSucceed):
* heap/GCTypeMap.h: Added.
(JSC::GCTypeMap::operator[]):
* heap/Heap.cpp:
(JSC::Heap::Heap):
(JSC::Heap::lastChanceToFinalize):
(JSC::Heap::finalizeUnconditionalFinalizers):
(JSC::Heap::markRoots):
(JSC::Heap::copyBackingStores):
(JSC::Heap::gatherStackRoots):
(JSC::Heap::gatherJSStackRoots):
(JSC::Heap::gatherScratchBufferRoots):
(JSC::Heap::clearLivenessData):
(JSC::Heap::visitSmallStrings):
(JSC::Heap::visitConservativeRoots):
(JSC::Heap::removeDeadCompilerWorklistEntries):
(JSC::Heap::gatherExtraHeapSnapshotData):
(JSC::Heap::removeDeadHeapSnapshotNodes):
(JSC::Heap::visitProtectedObjects):
(JSC::Heap::visitArgumentBuffers):
(JSC::Heap::visitException):
(JSC::Heap::visitStrongHandles):
(JSC::Heap::visitHandleStack):
(JSC::Heap::visitSamplingProfiler):
(JSC::Heap::traceCodeBlocksAndJITStubRoutines):
(JSC::Heap::converge):
(JSC::Heap::visitWeakHandles):
(JSC::Heap::updateObjectCounts):
(JSC::Heap::clearUnmarkedExecutables):
(JSC::Heap::deleteUnmarkedCompiledCode):
(JSC::Heap::collectAllGarbage):
(JSC::Heap::collect):
(JSC::Heap::collectWithoutAnySweep):
(JSC::Heap::collectImpl):
(JSC::Heap::suspendCompilerThreads):
(JSC::Heap::willStartCollection):
(JSC::Heap::flushOldStructureIDTables):
(JSC::Heap::flushWriteBarrierBuffer):
(JSC::Heap::stopAllocation):
(JSC::Heap::prepareForMarking):
(JSC::Heap::reapWeakHandles):
(JSC::Heap::pruneStaleEntriesFromWeakGCMaps):
(JSC::Heap::sweepArrayBuffers):
(JSC::MarkedBlockSnapshotFunctor::MarkedBlockSnapshotFunctor):
(JSC::MarkedBlockSnapshotFunctor::operator()):
(JSC::Heap::snapshotMarkedSpace):
(JSC::Heap::deleteSourceProviderCaches):
(JSC::Heap::notifyIncrementalSweeper):
(JSC::Heap::writeBarrierCurrentlyExecutingCodeBlocks):
(JSC::Heap::resetAllocators):
(JSC::Heap::updateAllocationLimits):
(JSC::Heap::didFinishCollection):
(JSC::Heap::resumeCompilerThreads):
(JSC::Zombify::visit):
(JSC::Heap::forEachCodeBlockImpl):
* heap/Heap.h:
(JSC::Heap::allocatorForObjectWithoutDestructor):
(JSC::Heap::allocatorForObjectWithDestructor):
(JSC::Heap::allocatorForAuxiliaryData):
(JSC::Heap::jitStubRoutines):
(JSC::Heap::codeBlockSet):
(JSC::Heap::storageAllocator): Deleted.
* heap/HeapCell.h:
(JSC::HeapCell::isZapped): Deleted.
* heap/HeapCellInlines.h: Added.
(JSC::HeapCell::isLargeAllocation):
(JSC::HeapCell::cellContainer):
(JSC::HeapCell::markedBlock):
(JSC::HeapCell::largeAllocation):
(JSC::HeapCell::heap):
(JSC::HeapCell::vm):
(JSC::HeapCell::cellSize):
(JSC::HeapCell::allocatorAttributes):
(JSC::HeapCell::destructionMode):
(JSC::HeapCell::cellKind):
* heap/HeapInlines.h:
(JSC::Heap::heap):
(JSC::Heap::isLive):
(JSC::Heap::isMarked):
(JSC::Heap::testAndSetMarked):
(JSC::Heap::setMarked):
(JSC::Heap::cellSize):
(JSC::Heap::forEachCodeBlock):
(JSC::Heap::allocateObjectOfType):
(JSC::Heap::subspaceForObjectOfType):
(JSC::Heap::allocatorForObjectOfType):
(JSC::Heap::allocateAuxiliary):
(JSC::Heap::tryAllocateAuxiliary):
(JSC::Heap::tryReallocateAuxiliary):
(JSC::Heap::isPointerGCObject): Deleted.
(JSC::Heap::isValueGCObject): Deleted.
* heap/HeapOperation.cpp: Added.
(WTF::printInternal):
* heap/HeapOperation.h:
* heap/HeapUtil.h: Added.
(JSC::HeapUtil::findGCObjectPointersForMarking):
(JSC::HeapUtil::isPointerGCObjectJSCell):
(JSC::HeapUtil::isValueGCObject):
* heap/IncrementalSweeper.cpp:
(JSC::IncrementalSweeper::sweepNextBlock):
* heap/IncrementalSweeper.h:
* heap/LargeAllocation.cpp: Added.
(JSC::LargeAllocation::tryCreate):
(JSC::LargeAllocation::LargeAllocation):
(JSC::LargeAllocation::lastChanceToFinalize):
(JSC::LargeAllocation::shrink):
(JSC::LargeAllocation::visitWeakSet):
(JSC::LargeAllocation::reapWeakSet):
(JSC::LargeAllocation::flip):
(JSC::LargeAllocation::isEmpty):
(JSC::LargeAllocation::sweep):
(JSC::LargeAllocation::destroy):
(JSC::LargeAllocation::dump):
* heap/LargeAllocation.h: Added.
(JSC::LargeAllocation::fromCell):
(JSC::LargeAllocation::cell):
(JSC::LargeAllocation::isLargeAllocation):
(JSC::LargeAllocation::heap):
(JSC::LargeAllocation::vm):
(JSC::LargeAllocation::weakSet):
(JSC::LargeAllocation::clearNewlyAllocated):
(JSC::LargeAllocation::isNewlyAllocated):
(JSC::LargeAllocation::isMarked):
(JSC::LargeAllocation::isMarkedOrNewlyAllocated):
(JSC::LargeAllocation::isLive):
(JSC::LargeAllocation::hasValidCell):
(JSC::LargeAllocation::cellSize):
(JSC::LargeAllocation::aboveLowerBound):
(JSC::LargeAllocation::belowUpperBound):
(JSC::LargeAllocation::contains):
(JSC::LargeAllocation::attributes):
(JSC::LargeAllocation::flipIfNecessary):
(JSC::LargeAllocation::flipIfNecessaryConcurrently):
(JSC::LargeAllocation::testAndSetMarked):
(JSC::LargeAllocation::setMarked):
(JSC::LargeAllocation::clearMarked):
(JSC::LargeAllocation::noteMarked):
(JSC::LargeAllocation::headerSize):
* heap/MarkedAllocator.cpp:
(JSC::MarkedAllocator::MarkedAllocator):
(JSC::MarkedAllocator::isPagedOut):
(JSC::MarkedAllocator::retire):
(JSC::MarkedAllocator::filterNextBlock):
(JSC::MarkedAllocator::setNextBlockToSweep):
(JSC::MarkedAllocator::tryAllocateWithoutCollectingImpl):
(JSC::MarkedAllocator::tryAllocateWithoutCollecting):
(JSC::MarkedAllocator::allocateSlowCase):
(JSC::MarkedAllocator::tryAllocateSlowCase):
(JSC::MarkedAllocator::allocateSlowCaseImpl):
(JSC::blockHeaderSize):
(JSC::MarkedAllocator::blockSizeForBytes):
(JSC::MarkedAllocator::tryAllocateBlock):
(JSC::MarkedAllocator::addBlock):
(JSC::MarkedAllocator::removeBlock):
(JSC::MarkedAllocator::stopAllocating):
(JSC::MarkedAllocator::reset):
(JSC::MarkedAllocator::lastChanceToFinalize):
(JSC::MarkedAllocator::setFreeList):
(JSC::isListPagedOut): Deleted.
(JSC::MarkedAllocator::tryAllocateHelper): Deleted.
(JSC::MarkedAllocator::tryPopFreeList): Deleted.
(JSC::MarkedAllocator::tryAllocate): Deleted.
(JSC::MarkedAllocator::allocateBlock): Deleted.
* heap/MarkedAllocator.h:
(JSC::MarkedAllocator::takeLastActiveBlock):
(JSC::MarkedAllocator::offsetOfFreeList):
(JSC::MarkedAllocator::offsetOfCellSize):
(JSC::MarkedAllocator::tryAllocate):
(JSC::MarkedAllocator::allocate):
(JSC::MarkedAllocator::forEachBlock):
(JSC::MarkedAllocator::offsetOfFreeListHead): Deleted.
(JSC::MarkedAllocator::MarkedAllocator): Deleted.
(JSC::MarkedAllocator::init): Deleted.
(JSC::MarkedAllocator::stopAllocating): Deleted.
* heap/MarkedBlock.cpp:
(JSC::MarkedBlock::tryCreate):
(JSC::MarkedBlock::Handle::Handle):
(JSC::MarkedBlock::Handle::~Handle):
(JSC::MarkedBlock::MarkedBlock):
(JSC::MarkedBlock::Handle::specializedSweep):
(JSC::MarkedBlock::Handle::sweep):
(JSC::MarkedBlock::Handle::sweepHelperSelectScribbleMode):
(JSC::MarkedBlock::Handle::sweepHelperSelectStateAndSweepMode):
(JSC::MarkedBlock::Handle::unsweepWithNoNewlyAllocated):
(JSC::SetNewlyAllocatedFunctor::SetNewlyAllocatedFunctor):
(JSC::SetNewlyAllocatedFunctor::operator()):
(JSC::MarkedBlock::Handle::stopAllocating):
(JSC::MarkedBlock::Handle::lastChanceToFinalize):
(JSC::MarkedBlock::Handle::resumeAllocating):
(JSC::MarkedBlock::Handle::zap):
(JSC::MarkedBlock::Handle::forEachFreeCell):
(JSC::MarkedBlock::flipIfNecessary):
(JSC::MarkedBlock::Handle::flipIfNecessary):
(JSC::MarkedBlock::flipIfNecessarySlow):
(JSC::MarkedBlock::flipIfNecessaryConcurrentlySlow):
(JSC::MarkedBlock::clearMarks):
(JSC::MarkedBlock::assertFlipped):
(JSC::MarkedBlock::needsFlip):
(JSC::MarkedBlock::Handle::needsFlip):
(JSC::MarkedBlock::Handle::willRemoveBlock):
(JSC::MarkedBlock::Handle::didConsumeFreeList):
(JSC::MarkedBlock::markCount):
(JSC::MarkedBlock::Handle::isEmpty):
(JSC::MarkedBlock::clearHasAnyMarked):
(JSC::MarkedBlock::noteMarkedSlow):
(WTF::printInternal):
(JSC::MarkedBlock::create): Deleted.
(JSC::MarkedBlock::destroy): Deleted.
(JSC::MarkedBlock::callDestructor): Deleted.
(JSC::MarkedBlock::specializedSweep): Deleted.
(JSC::MarkedBlock::sweep): Deleted.
(JSC::MarkedBlock::sweepHelper): Deleted.
(JSC::MarkedBlock::stopAllocating): Deleted.
(JSC::MarkedBlock::clearMarksWithCollectionType): Deleted.
(JSC::MarkedBlock::lastChanceToFinalize): Deleted.
(JSC::MarkedBlock::resumeAllocating): Deleted.
(JSC::MarkedBlock::didRetireBlock): Deleted.
* heap/MarkedBlock.h:
(JSC::MarkedBlock::VoidFunctor::returnValue):
(JSC::MarkedBlock::CountFunctor::CountFunctor):
(JSC::MarkedBlock::CountFunctor::count):
(JSC::MarkedBlock::CountFunctor::returnValue):
(JSC::MarkedBlock::Handle::hasAnyNewlyAllocated):
(JSC::MarkedBlock::Handle::isOnBlocksToSweep):
(JSC::MarkedBlock::Handle::setIsOnBlocksToSweep):
(JSC::MarkedBlock::Handle::state):
(JSC::MarkedBlock::needsDestruction):
(JSC::MarkedBlock::handle):
(JSC::MarkedBlock::Handle::block):
(JSC::MarkedBlock::firstAtom):
(JSC::MarkedBlock::atoms):
(JSC::MarkedBlock::isAtomAligned):
(JSC::MarkedBlock::Handle::cellAlign):
(JSC::MarkedBlock::blockFor):
(JSC::MarkedBlock::Handle::allocator):
(JSC::MarkedBlock::Handle::heap):
(JSC::MarkedBlock::Handle::vm):
(JSC::MarkedBlock::vm):
(JSC::MarkedBlock::Handle::weakSet):
(JSC::MarkedBlock::weakSet):
(JSC::MarkedBlock::Handle::shrink):
(JSC::MarkedBlock::Handle::visitWeakSet):
(JSC::MarkedBlock::Handle::reapWeakSet):
(JSC::MarkedBlock::Handle::cellSize):
(JSC::MarkedBlock::cellSize):
(JSC::MarkedBlock::Handle::attributes):
(JSC::MarkedBlock::attributes):
(JSC::MarkedBlock::Handle::needsDestruction):
(JSC::MarkedBlock::Handle::destruction):
(JSC::MarkedBlock::Handle::cellKind):
(JSC::MarkedBlock::Handle::markCount):
(JSC::MarkedBlock::Handle::size):
(JSC::MarkedBlock::atomNumber):
(JSC::MarkedBlock::flipIfNecessary):
(JSC::MarkedBlock::flipIfNecessaryConcurrently):
(JSC::MarkedBlock::Handle::flipIfNecessary):
(JSC::MarkedBlock::Handle::flipIfNecessaryConcurrently):
(JSC::MarkedBlock::Handle::flipForEdenCollection):
(JSC::MarkedBlock::assertFlipped):
(JSC::MarkedBlock::Handle::assertFlipped):
(JSC::MarkedBlock::isMarked):
(JSC::MarkedBlock::testAndSetMarked):
(JSC::MarkedBlock::Handle::isNewlyAllocated):
(JSC::MarkedBlock::Handle::setNewlyAllocated):
(JSC::MarkedBlock::Handle::clearNewlyAllocated):
(JSC::MarkedBlock::Handle::isMarkedOrNewlyAllocated):
(JSC::MarkedBlock::isMarkedOrNewlyAllocated):
(JSC::MarkedBlock::Handle::isLive):
(JSC::MarkedBlock::isAtom):
(JSC::MarkedBlock::Handle::isLiveCell):
(JSC::MarkedBlock::Handle::forEachCell):
(JSC::MarkedBlock::Handle::forEachLiveCell):
(JSC::MarkedBlock::Handle::forEachDeadCell):
(JSC::MarkedBlock::Handle::needsSweeping):
(JSC::MarkedBlock::Handle::isAllocated):
(JSC::MarkedBlock::Handle::isMarked):
(JSC::MarkedBlock::Handle::isFreeListed):
(JSC::MarkedBlock::hasAnyMarked):
(JSC::MarkedBlock::noteMarked):
(WTF::MarkedBlockHash::hash):
(JSC::MarkedBlock::FreeList::FreeList): Deleted.
(JSC::MarkedBlock::allocator): Deleted.
(JSC::MarkedBlock::heap): Deleted.
(JSC::MarkedBlock::shrink): Deleted.
(JSC::MarkedBlock::visitWeakSet): Deleted.
(JSC::MarkedBlock::reapWeakSet): Deleted.
(JSC::MarkedBlock::willRemoveBlock): Deleted.
(JSC::MarkedBlock::didConsumeFreeList): Deleted.
(JSC::MarkedBlock::markCount): Deleted.
(JSC::MarkedBlock::isEmpty): Deleted.
(JSC::MarkedBlock::destruction): Deleted.
(JSC::MarkedBlock::cellKind): Deleted.
(JSC::MarkedBlock::size): Deleted.
(JSC::MarkedBlock::capacity): Deleted.
(JSC::MarkedBlock::setMarked): Deleted.
(JSC::MarkedBlock::clearMarked): Deleted.
(JSC::MarkedBlock::isNewlyAllocated): Deleted.
(JSC::MarkedBlock::setNewlyAllocated): Deleted.
(JSC::MarkedBlock::clearNewlyAllocated): Deleted.
(JSC::MarkedBlock::isLive): Deleted.
(JSC::MarkedBlock::isLiveCell): Deleted.
(JSC::MarkedBlock::forEachCell): Deleted.
(JSC::MarkedBlock::forEachLiveCell): Deleted.
(JSC::MarkedBlock::forEachDeadCell): Deleted.
(JSC::MarkedBlock::needsSweeping): Deleted.
(JSC::MarkedBlock::isAllocated): Deleted.
(JSC::MarkedBlock::isMarkedOrRetired): Deleted.
* heap/MarkedSpace.cpp:
(JSC::MarkedSpace::initializeSizeClassForStepSize):
(JSC::MarkedSpace::MarkedSpace):
(JSC::MarkedSpace::~MarkedSpace):
(JSC::MarkedSpace::lastChanceToFinalize):
(JSC::MarkedSpace::allocate):
(JSC::MarkedSpace::tryAllocate):
(JSC::MarkedSpace::allocateLarge):
(JSC::MarkedSpace::tryAllocateLarge):
(JSC::MarkedSpace::sweep):
(JSC::MarkedSpace::sweepLargeAllocations):
(JSC::MarkedSpace::zombifySweep):
(JSC::MarkedSpace::resetAllocators):
(JSC::MarkedSpace::visitWeakSets):
(JSC::MarkedSpace::reapWeakSets):
(JSC::MarkedSpace::stopAllocating):
(JSC::MarkedSpace::prepareForMarking):
(JSC::MarkedSpace::resumeAllocating):
(JSC::MarkedSpace::isPagedOut):
(JSC::MarkedSpace::freeBlock):
(JSC::MarkedSpace::freeOrShrinkBlock):
(JSC::MarkedSpace::shrink):
(JSC::MarkedSpace::clearNewlyAllocated):
(JSC::VerifyMarked::operator()):
(JSC::MarkedSpace::flip):
(JSC::MarkedSpace::objectCount):
(JSC::MarkedSpace::size):
(JSC::MarkedSpace::capacity):
(JSC::MarkedSpace::addActiveWeakSet):
(JSC::MarkedSpace::didAddBlock):
(JSC::MarkedSpace::didAllocateInBlock):
(JSC::MarkedSpace::forEachAllocator): Deleted.
(JSC::VerifyMarkedOrRetired::operator()): Deleted.
(JSC::MarkedSpace::clearMarks): Deleted.
* heap/MarkedSpace.h:
(JSC::MarkedSpace::sizeClassToIndex):
(JSC::MarkedSpace::indexToSizeClass):
(JSC::MarkedSpace::version):
(JSC::MarkedSpace::blocksWithNewObjects):
(JSC::MarkedSpace::largeAllocations):
(JSC::MarkedSpace::largeAllocationsNurseryOffset):
(JSC::MarkedSpace::largeAllocationsOffsetForThisCollection):
(JSC::MarkedSpace::largeAllocationsForThisCollectionBegin):
(JSC::MarkedSpace::largeAllocationsForThisCollectionEnd):
(JSC::MarkedSpace::largeAllocationsForThisCollectionSize):
(JSC::MarkedSpace::forEachLiveCell):
(JSC::MarkedSpace::forEachDeadCell):
(JSC::MarkedSpace::allocatorFor):
(JSC::MarkedSpace::destructorAllocatorFor):
(JSC::MarkedSpace::auxiliaryAllocatorFor):
(JSC::MarkedSpace::allocateWithoutDestructor):
(JSC::MarkedSpace::allocateWithDestructor):
(JSC::MarkedSpace::allocateAuxiliary):
(JSC::MarkedSpace::tryAllocateAuxiliary):
(JSC::MarkedSpace::forEachBlock):
(JSC::MarkedSpace::forEachAllocator):
(JSC::MarkedSpace::optimalSizeFor):
(JSC::MarkedSpace::didAddBlock): Deleted.
(JSC::MarkedSpace::didAllocateInBlock): Deleted.
(JSC::MarkedSpace::objectCount): Deleted.
(JSC::MarkedSpace::size): Deleted.
(JSC::MarkedSpace::capacity): Deleted.
* heap/SlotVisitor.cpp:
(JSC::SlotVisitor::SlotVisitor):
(JSC::SlotVisitor::didStartMarking):
(JSC::SlotVisitor::reset):
(JSC::SlotVisitor::append):
(JSC::SlotVisitor::appendJSCellOrAuxiliary):
(JSC::SlotVisitor::setMarkedAndAppendToMarkStack):
(JSC::SlotVisitor::appendToMarkStack):
(JSC::SlotVisitor::markAuxiliary):
(JSC::SlotVisitor::noteLiveAuxiliaryCell):
(JSC::SlotVisitor::visitChildren):
* heap/SlotVisitor.h:
* heap/WeakBlock.cpp:
(JSC::WeakBlock::create):
(JSC::WeakBlock::WeakBlock):
(JSC::WeakBlock::visit):
(JSC::WeakBlock::reap):
* heap/WeakBlock.h:
(JSC::WeakBlock::disconnectContainer):
(JSC::WeakBlock::disconnectMarkedBlock): Deleted.
* heap/WeakSet.cpp:
(JSC::WeakSet::~WeakSet):
(JSC::WeakSet::sweep):
(JSC::WeakSet::shrink):
(JSC::WeakSet::addAllocator):
* heap/WeakSet.h:
(JSC::WeakSet::container):
(JSC::WeakSet::setContainer):
(JSC::WeakSet::WeakSet):
(JSC::WeakSet::visit):
(JSC::WeakSet::shrink): Deleted.
* heap/WeakSetInlines.h:
(JSC::WeakSet::allocate):
* inspector/InjectedScriptManager.cpp:
* inspector/JSGlobalObjectInspectorController.cpp:
* inspector/JSJavaScriptCallFrame.cpp:
* inspector/ScriptDebugServer.cpp:
* inspector/agents/InspectorDebuggerAgent.cpp:
* interpreter/CachedCall.h:
(JSC::CachedCall::CachedCall):
* interpreter/Interpreter.cpp:
(JSC::loadVarargs):
(JSC::StackFrame::sourceID): Deleted.
(JSC::StackFrame::sourceURL): Deleted.
(JSC::StackFrame::functionName): Deleted.
(JSC::StackFrame::computeLineAndColumn): Deleted.
(JSC::StackFrame::toString): Deleted.
* interpreter/Interpreter.h:
(JSC::StackFrame::isNative): Deleted.
* jit/AssemblyHelpers.h:
(JSC::AssemblyHelpers::emitAllocateWithNonNullAllocator):
(JSC::AssemblyHelpers::emitAllocate):
(JSC::AssemblyHelpers::emitAllocateJSCell):
(JSC::AssemblyHelpers::emitAllocateJSObject):
(JSC::AssemblyHelpers::emitAllocateJSObjectWithKnownSize):
(JSC::AssemblyHelpers::emitAllocateVariableSized):
* jit/GCAwareJITStubRoutine.cpp:
(JSC::GCAwareJITStubRoutine::GCAwareJITStubRoutine):
* jit/JIT.cpp:
(JSC::JIT::compileCTINativeCall):
(JSC::JIT::link):
* jit/JIT.h:
(JSC::JIT::compileCTINativeCall): Deleted.
* jit/JITExceptions.cpp:
(JSC::genericUnwind):
* jit/JITExceptions.h:
* jit/JITOpcodes.cpp:
(JSC::JIT::emit_op_new_object):
(JSC::JIT::emitSlow_op_new_object):
(JSC::JIT::emit_op_create_this):
(JSC::JIT::emitSlow_op_create_this):
* jit/JITOpcodes32_64.cpp:
(JSC::JIT::emit_op_new_object):
(JSC::JIT::emitSlow_op_new_object):
(JSC::JIT::emit_op_create_this):
(JSC::JIT::emitSlow_op_create_this):
* jit/JITOperations.cpp:
* jit/JITOperations.h:
* jit/JITPropertyAccess.cpp:
(JSC::JIT::emitWriteBarrier):
* jit/JITThunks.cpp:
* jit/JITThunks.h:
* jsc.cpp:
(functionDescribeArray):
(main):
* llint/LLIntData.cpp:
(JSC::LLInt::Data::performAssertions):
* llint/LLIntExceptions.cpp:
* llint/LLIntThunks.cpp:
* llint/LLIntThunks.h:
* llint/LowLevelInterpreter.asm:
* llint/LowLevelInterpreter.cpp:
* llint/LowLevelInterpreter32_64.asm:
* llint/LowLevelInterpreter64.asm:
* parser/ModuleAnalyzer.cpp:
* parser/NodeConstructors.h:
* parser/Nodes.h:
* profiler/ProfilerBytecode.cpp:
* profiler/ProfilerBytecode.h:
* profiler/ProfilerBytecodeSequence.cpp:
* runtime/ArrayConventions.h:
(JSC::indexingHeaderForArrayStorage):
(JSC::baseIndexingHeaderForArrayStorage):
(JSC::indexingHeaderForArray): Deleted.
(JSC::baseIndexingHeaderForArray): Deleted.
* runtime/ArrayPrototype.cpp:
(JSC::arrayProtoFuncSplice):
(JSC::concatAppendOne):
(JSC::arrayProtoPrivateFuncConcatMemcpy):
* runtime/ArrayStorage.h:
(JSC::ArrayStorage::vectorLength):
(JSC::ArrayStorage::totalSizeFor):
(JSC::ArrayStorage::totalSize):
(JSC::ArrayStorage::availableVectorLength):
(JSC::ArrayStorage::optimalVectorLength):
(JSC::ArrayStorage::sizeFor): Deleted.
* runtime/AuxiliaryBarrier.h: Added.
(JSC::AuxiliaryBarrier::AuxiliaryBarrier):
(JSC::AuxiliaryBarrier::clear):
(JSC::AuxiliaryBarrier::get):
(JSC::AuxiliaryBarrier::slot):
(JSC::AuxiliaryBarrier::operator bool):
(JSC::AuxiliaryBarrier::setWithoutBarrier):
* runtime/AuxiliaryBarrierInlines.h: Added.
(JSC::AuxiliaryBarrier<T>::AuxiliaryBarrier):
(JSC::AuxiliaryBarrier<T>::set):
* runtime/Butterfly.h:
* runtime/ButterflyInlines.h:
(JSC::Butterfly::availableContiguousVectorLength):
(JSC::Butterfly::optimalContiguousVectorLength):
(JSC::Butterfly::createUninitialized):
(JSC::Butterfly::growArrayRight):
* runtime/ClonedArguments.cpp:
(JSC::ClonedArguments::createEmpty):
* runtime/CommonSlowPathsExceptions.cpp:
* runtime/CommonSlowPathsExceptions.h:
* runtime/DataView.cpp:
* runtime/DirectArguments.h:
* runtime/ECMAScriptSpecInternalFunctions.cpp:
* runtime/Error.cpp:
* runtime/Error.h:
* runtime/ErrorInstance.cpp:
* runtime/ErrorInstance.h:
* runtime/Exception.cpp:
* runtime/Exception.h:
* runtime/GeneratorFrame.cpp:
* runtime/GeneratorPrototype.cpp:
* runtime/InternalFunction.cpp:
(JSC::InternalFunction::InternalFunction):
* runtime/IntlCollator.cpp:
* runtime/IntlCollatorConstructor.cpp:
* runtime/IntlCollatorPrototype.cpp:
* runtime/IntlDateTimeFormat.cpp:
* runtime/IntlDateTimeFormatConstructor.cpp:
* runtime/IntlDateTimeFormatPrototype.cpp:
* runtime/IntlNumberFormat.cpp:
* runtime/IntlNumberFormatConstructor.cpp:
* runtime/IntlNumberFormatPrototype.cpp:
* runtime/IntlObject.cpp:
* runtime/IteratorPrototype.cpp:
* runtime/JSArray.cpp:
(JSC::JSArray::tryCreateUninitialized):
(JSC::JSArray::setLengthWritable):
(JSC::JSArray::unshiftCountSlowCase):
(JSC::JSArray::setLengthWithArrayStorage):
(JSC::JSArray::appendMemcpy):
(JSC::JSArray::setLength):
(JSC::JSArray::pop):
(JSC::JSArray::push):
(JSC::JSArray::fastSlice):
(JSC::JSArray::shiftCountWithArrayStorage):
(JSC::JSArray::shiftCountWithAnyIndexingType):
(JSC::JSArray::unshiftCountWithArrayStorage):
(JSC::JSArray::fillArgList):
(JSC::JSArray::copyToArguments):
* runtime/JSArray.h:
(JSC::createContiguousArrayButterfly):
(JSC::createArrayButterfly):
(JSC::JSArray::create):
(JSC::JSArray::tryCreateUninitialized): Deleted.
* runtime/JSArrayBufferView.h:
* runtime/JSCInlines.h:
* runtime/JSCJSValue.cpp:
(JSC::JSValue::dumpInContextAssumingStructure):
* runtime/JSCallee.cpp:
(JSC::JSCallee::JSCallee):
* runtime/JSCell.cpp:
(JSC::JSCell::estimatedSize):
* runtime/JSCell.h:
(JSC::JSCell::cellStateOffset): Deleted.
* runtime/JSCellInlines.h:
(JSC::ExecState::vm):
(JSC::JSCell::classInfo):
(JSC::JSCell::callDestructor):
(JSC::JSCell::vm): Deleted.
* runtime/JSFunction.cpp:
(JSC::JSFunction::create):
(JSC::JSFunction::allocateAndInitializeRareData):
(JSC::JSFunction::initializeRareData):
(JSC::JSFunction::getOwnPropertySlot):
(JSC::JSFunction::put):
(JSC::JSFunction::deleteProperty):
(JSC::JSFunction::defineOwnProperty):
(JSC::JSFunction::setFunctionName):
(JSC::JSFunction::reifyLength):
(JSC::JSFunction::reifyName):
(JSC::JSFunction::reifyLazyPropertyIfNeeded):
(JSC::JSFunction::reifyBoundNameIfNeeded):
* runtime/JSFunction.h:
* runtime/JSFunctionInlines.h:
(JSC::JSFunction::createWithInvalidatedReallocationWatchpoint):
(JSC::JSFunction::JSFunction):
* runtime/JSGenericTypedArrayViewInlines.h:
(JSC::JSGenericTypedArrayView<Adaptor>::slowDownAndWasteMemory):
* runtime/JSInternalPromise.cpp:
* runtime/JSInternalPromiseConstructor.cpp:
* runtime/JSInternalPromiseDeferred.cpp:
* runtime/JSInternalPromisePrototype.cpp:
* runtime/JSJob.cpp:
* runtime/JSMapIterator.cpp:
* runtime/JSModuleNamespaceObject.cpp:
* runtime/JSModuleRecord.cpp:
* runtime/JSObject.cpp:
(JSC::JSObject::visitButterfly):
(JSC::JSObject::notifyPresenceOfIndexedAccessors):
(JSC::JSObject::createInitialIndexedStorage):
(JSC::JSObject::createInitialUndecided):
(JSC::JSObject::createInitialInt32):
(JSC::JSObject::createInitialDouble):
(JSC::JSObject::createInitialContiguous):
(JSC::JSObject::createArrayStorage):
(JSC::JSObject::createInitialArrayStorage):
(JSC::JSObject::convertUndecidedToInt32):
(JSC::JSObject::convertUndecidedToContiguous):
(JSC::JSObject::convertUndecidedToArrayStorage):
(JSC::JSObject::convertInt32ToDouble):
(JSC::JSObject::convertInt32ToArrayStorage):
(JSC::JSObject::convertDoubleToArrayStorage):
(JSC::JSObject::convertContiguousToArrayStorage):
(JSC::JSObject::putByIndexBeyondVectorLength):
(JSC::JSObject::putDirectIndexBeyondVectorLength):
(JSC::JSObject::getNewVectorLength):
(JSC::JSObject::increaseVectorLength):
(JSC::JSObject::ensureLengthSlow):
(JSC::JSObject::growOutOfLineStorage):
(JSC::JSObject::copyButterfly): Deleted.
(JSC::JSObject::copyBackingStore): Deleted.
* runtime/JSObject.h:
(JSC::JSObject::globalObject):
(JSC::JSObject::putDirectInternal):
(JSC::JSObject::setStructureAndReallocateStorageIfNecessary): Deleted.
* runtime/JSObjectInlines.h:
* runtime/JSPromise.cpp:
* runtime/JSPromiseConstructor.cpp:
* runtime/JSPromiseDeferred.cpp:
* runtime/JSPromisePrototype.cpp:
* runtime/JSPropertyNameIterator.cpp:
* runtime/JSScope.cpp:
(JSC::JSScope::resolve):
* runtime/JSScope.h:
(JSC::JSScope::globalObject):
(JSC::JSScope::vm): Deleted.
* runtime/JSSetIterator.cpp:
* runtime/JSStringIterator.cpp:
* runtime/JSTemplateRegistryKey.cpp:
* runtime/JSTypedArrayViewConstructor.cpp:
* runtime/JSTypedArrayViewPrototype.cpp:
* runtime/JSWeakMap.cpp:
* runtime/JSWeakSet.cpp:
* runtime/MapConstructor.cpp:
* runtime/MapIteratorPrototype.cpp:
* runtime/MapPrototype.cpp:
* runtime/NativeErrorConstructor.cpp:
* runtime/NativeStdFunctionCell.cpp:
* runtime/Operations.h:
(JSC::scribbleFreeCells):
(JSC::scribble):
* runtime/Options.h:
* runtime/PropertyTable.cpp:
* runtime/ProxyConstructor.cpp:
* runtime/ProxyObject.cpp:
* runtime/ProxyRevoke.cpp:
* runtime/RegExp.cpp:
(JSC::RegExp::match):
(JSC::RegExp::matchConcurrently):
(JSC::RegExp::matchCompareWithInterpreter):
* runtime/RegExp.h:
* runtime/RegExpConstructor.h:
* runtime/RegExpInlines.h:
(JSC::RegExp::matchInline):
* runtime/RegExpMatchesArray.h:
(JSC::tryCreateUninitializedRegExpMatchesArray):
(JSC::createRegExpMatchesArray):
* runtime/RegExpPrototype.cpp:
(JSC::genericSplit):
* runtime/RuntimeType.cpp:
* runtime/SamplingProfiler.cpp:
(JSC::SamplingProfiler::processUnverifiedStackTraces):
* runtime/SetConstructor.cpp:
* runtime/SetIteratorPrototype.cpp:
* runtime/SetPrototype.cpp:
* runtime/StackFrame.cpp: Added.
(JSC::StackFrame::sourceID):
(JSC::StackFrame::sourceURL):
(JSC::StackFrame::functionName):
(JSC::StackFrame::computeLineAndColumn):
(JSC::StackFrame::toString):
* runtime/StackFrame.h: Added.
(JSC::StackFrame::isNative):
* runtime/StringConstructor.cpp:
* runtime/StringIteratorPrototype.cpp:
* runtime/StructureInlines.h:
(JSC::Structure::propertyTable):
* runtime/TemplateRegistry.cpp:
* runtime/TestRunnerUtils.cpp:
(JSC::finalizeStatsAtEndOfTesting):
* runtime/TestRunnerUtils.h:
* runtime/TypeProfilerLog.cpp:
* runtime/TypeSet.cpp:
* runtime/VM.cpp:
(JSC::VM::VM):
(JSC::VM::ensureStackCapacityForCLoop):
(JSC::VM::isSafeToRecurseSoftCLoop):
* runtime/VM.h:
* runtime/VMEntryScope.h:
* runtime/VMInlines.h:
(JSC::VM::ensureStackCapacityFor):
(JSC::VM::isSafeToRecurseSoft):
* runtime/WeakMapConstructor.cpp:
* runtime/WeakMapData.cpp:
* runtime/WeakMapPrototype.cpp:
* runtime/WeakSetConstructor.cpp:
* runtime/WeakSetPrototype.cpp:
* testRegExp.cpp:
(testOneRegExp):
* tools/JSDollarVM.cpp:
* tools/JSDollarVMPrototype.cpp:
(JSC::JSDollarVMPrototype::isInObjectSpace):

Source/WebCore:

No new tests because no new WebCore behavior.

Just rewiring #includes.

* ForwardingHeaders/heap/HeapInlines.h: Added.
* ForwardingHeaders/interpreter/Interpreter.h: Removed.
* ForwardingHeaders/runtime/AuxiliaryBarrierInlines.h: Added.
* Modules/indexeddb/IDBCursorWithValue.cpp:
* Modules/indexeddb/client/TransactionOperation.cpp:
* Modules/indexeddb/server/SQLiteIDBBackingStore.cpp:
* Modules/indexeddb/server/UniqueIDBDatabase.cpp:
* bindings/js/JSApplePayPaymentAuthorizedEventCustom.cpp:
* bindings/js/JSApplePayPaymentMethodSelectedEventCustom.cpp:
* bindings/js/JSApplePayShippingContactSelectedEventCustom.cpp:
* bindings/js/JSApplePayShippingMethodSelectedEventCustom.cpp:
* bindings/js/JSClientRectCustom.cpp:
* bindings/js/JSDOMBinding.cpp:
* bindings/js/JSDOMBinding.h:
* bindings/js/JSDeviceMotionEventCustom.cpp:
* bindings/js/JSDeviceOrientationEventCustom.cpp:
* bindings/js/JSErrorEventCustom.cpp:
* bindings/js/JSIDBCursorWithValueCustom.cpp:
* bindings/js/JSIDBIndexCustom.cpp:
* bindings/js/JSPopStateEventCustom.cpp:
* bindings/js/JSWebGL2RenderingContextCustom.cpp:
* bindings/js/JSWorkerGlobalScopeCustom.cpp:
* bindings/js/WorkerScriptController.cpp:
* contentextensions/ContentExtensionParser.cpp:
* dom/ErrorEvent.cpp:
* html/HTMLCanvasElement.cpp:
* html/MediaDocument.cpp:
* inspector/CommandLineAPIModule.cpp:
* loader/EmptyClients.cpp:
* page/CaptionUserPreferences.cpp:
* page/Frame.cpp:
* page/PageGroup.cpp:
* page/UserContentController.cpp:
* platform/mock/mediasource/MockBox.cpp:
* testing/GCObservation.cpp:

Source/WebKit2:

Just rewiring some #includes.

* UIProcess/ViewGestureController.cpp:
* UIProcess/WebPageProxy.cpp:
* UIProcess/WebProcessPool.cpp:
* UIProcess/WebProcessProxy.cpp:
* WebProcess/InjectedBundle/DOM/InjectedBundleRangeHandle.cpp:
* WebProcess/Plugins/Netscape/JSNPObject.cpp:

Source/WTF:

I needed tryFastAlignedMalloc() so I added it.

* wtf/FastMalloc.cpp:
(WTF::tryFastAlignedMalloc):
* wtf/FastMalloc.h:
* wtf/ParkingLot.cpp:
(WTF::ParkingLot::forEachImpl):
(WTF::ParkingLot::forEach): Deleted.
* wtf/ParkingLot.h:
(WTF::ParkingLot::parkConditionally):
(WTF::ParkingLot::unparkOne):
(WTF::ParkingLot::forEach):
* wtf/ScopedLambda.h:
(WTF::scopedLambdaRef):
* wtf/SentinelLinkedList.h:
(WTF::SentinelLinkedList::forEach):
(WTF::RawNode>::takeFrom):
* wtf/SimpleStats.h:
(WTF::SimpleStats::operator bool):
(WTF::SimpleStats::operator!): Deleted.

Tools:

* DumpRenderTree/TestRunner.cpp:
* DumpRenderTree/mac/DumpRenderTree.mm:
(DumpRenderTreeMain):
* Scripts/run-jsc-stress-tests:
* TestWebKitAPI/Tests/WTF/Vector.cpp:
(TestWebKitAPI::TEST):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205462 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago"Redundant argument in sprintf" warning spam from prepare-ChangeLog
mcatanzaro@igalia.com [Tue, 6 Sep 2016 00:03:08 +0000 (00:03 +0000)]
"Redundant argument in sprintf" warning spam from prepare-ChangeLog
https://bugs.webkit.org/show_bug.cgi?id=161606

Reviewed by Darin Adler.

Avoid extra arguments in call to sprintf.

* Scripts/prepare-ChangeLog:
(statusDescription):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205461 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed error message typo fix
mcatanzaro@igalia.com [Mon, 5 Sep 2016 23:57:17 +0000 (23:57 +0000)]
Unreviewed error message typo fix

* WebProcess/UserContent/WebUserContentController.cpp:
(WebKit::WebUserContentController::removeUserContentWorlds):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205460 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Cocoa] Drop ".Development" suffix from XPC service bundle IDs, leave it only on...
mitz@apple.com [Mon, 5 Sep 2016 20:54:06 +0000 (20:54 +0000)]
[Cocoa] Drop ".Development" suffix from XPC service bundle IDs, leave it only on the executable
https://bugs.webkit.org/show_bug.cgi?id=161575

Reviewed by Sam Weinig.

Append the ".Development" suffix to the executable names, but not to the bundle names and
bundle IDs.

* Configurations/BaseTarget.xcconfig: No need for a preprocessor definition anymore.
* Configurations/BaseXPCService.xcconfig: Set EXECUTABLE_SUFFIX.
* Configurations/DatabaseService.xcconfig: No need to suffix the service name and bundle ID.
* Configurations/NetworkService.xcconfig: Ditto.
* Configurations/PluginService.32.xcconfig: Ditto.
* Configurations/PluginService.64.xcconfig: Ditto.
* Configurations/WebContentService.xcconfig: Ditto.
* UIProcess/Launcher/mac/ProcessLauncherMac.mm:
(WebKit::serviceName): Service names are no longer suffixed.
* WebKit2.xcodeproj/project.pbxproj: Ditto.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205459 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agorun-bindings-tests fails since r205422
commit-queue@webkit.org [Mon, 5 Sep 2016 17:25:52 +0000 (17:25 +0000)]
run-bindings-tests fails since r205422
https://bugs.webkit.org/show_bug.cgi?id=161595

Patch by Fujii Hironori <Hironori.Fujii@sony.com> on 2016-09-05
Reviewed by Darin Adler.

Rebaseline binding tests after r205422.

* bindings/scripts/test/JS/JSInterfaceName.cpp:
* bindings/scripts/test/JS/JSInterfaceName.h:
* bindings/scripts/test/JS/JSTestActiveDOMObject.cpp:
* bindings/scripts/test/JS/JSTestActiveDOMObject.h:
* bindings/scripts/test/JS/JSTestClassWithJSBuiltinConstructor.cpp:
* bindings/scripts/test/JS/JSTestClassWithJSBuiltinConstructor.h:
* bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.cpp:
* bindings/scripts/test/JS/JSTestCustomConstructorWithNoInterfaceObject.h:
* bindings/scripts/test/JS/JSTestCustomNamedGetter.cpp:
* bindings/scripts/test/JS/JSTestCustomNamedGetter.h:
* bindings/scripts/test/JS/JSTestEventConstructor.cpp:
* bindings/scripts/test/JS/JSTestEventConstructor.h:
* bindings/scripts/test/JS/JSTestEventTarget.cpp:
* bindings/scripts/test/JS/JSTestEventTarget.h:
* bindings/scripts/test/JS/JSTestException.cpp:
* bindings/scripts/test/JS/JSTestException.h:
* bindings/scripts/test/JS/JSTestGenerateIsReachable.cpp:
* bindings/scripts/test/JS/JSTestGenerateIsReachable.h:
* bindings/scripts/test/JS/JSTestGlobalObject.cpp:
* bindings/scripts/test/JS/JSTestGlobalObject.h:
* bindings/scripts/test/JS/JSTestInterface.cpp:
* bindings/scripts/test/JS/JSTestInterface.h:
* bindings/scripts/test/JS/JSTestIterable.cpp:
* bindings/scripts/test/JS/JSTestIterable.h:
* bindings/scripts/test/JS/JSTestMediaQueryListListener.cpp:
* bindings/scripts/test/JS/JSTestMediaQueryListListener.h:
* bindings/scripts/test/JS/JSTestNamedConstructor.cpp:
* bindings/scripts/test/JS/JSTestNamedConstructor.h:
* bindings/scripts/test/JS/JSTestNode.cpp:
* bindings/scripts/test/JS/JSTestNode.h:
* bindings/scripts/test/JS/JSTestNondeterministic.cpp:
* bindings/scripts/test/JS/JSTestNondeterministic.h:
* bindings/scripts/test/JS/JSTestObj.cpp:
* bindings/scripts/test/JS/JSTestObj.h:
* bindings/scripts/test/JS/JSTestOverloadedConstructors.cpp:
* bindings/scripts/test/JS/JSTestOverloadedConstructors.h:
* bindings/scripts/test/JS/JSTestOverloadedConstructorsWithSequence.cpp:
* bindings/scripts/test/JS/JSTestOverloadedConstructorsWithSequence.h:
* bindings/scripts/test/JS/JSTestOverrideBuiltins.cpp:
* bindings/scripts/test/JS/JSTestOverrideBuiltins.h:
* bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.cpp:
* bindings/scripts/test/JS/JSTestSerializedScriptValueInterface.h:
* bindings/scripts/test/JS/JSTestTypedefs.cpp:
* bindings/scripts/test/JS/JSTestTypedefs.h:
* bindings/scripts/test/JS/JSattribute.cpp:
* bindings/scripts/test/JS/JSattribute.h:
* bindings/scripts/test/JS/JSreadonly.cpp:
* bindings/scripts/test/JS/JSreadonly.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205458 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[CMake] Duplicated IDL files in WebCore_IDL_FILES
commit-queue@webkit.org [Mon, 5 Sep 2016 17:21:33 +0000 (17:21 +0000)]
[CMake] Duplicated IDL files in WebCore_IDL_FILES
https://bugs.webkit.org/show_bug.cgi?id=161592

Patch by Fujii Hironori <Hironori.Fujii@sony.com> on 2016-09-05
Reviewed by Darin Adler.

* CMakeLists.txt: Removed duplicated IDL files.
* DerivedSources.cpp: Added #if ENABLE(USER_MESSAGE_HANDLERS)

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205457 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Mac] Static Analyzer warnings about unused values in CDMSessionAVStreamSession.mm
commit-queue@webkit.org [Mon, 5 Sep 2016 17:19:44 +0000 (17:19 +0000)]
[Mac] Static Analyzer warnings about unused values in CDMSessionAVStreamSession.mm
https://bugs.webkit.org/show_bug.cgi?id=161598

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2016-09-05
Reviewed by Darin Adler.

* platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.h:
* platform/graphics/avfoundation/objc/CDMSessionAVStreamSession.mm:
(WebCore::CDMSessionAVStreamSession::generateKeyReleaseMessage):
Make this parameter a reference again, accidentally lost in r195410.
Now assignments to the parameter make sense.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205456 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoClean CSS stylesheets should be accessible from JavaScript
commit-queue@webkit.org [Mon, 5 Sep 2016 16:58:48 +0000 (16:58 +0000)]
Clean CSS stylesheets should be accessible from JavaScript
https://bugs.webkit.org/show_bug.cgi?id=158728

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-05
Reviewed by Darin Adler.

Source/WebCore:

Covered by updated tests.

Making use of the resource response type to evaluate whether to make the CSS stylesheets visible
for stylesheet link elements and for stylesheet processing instructions.
Ideally, the origin flag should be a boolean.

To keep behavior consistent in cors-check-unaware cases, the flag might remain unset.
In that case, the behavior remains the same (checking of the stylesheet URL).
The origin flag is set to true or false only in case of fetch mode being set to cors using the crossorigin attribute.

Updated CSSStyleSheet::create to take a Node reference. Updated callers accordingly.

* contentextensions/ContentExtensionStyleSheet.cpp:
(WebCore::ContentExtensions::ContentExtensionStyleSheet::ContentExtensionStyleSheet): Updated to pass a Node reference and not pointer.
* css/CSSStyleSheet.cpp:
(WebCore::CSSStyleSheet::create): Makes origin flag undefined if boolean parameter is not passed.
Otherwise origin flag is set according given boolean value.
(WebCore::CSSStyleSheet::createInline): Makes origin flag set to true.
(WebCore::CSSStyleSheet::CSSStyleSheet): Adding origin flag initialization.
(WebCore::CSSStyleSheet::canAccessRules): Returning according the origin flag if set.
Returning as before in case the flag is not set.
* css/CSSStyleSheet.h: Adding origin clean flag.
* dom/ExtensionStyleSheets.cpp:
(WebCore::ExtensionStyleSheets::addUserStyleSheet):
(WebCore::ExtensionStyleSheets::addAuthorStyleSheetForTesting):
(WebCore::ExtensionStyleSheets::maybeAddContentExtensionSheet):
* html/HTMLLinkElement.cpp:
(WebCore::HTMLLinkElement::initializeStyleSheet): Helper routine used in setCSSStyleSheet.
Sets origin clean flag if the resource is clean.
(WebCore::HTMLLinkElement::setCSSStyleSheet): Making use of initializeStyleSheet.
* html/HTMLLinkElement.h:

LayoutTests:

* http/tests/security/cannot-read-cssrules-expected.txt:
* http/tests/security/cannot-read-cssrules-redirect-expected.txt:
* http/tests/security/cannot-read-cssrules-redirect.html:
* http/tests/security/cannot-read-cssrules.html:
* http/tests/security/cross-origin-css-9.html:
* http/tests/security/resources/xorigincss1-allow-star.php: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205455 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] Skip several failing API tests
mcatanzaro@igalia.com [Mon, 5 Sep 2016 16:42:23 +0000 (16:42 +0000)]
[GTK] Skip several failing API tests

Unfortunately we don't have any form of test expectations for API tests, so we'll just skip
them after bugs are filed.

* Scripts/run-gtk-tests:
(TestRunner):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoFix W3C test importer unit tests after r205447
commit-queue@webkit.org [Mon, 5 Sep 2016 14:34:09 +0000 (14:34 +0000)]
Fix W3C test importer unit tests after r205447
https://bugs.webkit.org/show_bug.cgi?id=161604

Unreviewed.

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-05

* Scripts/webkitpy/w3c/test_importer.py:
(TestImporter.__init__): Skip cleaning test_resource_files files if resource-files.json cannot be read.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205453 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] GL_PACK_ROW_LENGTH is not available in GLES2
gns@gnome.org [Mon, 5 Sep 2016 14:22:07 +0000 (14:22 +0000)]
[GTK] GL_PACK_ROW_LENGTH is not available in GLES2
https://bugs.webkit.org/show_bug.cgi?id=161484

Reviewed by Carlos Garcia Campos.

* UIProcess/gtk/AcceleratedBackingStoreWayland.cpp:
(WebKit::AcceleratedBackingStoreWayland::paint): when under GLES2 we cannot rely on
GL_PACK_ROW_LENGTH; use glReadPixel directly when stride matches width, read line
by line manually otherwise. Colour conversion is also required to get the data out
correctly.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205452 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWebRTC: Several media stream tests crashing
philn@webkit.org [Mon, 5 Sep 2016 13:54:25 +0000 (13:54 +0000)]
WebRTC: Several media stream tests crashing
https://bugs.webkit.org/show_bug.cgi?id=161582

Reviewed by Alejandro G. Castro.

Tools:

* gtk/jhbuild.modules: Bump to latest OpenWebRTC master fixing the crash on missing
glvideoflip GStreamer element.

LayoutTests:

* platform/gtk/TestExpectations: Unskip now-passing tests.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205451 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoCachedResourceLoader is not taking into account fetch options to use or not cached...
commit-queue@webkit.org [Mon, 5 Sep 2016 11:02:37 +0000 (11:02 +0000)]
CachedResourceLoader is not taking into account fetch options to use or not cached resources
https://bugs.webkit.org/show_bug.cgi?id=161389

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-05
Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Updated as new console log messages appear now that cors checks are done at SubresourceLoader level.

* web-platform-tests/XMLHttpRequest/security-consideration.sub-expected.txt:
* web-platform-tests/fetch/api/cors/cors-basic-expected.txt:
* web-platform-tests/fetch/api/cors/cors-basic-worker-expected.txt:
* web-platform-tests/fetch/api/cors/cors-multiple-origins-expected.txt:
* web-platform-tests/fetch/api/cors/cors-multiple-origins-worker-expected.txt:
* web-platform-tests/fetch/api/cors/cors-origin-expected.txt:
* web-platform-tests/fetch/api/cors/cors-origin-worker-expected.txt:
* web-platform-tests/fetch/api/cors/cors-origin.js:
(corsOrigin): Fixing a typo in the test making the tests always passing, since the fetch promise was not taken into account.
* web-platform-tests/fetch/api/cors/cors-redirect-credentials-expected.txt:
* web-platform-tests/fetch/api/cors/cors-redirect-credentials-worker-expected.txt:

Source/WebCore:

Tests: http/tests/fetch/fetching-same-resource-with-diffferent-options.html
       http/tests/security/cross-origin-cached-resource-parallel.html
       http/tests/security/cross-origin-cached-resource.html
       http/tests/security/load-image-after-redirection-2.html
       http/tests/security/shape-outside-and-cached-resources.html

Adding CORS checks for the response in case of CORS fetch mode, in SubresourceLoader.
Removing the CORS checks in Image and DocumentThreadableLoader.

The direction of this patch is to make CachedResource origin-specific/fetch mode specific.

This will remove the need for CachedResource clients to do CORS checks when receiving the notifyFinished call.
This will also make the computation of whether a resource is clean or not much easier since the CachedResource knowd its origin and its response tainting.

Removing the CORS checks at ImageLoader creates the risk of using some cached resources loaded from previously no-cors mode without doing the actual CORS check.
Note that the risk was already there in case of a resource loaded through redirections.
Reusing a cached resource for a load with different options also leads to bad computation of the resource tainting.

As a first step, improvements are done but only for CachedImage resources.

This patch limits the direct reuse of cached resources as follow:
- If the request and existing resources have different origins.
- If the fetch mode is different between request and existing resource.

In those cases, a new CachedResource is created with the correct options and origin.
The data and response of the CachedResource found in the cache are copied efficiently in the new CachedResource, if the matching CachedResource finished loading (CachedImage specific).

If the matching CachedResource is still loading, we trigger a reload (with caching=false to not disturb the being loaded resource).
This should be made more efficient at some point, especially if the matching CachedResource already has its response set.

This triggers a change of behavior: previously, the CORS checks were done by the ImageLoader when the resource was finished loading.
The CORS checks were controlled by the crossOrigin attribute, which may be set or unset between the load start and the load end.

Now the crossOrigin attribute is checked at load start. If it is set, the CORS checks will happen even if the attribute is unset before the end of the load.
This is more consistent as the actual request was built with CORS enabled.

* loader/CrossOriginPreflightChecker.cpp:
(WebCore::CrossOriginPreflightChecker::startPreflight): Setting correctly the preflight options as per fetch spec.
* loader/DocumentThreadableLoader.cpp:
(WebCore::DocumentThreadableLoader::didReceiveResponse): Removing CORS check.
(WebCore::DocumentThreadableLoader::loadRequest): Adding CORS check in sync mode.
* loader/ImageLoader.cpp:
(WebCore::ImageLoader::updateFromElement):
(WebCore::ImageLoader::notifyFinished):
* loader/SubresourceLoader.cpp:
(WebCore::SubresourceLoader::didReceiveResponse): Adding CORS checks to the response
(WebCore::SubresourceLoader::checkResponseCrossOriginAccessControl): Helper routine to do CORS checks
* loader/SubresourceLoader.h:
* loader/cache/CachedImage.cpp:
(WebCore::CachedImage::cloneData): Responsible to set image content from another CachedImage.
* loader/cache/CachedImage.h:
* loader/cache/CachedResource.cpp:
(WebCore::CachedResource::computeOrigin): Helper routine to set the origin and whether the resource is cross-origin or not.
(WebCore::CachedResource::load): Using computeOrigin.
(WebCore::CachedResource::loadFrom): Loading from a CachedResource from the same type and which finished loading.
* loader/cache/CachedResource.h:
(WebCore::CachedResource::cloneData):
* loader/cache/CachedResourceLoader.cpp:
(WebCore::CachedResourceLoader::updateCachedResourceWithCurrentRequest): Helper routine responsible to adapt the CachedResource
that can be reused to the origin and options of a new request.
(WebCore::CachedResourceLoader::requestResource): Calling updateCachedResourceWithCurrentRequest before actually returning the resource.
(WebCore::CachedResourceLoader::determineRevalidationPolicy): Space clean-up.
* loader/cache/CachedResourceLoader.h:
* loader/cache/CachedResourceRequest.h:
(WebCore::CachedResourceRequest::setCachingPolicy):
* style/StylePendingResources.cpp:
(WebCore::Style::loadPendingImage): Allowing data URLs for ShapeOutside data.

LayoutTests:

Added specific expectations for fetch cors-origin* tests for mac-wk2 and ios-simulator-wk2 as these tests use
HTTPS, and the connection is refused.

* http/tests/eventsource/eventsource-cors-basic-expected.txt:
* http/tests/eventsource/eventsource-cors-with-credentials-expected.txt:
* http/tests/fetch/fetching-same-resource-with-diffferent-options-expected.txt: Added.
* http/tests/fetch/fetching-same-resource-with-diffferent-options.html: Added.
* http/tests/loading/cross-origin-XHR-willLoadRequest-expected.txt:
* http/tests/resources/download-json-with-delay.php:
* http/tests/resources/redirect.php:
* http/tests/security/cross-origin-cached-resource-expected.txt: Added.
* http/tests/security/cross-origin-cached-resource-parallel-expected.txt: Added.
* http/tests/security/cross-origin-cached-resource-parallel.html: Added.
* http/tests/security/cross-origin-cached-resource.html: Added.
* http/tests/security/img-with-failed-cors-check-fails-to-load-expected.txt:
* http/tests/security/load-image-after-redirection-2-expected.txt: Added.
* http/tests/security/load-image-after-redirection-2.html: Added.
* http/tests/security/resources/abe-allow-star.php:
* http/tests/security/resources/allow-if-origin.php: Added.
* http/tests/security/resources/cross-origin-cached-resource-iframe.html: Added.
* http/tests/security/resources/rgbalpha.png: Added.
* http/tests/security/shape-outside-and-cached-resources-expected.html: Added.
* http/tests/security/shape-outside-and-cached-resources.html: Added.
* http/tests/security/video-poster-cross-origin-crash-expected.txt:
* http/tests/security/video-poster-cross-origin-crash2-expected.txt:
* http/tests/xmlhttprequest/access-control-and-redirects-async-same-origin-expected.txt:
* http/tests/xmlhttprequest/access-control-repeated-failed-preflight-crash-expected.txt:
* http/tests/xmlhttprequest/cross-origin-no-authorization-expected.txt:
* http/tests/xmlhttprequest/cross-origin-no-credential-prompt-expected.txt:
* http/tests/xmlhttprequest/cross-site-denied-response-expected.txt:
* http/tests/xmlhttprequest/onerror-event-expected.txt:
* http/tests/xmlhttprequest/origin-whitelisting-https-expected.txt:
* http/tests/xmlhttprequest/origin-whitelisting-ip-addresses-with-subdomains-expected.txt:
* http/tests/xmlhttprequest/post-blob-content-type-async-expected.txt:
* http/tests/xmlhttprequest/redirect-cross-origin-2-expected.txt:
* http/tests/xmlhttprequest/redirect-cross-origin-expected.txt:
* http/tests/xmlhttprequest/simple-cross-origin-denied-events-expected.txt:
* http/tests/xmlhttprequest/simple-cross-origin-progress-events-expected.txt:
* http/tests/xmlhttprequest/xmlhttprequest-unsafe-redirect-expected.txt:
* platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-expected.txt:
* platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-worker-expected.txt:
* platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin-expected.txt: Added.
* platform/ios-simulator-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin-worker-expected.txt: Added.
* platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-expected.txt:
* platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-basic-worker-expected.txt:
* platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin-expected.txt: Added.
* platform/mac-wk2/imported/w3c/web-platform-tests/fetch/api/cors/cors-origin-worker-expected.txt: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoW3C test importer should generate the list of resource files
commit-queue@webkit.org [Mon, 5 Sep 2016 10:30:07 +0000 (10:30 +0000)]
W3C test importer should generate the list of resource files
https://bugs.webkit.org/show_bug.cgi?id=161578

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-05
Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

* resources/resource-files.json: Reformatting the file.

Tools:

Generating LayoutTests/imported/w3c/resources/resource-files.json when importing tests.

* Scripts/webkitpy/w3c/test_importer.py:
(TestImporter.__init__): Reading test resource files configuration file.
(TestImporter.find_importable_tests):
(TestImporter.import_tests): Update the test resource files based on test parsed files.
(TestImporter):
(TestImporter._already_identified_as_resource_file): Helper routine to know whether a file is already set as a resource file.
(TestImporter._is_in_resources_directory: Helper routine to know whether a file is in a resources directory.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205447 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMore refactoring of RenderMathMLScripts
fred.wang@free.fr [Mon, 5 Sep 2016 09:19:48 +0000 (09:19 +0000)]
More refactoring of RenderMathMLScripts
https://bugs.webkit.org/show_bug.cgi?id=161371

Patch by Frederic Wang <fwang@igalia.com> on 2016-09-05
Reviewed by Darin Adler.

This is a follow-up of bug 161084. The function getScriptMetricsAndLayoutIfNeeded was quite
complicated and it was not obvious that we have to call it twice with the same reference
to a struture holding vertical metrics. We extract the part retrieving layout parameters
into verticalParameters and move its layoutIfNeeded calls into layoutBlock. Then it can
be reduced to a simple function that retrieve the vertical metrics in one call.
We also improve getBaseAndScripts to make clear that it is performing validation. It returns
a ReferenceChildren structure encapsulating pointers to important children so that we no
longer pass these pointers as function parameters. We continue to need them to browse the
list of prescripts & postscripts but we refactor a bit the loop to avoid explicit mention
of RenderBox*.

No new tests, already covered by existing tests.

* rendering/mathml/RenderMathMLScripts.cpp:
(WebCore::RenderMathMLScripts::validateAndGetReferenceChildren): We now store the pointers to
the base, firstPostScript and firstPreScript children in the ReferenceChildren structure. We
also add a pointer to the prescriptDelimiter for convenience.
(WebCore::RenderMathMLScripts::italicCorrection): Use the ReferenceChildren structure so that
we are sure the base has been validated before calling this function.
(WebCore::RenderMathMLScripts::computePreferredLogicalWidths): Retrieve the reference
children with validateAndGetReferenceChildren instead of calling getBaseAndScripts and use
ReferenceChildren to handle these children and to call italicCorrection. The loops for
SubSup, UnderOver, Multiscripts are also rewritten a bit to avoid declaring a null RenderBox*
outside of them and hence allow to use auto.
(WebCore::RenderMathMLScripts::verticalParameters): This part to extract the layout
parameters is extracted from getScriptMetricsAndLayoutIfNeeded. The parameters are returned
as a VerticalParameters struct.
(WebCore::RenderMathMLScripts::verticalMetrics): This is the remaining part of
getScriptMetricsAndLayoutIfNeeded It used to call layoutIfNeeded on children and to
calculate maximum vertical metrics. For Multiscripts it was called twice: We did a first
call to handle the prescripts and then pass the result again in the second call to handle
the postscripts. We modify a bit the loop so that all the scripts are handled in one call and
hence we can directly return a VerticalMetrics. Again, the reference children are now handled
using the ReferenceChildren structure passed as a parameter.
(WebCore::RenderMathMLScripts::layoutBlock): We retrieve the reference children with
validateAndGetReferenceChildren instead of calling getBaseAndScripts and use
ReferenceChildren to handle these children and to call italicCorrection. We layout all the
children if needed in one loop at the beginning instead of doing that when their vertical
metrics are needed. We can now also retrieve vertical metrics with a single call.
(WebCore::RenderMathMLScripts::getBaseAndScripts): Renamed validateAndGetReferenceChildren.
(WebCore::RenderMathMLScripts::getScriptMetricsAndLayoutIfNeeded): Deleted. Split into
verticalParameters and verticalMetrics.
* rendering/mathml/RenderMathMLScripts.h: New structure to handle the pointers to reference
children. Update the signature of getBaseAndScripts to use this struture and give a clearer
name. Update the signature of italicCorrection to use this structure too. Add a new structure
VerticalParameters and declare the helper function to retrieve them. Rename ScriptMetrics
to VerticalMetrics and update the signature of the function needed to retrieve it.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205441 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] SetNativeSurfaceHandleForCompositing should not be compiled in for wayland...
carlosgc@webkit.org [Mon, 5 Sep 2016 07:57:57 +0000 (07:57 +0000)]
[GTK] SetNativeSurfaceHandleForCompositing should not be compiled in for wayland build
https://bugs.webkit.org/show_bug.cgi?id=161483

Reviewed by Žan Doberšek.

SetNativeSurfaceHandleForCompositing and DestroyNativeSurfaceHandleForCompositing messages should only be used
for GTK+ port when compiled with OpenGL and X11 enabled and redirected XComposite window disabled. So, we should
use USE(TEXTURE_MAPPER) && PLATFORM(GTK) && PLATFORM(X11) && !USE(REDIRECTED_XCOMPOSITE_WINDOW) consistently.
This fixes the build when building the GTK+ port with Wayland enabled and X11 disabled.

* UIProcess/AcceleratedDrawingAreaProxy.cpp:
(WebKit::AcceleratedDrawingAreaProxy::didUpdateBackingStoreState):
* UIProcess/AcceleratedDrawingAreaProxy.h:
* WebProcess/WebPage/AcceleratedDrawingArea.cpp:
(WebKit::AcceleratedDrawingArea::enterAcceleratedCompositingMode):
* WebProcess/WebPage/AcceleratedDrawingArea.h:
* WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.cpp:
* WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.h:
* WebProcess/WebPage/DrawingArea.h:
* WebProcess/WebPage/DrawingArea.messages.in:
* WebProcess/WebPage/gtk/LayerTreeHostGtk.cpp:
* WebProcess/WebPage/gtk/LayerTreeHostGtk.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205431 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMediaPlayerPrivateGStreamerBase: improve build guards in nativeImageForCurrentTime()
zandobersek@gmail.com [Mon, 5 Sep 2016 07:15:50 +0000 (07:15 +0000)]
MediaPlayerPrivateGStreamerBase: improve build guards in nativeImageForCurrentTime()
https://bugs.webkit.org/show_bug.cgi?id=161594

Reviewed by Philippe Normand.

Guard the uses of cairo-gl API in nativeImageForCurrentTime() with the
USE(CAIRO) && ENABLE(ACCELERATED_2D_CANVAS) pair of build guards. This
API is only made available when the ACCELERATED_2D_CANVAS option is
enabled. Placing the guards this way thus avoids compilation errors
when compiling USE_GSTREAMER_GL code without the accelerated 2D canvas
support enabled.

* platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:
(WebCore::MediaPlayerPrivateGStreamerBase::nativeImageForCurrentTime):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205430 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GObject bindings] override handleEvent() method in GObjectEventListener
zandobersek@gmail.com [Mon, 5 Sep 2016 07:14:41 +0000 (07:14 +0000)]
[GObject bindings] override handleEvent() method in GObjectEventListener
https://bugs.webkit.org/show_bug.cgi?id=161593

Reviewed by Carlos Garcia Campos.

* WebProcess/InjectedBundle/API/gtk/DOM/GObjectEventListener.h: Override
the handleMethod() which is inherited from the WebCore EventListener class.
Suppresses the -Winconsistent-missing-override warning that Clang propagates.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205429 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agohttp/tests/fetch/clone-response-body.html uses an incorrect URL
commit-queue@webkit.org [Mon, 5 Sep 2016 06:49:33 +0000 (06:49 +0000)]
http/tests/fetch/clone-response-body.html uses an incorrect URL
https://bugs.webkit.org/show_bug.cgi?id=161382

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-04
Reviewed by Darin Adler.

* http/tests/fetch/clone-response-body.html: Fixing url, increasing delay to ensure non-flakiness.
Adding assertions to ensure load is successful.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205428 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove unnecessary RetainPtr.
joepeck@webkit.org [Mon, 5 Sep 2016 06:41:59 +0000 (06:41 +0000)]
Remove unnecessary RetainPtr.

Review comments on earlier change.

Source/WebKit/mac:

* WebView/WebImmediateActionController.mm:
(-[WebImmediateActionController _defaultAnimationController]):

Source/WebKit2:

* UIProcess/mac/WKImmediateActionController.mm:
(-[WKImmediateActionController _defaultAnimationController]):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205427 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Enable and enforce some recommended and stylistic ESLint rules
joepeck@webkit.org [Mon, 5 Sep 2016 05:02:38 +0000 (05:02 +0000)]
Web Inspector: Enable and enforce some recommended and stylistic ESLint rules
https://bugs.webkit.org/show_bug.cgi?id=161567

Reviewed by Brian Burg.

* .eslintrc:
* UserInterface/Base/EventListener.js:
* UserInterface/Base/EventListenerSet.js:
* UserInterface/Base/MIMETypeUtilities.js:
* UserInterface/Base/Main.js:
* UserInterface/Controllers/CodeMirrorTokenTrackingController.js:
* UserInterface/Controllers/DOMTreeManager.js:
* UserInterface/Controllers/TimelineManager.js:
* UserInterface/Models/CSSKeywordCompletions.js:
* UserInterface/Models/DOMNode.js:
* UserInterface/Models/DOMNodeStyles.js:
* UserInterface/Models/Frame.js:
* UserInterface/Models/Geometry.js:
* UserInterface/Models/ProfileNode.js:
* UserInterface/Models/RenderingFrameTimelineRecord.js:
* UserInterface/Models/SourceCodeLocation.js:
* UserInterface/Protocol/RemoteObject.js:
* UserInterface/Views/ApplicationCacheFrameContentView.js:
* UserInterface/Views/ContextMenu.js:
* UserInterface/Views/DOMTreeContentView.js:
* UserInterface/Views/DOMTreeElement.js:
* UserInterface/Views/DataGrid.js:
* UserInterface/Views/GeneralTreeElement.js:
* UserInterface/Views/HeapAllocationsTimelineDataGridNode.js:
* UserInterface/Views/LogContentView.js:
* UserInterface/Views/OpenResourceDialog.js:
* UserInterface/Views/StyleDetailsPanel.js:
* UserInterface/Views/VisualStyleDetailsPanel.js:
* UserInterface/Views/VisualStyleKeywordPicker.js:
* UserInterface/Views/VisualStylePropertyEditor.js:
* UserInterface/Workers/Formatter/FormatterWorker.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205426 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Run ESLint over UserInterface/Views
joepeck@webkit.org [Mon, 5 Sep 2016 05:02:23 +0000 (05:02 +0000)]
Web Inspector: Run ESLint over UserInterface/Views
https://bugs.webkit.org/show_bug.cgi?id=161564

Reviewed by Brian Burg.

* .eslintrc:
* UserInterface/Views/BoxModelDetailsSectionRow.js:
* UserInterface/Views/ButtonNavigationItem.js:
* UserInterface/Views/CSSStyleDeclarationSection.js:
* UserInterface/Views/CSSStyleDeclarationTextEditor.js:
* UserInterface/Views/CodeMirrorAdditions.js:
* UserInterface/Views/CodeMirrorEditor.js:
* UserInterface/Views/CodeMirrorFormatters.js:
* UserInterface/Views/ColorWheel.js:
* UserInterface/Views/ComputedStyleDetailsPanel.js:
* UserInterface/Views/ConsoleMessageView.js:
* UserInterface/Views/ContentViewContainer.js:
* UserInterface/Views/CookieStorageContentView.js:
* UserInterface/Views/DOMNodeDetailsSidebarPanel.js:
* UserInterface/Views/DOMStorageContentView.js:
* UserInterface/Views/DOMTreeElement.js:
* UserInterface/Views/DOMTreeOutline.js:
* UserInterface/Views/DashboardContainerView.js:
* UserInterface/Views/DataGrid.js:
* UserInterface/Views/DataGridNode.js:
* UserInterface/Views/DebuggerSidebarPanel.js:
* UserInterface/Views/DefaultDashboardView.js:
* UserInterface/Views/FontResourceContentView.js:
* UserInterface/Views/FormattedValue.js:
* UserInterface/Views/GradientEditor.js:
* UserInterface/Views/GradientSlider.js:
* UserInterface/Views/HeapSnapshotContentView.js:
* UserInterface/Views/HeapSnapshotDataGridTree.js:
* UserInterface/Views/HeapSnapshotInstanceDataGridNode.js:
* UserInterface/Views/HoverMenu.js:
* UserInterface/Views/LogContentView.js:
* UserInterface/Views/MemoryTimelineOverviewGraph.js:
* UserInterface/Views/MemoryTimelineView.js:
* UserInterface/Views/NavigationSidebarPanel.js:
* UserInterface/Views/NetworkTimelineOverviewGraph.js:
* UserInterface/Views/NewTabContentView.js:
* UserInterface/Views/ObjectPreviewView.js:
* UserInterface/Views/ObjectTreeBaseTreeElement.js:
* UserInterface/Views/ObjectTreeMapEntryTreeElement.js:
* UserInterface/Views/OverviewTimelineView.js:
* UserInterface/Views/Popover.js:
* UserInterface/Views/ProbeSetDetailsSection.js:
* UserInterface/Views/ProfileDataGridNode.js:
* UserInterface/Views/ProfileDataGridTree.js:
* UserInterface/Views/ProfileView.js:
* UserInterface/Views/ResourceSidebarPanel.js:
* UserInterface/Views/RulesStyleDetailsPanel.js:
* UserInterface/Views/ScopeRadioButtonNavigationItem.js:
* UserInterface/Views/ScriptClusterTimelineView.js:
* UserInterface/Views/ScriptDetailsTimelineView.js:
* UserInterface/Views/ScriptProfileTimelineView.js:
* UserInterface/Views/StackedLineChart.js:
* UserInterface/Views/StorageSidebarPanel.js:
* UserInterface/Views/TabBrowser.js:
* UserInterface/Views/TextEditor.js:
* UserInterface/Views/TimelineDataGridNode.js:
* UserInterface/Views/TimelineOverview.js:
* UserInterface/Views/TimelineRecordFrame.js:
* UserInterface/Views/TimelineRecordingContentView.js:
* UserInterface/Views/TimelineTabContentView.js:
* UserInterface/Views/TreeOutline.js:
* UserInterface/Views/VisualStyleCommaSeparatedKeywordEditor.js:
* UserInterface/Views/VisualStylePropertyEditor.js:
* UserInterface/Views/VisualStyleSelectorSection.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205425 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Update ESLint rules and run it over most of UserInterface
joepeck@webkit.org [Mon, 5 Sep 2016 05:01:57 +0000 (05:01 +0000)]
Web Inspector: Update ESLint rules and run it over most of UserInterface
https://bugs.webkit.org/show_bug.cgi?id=161561

Reviewed by Brian Burg.

* .eslintrc:
Updated ESLint rules.

* UserInterface/Base/InspectorFrontendHostStub.js:
* UserInterface/Base/Main.js:
* UserInterface/Base/Object.js:
* UserInterface/Base/Platform.js:
* UserInterface/Base/Setting.js:
* UserInterface/Controllers/CSSStyleManager.js:
* UserInterface/Controllers/CodeMirrorCompletionController.js:
* UserInterface/Controllers/CodeMirrorTextKillController.js:
* UserInterface/Controllers/DebuggerManager.js:
* UserInterface/Controllers/DragToAdjustController.js:
* UserInterface/Controllers/ReplayManager.js:
* UserInterface/Controllers/SourceMapManager.js:
* UserInterface/Controllers/TimelineManager.js:
* UserInterface/Controllers/TypeTokenAnnotator.js:
* UserInterface/Controllers/VisualStyleCompletionsController.js:
* UserInterface/Debug/Bootstrap.js:
* UserInterface/Debug/UncaughtExceptionReporter.js:
* UserInterface/Models/CSSRule.js:
* UserInterface/Models/CallingContextTree.js:
* UserInterface/Models/CallingContextTreeNode.js:
* UserInterface/Models/ConsoleCommandResultMessage.js:
* UserInterface/Models/CookieStorageObject.js:
* UserInterface/Models/Frame.js:
* UserInterface/Models/Geometry.js:
* UserInterface/Models/IssueMessage.js:
* UserInterface/Models/LayoutTimelineRecord.js:
* UserInterface/Models/MemoryTimeline.js:
* UserInterface/Models/NativeFunctionParameters.js:
* UserInterface/Models/ProfileNode.js:
* UserInterface/Models/RenderingFrameTimelineRecord.js:
* UserInterface/Models/Resource.js:
* UserInterface/Models/ScriptSyntaxTree.js:
* UserInterface/Models/ScriptTimelineRecord.js:
* UserInterface/Models/StackTrace.js:
* UserInterface/Models/TypeSet.js:
* UserInterface/Models/WrappedPromise.js:
* UserInterface/Protocol/InspectorBackend.js:
* UserInterface/Protocol/InspectorFrontendAPI.js:
* UserInterface/Protocol/RemoteObject.js:
* UserInterface/Protocol/ScriptProfilerObserver.js:
* UserInterface/Test/FrontendTestHarness.js:
* UserInterface/Test/InspectorProtocol.js:
* UserInterface/Test/ProtocolTestHarness.js:
* UserInterface/Test/Test.js:
* UserInterface/Test/TestHarness.js:
* UserInterface/Test/TestSuite.js:
* UserInterface/Workers/Formatter/EsprimaFormatter.js:
* UserInterface/Workers/HeapSnapshot/HeapSnapshotWorker.js:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205424 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMiscellaneous unreviewed GTK test gardening
mcatanzaro@igalia.com [Mon, 5 Sep 2016 03:13:00 +0000 (03:13 +0000)]
Miscellaneous unreviewed GTK test gardening

* platform/gtk/TestExpectations:
* platform/gtk/accessibility/math-multiscript-attributes-expected.txt:
* platform/gtk/css1/box_properties/float_on_text_elements-expected.txt:
* platform/gtk/editing/pasteboard/image-in-iframe-expected.txt: Added.
* platform/gtk/editing/pasteboard/onpaste-text-html-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205423 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove the CREATE_DOM_WRAPPER macro
weinig@apple.com [Mon, 5 Sep 2016 03:11:55 +0000 (03:11 +0000)]
Remove the CREATE_DOM_WRAPPER macro
https://bugs.webkit.org/show_bug.cgi?id=161580

Reviewed by Dan Bernstein.

The CREATE_DOM_WRAPPER has irritated me for too long! Replace it
with direct calls to createWrapper<ImplementationType>, which has
been upgraded to not require specifying both the implementation type
and the wrapper type by way of the new JSDOMWrapperConverterTraits
struct which provides mapping from ImplementationType -> JSWrapper.
createWrapper<ImplementationType> has also been upgraded to have a
specialization for when the object being passed in needs to be casted.

* bindings/js/JSAnimationTimelineCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSBlobCustom.cpp:
(WebCore::toJSNewlyCreated):
(WebCore::constructJSBlob):
* bindings/js/JSCSSRuleCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSCSSValueCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSDataCueCustom.cpp:
(WebCore::constructJSDataCue):
* bindings/js/JSDocumentCustom.cpp:
(WebCore::createNewDocumentWrapper):
* bindings/js/JSDocumentFragmentCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSElementCustom.cpp:
(WebCore::createNewElementWrapper):
* bindings/js/JSEventCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSFileCustom.cpp:
(WebCore::constructJSFile):
* bindings/js/JSHTMLCollectionCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSHTMLDocumentCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSIDBCursorCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSImageDataCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSNodeCustom.cpp:
(WebCore::createWrapperInline):
* bindings/js/JSNodeListCustom.cpp:
(WebCore::createWrapper):
* bindings/js/JSPerformanceEntryCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSSVGPathSegCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSStyleSheetCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSTextCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSTextTrackCueCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSWebGLRenderingContextBaseCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSXMLDocumentCustom.cpp:
(WebCore::toJSNewlyCreated):
* dom/make_names.pl:
(printWrapperFunctions):
(printWrapperFactoryCppFile):
Replace CREATE_DOM_WRAPPER with direct calls to createWrapper.

* bindings/js/JSWorkerGlobalScopeBase.h:
Add #include of JSDOMWrapper.h to allow generated subclasses to use
JSDOMWrapperConverterTraits.

* bindings/js/JSDOMBinding.h:
(WebCore::castDOMObjectForWrapperCreation): Deleted.
Remove CREATE_DOM_WRAPPER and castDOMObjectForWrapperCreation and
specialize createWrapper.

* bindings/js/JSDOMWrapper.h:
Forward declare JSDOMWrapperConverterTraits.

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateHeader):
Add specialization of JSDOMWrapperConverterTraits for each header.

(GenerateImplementation):
(GenerateConstructorDefinition):
Replace CREATE_DOM_WRAPPER with direct calls to createWrapper.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205422 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRemove Style::PendingResources
antti@apple.com [Sun, 4 Sep 2016 18:50:21 +0000 (18:50 +0000)]
Remove Style::PendingResources
https://bugs.webkit.org/show_bug.cgi?id=161574

Reviewed by Andreas Kling.

RenderStyle contains all the information needed to initialize resource loads. There is no need for this side structure.

* css/CSSFilterImageValue.cpp:
(WebCore::CSSFilterImageValue::loadSubimages):

    Load external SVG resources along with any image resources.

* css/CSSToStyleMap.cpp:
(WebCore::CSSToStyleMap::styleImage):
(WebCore::CSSToStyleMap::mapFillImage):
(WebCore::CSSToStyleMap::mapNinePieceImage):
* css/CSSToStyleMap.h:
* css/StyleBuilderConverter.h:
(WebCore::StyleBuilderConverter::convertStyleImage):
(WebCore::StyleBuilderConverter::convertShapeValue):
* css/StyleBuilderCustom.h:
(WebCore::StyleBuilderCustom::applyValueCursor):
(WebCore::StyleBuilderCustom::applyValueContent):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::State::clear):
(WebCore::StyleResolver::styleImage):
(WebCore::StyleResolver::createFilterOperations):
(WebCore::StyleResolver::loadPendingResources):
(WebCore::StyleResolver::State::ensurePendingResources): Deleted.
(WebCore::StyleResolver::styleCachedImageFromValue): Deleted.
(WebCore::StyleResolver::styleGeneratedImageFromValue): Deleted.
* css/StyleResolver.h:
(WebCore::StyleResolver::State::takePendingResources): Deleted.
* platform/graphics/filters/FilterOperation.cpp:
(WebCore::ReferenceFilterOperation::loadExternalDocumentIfNeeded):
(WebCore::ReferenceFilterOperation::getOrCreateCachedSVGDocumentReference): Deleted.
* platform/graphics/filters/FilterOperation.h:
(WebCore::ReferenceFilterOperation::cachedSVGDocumentReference):
* rendering/style/StyleCachedImage.cpp:
(WebCore::StyleCachedImage::StyleCachedImage):
* style/StylePendingResources.cpp:
(WebCore::Style::loadPendingResources):

    Trigger resource loads by checking pending resources in RenderStyle unconditionally. Keeping track of them
    separately wasn't necessary or a meaningful optimization.

(WebCore::Style::loadPendingImages): Deleted.
(WebCore::Style::loadPendingSVGFilters): Deleted.
* style/StylePendingResources.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205421 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Cocoa] Wrapping XPC service execution in os_activity is no longer needed
mitz@apple.com [Sun, 4 Sep 2016 18:11:57 +0000 (18:11 +0000)]
[Cocoa] Wrapping XPC service execution in os_activity is no longer needed
https://bugs.webkit.org/show_bug.cgi?id=161577

Reviewed by Sam Weinig.

Removed code that created an os_activity around the execution of the XPC service
code, because this workaround is no longer needed.

* DatabaseProcess/EntryPoint/mac/XPCService/DatabaseServiceEntryPoint.mm:
(DatabaseServiceInitializer):
* NetworkProcess/EntryPoint/mac/XPCService/NetworkServiceEntryPoint.mm:
(NetworkServiceInitializer):
* PluginProcess/EntryPoint/mac/XPCService/PluginServiceEntryPoint.mm:
(PluginServiceInitializer):
* WebProcess/EntryPoint/mac/XPCService/WebContentServiceEntryPoint.mm:
(WebContentServiceInitializer):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205420 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoReverse ownership relation of StyleCachedImage and CSSImageValue
antti@apple.com [Sun, 4 Sep 2016 13:42:42 +0000 (13:42 +0000)]
Reverse ownership relation of StyleCachedImage and CSSImageValue
https://bugs.webkit.org/show_bug.cgi?id=161447

Reviewed by Andreas Kling.

Currently StyleCachedImage (which represents an image in RenderStyle) has a weak ref to the
underlying CSSImageValue/CSSImageSetValue which actually owns it. This is awkwards especially since
StyleGeneratedImage, the other StyleImage subclass has reversed relationship where it refs
the underlying CSSImageGeneratorValue.

This patch makes StyleCachedImage similar to StyleGeneratedImage. StyleCachedImage now refs the
underlying CSSImageValue/CSSImageSetValue. CSSImageValues no longer need to know about StyleCachedImage.
Instead they reference CachedImages (memory cache objects) directly. StyleCachedImage instances are now
conceptually unique to RenderStyle instances. Actual resources are shared as before by sharing CachedImages.

* css/CSSCursorImageValue.cpp:
(WebCore::CSSCursorImageValue::loadImage):
(WebCore::CSSCursorImageValue::cachedImage):
(WebCore::CSSCursorImageValue::styleImage): Deleted.
* css/CSSCursorImageValue.h:
* css/CSSImageGeneratorValue.cpp:
(WebCore::CSSImageGeneratorValue::cachedImageForCSSValue):
* css/CSSImageSetValue.cpp:
(WebCore::CSSImageSetValue::~CSSImageSetValue):
(WebCore::CSSImageSetValue::loadBestFitImage):
(WebCore::CSSImageSetValue::traverseSubresources):
(WebCore::CSSImageSetValue::styleImage): Deleted.
* css/CSSImageSetValue.h:
* css/CSSImageValue.cpp:
(WebCore::CSSImageValue::CSSImageValue):
(WebCore::CSSImageValue::~CSSImageValue):
(WebCore::CSSImageValue::isPending):
(WebCore::CSSImageValue::loadImage):
(WebCore::CSSImageValue::traverseSubresources):
(WebCore::CSSImageValue::knownToBeOpaque):
(WebCore::CSSImageValue::styleImage): Deleted.
* css/CSSImageValue.h:
* css/StyleBuilderCustom.h:
(WebCore::StyleBuilderCustom::applyValueContent):
* css/StyleResolver.cpp:
(WebCore::StyleResolver::styleImage):
(WebCore::StyleResolver::styleCachedImageFromValue):
(WebCore::StyleResolver::styleGeneratedImageFromValue):
(WebCore::StyleResolver::cachedOrPendingFromValue): Deleted.
(WebCore::StyleResolver::generatedOrPendingFromValue): Deleted.
(WebCore::StyleResolver::setOrPendingFromValue): Deleted.
(WebCore::StyleResolver::cursorOrPendingFromValue): Deleted.
* css/StyleResolver.h:
* editing/TextIterator.cpp:
(WebCore::fullyClipsContents):
* page/PageSerializer.cpp:
(WebCore::PageSerializer::retrieveResourcesForProperties):
* rendering/style/FillLayer.cpp:
(WebCore::FillLayer::imagesIdentical):

    Compare data equality instead of pointer equality for StyleImages (since StyleImages are no longer shared).

(WebCore::layerImagesIdentical): Deleted.
* rendering/style/StyleCachedImage.cpp:
(WebCore::StyleCachedImage::StyleCachedImage):
(WebCore::StyleCachedImage::~StyleCachedImage):
(WebCore::StyleCachedImage::cachedImage):
(WebCore::StyleCachedImage::cssValue):
(WebCore::StyleCachedImage::canRender):
(WebCore::StyleCachedImage::isPending):
(WebCore::StyleCachedImage::isLoaded):
(WebCore::StyleCachedImage::errorOccurred):
(WebCore::StyleCachedImage::imageSize):
(WebCore::StyleCachedImage::imageHasRelativeWidth):
(WebCore::StyleCachedImage::imageHasRelativeHeight):
(WebCore::StyleCachedImage::computeIntrinsicDimensions):
(WebCore::StyleCachedImage::usesImageContainerSize):
(WebCore::StyleCachedImage::setContainerSizeForRenderer):
(WebCore::StyleCachedImage::addClient):
(WebCore::StyleCachedImage::removeClient):
(WebCore::StyleCachedImage::image):
(WebCore::StyleCachedImage::knownToBeOpaque):
(WebCore::StyleCachedImage::setCachedImage): Deleted.
* rendering/style/StyleCachedImage.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205419 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed, rolling out r205415.
commit-queue@webkit.org [Sun, 4 Sep 2016 07:10:53 +0000 (07:10 +0000)]
Unreviewed, rolling out r205415.
https://bugs.webkit.org/show_bug.cgi?id=161573

Many bots see inspector test failures, rolling out now and
investigating later. (Requested by brrian on #webkit).

Reverted changeset:

"Web Inspector: unify Main.html and Test.html sources and
generate different copies with the preprocessor"
https://bugs.webkit.org/show_bug.cgi?id=161212
http://trac.webkit.org/changeset/205415

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205418 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoMedia controls behave strangely when videos mute from within a playing handler
wenson_hsieh@apple.com [Sun, 4 Sep 2016 05:45:10 +0000 (05:45 +0000)]
Media controls behave strangely when videos mute from within a playing handler
https://bugs.webkit.org/show_bug.cgi?id=161559
<rdar://problem/28018438>

Reviewed by Darin Adler.

Source/WebCore:

Defer showing media controls until after the media element has fired its onplaying handler. This handles cases
where videos that autoplay may initially meet the criteria for main content, but once the video begins to play,
the page may change the media in some way (e.g. muting) that makes the video no longer main content. This causes
media controls to flicker in and out.

These changes are covered by existing unit tests, which have been refactored to check media controller state
after all autoplaying videos have begun playing. Also adds an additional unit test.

* html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::notifyAboutPlaying):
(WebCore::HTMLMediaElement::hasEverNotifiedAboutPlaying):
* html/HTMLMediaElement.h:
* html/MediaElementSession.cpp:
(WebCore::MediaElementSession::canShowControlsManager):

Tools:

Accounts for changes when determining whether or not to show media controls for autoplaying videos that have not
begun playing yet. Rather than check for a controlled media element upon page load, we force tests to wait until
all autoplaying videos have actually begun playing. This extends to tests that involve interaction, such as
clicking or scrolling.

* TestWebKitAPI/Tests/WebKit2Cocoa/VideoControlsManager.mm:
(-[VideoControlsManagerTestWebView callJavascriptFunction:]):
(-[VideoControlsManagerTestWebView expectControlsManager:afterReceivingMessage:]):
(-[VideoControlsManagerTestWebView performAfterReceivingMessage:action:]):
(-[VideoControlsManagerTestWebView waitForPageToLoadWithAutoplayingVideos:]):
(TestWebKitAPI::TEST):
(-[VideoControlsManagerTestWebView loadTestPageNamed:andExpectControlsManager:afterReceivingMessage:]): Deleted.
* TestWebKitAPI/Tests/WebKit2Cocoa/autoplaying-video-with-audio.html:
* TestWebKitAPI/Tests/WebKit2Cocoa/large-video-hides-controls-after-seek-to-end.html:
* TestWebKitAPI/Tests/WebKit2Cocoa/large-video-playing-scroll-away.html:
* TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-autoplaying-click-to-pause.html:
* TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-autoplaying-scroll-to-video.html:
* TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-paused-video-hides-controls.html:
* TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-playing-muted-video-hides-controls.html:
* TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-playing-video-keeps-controls.html:
* TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-with-audio-autoplay.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205417 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUpdate the semantics of defined-ness of custom elements per spec changes
rniwa@webkit.org [Sun, 4 Sep 2016 05:09:28 +0000 (05:09 +0000)]
Update the semantics of defined-ness of custom elements per spec changes
https://bugs.webkit.org/show_bug.cgi?id=161570

Reviewed by Darin Adler.

Source/WebCore:

This patch adds the notion of a custom element that failed to construct or upgrade so that :defined
doesn't apply to such an element. We also set the defined flag inside the HTMLElement constructor in
the case of synchronous construction instead of waiting for the custom element constructor to finish.
https://dom.spec.whatwg.org/#concept-create-element

Conceptually, there are four distinct states for an element:
1. The element is a built-in element
2. The element is a custom element yet to be defined (an upgrade candidate).
3. The element is a well-defined custom element (constructed or upgraded).
4. The element has failed to construct or upgrade as a custom element (because the custom element
constructor threw an exception or returned an unexpected object).

In the latest DOM/HTML specifications, these states are called as 1. "uncustomized", 2. "undefined",
3. "custom", and 4. "failed": https://dom.spec.whatwg.org/#concept-element-defined

This patch refactors Node flags to introduce these distinct states as the following:
1. Neither IsCustomElement nor IsEditingTextOrUnresolvedCustomElementFlag is set.
2. IsCustomElement and IsEditingTextOrUnresolvedCustomElementFlag are set.
isCustomElementUpgradeCandidate() and isUndefinedCustomElement() return true.
3. IsCustomElement is set and IsEditingTextOrUnresolvedCustomElementFlag is unset.
isDefinedCustomElement() returns true.
4. IsCustomElement is unset and IsEditingTextOrUnresolvedCustomElementFlag is set.
isFailedCustomElement() and isUndefinedCustomElement() return true.

Per a spec change, this patch also makes :defined applied to a synchronously constructed custom element
immediately after super() call in the constructor. When the constructor throws an exception or fails to
return the right element, the HTML parser marks the fallback element with setIsUndefinedCustomElement.

Tests: fast/custom-elements/defined-pseudo-class.html
       fast/custom-elements/defined-rule.html
       fast/custom-elements/upgrading/Node-cloneNode.html

* bindings/js/JSCustomElementInterface.cpp:
(WebCore::JSCustomElementInterface::constructElement): Don't set :defined flag here since that's done
in the HTMLElement constructor now.
(WebCore::JSCustomElementInterface::upgradeElement): Mark the element as failed-to-upgrade as needed.
* bindings/js/JSElementCustom.cpp:
(WebCore::toJSNewlyCreated):
* bindings/js/JSHTMLElementCustom.cpp:
(WebCore::constructJSHTMLElement):
* css/SelectorCheckerTestFunctions.h:
(WebCore::isDefinedElement):
* dom/CustomElementReactionQueue.cpp:
(WebCore::CustomElementReactionQueue::enqueueElementUpgradeIfDefined): Enqueue custom element reactions
only if the element is well defined (successfully constructed or upgraded).
(WebCore::CustomElementReactionQueue::enqueueConnectedCallbackIfNeeded): Ditto.
(WebCore::CustomElementReactionQueue::enqueueDisconnectedCallbackIfNeeded): Ditto.
(WebCore::CustomElementReactionQueue::enqueueAdoptedCallbackIfNeeded): Ditto.
(WebCore::CustomElementReactionQueue::enqueueAttributeChangedCallbackIfNeeded): Ditto.
* dom/CustomElementRegistry.cpp:
(WebCore::enqueueUpgradeInShadowIncludingTreeOrder):
* dom/Document.cpp:
(WebCore::createUpgradeCandidateElement):
(WebCore::createFallbackHTMLElement):
* dom/Element.cpp:
(WebCore::Element::attributeChanged):
(WebCore::Element::didMoveToNewDocument):
(WebCore::Element::insertedInto):
(WebCore::Element::removedFrom):
(WebCore::Element::setCustomElementIsResolved): Deleted.
(WebCore::Element::setIsDefinedCustomElement): Renamed from setCustomElementIsResolved.
(WebCore::Element::setIsFailedCustomElement): Added.
(WebCore::Element::setIsCustomElementUpgradeCandidate): Added.
(WebCore::Element::customElementInterface):
* dom/Element.h:
* dom/Node.h:
(WebCore::Node::setIsCustomElement): Deleted.
(WebCore::Node::isUndefinedCustomElement): Renamed from isUnresolvedCustomElement.
(WebCore::Node::setIsUnresolvedCustomElement): Deleted.
(WebCore::Node::isCustomElementUpgradeCandidate): Added.
(WebCore::Node::isDefinedCustomElement): Renamed from isCustomElement.
(WebCore::Node::isFailedCustomElement): Added.
* dom/make_names.pl:
(printWrapperFactoryCppFile): Use the HTMLElement wrapper on upgrade candidates. When a custom element
failed to upgrade, the HTMLElement constructor would have created the wrapper so we never run this code.
* html/parser/HTMLConstructionSite.cpp:
(WebCore::HTMLConstructionSite::createHTMLElementOrFindCustomElementInterface):
* html/parser/HTMLDocumentParser.cpp:
(WebCore::HTMLDocumentParser::runScriptsForPausedTreeBuilder): Mark the HTMLUnknownElement created when
the custom element constructor failed to run successfully as a failed custom element so that :define
wouldn't apply to this element.

LayoutTests:

Added a new test cases to defined-pseudo-class.html, defined-rule.html, and Node-cloneNode.html
and rebaselined the tests.

* fast/custom-elements/defined-pseudo-class-expected.txt:
* fast/custom-elements/defined-pseudo-class.html:
(MyElement): Made matchInsideConstructor an instance variable so that there won't be inter-test dependency.
Added test cases for :defined not being not applying to a failed-to-upgrade custom element. Finally, updated
test expectation to reflect the fact :defined now applies inside custom element constructors immediately after
super() call.
* fast/custom-elements/defined-rule.html: Added a test case for :defined not applying to a failed-to-upgrade
custom element. Also adjusted the height of the last box so that the green box is still 100px by 100px.
* fast/custom-elements/upgrading/Node-cloneNode-expected.txt:
* fast/custom-elements/upgrading/Node-cloneNode.html: Added a test to make sure we don't try to upgrade
a custom element for the second time when the first attempt resulted in the constructor throwing an exception.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205416 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: unify Main.html and Test.html sources and generate different copies...
bburg@apple.com [Sun, 4 Sep 2016 04:34:57 +0000 (04:34 +0000)]
Web Inspector: unify Main.html and Test.html sources and generate different copies with the preprocessor
https://bugs.webkit.org/show_bug.cgi?id=161212
<rdar://problem/28017961>

Reviewed by Joseph Pecoraro.

.:

Rearrange some CMake rules so most Inspector UI work is done in WebInspectorUI.

* Source/CMakeLists.txt: Add 'WebInspectorUI' subdirectory.
* Source/PlatformEfl.cmake:
- Move the rule to copy InspectorBackendCommands.js into WebInspectorUI/CMakeLists.txt.
- Add a FIXME to use the cross-port list of Inspector resources instead of copying everything.
- Add new dependency so generated WebInspectorUI files are made by web-inspector-resources.
- Copy over generated files Main.html and Test.html.

* Source/PlatformWin.cmake:
- Add a FIXME to use the cross-port list of Inspector resources instead of copying everything.
- Add new dependency so generated WebInspectorUI files are made by web-inspector-resources.
- Copy over generated files Main.html and Test.html.

* Source/cmake/WebKitFS.cmake:
- Set up WEBINSPECTORUI_DIR and use it.
- Move directory creation commands here from JavaScriptCore.

* Source/cmake/WebKitMacros.cmake:
Add a helper to turn a CMake list into a space-delimited string of elements.

Source/JavaScriptCore:

* CMakeLists.txt: Remove some unnecessary MAKE_DIRECTORY commands.

Source/WebInspectorUI:

Add a Derived Sources build phase to WebInspectorUI project. Generate
Test.html and Main.html from a combined Inspector.html.in which has preprocessor
macros to include/exclude files not needed by all main resource versions.

Similarly, start generating these Inspector files in WebInspectorUI/CMakeLists.txt.
Move platform-specific bundling commands into PlatformGTK.cmake.

* CMakeLists.txt: Added.
Set up a list of common frontend resources that specific ports can extend, such as
with their own port-specific image resources. This list is the input to port-specific
packaging/bundling scripts. Eventually, minification and concatenation should happen
independently of the specific port by constructing the list of resources dynamically.

To ensure resources are always generated in WebInspectorUI and accessible from WebKit2,
add a dummy target that is always out of date and depends on generated files, causing
them to be built.

Also create a macro to run the preprocessor over Inspector.html.in using various
preprocessor macro definitions. These are customizable by ports to control the
appearance of ENGINEERING_BUILD, which guards resources not meant for shipping builds.

* Configurations/WebInspectorUIFramework.xcconfig:
We need to use preprocessor.pm from WebCore. On Mac, this is a private header.
Teach xcodebuild how to compute WEBCORE_PRIVATE_HEADERS_DIR. This is copied
from WebKit2's configuration files.

* PlatformGTK.cmake: Added.
Add GTK image resources to the resource list. Generate GResource catalog and embedded
C file into DerivedSources. WebKit2 will copy over this file and compile it.

* Scripts/combine-resources.pl:
(concatenateFiles):
Remove the --strip option as this patch removes the only use of it.

* DerivedSources.make: Added.
Generate Test.html and Main.html from the new combined Inspector.html.in.
The 'preprocess_main_resource' rule was copied from WebCore's DerivedSources.make.

* Scripts/cssmin.py: Removed.
* Scripts/jsmin.py: Removed.
These scripts are copied from JavaScriptCore but nobody uses these copies. Remove them.

* Scripts/copy-user-interface-resources-dryrun.rb: Do some cleanup.
- Stage scripts from SRCROOT into the tmpdir so dryrun doesn't require
  a previous build to process WebInspectorUI resources.
- Run DerivedSources.make before copying/processing resources.
- Add some environment variables for new phase and group by script affected.
- Add a comment to clarify what this script is simulating.

* Scripts/copy-user-interface-resources.pl:
- Use Main.html and Test.html from DerivedSources/ instead of SRCROOT.
- Copy over Main.html and Test.html manually if not combining resources.
- Remove the command to strip files from Debug/ for production. This is
  now redundant with ENGINEERING_BUILD guards in Inspector.html.in.
- Use jsmin.py from JavaScriptCore instead of the local copy.
- Wrap all multi-argument 'system' invocations so they are readable.

* Scripts/generate-webinspectorui-derived-sources: Added.
Added boilerplate script to run DerivedSources.make for Mac port.

* Scripts/preprocess-main-resource.pl: Added.
Trivially invoke the preprocessor on $0 using the given defines.

* UserInterface/Inspector.html.in: Renamed from Source/WebInspectorUI/UserInterface/Main.html.
* UserInterface/Test.html: Removed.
Combine Test.html and Main.html into Inspector.html.in. Add these guards:
- INCLUDE_TEST_RESOURCES: for resources excluded from Main.html.
- INCLUDE_UI_RESOURCES: for resources excluded from model tests.
- ENGINEERING_BUILD: for resources not to be shipped (Debug/ directory).

* WebInspectorUI.xcodeproj/project.pbxproj:
- Add new aggregate target 'Derived Sources' to project 'WebInspectorUI'.
- Add dependency on 'Derived Sources' to WebInspectorUI.framework.
- Remove unused copies of jsmin.py and cssmin.py.

Source/WebKit2:

Rearrange CMake rules so that most Inspector UI work is done in WebInspectorUI.

* PlatformGTK.cmake:
- Move the list of Inspector resources into WebInspectorUI/CMakeLists.txt.
- Move generation of InspectorGResourceBundle into WebInspectorUI.
- Copy over InspectorGResourceBundle.c into WebKit2's Derived Sources before compiling.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205415 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Change Cmd-D from kill line to selecting next occurrence
commit-queue@webkit.org [Sun, 4 Sep 2016 00:38:25 +0000 (00:38 +0000)]
Web Inspector: Change Cmd-D from kill line to selecting next occurrence
https://bugs.webkit.org/show_bug.cgi?id=161514

Patch by Devin Rousso <dcrousso+webkit@gmail.com> on 2016-09-03
Reviewed by Brian Burg.

* UserInterface/Controllers/CodeMirrorTextKillController.js:
(WebInspector.CodeMirrorTextKillController):
(WebInspector.CodeMirrorTextKillController.prototype._handleTextChange):
Remove Cmd-D mapping.

* UserInterface/External/CodeMirror/sublime.js:
* UserInterface/Main.html:
Add Sublime Text keybinding support.

* UserInterface/Views/CodeMirrorAdditions.js:
Use Sublime text selectNextOccurrence for Cmd-D instead of deleteLine.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205414 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Add keyboard shortcut for switching to last used dock configuration
commit-queue@webkit.org [Sun, 4 Sep 2016 00:27:40 +0000 (00:27 +0000)]
Web Inspector: Add keyboard shortcut for switching to last used dock configuration
https://bugs.webkit.org/show_bug.cgi?id=110328

Patch by Devin Rousso <dcrousso+webkit@gmail.com> on 2016-09-03
Reviewed by Brian Burg.

* UserInterface/Base/Main.js:
(WebInspector.contentLoaded):
(WebInspector.updateDockedState):
(WebInspector._togglePreviousDockConfiguration):
Added variable for saving the previous dock state, which is used by the new Cmd+Shift+D
shortcut to toggle the docked state between the current and previous states.

(WebInspector._undock):
(WebInspector._dockBottom):
(WebInspector._dockRight):
(WebInspector._updateDockNavigationItems):
(WebInspector._dockedResizerMouseDown):
(WebInspector._dockedResizerMouseDown.dockedResizerDrag):
Make use of new WebInspector.DockConfiguration instead of hardcoded strings.

(WebInspector.DockConfiguration):
Create enum for different WebInspector dock modes:
 - Right
 - Bottom
 - Undocked

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205413 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoRefactor the heuristic for showing media controls to take all media sessions into...
wenson_hsieh@apple.com [Sun, 4 Sep 2016 00:24:14 +0000 (00:24 +0000)]
Refactor the heuristic for showing media controls to take all media sessions into account
https://bugs.webkit.org/show_bug.cgi?id=161503
<rdar://problem/28033783>

Reviewed by Darin Adler.

Source/WebCore:

Currently, when selecting a media session to show playback controls for, we grab the first media session that
passes our heuristic. Using this method, we are unable to take additional factors into account, such as whether
another media session's element is scrolled in view, or if another media session has been interacted with more
recently. To address this, we make the following changes:

    1.  Consider the list of all MediaElementSessions.

    2.  Select only the MediaElementSessions capable of showing media controls and sort the list by a special
        heuristic that takes visibility and time of last user interaction into account. The first element on
        this list is the strongest candidate for main content.

    3.  If this strongest candidate is visible in the viewport, or it is playing with audio, we return this
        as the chosen candidate. Otherwise, we return this session only if no other non-candidate video could be
        confused as the main content (i.e. the non-candidate video is not only visible in the viewport, but also
        large enough to be considered main content).

Using this new method of determining the video to show controls for, we retain previous behavior for pages with
a single video. On pages with multiple videos, the above logic ensures that if the current controlled video is
paused, scrolled out of view, and then a new video is scrolled into view, we will either hide media controls to
avoid confusion if that video could be confused for main content (using the mechanism in step 3), or we
hook up the media controls to the new video if it satisfies main content (using the mechanism in step 2).

This patch also adds 6 new TestWebKitAPI unit tests.

* html/HTMLMediaElement.cpp:
(WebCore::mediaElementSessionInfoForSession):
(WebCore::preferMediaControlsForCandidateSessionOverOtherCandidateSession):
(WebCore::mediaSessionMayBeConfusedWithMainContent):
(WebCore::bestMediaSessionForShowingPlaybackControlsManager):
(WebCore::HTMLMediaElement::didAttachRenderers):
(WebCore::HTMLMediaElement::layoutSizeChanged):
(WebCore::HTMLMediaElement::isVisibleInViewportChanged):
(WebCore::HTMLMediaElement::resetPlaybackSessionState):
(WebCore::HTMLMediaElement::isVisibleInViewport):
(WebCore::HTMLMediaElement::updatePlaybackControlsManager):
* html/HTMLMediaElement.h:
* html/MediaElementSession.cpp:
(WebCore::MediaElementSession::removeBehaviorRestriction):
(WebCore::MediaElementSession::canShowControlsManager):
(WebCore::MediaElementSession::isLargeEnoughForMainContent):
(WebCore::MediaElementSession::mostRecentUserInteractionTime):
(WebCore::MediaElementSession::wantsToObserveViewportVisibilityForMediaControls):
(WebCore::MediaElementSession::wantsToObserveViewportVisibilityForAutoplay):
(WebCore::MediaElementSession::resetPlaybackSessionState):
(WebCore::MediaElementSession::canControlControlsManager): Deleted.
* html/MediaElementSession.h:
* platform/audio/PlatformMediaSession.h:
(WebCore::PlatformMediaSession::resetPlaybackSessionState):
(WebCore::PlatformMediaSession::canControlControlsManager): Deleted.
* platform/audio/PlatformMediaSessionManager.cpp:
(WebCore::PlatformMediaSessionManager::currentSessionsMatching):
(WebCore::PlatformMediaSessionManager::currentSessionMatching): Deleted.
* platform/audio/PlatformMediaSessionManager.h:
* platform/cocoa/WebPlaybackSessionModelMediaElement.mm:
(WebPlaybackSessionModelMediaElement::setMediaElement):

Source/WebKit2:

Adds an SPI testing hook for sending the element ID of the currently controlled video element from the web
process to the UI process. See VideoControlsManager.mm in Tools/TestWebKitAPI/ for usage.

* UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _requestControlledElementID]):
(-[WKWebView _handleControlledElementIDResponse:]):
(-[WKWebView _hasActiveVideoForControlsManager]): Deleted.
* UIProcess/API/Cocoa/WKWebViewPrivate.h:
* UIProcess/Cocoa/WebPlaybackSessionManagerProxy.h:
* UIProcess/Cocoa/WebPlaybackSessionManagerProxy.messages.in:
* UIProcess/Cocoa/WebPlaybackSessionManagerProxy.mm:
(WebKit::WebPlaybackSessionManagerProxy::handleControlledElementIDResponse):
(WebKit::WebPlaybackSessionManagerProxy::requestControlledElementID):
* UIProcess/PageClient.h:
* UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::requestControlledElementID):
(WebKit::WebPageProxy::handleControlledElementIDResponse):
* UIProcess/WebPageProxy.h:
* UIProcess/mac/PageClientImpl.h:
* UIProcess/mac/PageClientImpl.mm:
(WebKit::PageClientImpl::handleControlledElementIDResponse):
* WebProcess/cocoa/WebPlaybackSessionManager.h:
* WebProcess/cocoa/WebPlaybackSessionManager.messages.in:
* WebProcess/cocoa/WebPlaybackSessionManager.mm:
(WebKit::WebPlaybackSessionManager::handleControlledElementIDRequest):

Tools:

Adds new unit tests verifying the behavior of media playback controls when scrolling another video into view.
Please see the WebCore ChangeLog for more details about this change. Also refactors existing
VideoControlsManager tests by folding duplicated setup and testing logic into helper methods to make the unit
tests more readable.

* TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
* TestWebKitAPI/Tests/WebKit2Cocoa/VideoControlsManager.mm:
(-[MessageHandler initWithMessage:handler:]):
(-[MessageHandler userContentController:didReceiveScriptMessage:]):
(-[VideoControlsManagerTestWebView performAfterLoading:]):
(-[VideoControlsManagerTestWebView loadTestPageNamed:]):
(-[VideoControlsManagerTestWebView loadTestPageNamed:andExpectControlsManager:afterReceivingMessage:]):
(-[VideoControlsManagerTestWebView performAfterReceivingMessage:action:]):
(-[VideoControlsManagerTestWebView controlledElementID]):
(-[VideoControlsManagerTestWebView _handleControlledElementIDResponse:]):
(TestWebKitAPI::setUpWebViewForTestingVideoControlsManager):
(TestWebKitAPI::TEST):
(-[MediaPlaybackMessageHandler initWithWKWebView:finalMessageString:]): Deleted.
(-[MediaPlaybackMessageHandler userContentController:didReceiveScriptMessage:]): Deleted.
(-[OnLoadMessageHandler initWithWKWebView:handler:]): Deleted.
(-[OnLoadMessageHandler userContentController:didReceiveScriptMessage:]): Deleted.
(-[WKWebView performAfterLoading:]): Deleted.
* TestWebKitAPI/Tests/WebKit2Cocoa/large-video-playing-scroll-away.html: Added.
* TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-autoplaying-click-to-pause.html: Added.
* TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-autoplaying-scroll-to-video.html: Added.
* TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-paused-video-hides-controls.html: Added.
* TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-playing-muted-video-hides-controls.html: Added.
* TestWebKitAPI/Tests/WebKit2Cocoa/large-videos-playing-video-keeps-controls.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205412 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoStreamline DOMImplementation, and move it to our new DOM exception system
darin@apple.com [Sat, 3 Sep 2016 23:34:00 +0000 (23:34 +0000)]
Streamline DOMImplementation, and move it to our new DOM exception system
https://bugs.webkit.org/show_bug.cgi?id=161295

Reviewed by Ryosuke Niwa.

Source/WebCore:

* WebCore.xcodeproj/project.pbxproj: Added new headers to project.

* bindings/js/JSDOMBinding.h:
(WebCore::toJS): Added an overload for ExceptionOr<>; this handles the
exception case here so it doesn't need to be handled in generated code
for the binding. Implemented here so that ExceptionOr.h does not know
about bindings. But since this is a template, it will only compile when
instantiated and there is no need to include ExceptionOr.h and indirectly
the Variant.h header in this header.
(WebCore::toJSNewlyCreated): Ditto.

* bindings/scripts/CodeGeneratorJS.pm:
(GenerateCallbackImplementation): Refer to JSC::Exception with explicit
namespace to avoid ambiguity with WebCore::Exception.

* dom/DOMImplementation.cpp:
(WebCore::DOMImplementation::createDocumentType): Changed to return
ExceptionOr.
(WebCore::createXMLDocument): Added. Helper used in createDocument.
(WebCore::DOMImplementation::getInterface): Deleted. This was unused.
(WebCore::DOMImplementation::createDocument): Changed to return
ExceptionOr.
(WebCore::DOMImplementation::createCSSStyleSheet): Removed the unused
ExceptionCode out argument.
(WebCore::isValidXMLMIMETypeChar): Deleted. Moved to MIMETypeRegistry.
(WebCore::DOMImplementation::isXMLMIMEType): Ditto.
(WebCore::DOMImplementation::isTextMIMEType): Ditto.

* dom/DOMImplementation.h: Changed functions as described above to
return ExceptionOr values. Also removed unused getInterface function,
and isXMLMIMEType and isTextMIMEType, which both moved to the
MIMETypeRegistry class alongside all the other similar MIME type
functions.

* dom/DOMImplementation.idl: Reorganized this to match the IDL files
in the specifications a little better. Also removed [RaisesException]
since that is only needed for the old legacy ExceptionCode& style.

* dom/Document.cpp:
(WebCore::Document::setXMLVersion): Removed call to the
DOMImplementation::hasFeature function since the values passed in
unconditionally result in the return value "true". This is left over
either from specification language, or from an ancient version of this
code that worked in a "no XML supported" mode.
(WebCore::Document::setXMLStandalone): Ditto.

* dom/Document.h: Removed the ExceptionCode& out argument from setXMLStandalone.
* dom/Document.idl: Removed [SetterRaisesException] from xmlStandalone.

* dom/Exception.h: Added.
* dom/ExceptionOr.h: Added.

* html/HTMLTemplateElement.cpp: Removed unneeded include of DOMImplementation.h.

* inspector/InspectorPageAgent.cpp:
(WebCore::createXHRTextDecoder): Use isXMLMIMEType in its new location in
MIMETypeRegistry.
* inspector/NetworkResourcesData.cpp:
(WebCore::createOtherResourceTextDecoder): Ditto.

* loader/FrameLoader.cpp: Removed unneeded include of DOMImplementation.h.

* loader/TextResourceDecoder.cpp:
(WebCore::TextResourceDecoder::determineContentType): Use isXMLMIMEType in its
new location in MIMETypeRegistry.

* platform/MIMETypeRegistry.cpp:
(WebCore::MIMETypeRegistry::isTextMIMEType): Added. Moved here from
DOMImplementation.
(WebCore::isValidXMLMIMETypeChar): Ditto.
(WebCore::MIMETypeRegistry::isXMLMIMEType): Ditto.

* platform/MIMETypeRegistry.h: Added isXMLMIMEType and isTextMIMEType.
Made isUnsupportedTextMIMEType private.

* svg/SVGElement.cpp:
(WebCore::SVGElement::isSupported): Deleted. This function was never called.
* svg/SVGElement.h: Updated for the above change.

* xml/XMLHttpRequest.cpp:
(WebCore::XMLHttpRequest::responseIsXML): Use isXMLMIMEType in its new
location in MIMETypeRegistry.

* xml/parser/XMLDocumentParserLibxml2.cpp:
(WebCore::XMLDocumentParser::startDocument): Updated since setXMLStandalone
no longer can raise an exception.

Source/WebKit/mac:

* DOM/DOMDOMImplementation.mm:
(unwrap): Added. Use this instead of the IMPL macro.
(-[DOMImplementation dealloc]): Updated to use unwrap.
(-[DOMImplementation hasFeature:version:]): Ditto.
(-[DOMImplementation createDocumentType:publicId:systemId:]): Updated to work with
ExceptionOr.
(-[DOMImplementation createDocument:qualifiedName:doctype:]): Ditto.
(-[DOMImplementation createCSSStyleSheet:media:]): Removed exception logic since
this function can no longer raise an exception.
(-[DOMImplementation createHTMLDocument:]): Updated to use unwrap.
(-[DOMImplementation hasFeature::]): Changed to call the non-deprecated version
rather than duplicating its implementation. Also moved into the category as defined
in the header.
(-[DOMImplementation createDocumentType:::]): Ditto.
(-[DOMImplementation createDocument:::]): Ditto.
(-[DOMImplementation createCSSStyleSheet::]): Ditto.

* DOM/DOMDocument.mm: Removed unneeded include of DOMImplementation.h.
(-[DOMDocument setXmlStandalone:]): Updated since setXMLStandalone no longer can
raise an exception.

* WebView/WebFrame.mm:
(-[WebFrame _canProvideDocumentSource]): Updated to use isTextMIMEType in its new
location in MIMETypeRegistry instead of in DOMImplementation.

Source/WebKit/win:

* WebFrame.cpp:
(WebFrame::canProvideDocumentSource): Updated to use isXMLMIMEType in its new
location in MIMETypeRegistry instead of in DOMImplementation.

Source/WebKit2:

* UIProcess/WebFrameProxy.cpp:
(WebKit::WebFrameProxy::isDisplayingMarkupDocument): Use isXMLMIMEType in its
new location in MIMETypeRegistry rather than in DOMImplementation.
(WebKit::WebFrameProxy::isDisplayingPDFDocument): Removed unneeded redundant
check for empty string, already done by MIMETypeRegistry.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205411 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnbreak customElements.whenDefined after r205383 with a crash fix
rniwa@webkit.org [Sat, 3 Sep 2016 23:25:47 +0000 (23:25 +0000)]
Unbreak customElements.whenDefined after r205383 with a crash fix
https://bugs.webkit.org/show_bug.cgi?id=161562

Reviewed by Darin Adler.

Source/WebCore:

The crash was caused by DeferredWrapper::contextDestroyed not calling ContextDestructionObserver::contextDestroyed.

This caused m_scriptExecutionContext to not being set to nullptr when the Document was destroyed before DOMWindow
during a single GC sweeping, and resulted in a use-after-free in ContextDestructionObserver's destructor.

Fixed the crash and reverted r205383.

Tests: fast/custom-elements/CustomElementRegistry.html

* bindings/js/JSCustomElementRegistryCustom.cpp:
(WebCore::whenDefinedPromise):
* bindings/js/JSDOMPromise.cpp:
(WebCore::DeferredWrapper::contextDestroyed): Fixed the crash.
* dom/CustomElementRegistry.cpp:
(WebCore::CustomElementRegistry::addElementDefinition):
* dom/CustomElementRegistry.h:
(WebCore::CustomElementRegistry::promiseMap):

LayoutTests:

Revert r205383 now that all test cases pass.

* fast/custom-elements/CustomElementRegistry-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205410 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAlign cross-Origin Object.getOwnPropertyNames() with the HTML specification
cdumez@apple.com [Sat, 3 Sep 2016 22:50:55 +0000 (22:50 +0000)]
Align cross-Origin Object.getOwnPropertyNames() with the HTML specification
https://bugs.webkit.org/show_bug.cgi?id=161457

Reviewed by Darin Adler.

Source/WebCore:

Align cross-Origin Object.getOwnPropertyNames() with the HTML specification:
- https://html.spec.whatwg.org/#windowproxy-ownpropertykeys
- https://html.spec.whatwg.org/#location-ownpropertykeys
- https://html.spec.whatwg.org/#crossoriginproperties-(-o-)

We should list cross origin properties.

Firefox complies with the specification. However, WebKit was returning an
empty array and logs a security error message.

No new tests, updated existing test.

* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::addCrossOriginPropertyNames):
(WebCore::JSDOMWindow::getOwnPropertyNames):
* bindings/js/JSLocationCustom.cpp:
(WebCore::addCrossOriginPropertyNames):
(WebCore::JSLocation::getOwnPropertyNames):

LayoutTests:

Add test coverage.

* http/tests/security/cross-frame-access-enumeration-expected.txt:
* http/tests/security/cross-frame-access-enumeration.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205409 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoConstructors of MathML renderers should only accept MathMLPresentationElement-derived...
commit-queue@webkit.org [Sat, 3 Sep 2016 21:32:45 +0000 (21:32 +0000)]
Constructors of MathML renderers should only accept MathMLPresentationElement-derived classes
https://bugs.webkit.org/show_bug.cgi?id=161378

Patch by Frédéric Wang <fwang@igalia.com> on 2016-09-03
Reviewed by Darin Adler.

We update constructors of RenderMathMLBlock, to only accept MathMLPresentationElement
instances as a parameter. Similarly, we make the constructor of RenderMathMLToken only
accept MathMLTokenElement instances.

No new tests, behavior is unchanged.

* rendering/mathml/RenderMathMLBlock.cpp:
(WebCore::RenderMathMLBlock::RenderMathMLBlock):
* rendering/mathml/RenderMathMLBlock.h:
* rendering/mathml/RenderMathMLToken.cpp:
(WebCore::RenderMathMLToken::RenderMathMLToken):
* rendering/mathml/RenderMathMLToken.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205408 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSource/WebCore:
bweinstein@apple.com [Sat, 3 Sep 2016 19:58:28 +0000 (19:58 +0000)]
Source/WebCore:
Consult with the FrameLoaderClient about whether or not content extensions should be enabled when loading this URL.
https://bugs.webkit.org/show_bug.cgi?id=161441

Reviewed by Darin Adler.

* loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::startLoadingMainResource): If content extensions aren't already disabled, consult with the
FrameLoaderClient about whether or not we should use content extensions for this URL.
* loader/FrameLoaderClient.h: Add the FrameLoaderClient call to determine if we should use content extensions for a given
URL...
* loader/EmptyClients.h: ... And add a stub implementation.

Source/WebKit/mac:
Implement a stub version of FrameLoaderClient::shouldUseContentExtensionsForURL.
https://bugs.webkit.org/show_bug.cgi?id=16144

Reviewed by Darin Adler.

* WebCoreSupport/WebFrameLoaderClient.h:

Source/WebKit2:
Implement WebFrameLoaderClient::shouldUseContentExtensionsForURL and consult the InjectedBundlePageLoaderClient.
https://bugs.webkit.org/show_bug.cgi?id=161441

Reviewed by Darin Adler.

WebFrameLoaderClient::shouldUseContentExtensionsForURL only consults the injected bundle, because we don't want to
defer the loading of every main resource to consult with the UI Process about whether or not we should use content
extensions for the load.

* WebProcess/InjectedBundle/API/c/WKBundlePageLoaderClient.h: Bump the latest version to WKBundlePageLoaderClientV9 and
add WKBundlePageShouldUseContentExtensionsForURLCallback.
* WebProcess/InjectedBundle/InjectedBundlePageLoaderClient.cpp:
(WebKit::InjectedBundlePageLoaderClient::shouldUseContentExtensionsForURL): Ask the client if we should use content
extensions for this URL.
* WebProcess/InjectedBundle/InjectedBundlePageLoaderClient.h:
* WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::shouldUseContentExtensionsForURL): Only consult with the injected bundle about whether
or not we should use content extensions for this URL.
* WebProcess/WebCoreSupport/WebFrameLoaderClient.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205407 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoSilence -Wparentheses warning triggered by r205266
mcatanzaro@igalia.com [Sat, 3 Sep 2016 15:21:54 +0000 (15:21 +0000)]
Silence -Wparentheses warning triggered by r205266

Unreviewed

* platform/URL.cpp:
(WebCore::URL::URL):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205406 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUse ASCIILiteral in some more places
commit-queue@webkit.org [Sat, 3 Sep 2016 14:52:33 +0000 (14:52 +0000)]
Use ASCIILiteral in some more places
https://bugs.webkit.org/show_bug.cgi?id=161557

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2016-09-03
Reviewed by Darin Adler.

Source/JavaScriptCore:

* runtime/TypeSet.h:
(JSC::StructureShape::setConstructorName):

Source/WebCore:

* Modules/indexeddb/IDBDatabaseException.cpp:
(WebCore::IDBDatabaseException::getErrorName):
(WebCore::IDBDatabaseException::getErrorDescription):
* Modules/websockets/WebSocket.cpp:
(WebCore::WebSocket::binaryType):
* css/FontFace.cpp:
(WebCore::FontFace::stretch):
(WebCore::FontFace::unicodeRange):
(WebCore::FontFace::featureSettings):
* html/canvas/WebGLRenderingContextBase.cpp:
* html/parser/HTMLPreloadScanner.cpp:
(WebCore::TokenPreloadScanner::initiatorFor):
* loader/FormSubmission.cpp:
(WebCore::FormSubmission::Attributes::parseEncodingType):
* page/SecurityOrigin.cpp:
(WebCore::SecurityOrigin::toRawString):
* platform/graphics/cg/ImageBufferCG.cpp:
(WebCore::CGImageToDataURL):
(WebCore::ImageBuffer::toDataURL):
(WebCore::ImageDataToDataURL):
* svg/graphics/SVGImage.cpp:
(WebCore::SVGImage::filenameExtension):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205405 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoObject.preventExtensions(window) should throw a TypeError
cdumez@apple.com [Sat, 3 Sep 2016 14:44:59 +0000 (14:44 +0000)]
Object.preventExtensions(window) should throw a TypeError
https://bugs.webkit.org/show_bug.cgi?id=161554

Reviewed by Darin Adler.

Source/WebCore:

Object.preventExtensions(window) should throw a TypeError.

[[PreventExtensions]] should return false for Window:
- https://html.spec.whatwg.org/#windowproxy-preventextensions

EcmaScript says that Object.preventExtensions() should throw a TypeError
if [[PreventExtension]] returns false:
- https://tc39.github.io/ecma262/#sec-object.preventextensions

No new tests, updated existing test.

* bindings/js/JSDOMWindowCustom.cpp:
(WebCore::JSDOMWindow::preventExtensions):

LayoutTests:

* http/tests/security/preventExtensions-window-location-expected.txt:
* http/tests/security/preventExtensions-window-location.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205404 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Move WebKit2 WebInspector files to #pragma once
commit-queue@webkit.org [Sat, 3 Sep 2016 14:42:18 +0000 (14:42 +0000)]
Web Inspector: Move WebKit2 WebInspector files to #pragma once
https://bugs.webkit.org/show_bug.cgi?id=161550

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2016-09-03
Reviewed by Darin Adler.

* UIProcess/InspectorServer/WebInspectorServer.h:
* UIProcess/WebInspectorProxy.h:
* UIProcess/gtk/WebInspectorClientGtk.h:
* WebProcess/WebCoreSupport/WebInspectorClient.h:
* WebProcess/WebPage/WebInspector.h:
* WebProcess/WebPage/WebInspectorFrontendAPIDispatcher.h:
* WebProcess/WebPage/WebInspectorUI.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205403 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Address ESLint undefined variable errors
commit-queue@webkit.org [Sat, 3 Sep 2016 14:40:36 +0000 (14:40 +0000)]
Web Inspector: Address ESLint undefined variable errors
https://bugs.webkit.org/show_bug.cgi?id=161563

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2016-09-03
Reviewed by Darin Adler.

Source/WebInspectorUI:

* UserInterface/Controllers/CSSStyleManager.js:
(WebInspector.CSSStyleManager.protocolMediaSourceToEnum):
* UserInterface/Models/IssueMessage.js:
(WebInspector.IssueMessage):
* UserInterface/Protocol/InspectorBackend.js:

LayoutTests:

* inspector/protocol/inspector-backend-invocation-return-value-expected.txt:
* inspector/protocol/inspector-backend-invocation-return-value.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205402 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoWeb Inspector: Address ESLint undefined variable errors in UserInterface/Views
commit-queue@webkit.org [Sat, 3 Sep 2016 14:38:38 +0000 (14:38 +0000)]
Web Inspector: Address ESLint undefined variable errors in UserInterface/Views
https://bugs.webkit.org/show_bug.cgi?id=161565

Patch by Joseph Pecoraro <pecoraro@apple.com> on 2016-09-03
Reviewed by Darin Adler.

* UserInterface/Views/DataGrid.js:
(WebInspector.DataGrid.prototype.columnWidthsMap):
* UserInterface/Views/SourceCodeTextEditor.js:
(WebInspector.SourceCodeTextEditor.prototype.textEditorGutterContextMenu):
* UserInterface/Views/TextContentView.js:
(WebInspector.TextContentView.prototype._togglePrettyPrint):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205401 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoAlign meta element http-equiv="refresh" parsing with the HTML specification
cdumez@apple.com [Sat, 3 Sep 2016 14:09:58 +0000 (14:09 +0000)]
Align meta element http-equiv="refresh" parsing with the HTML specification
https://bugs.webkit.org/show_bug.cgi?id=161543

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Import corresponding test from W3C.

* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/parsing-expected.txt: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/parsing.html: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/;url=foo: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/__dir__.headers: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/foo: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/foo'bar: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/refresh.sub.html: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/ufoo: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/urfoo: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/url foo: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/urlfoo: Added.
* web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/support/x;url=foo: Added.

Source/WebCore:

Align meta element http-equiv="refresh" parsing with the HTML specification:
- https://html.spec.whatwg.org/multipage/semantics.html#attr-meta-http-equiv-refresh

Tests: imported/w3c/web-platform-tests/html/semantics/document-metadata/the-meta-element/pragma-directives/attr-meta-http-equiv-refresh/parsing.html

* dom/Document.cpp:
(WebCore::Document::processHttpEquiv):
* html/parser/HTMLParserIdioms.cpp:
(WebCore::parseHTTPRefreshInternal):
(WebCore::parseMetaHTTPEquivRefresh):
* html/parser/HTMLParserIdioms.h:
* loader/FrameLoader.cpp:
(WebCore::FrameLoader::receivedFirstData):
* platform/network/HTTPParsers.cpp:
(WebCore::skipWhiteSpace):
(WebCore::skipEquals):
(WebCore::parseHTTPRefresh):
(WebCore::parseXSSProtectionHeader):
(WebCore::skipValue): Deleted.
* platform/network/HTTPParsers.h:

LayoutTests:

Update layout test to only use HTML spaces in http-equiv="refresh"
content value.

* http/tests/misc/refresh-meta-with-newline.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205400 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agorun-webkit-tests should detect w3c test resource files
commit-queue@webkit.org [Sat, 3 Sep 2016 09:44:35 +0000 (09:44 +0000)]
run-webkit-tests should detect w3c test resource files
https://bugs.webkit.org/show_bug.cgi?id=161307

Patch by Youenn Fablet <youenn@apple.com> on 2016-09-03
Reviewed by Ryosuke Niwa.

LayoutTests/imported/w3c:

* resources/resource-files.json: Added.

Tools:

* Scripts/webkitpy/port/base.py:
(Port.__init__): Adding self._w3c_resource_files to store the list of resource files.
(Port.potential_test_names_from_expected_file):
(Port._real_tests):
(Port.is_w3c_resource_file): Computing whether a file is a resource file based on imported/w3c/resources/resource-files.json.
(Port._is_test_file): Updated to call Port.is_w3c_resource_file. _is_test_file is no longer static.
(Port): Deleted.
* Scripts/webkitpy/port/base_unittest.py:
(PortTest.test_additional_platform_directory):
(PortTest.test_find_no_paths_specified):
(PortTest.test_is_test_file): Updated to use non-static version of _is_test_file.
(PortTest.test_is_w3c_resource_file): Adding tests.

LayoutTests:

* TestExpectations: Removing resource files skip expectations.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205399 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Cocoa] Distinguish between paint advances and base advances
mmaxfield@apple.com [Sat, 3 Sep 2016 07:40:26 +0000 (07:40 +0000)]
[Cocoa] Distinguish between paint advances and base advances
https://bugs.webkit.org/show_bug.cgi?id=160892

Unreviewed.

Update test results.

* platform/mac-elcapitan/css2.1/t051202-c26-psudo-nest-00-c-expected.png: Removed.
* platform/mac-elcapitan/css2.1/t051202-c26-psudo-nest-00-c-expected.txt: Removed.
* platform/mac-elcapitan/css2.1/t1508-c527-font-00-b-expected.png: Removed.
* platform/mac-elcapitan/css2.1/t1508-c527-font-00-b-expected.txt: Removed.
* platform/mac-elcapitan/fast/inline/absolute-positioned-inline-in-centred-block-expected.png: Removed.
* platform/mac-elcapitan/fast/inline/absolute-positioned-inline-in-centred-block-expected.txt: Removed.
* platform/mac/css2.1/t051202-c26-psudo-nest-00-c-expected.png:
* platform/mac/css2.1/t051202-c26-psudo-nest-00-c-expected.txt:
* platform/mac/css2.1/t1508-c527-font-00-b-expected.png:
* platform/mac/css2.1/t1508-c527-font-00-b-expected.txt:
* platform/mac/fast/inline/absolute-positioned-inline-in-centred-block-expected.png:
* platform/mac/fast/inline/absolute-positioned-inline-in-centred-block-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205398 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[GTK] -Wmissing-field-initializers on WaylandCompositor.cpp:295
mcatanzaro@igalia.com [Sat, 3 Sep 2016 06:04:04 +0000 (06:04 +0000)]
[GTK] -Wmissing-field-initializers on WaylandCompositor.cpp:295
https://bugs.webkit.org/show_bug.cgi?id=161524

Reviewed by Carlos Garcia Campos.

* UIProcess/gtk/WaylandCompositor.cpp:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205397 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Cocoa] Distinguish between paint advances and base advances
mmaxfield@apple.com [Sat, 3 Sep 2016 05:52:44 +0000 (05:52 +0000)]
[Cocoa] Distinguish between paint advances and base advances
https://bugs.webkit.org/show_bug.cgi?id=160892

Reviewed by Simon Fraser.

Source/WebCore:

This patch introduces the concept of a layout (or "base") advance which is distinct
from a painting advance. In extremely complicated scripts such as Urdu, it is common
for a glyph advance to be negative in the horizontal direction, and have large advances
in the vertical direction. In particular, in cursive scripts, the glyph placement is
only indirectly related to where the actual characters lie. Conceptually, these glyph
locations are correct for painting, but are not correct when performing width
measurements.

In many text engines, glyph shaping actually can be split into two phases: adjusting
advances, and then placing glyphs relative to those advances. The secondary glyph
placement step is much more context-sensitive than the first step. In addition, when
multiple glyphs combine to form a character, it is common for one glyph to own the
full base advance for the character, and for the other glyphs in the character to
have zero base advances. (Then, in the glyph placement phase, the other glyphs get
placed all around.)

Because of the context-insensitivity of the base advances, it is valuable to use
these for text measurement. Then, when we want to paint, we should add in the extra
origins. This dramatically improves the layout of complex fonts like Noto Nastaliq.

This patch migrates WebKit to use this two-phase shaping.

No new tests just yet, because I have to create a font which exercises the
advanced glyph placement support.

* platform/graphics/GlyphBuffer.h:
(WebCore::GlyphBufferAdvance::setHeight):
(WebCore::GlyphBufferAdvance::setWidth): Deleted.
* platform/graphics/TextRun.h:
(WebCore::TextRun::TextRun):
(WebCore::TextRun::shouldDisableLayoutSpecificAdvances):
(WebCore::TextRun::setShouldDisableLayoutSpecificAdvances):
(WebCore::TextRun::spacingDisabled): Deleted.
(WebCore::TextRun::setCharacterScanForCodePath): Deleted.
* platform/graphics/cocoa/FontCascadeCocoa.mm:
(WebCore::FontCascade::getGlyphsAndAdvancesForComplexText):
* platform/graphics/mac/ComplexTextController.cpp:
(WebCore::ComplexTextController::ComplexTextController):
(WebCore::ComplexTextController::offsetForPosition):
(WebCore::ComplexTextController::collectComplexTextRuns):
(WebCore::ComplexTextController::ComplexTextRun::setIsNonMonotonic):
(WebCore::ComplexTextController::runWidthSoFarFraction):
(WebCore::ComplexTextController::advance):
(WebCore::ComplexTextController::adjustGlyphsAndAdvances):
* platform/graphics/mac/ComplexTextController.h:
(WebCore::ComplexTextController::ComplexTextRun::create):
(WebCore::ComplexTextController::ComplexTextRun::baseAdvances):
(WebCore::ComplexTextController::ComplexTextRun::glyphOrigins):
(WebCore::ComplexTextController::useLayoutSpecificAdvances):
(WebCore::ComplexTextController::finalRoundingWidth): Deleted.
(WebCore::ComplexTextController::ComplexTextRun::advances): Deleted.
* platform/graphics/mac/ComplexTextControllerCoreText.mm:
(SOFT_LINK):
(WebCore::ComplexTextController::ComplexTextRun::ComplexTextRun):
(WebCore::ComplexTextController::collectComplexTextRunsForCharacters):
* platform/spi/cocoa/CoreTextSPI.h:

LayoutTests:

Update tests. There are some expected (small) changes in metrics due to this patch.

* platform/mac-elcapitan/css2.1/t051202-c26-psudo-nest-00-c-expected.png: Copied from LayoutTests/platform/mac/css2.1/t051202-c26-psudo-nest-00-c-expected.png.
* platform/mac-elcapitan/css2.1/t051202-c26-psudo-nest-00-c-expected.txt: Copied from LayoutTests/platform/mac/css2.1/t051202-c26-psudo-nest-00-c-expected.txt.
* platform/mac-elcapitan/css2.1/t1508-c527-font-00-b-expected.png: Copied from LayoutTests/platform/mac/css2.1/t1508-c527-font-00-b-expected.png.
* platform/mac-elcapitan/css2.1/t1508-c527-font-00-b-expected.txt: Added.
* platform/mac-elcapitan/fast/inline/absolute-positioned-inline-in-centred-block-expected.png: Copied from LayoutTests/platform/mac/fast/inline/absolute-positioned-inline-in-centred-block-expected.png.
* platform/mac-elcapitan/fast/inline/absolute-positioned-inline-in-centred-block-expected.txt: Copied from LayoutTests/platform/mac/fast/inline/absolute-positioned-inline-in-centred-block-expected.txt.
* platform/mac/css2.1/t051202-c26-psudo-nest-00-c-expected.png:
* platform/mac/css2.1/t051202-c26-psudo-nest-00-c-expected.txt:
* platform/mac/css2.1/t1508-c527-font-00-b-expected.png:
* platform/mac/css2.1/t1508-c527-font-00-b-expected.txt: Added.
* platform/mac/fast/inline/absolute-positioned-inline-in-centred-block-expected.png:
* platform/mac/fast/inline/absolute-positioned-inline-in-centred-block-expected.txt:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205396 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years ago[Threaded Compositor] Move the viewport controller off the compositing thread
carlosgc@webkit.org [Sat, 3 Sep 2016 05:45:10 +0000 (05:45 +0000)]
[Threaded Compositor] Move the viewport controller off the compositing thread
https://bugs.webkit.org/show_bug.cgi?id=161532

Reviewed by Michael Catanzaro.

While working on bug #161242 I've realized that having the view port controller in the compositing thread makes
everything more complex. The viewport controller receives changes about things like contents size, viewport
size, etc. and uses that information to compute the visible contents rect and page scale factor. Then it
notifies back to main thread about the computed visible contents rect and page scale. Those computations are not
heave at all, so they could be done in the main thread and we would avoid communications between the main and
compositing thread in both directions. The main thread needs the visible contents rect to notify the compositing
coordinator and the page cale to scale the page in case of pixed layout. But the compositing thread only needs
to know the effective scale and scroll position. So, instead of going to the compositing thread after every
change that might update the visible contents rect and page scale factor, we could do those calculations in the
main thread and only notify the compositing thread about the actual changes in the scroll position and effective scale.

* Shared/CoordinatedGraphics/CoordinatedGraphicsScene.cpp:
(WebKit::CoordinatedGraphicsScene::createTilesIfNeeded): Return early if backingStore is nullptr, which can
happen if the layer shouldn't have a backing store and was removed by the previous call to prepareContentBackingStore().
(WebKit::CoordinatedGraphicsScene::updateTilesIfNeeded): Ditto.
* Shared/CoordinatedGraphics/SimpleViewportController.cpp:
(WebKit::SimpleViewportController::SimpleViewportController): Remove the client since we no longer need to
notify about changes.
(WebKit::SimpleViewportController::didChangeViewportSize): Remove call to syncVisibleContents().
(WebKit::SimpleViewportController::didChangeContentsSize): Ditto.
(WebKit::SimpleViewportController::didChangeViewportAttributes): Ditto.
(WebKit::SimpleViewportController::didScroll): Removed unused scrollBy methods and renamed scrollTo as
didiScroll for consistency. Save the position without calling boundContentsPosition, because that's already
donde when the position is used to compute the contents visible rectangle.
(WebKit::SimpleViewportController::visibleContentsRect): No need to notify about the changes.
(WebKit::SimpleViewportController::visibleContentsSize): Deleted.
* Shared/CoordinatedGraphics/SimpleViewportController.h:
* Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.cpp:
(WebKit::ThreadedCompositor::create): Pass a reference to the client instead of a pointer. It's no longer
possible to have a valid pointer when the object has been destroyed, so we can better use a reference now.
(WebKit::ThreadedCompositor::ThreadedCompositor): Ditto.
(WebKit::ThreadedCompositor::~ThreadedCompositor): Remove assert.
(WebKit::ThreadedCompositor::invalidate): No need to invalidate the client.
(WebKit::ThreadedCompositor::setScaleFactor): Set the effective scale factor that should be used for rendering.
(WebKit::ThreadedCompositor::setScrollPosition): Set the current scroll position and effective scale factor.
(WebKit::ThreadedCompositor::setViewportSize): Set the viewport size and effective scale factor.
(WebKit::ThreadedCompositor::renderNextFrame): Update m_client use that is no longer a pointer.
(WebKit::ThreadedCompositor::commitScrollOffset): Ditto.
(WebKit::ThreadedCompositor::renderLayerTree): Call glViewport after a resize and use m_viewportSize,
m_scrollPosition and m_scaleFactor members.
(WebKit::ThreadedCompositor::didChangeVisibleRect): Deleted.
* Shared/CoordinatedGraphics/threadedcompositor/ThreadedCompositor.h:
* WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.cpp:
(WebKit::ThreadedCoordinatedLayerTreeHost::ThreadedCoordinatedLayerTreeHost): Pass the compositor client as a
reference to ThreadedCompositor constructor.
(WebKit::ThreadedCoordinatedLayerTreeHost::scrollNonCompositedContents): Update the viewport and call didChangeViewport().
(WebKit::ThreadedCoordinatedLayerTreeHost::contentsSizeChanged): Ditto.
(WebKit::ThreadedCoordinatedLayerTreeHost::deviceOrPageScaleFactorChanged): Pass the effective scale factor to
the compositor.
(WebKit::ThreadedCoordinatedLayerTreeHost::sizeDidChange): Update the viewport, the compositor and call didChangeViewport().
(WebKit::ThreadedCoordinatedLayerTreeHost::didChangeViewportProperties): Update the viewport and call didChangeViewport().
(WebKit::ThreadedCoordinatedLayerTreeHost::didChangeViewport): Notify the compositing coordinator about the new
visible contents rectangle, and update the threaded compositor if needed.
* WebProcess/WebPage/CoordinatedGraphics/ThreadedCoordinatedLayerTreeHost.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205395 268f45cc-cd09-0410-ab3c-d52691b4dbfc

3 years agoUnreviewed. Fix archive-built-product step in GTK+ bots after r205280.
carlosgc@webkit.org [Sat, 3 Sep 2016 05:24:13 +0000 (05:24 +0000)]
Unreviewed. Fix archive-built-product step in GTK+ bots after r205280.

* BuildSlaveSupport/built-product-archive:
(archiveBuiltProduct): Do not copy to the archive the files needed to run GObject DOM bindings API breaks tests
that no longer exist.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@205394 268f45cc-cd09-0410-ab3c-d52691b4dbfc