ChangeLog

2011-02-18  Gavin Barraclough  <barraclough@apple.com>

        Reviewed by Sam Weinig.

        Bug 54786 - Devirtualize JSCell::classInfo()

        Instead of making a virtual function call, add a pointer to the ClassInfo
        onto Structure.

        This removes a virtual function call, and paves the way towards removing all
        the createStructure methods, and StructureFlags/AnonymousSlotCount properties
        (these should be able to move onto ClassInfo).

        Calls to Structure::create must now pass a pointer to the ClassInfo for the
        structure. All objects now have a ClassInfo pointer, non-object cell types
        still do not.

        Changes are most mechanical, involving three steps:
            * Remove virtual classInfo() methods.
            * Add &s_info parameter to calls to Structure::create.
            * Rename ClassInfo static members on classes from 'info' to 's_info',
              for consistency.

        * API/JSCallbackConstructor.cpp:
        * API/JSCallbackConstructor.h:
        * API/JSCallbackFunction.cpp:
        * API/JSCallbackFunction.h:
        * API/JSCallbackObject.cpp:
        * API/JSCallbackObject.h:
        * API/JSCallbackObjectFunctions.h:
        * API/JSObjectRef.cpp:
        * API/JSValueRef.cpp:
        * API/JSWeakObjectMapRefPrivate.cpp:
        * JavaScriptCore.exp:
        * debugger/Debugger.cpp:
        * debugger/DebuggerActivation.h:
        * debugger/DebuggerCallFrame.cpp:
        * interpreter/Interpreter.cpp:
        * jit/JITCall32_64.cpp:
        * jit/JITOpcodes.cpp:
        * jit/JITStubs.cpp:
        * profiler/Profiler.cpp:
        * runtime/Arguments.cpp:
        * runtime/Arguments.h:
        * runtime/ArrayConstructor.cpp:
        * runtime/ArrayPrototype.cpp:
        * runtime/ArrayPrototype.h:
        * runtime/BooleanObject.cpp:
        * runtime/BooleanObject.h:
        * runtime/BooleanPrototype.cpp:
        * runtime/DateConstructor.cpp:
        * runtime/DateInstance.cpp:
        * runtime/DateInstance.h:
        * runtime/DatePrototype.cpp:
        * runtime/DatePrototype.h:
        * runtime/ErrorInstance.cpp:
        * runtime/ErrorInstance.h:
        * runtime/ErrorPrototype.cpp:
        * runtime/FunctionPrototype.cpp:
        * runtime/FunctionPrototype.h:
        * runtime/GetterSetter.h:
        * runtime/GlobalEvalFunction.h:
        * runtime/InternalFunction.cpp:
        * runtime/InternalFunction.h:
        * runtime/JSAPIValueWrapper.h:
        * runtime/JSActivation.cpp:
        * runtime/JSActivation.h:
        * runtime/JSArray.cpp:
        * runtime/JSArray.h:
        * runtime/JSByteArray.cpp:
        * runtime/JSByteArray.h:
        * runtime/JSCell.cpp:
        * runtime/JSCell.h:
        * runtime/JSFunction.cpp:
        * runtime/JSFunction.h:
        * runtime/JSGlobalData.cpp:
        * runtime/JSGlobalObject.cpp:
        * runtime/JSGlobalObject.h:
        * runtime/JSNotAnObject.h:
        * runtime/JSONObject.cpp:
        * runtime/JSONObject.h:
        * runtime/JSObject.cpp:
        * runtime/JSObject.h:
        * runtime/JSObjectWithGlobalObject.h:
        * runtime/JSPropertyNameIterator.h:
        * runtime/JSStaticScopeObject.h:
        * runtime/JSString.h:
        * runtime/JSVariableObject.h:
        * runtime/JSWrapperObject.h:
        * runtime/JSZombie.cpp:
        * runtime/JSZombie.h:
        * runtime/Lookup.cpp:
        * runtime/MathObject.cpp:
        * runtime/MathObject.h:
        * runtime/NativeErrorConstructor.cpp:
        * runtime/NativeErrorConstructor.h:
        * runtime/NumberConstructor.cpp:
        * runtime/NumberConstructor.h:
        * runtime/NumberObject.cpp:
        * runtime/NumberObject.h:
        * runtime/NumberPrototype.cpp:
        * runtime/ObjectConstructor.cpp:
        * runtime/ObjectPrototype.cpp:
        * runtime/RegExpConstructor.cpp:
        * runtime/RegExpConstructor.h:
        * runtime/RegExpObject.cpp:
        * runtime/RegExpObject.h:
        * runtime/RegExpPrototype.cpp:
        * runtime/ScopeChain.cpp:
        * runtime/StringConstructor.cpp:
        * runtime/StringObject.cpp:
        * runtime/StringObject.h:
        * runtime/StringObjectThatMasqueradesAsUndefined.h:
        * runtime/StringPrototype.cpp:
        * runtime/StringPrototype.h:
        * runtime/Structure.cpp:
        * runtime/Structure.h:

2011-02-19  David Kilzer  <ddkilzer@apple.com>

        <http://webkit.org/b/54808> Change jsc target to build directly into JavaScriptCore.framework/Resources/jsc

        Reviewed by Dan Bernstein.

        * Configurations/Base.xcconfig: Added
        JAVASCRIPTCORE_FRAMEWORKS_DIR variable.
        * Configurations/JavaScriptCore.xcconfig: Used
        JAVASCRIPTCORE_FRAMEWORKS_DIR to define INSTALL_PATH.
        * JavaScriptCore.xcodeproj/project.pbxproj: Set the INSTALL_PATH
        for Production configuration of jsc target.
        (Copy Into Framework): Removed old build phase.
        (Fix Framework Reference): Renamed build phase to "Copy Into
        Framework".  Added "set -x" call to make the script print the
        commands it is running.  Added code to exit early for Production
        builds since this was never intended for them.  Added code to
        copy jsc into the JavaScriptCore.framework/Resources directory.

2011-02-19  Siddharth Mathur  <siddharth.mathur@nokia.com>

        Reviewed by Laszlo Gombos.

        [Symbian] OSAllocator implementation for Symbian OS. 
        Manages both data and code region requests. V8 and Sunspider tested
        OK with interpreter. Not tested with JSC JIT yet as it has unrelated
        failures. Also no thread safety yet.
        https://bugs.webkit.org/show_bug.cgi?id=51128

        * JavaScriptCore.pri: removed HAL linkage
        * wtf/Bitmap.h:
        (WTF::::findRunOfZeros): find run of zeros in a bitmap. quick n dirty
        * wtf/OSAllocator.h:
        (WTF::OSAllocator::decommitAndRelease): decommit explicitly 
        * wtf/OSAllocatorSymbian.cpp: Impl. of OSAllocator interface 
        (WTF::allocateCodeChunk): utility for code chunks
        (WTF::deallocateCodeChunk): utility for code chunks
        (WTF::dataAllocatorInstance): getter for data allocator instance
        (WTF::OSAllocator::reserveUncommitted):
        (WTF::OSAllocator::releaseDecommitted):
        (WTF::OSAllocator::commit):
        (WTF::OSAllocator::decommit):
        (WTF::OSAllocator::reserveAndCommit):
        (WTF::PageAllocatorSymbian::PageAllocatorSymbian): maps requests 
        to one underlying Symbian chunk
        (WTF::PageAllocatorSymbian::~PageAllocatorSymbian):
        (WTF::PageAllocatorSymbian::reserve):
        (WTF::PageAllocatorSymbian::release):
        (WTF::PageAllocatorSymbian::commit):
        (WTF::PageAllocatorSymbian::decommit):
        (WTF::PageAllocatorSymbian::contains):
        * wtf/PageAllocatorSymbian.h: Added.
        (WTF::SymbianChunk::SymbianChunk): wrapper around RChunk  
        (WTF::SymbianChunk::~SymbianChunk):
        (WTF::SymbianChunk::contains):
        
2011-02-19  Yong Li  <yoli@rim.com>

        Reviewed by Eric Seidel.

        https://bugs.webkit.org/show_bug.cgi?id=54687
        When being built with armcc, "int" bit fields are treated as
        unsigned integers, which will fail the comparisons like "m_offset == -1".
        Using "signed" fixes the problem.

        * assembler/ARMAssembler.h:
        * assembler/ARMv7Assembler.h:

2011-02-18  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Made MarkedSpace block iteration size-class agnostic
        https://bugs.webkit.org/show_bug.cgi?id=54792
        
        SunSpider reports no change.

        * runtime/MarkedSpace.cpp:
        (JSC::MarkedSpace::clearMarks):
        (JSC::MarkedSpace::sweep):
        (JSC::MarkedSpace::objectCount):
        (JSC::MarkedSpace::size):
        (JSC::MarkedSpace::capacity):
        * runtime/MarkedSpace.h:
        (JSC::MarkedSpace::forEach): Iterate blocks in hashing order instead of
        size class list order. This is a much simpler convention in a world
        of many different size classes.

2011-02-18  Kristian Amlie  <kristian.amlie@nokia.com>

        Reviewed by Andreas Kling.

        Added friend exception to Qt platform, which also compiles Windows.

        QtWebKit fails to compile on Windows XP with msvc-2008
        https://bugs.webkit.org/show_bug.cgi?id=54746

        * bytecode/CodeBlock.h:
        * runtime/RegExpObject.h:

2011-02-18  Geoffrey Garen  <ggaren@apple.com>

        (Rolled back in r79022 with crash fixed.)

        Reviewed by Sam Weinig.

        Use hashing instead of linear search in the conservative pointer test
        https://bugs.webkit.org/show_bug.cgi?id=54767
        
        SunSpider reports no change.

        * runtime/MarkedSpace.cpp:
        (JSC::MarkedSpace::destroy): No need to explicitly clear the blocks array,
        since freeBlock removes items for us.

        (JSC::MarkedSpace::freeBlock): Fixed a typo that always removed the last
        block from the block set instead of the block being freed. Changed to
        remove a block from our data structures before deallocating it, since
        this is slightly cleaner.

        * runtime/MarkedSpace.h:
        (JSC::MarkedSpace::contains): Variable-sized objects will use more,
        smaller blocks, so it's important for the contains check not to be O(n)
        in the number of blocks.

2011-02-18  chris reiss  <christopher.reiss@nokia.com>

        Reviewed by Andreas Kling.

        REGRESSION: Date.parse("Tue Nov 23 20:40:05 2010 GMT") returns NaN
        https://bugs.webkit.org/show_bug.cgi?id=49989

        updated test fast/js/script-tests/date-parse-test.js

        * wtf/DateMath.cpp:
        (WTF::parseDateFromNullTerminatedCharacters):

2011-02-18  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r79022.
        http://trac.webkit.org/changeset/79022
        https://bugs.webkit.org/show_bug.cgi?id=54775

        It broke the whole world (Requested by Ossy on #webkit).

        * runtime/MarkedSpace.h:
        (JSC::MarkedSpace::contains):

2011-02-18  Yael Aharon  <yael.aharon@nokia.com>

        Reviewed by Dave Hyatt.

        Add support for dir=auto
        https://bugs.webkit.org/show_bug.cgi?id=50916

        Change defaultWritingDirection() to return if the writing direction
        was determined from a letter with strong directionality or not.

        * JavaScriptCore.exp:
        * JavaScriptCore.order:
        * wtf/text/StringImpl.cpp:
        (WTF::StringImpl::defaultWritingDirection):
        * wtf/text/StringImpl.h:
        * wtf/text/WTFString.h:
        (WTF::String::defaultWritingDirection):

2011-02-18  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Use hashing instead of linear search in the conservative pointer test
        https://bugs.webkit.org/show_bug.cgi?id=54767
        
        SunSpider reports no change.

        * runtime/MarkedSpace.h:
        (JSC::MarkedSpace::contains): Variable-sized objects will use more,
        smaller blocks, so it's important for the contains check not to be O(n)
        in the number of blocks.

2011-02-18  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Tightened some usage accounting code in MarkedSpace
        https://bugs.webkit.org/show_bug.cgi?id=54761
        
        SunSpider reports no change.

        * runtime/Heap.cpp:
        (JSC::Heap::Heap): Initialize the marked space high water mark on
        construction, instead of relying on some implicit subtleties to make
        not initializing it work out OK.

        * runtime/Heap.h: Fixed up includes.

        * runtime/MarkedBlock.h: Made firstAtom() static so clients can call it
        even without having allocated a block.

        * runtime/MarkedSpace.cpp:
        (JSC::MarkedSpace::MarkedSpace): Don't pre-allocate a block, since this
        would be prohibitively expensive with multiple size classes.

        (JSC::MarkedSpace::allocateBlock):
        (JSC::MarkedSpace::freeBlock): Track allocated blocks in a hash set,
        since linear search in the contains check will be prohibitively
        expensive once we're using lots of smaller blocks.

        (JSC::MarkedSpace::allocate): Don't assume that we always have a block
        allocated, since we don't anymore. (See above.)

        (JSC::MarkedSpace::reset):
        * runtime/MarkedSpace.h: Updated for changes mentioned above.

2011-02-17  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Made object allocation secretly variable-sized (Shhhh!)
        https://bugs.webkit.org/show_bug.cgi?id=54721
        
        SunSpider reports no change.
        
        Internally, MarkedBlock now makes variable-sized allocations, even
        though MarkedSpace doesn't take advantage of this yet.

        * runtime/MarkedBlock.cpp:
        (JSC::MarkedBlock::MarkedBlock): No need to ASSERT that allocations are
        fixed-sized.

        * runtime/MarkedBlock.h: Shrunk the atom size so we can allocate things
        that are not multiples of 64 bytes.

2011-02-17  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Fixed some math errors when when using variable-sized cells
        https://bugs.webkit.org/show_bug.cgi?id=54717
        
        SunSpider reports no change.
        
        Computer Science Barbie says, "Math is not so hard afterall!"

        * runtime/JSCell.h:
        (JSC::JSCell::MarkedBlock::allocate): Round up when calculating the
        minimum number of atoms required for a cell, since rounding down
        will get you splinched.

        * runtime/MarkedBlock.cpp:
        (JSC::MarkedBlock::MarkedBlock):
        (JSC::MarkedBlock::sweep):
        * runtime/MarkedBlock.h:
        (JSC::MarkedBlock::forEach): Changed a bunch of != tests to < tests
        because m_endAtom is actually a fuzzy end -- iterating from firstAtom()
        may not hit m_endAtom exactly.

2011-02-17  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        A little more abstraction for MarkedSpace::contains
        https://bugs.webkit.org/show_bug.cgi?id=54715

        * runtime/MarkedBlock.h:
        (JSC::MarkedBlock::contains): Added a contains function, so MarkedSpace
        doesn't have to know how MarkedBlock tracks containment internally.

        * runtime/MarkedSpace.h:
        (JSC::MarkedSpace::contains): Call through to MarkedBlock to figure out
        if a cell that seems to be in a block is valid.

2011-02-17  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Sam Weinig.

        Removed the invariant that the last cell in a block is always marked
        https://bugs.webkit.org/show_bug.cgi?id=54713
        
        SunSpider reports no change.
        
        This adds one branch to allocation, but simplifies the mark invariant,
        especially in a world of variable-sized cells. Now, it really is true
        that any cell whose mark bit is set is a valid, live cell whose
        constructor has run and whose destructor has not run.

        * runtime/JSCell.h: 
        (JSC::JSCell::MarkedBlock::allocate): Changed this do-while into a while
        since we can no longer rely on a set mark bit to break out of this loop
        before it reaches the end of the block.

        * runtime/MarkedBlock.cpp:
        (JSC::MarkedBlock::MarkedBlock):
        (JSC::MarkedBlock::sweep): 
        * runtime/MarkedBlock.h:
        (JSC::MarkedBlock::isEmpty):
        (JSC::MarkedBlock::clearMarks):
        (JSC::MarkedBlock::markCount):
        (JSC::MarkedBlock::forEach): No need to set a special last mark bit.

2011-02-17  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r78856 and r78907.
        http://trac.webkit.org/changeset/78856
        http://trac.webkit.org/changeset/78907
        https://bugs.webkit.org/show_bug.cgi?id=54705

        These seem to break tests on 32-bit builds. (Requested by
        aroben on #webkit).

        * JavaScriptCore.xcodeproj/project.pbxproj:
        * collector/handles/Global.h:
        (JSC::Global::internalSet):
        * collector/handles/Handle.h:
        (JSC::HandleTypes::getFromSlot):
        (JSC::HandleTypes::toJSValue):
        (JSC::HandleTypes::validateUpcast):
        (JSC::HandleConverter::operator->):
        (JSC::HandleConverter::operator*):
        (JSC::Handle::Handle):
        (JSC::Handle::get):
        * runtime/JSArray.cpp:
        (JSC::JSArray::sortNumeric):
        * runtime/JSObject.h:
        (JSC::JSObject::inlineGetOwnPropertySlot):
        * runtime/SlotAccessor.h: Removed.
        * runtime/WeakGCPtr.h:
        (JSC::WeakGCPtr::get):
        (JSC::WeakGCPtr::internalSet):
        * runtime/WriteBarrier.h:
        (JSC::DeprecatedPtr::DeprecatedPtr):
        (JSC::DeprecatedPtr::get):
        (JSC::DeprecatedPtr::operator*):
        (JSC::DeprecatedPtr::operator->):
        (JSC::DeprecatedPtr::slot):
        (JSC::DeprecatedPtr::operator UnspecifiedBoolType*):
        (JSC::DeprecatedPtr::operator!):
        (JSC::WriteBarrierBase::set):
        (JSC::WriteBarrierBase::get):
        (JSC::WriteBarrierBase::operator*):
        (JSC::WriteBarrierBase::operator->):
        (JSC::WriteBarrierBase::clear):
        (JSC::WriteBarrierBase::slot):
        (JSC::WriteBarrierBase::operator UnspecifiedBoolType*):
        (JSC::WriteBarrierBase::operator!):
        (JSC::WriteBarrierBase::setWithoutWriteBarrier):
        (JSC::WriteBarrier::WriteBarrier):

2011-02-17  Csaba Osztrogonác  <ossy@webkit.org>

        Unreviewed.

        [Qt] Buildfix.

        * wtf/RetainPtr.h: Add missing PLATFORM(CF) guard.

2011-02-17  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Oliver Hunt.

        Made MarkedBlock variable-sized
        https://bugs.webkit.org/show_bug.cgi?id=54692
        
        SunSpider reports no change.
        
        Each MarkedBlock is now composed of a set of fixed-sized atoms, with one
        mark bit per atom. A given cell may be composed of one or more atoms.
        
        * runtime/Heap.cpp:
        (JSC::Heap::allocate): Made fixed-sizedness a property of MarkedSpace,
        bubbling it up from MarkedBlock, since MarkedBlock now supports variable-
        sizedness.

        * runtime/JSCell.h:
        (JSC::JSCell::MarkedBlock::allocate): Removed use of CELLS_PER_BLOCK and
        (implicit) one constants -- these quantities are not constant anymore.
        Updated for switch from cell to atom.

        * runtime/MarkedBlock.cpp:
        (JSC::MarkedBlock::create):
        (JSC::MarkedBlock::destroy):
        (JSC::MarkedBlock::MarkedBlock):
        (JSC::MarkedBlock::sweep):
        * runtime/MarkedBlock.h:
        (JSC::MarkedBlock::firstAtom):
        (JSC::MarkedBlock::atoms):
        (JSC::MarkedBlock::isAtomAligned):
        (JSC::MarkedBlock::blockFor):
        (JSC::MarkedBlock::isEmpty):
        (JSC::MarkedBlock::clearMarks):
        (JSC::MarkedBlock::size):
        (JSC::MarkedBlock::capacity):
        (JSC::MarkedBlock::atomNumber):
        (JSC::MarkedBlock::isMarked):
        (JSC::MarkedBlock::testAndSetMarked):
        (JSC::MarkedBlock::setMarked):
        (JSC::MarkedBlock::forEach): Same as above. Also removed use of CELL_SIZE
        and BLOCK_SIZE, and switched away from calling arbitrary pointers cells.

        * runtime/MarkedSpace.cpp:
        (JSC::MarkedSpace::MarkedSpace):
        (JSC::MarkedSpace::allocateBlock):
        (JSC::MarkedSpace::allocate):
        (JSC::MarkedSpace::reset):
        * runtime/MarkedSpace.h:
        (JSC::CollectorHeap::CollectorHeap):
        (JSC::MarkedSpace::contains): Updated for renames. Made fixed-sizedness
        a property of MarkedSpace.

2011-02-17  Oliver Hunt  <oliver@apple.com>

        Attempt to fix windows build

        * runtime/WriteBarrier.h:

2011-02-17  Oliver Hunt  <oliver@apple.com>

        Reviewed by Geoffrey Garen.

        Refactor WriteBarrier and DeprecatedPtr to have less code duplication.
        https://bugs.webkit.org/show_bug.cgi?id=54608

        Make use of the tricks used for Handle, et al to avoid duplicating all
        of the logic for DeprecatedPtr and WriteBarrier simply to support known
        vs. unknown types.

        * JavaScriptCore.xcodeproj/project.pbxproj:
        * collector/handles/Global.h:
        (JSC::Global::internalSet):
        * collector/handles/Handle.h:
        (JSC::Handle::Handle):
        (JSC::Handle::get):
        * runtime/JSArray.cpp:
        (JSC::JSArray::sortNumeric):
        * runtime/JSObject.h:
        (JSC::JSObject::inlineGetOwnPropertySlot):
        * runtime/SlotAccessor.h: Added.
        (JSC::SlotTypes::getFromBaseType):
        (JSC::SlotTypes::convertToBaseType):
        (JSC::SlotTypes::getFromSlot):
        (JSC::SlotTypes::toJSValue):
        (JSC::SlotTypes::validateUpcast):
        (JSC::SlotAccessor::operator->):
        (JSC::SlotAccessor::operator*):
        * runtime/WeakGCPtr.h:
        (JSC::WeakGCPtr::get):
        (JSC::WeakGCPtr::internalSet):
        * runtime/WriteBarrier.h:
        (JSC::DeprecatedPtr::DeprecatedPtr):
        (JSC::DeprecatedPtr::get):
        (JSC::DeprecatedPtr::slot):
        (JSC::DeprecatedPtr::operator=):
        (JSC::WriteBarrierTranslator::convertToStorage):
        (JSC::WriteBarrierTranslator::convertFromStorage):
        (JSC::WriteBarrierBase::set):
        (JSC::WriteBarrierBase::get):
        (JSC::WriteBarrierBase::clear):
        (JSC::WriteBarrierBase::slot):
        (JSC::WriteBarrierBase::operator UnspecifiedBoolType*):
        (JSC::WriteBarrierBase::operator!):
        (JSC::WriteBarrierBase::setWithoutWriteBarrier):
        (JSC::WriteBarrier::WriteBarrier):

2011-02-17  Kevin Ollivier  <kevino@theolliviers.com>

        [wx] Revert incorrect blind fix and restore previous working code.

        * wtf/wx/StringWx.cpp:
        (WTF::String::String):

2011-02-16  Geoffrey Garen  <ggaren@apple.com>

        Reviewed by Maciej Stachowiak.

        Intermittent crashes beneath MarkStack::drain
        https://bugs.webkit.org/show_bug.cgi?id=54614
        <rdar://problem/8971070>
        
        The crashes were caused by a GC happening after the global object's
        property table had grown (due to compilation), but before the properties
        had been fully initialized by program execution.

        * bytecompiler/BytecodeGenerator.cpp:
        (JSC::BytecodeGenerator::BytecodeGenerator): Explicitly resize the global
        object's register storage immediately, without waiting for program
        execution to do it for us. This ensures that the global object's count
        of global variables is consistent with the size of its global variable
        storage at all times, and it ensures that all global variables are
        properly initialized from the get-go.

        * runtime/JSGlobalObject.cpp:
        (JSC::JSGlobalObject::resizeRegisters):
        * runtime/JSGlobalObject.h: Added a helper function for growing the
        global object's register storage, and initializing new registers.

== Rolled over to ChangeLog-2011-02-16 ==