2011-02-15 Gavin Barraclough <barraclough@apple.com>

Reviewed by Geoff Garen.

Bug 54524 - Allow JSObject to fully utilize cell's capacity for inline storage.

Currently JSObject is both directly instantiated for regular JS objects, and

derived to implement subtypes. A consequence of this is that we need to ensure

that sufficient space from the cell is left unused and available for any data

members that will be introduced by subclasses of JSObject. By restructuring

the internal storage array out of JSObject we can increase the size in the

internal storage for regular objects.

Add classes JSFinalObject and JSNonFinalObject. JSNonFinalObject retains as

much additional capacity as is currently available to allow for data members

in subclasses. JSFinalObject utilizes all available space for internal storage,

and only allows construction through JSFinalObject::create().

The additional storage made available in the JSObject means that we need no

longer rely on a union of the internal storage with a pointer to storage that

is only valid for external storage. This means we can go back to always having

a valid pointer to property storage, regardless of whether this is internal or

external. This simplifies some cases of access to the array from C code, and

significantly simplifies JIT access, since repatching no longer needs to be

able to change between a load of the storage pointer / a LEA of the internal

storage.

* API/JSObjectRef.cpp:

(JSObjectMake):

* assembler/ARMAssembler.h:

* assembler/ARMv7Assembler.h:

* assembler/AbstractMacroAssembler.h:

(JSC::AbstractMacroAssembler::repatchPointer):

* assembler/MIPSAssembler.h:

* assembler/MacroAssemblerARM.h:

* assembler/MacroAssemblerARMv7.h:

* assembler/MacroAssemblerMIPS.h:

* assembler/MacroAssemblerX86.h:

* assembler/MacroAssemblerX86_64.h:

* assembler/RepatchBuffer.h:

* assembler/X86Assembler.h:

* debugger/DebuggerActivation.cpp:

(JSC::DebuggerActivation::DebuggerActivation):

* debugger/DebuggerActivation.h:

* interpreter/Interpreter.cpp:

(JSC::Interpreter::privateExecute):

* jit/JIT.h:

* jit/JITOpcodes.cpp:

(JSC::JIT::emit_op_resolve_global):

* jit/JITOpcodes32_64.cpp:

(JSC::JIT::emit_op_resolve_global):

* jit/JITPropertyAccess.cpp:

(JSC::JIT::compileGetDirectOffset):

(JSC::JIT::emit_op_get_by_pname):

(JSC::JIT::compileGetByIdHotPath):

(JSC::JIT::emit_op_put_by_id):

(JSC::JIT::compilePutDirectOffset):

(JSC::JIT::patchGetByIdSelf):

(JSC::JIT::patchPutByIdReplace):

(JSC::JIT::privateCompileGetByIdProto):

(JSC::JIT::privateCompileGetByIdSelfList):

(JSC::JIT::privateCompileGetByIdProtoList):

(JSC::JIT::privateCompileGetByIdChainList):

(JSC::JIT::privateCompileGetByIdChain):

* jit/JITPropertyAccess32_64.cpp:

(JSC::JIT::compileGetByIdHotPath):

(JSC::JIT::emit_op_put_by_id):

(JSC::JIT::compilePutDirectOffset):

(JSC::JIT::compileGetDirectOffset):

(JSC::JIT::patchGetByIdSelf):

(JSC::JIT::patchPutByIdReplace):

(JSC::JIT::privateCompileGetByIdProto):

(JSC::JIT::privateCompileGetByIdSelfList):

(JSC::JIT::privateCompileGetByIdProtoList):

(JSC::JIT::privateCompileGetByIdChainList):

(JSC::JIT::privateCompileGetByIdChain):

(JSC::JIT::emit_op_get_by_pname):

* jit/JITStubs.cpp:

(JSC::DEFINE_STUB_FUNCTION):

* runtime/Arguments.h:

(JSC::Arguments::Arguments):

* runtime/ErrorInstance.cpp:

(JSC::ErrorInstance::ErrorInstance):

* runtime/ErrorInstance.h:

* runtime/ExceptionHelpers.cpp:

(JSC::InterruptedExecutionError::InterruptedExecutionError):

(JSC::TerminatedExecutionError::TerminatedExecutionError):

* runtime/JSArray.cpp:

(JSC::JSArray::JSArray):

* runtime/JSArray.h:

* runtime/JSByteArray.cpp:

(JSC::JSByteArray::JSByteArray):

* runtime/JSByteArray.h:

(JSC::JSByteArray::JSByteArray):

* runtime/JSFunction.cpp:

(JSC::JSFunction::getOwnPropertySlot):

* runtime/JSGlobalData.cpp:

(JSC::JSGlobalData::JSGlobalData):

* runtime/JSGlobalObject.h:

(JSC::constructEmptyObject):

* runtime/JSNotAnObject.h:

(JSC::JSNotAnObject::JSNotAnObject):

* runtime/JSObject.cpp:

(JSC::JSObject::createInheritorID):

(JSC::JSObject::allocatePropertyStorage):

* runtime/JSObject.h:

(JSC::JSObject::propertyStorage):

(JSC::JSNonFinalObject::JSNonFinalObject):

(JSC::JSNonFinalObject::createStructure):

(JSC::JSFinalObject::create):

(JSC::JSFinalObject::createStructure):

(JSC::JSFinalObject::JSFinalObject):

(JSC::JSObject::offsetOfInlineStorage):

(JSC::constructEmptyObject):

(JSC::createEmptyObjectStructure):

(JSC::JSObject::JSObject):

(JSC::JSObject::~JSObject):

(JSC::Structure::isUsingInlineStorage):

* runtime/JSObjectWithGlobalObject.cpp:

(JSC::JSObjectWithGlobalObject::JSObjectWithGlobalObject):

* runtime/JSObjectWithGlobalObject.h:

(JSC::JSObjectWithGlobalObject::JSObjectWithGlobalObject):

* runtime/JSTypeInfo.h:

(JSC::TypeInfo::TypeInfo):

(JSC::TypeInfo::isVanilla):

* runtime/JSVariableObject.h:

(JSC::JSVariableObject::JSVariableObject):

* runtime/JSWrapperObject.h:

(JSC::JSWrapperObject::JSWrapperObject):

* runtime/ObjectConstructor.cpp:

(JSC::constructObject):

* runtime/ObjectPrototype.cpp:

(JSC::ObjectPrototype::ObjectPrototype):

* runtime/ObjectPrototype.h:

* runtime/StrictEvalActivation.cpp:

(JSC::StrictEvalActivation::StrictEvalActivation):

* runtime/StrictEvalActivation.h:

* runtime/Structure.cpp:

(JSC::Structure::Structure):

(JSC::Structure::growPropertyStorageCapacity):

2011-02-16 Oliver Hunt <oliver@apple.com>

Reviewed by Geoff Garen.

Incorrect handling of global writes in dynamic contexts

https://bugs.webkit.org/show_bug.cgi?id=49383

* interpreter/Interpreter.cpp:

(JSC::Interpreter::privateExecute):

Can't use the existing callframe to return an uncaught exception

as by definition that callframe has already been torn down.

* parser/ASTBuilder.h:

(JSC::ASTBuilder::ASTBuilder):

(JSC::ASTBuilder::varDeclarations):

(JSC::ASTBuilder::funcDeclarations):

(JSC::ASTBuilder::features):

(JSC::ASTBuilder::numConstants):

(JSC::ASTBuilder::createFuncDeclStatement):

(JSC::ASTBuilder::addVar):

(JSC::ASTBuilder::incConstants):

(JSC::ASTBuilder::usesThis):

(JSC::ASTBuilder::usesCatch):

(JSC::ASTBuilder::usesClosures):

(JSC::ASTBuilder::usesArguments):

(JSC::ASTBuilder::usesAssignment):

(JSC::ASTBuilder::usesWith):

(JSC::ASTBuilder::usesEval):

Don't need a vector of scopes in the ASTBuilder

* runtime/Operations.h:

(JSC::resolveBase):

In strict mode the optimisation that we use to skip a lookup

on the global object is incorrect and lead to us always

disallowing global writes when we needed to do a dynamic slot

lookup. Now the strict mode path actually checks for the

property.

2011-02-15 Jon Honeycutt <jhoneycutt@apple.com>

Windows build fix for

https://bugs.webkit.org/show_bug.cgi?id=54415

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

Remove deleted files.

2011-02-15 Oliver Hunt <oliver@apple.com>

Fix EFL build for

https://bugs.webkit.org/show_bug.cgi?id=54415

* CMakeLists.txt:

2011-02-14 Oliver Hunt <oliver@apple.com>

Reviewed by Gavin Barraclough and Geoff Garen.

Refactor handles and weak pointers to become nicer and more automatic

https://bugs.webkit.org/show_bug.cgi?id=54415

Move to a true handle based mechanism for GC value protection. This

also allows us to switch to a more sensible behaviour for weak pointers

in which weak pointers are automatically updated.

This allows us to remove the old (and convoluted) that required all

objects that may be held by a weak reference to be aware of the reference

and manually clear them in their destructors.

This also adds a few new data types to JSC that we use to efficiently

allocate and return the underlying handle storage.

This patch is largely renaming and removing now unnecessary destructors

from objects.

* API/JSClassRef.cpp:

(OpaqueJSClass::create):

(OpaqueJSClassContextData::OpaqueJSClassContextData):

(OpaqueJSClass::contextData):

(OpaqueJSClass::prototype):

* API/JSClassRef.h:

* CMakeLists.txt:

* GNUmakefile.am:

* JavaScriptCore.exp:

* JavaScriptCore.gypi:

* JavaScriptCore.pro:

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops:

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.make:

* JavaScriptCore.vcproj/WTF/WTF.vcproj:

* JavaScriptCore.vcproj/jsc/jscCommon.vsprops:

* JavaScriptCore.xcodeproj/project.pbxproj:

* collector/handles/Global.h: Added.

New Global handle type used to keep gc objects live, even if they're not

marked.

(JSC::Global::Global):

(JSC::Global::~Global):

(JSC::Global::set):

We can only assign directly to a global from another global.

In all other cases we need the JSGlobalData to be provided

explicitly so we use a set function.

(JSC::Global::operator=):

(JSC::Global::clear):

(JSC::Global::isHashTableDeletedValue):

(JSC::Global::internalSet):

* collector/handles/Handle.h: Added.

Root "Handle" type used for immutable handles and to provide the basic

APIs needed for pointer-like behaviour.

(JSC::HandleBase::operator!):

(JSC::HandleBase::operator UnspecifiedBoolType*):

(JSC::HandleBase::isEmpty):

(JSC::HandleBase::HandleBase):

(JSC::HandleBase::slot):

(JSC::HandleBase::invalidate):

(JSC::HandleBase::setSlot):

(JSC::HandleTypes::getFromSlot):

(JSC::HandleTypes::toJSValue):

(JSC::HandleTypes::validateUpcast):

(JSC::HandleConverter::operator->):

(JSC::HandleConverter::operator*):

(JSC::Handle::Handle):

(JSC::Handle::get):

(JSC::Handle::wrapSlot):

(JSC::operator==):

(JSC::operator!=):

* collector/handles/HandleHeap.cpp: Added.

New heap for global handles.

(JSC::HandleHeap::HandleHeap):

(JSC::HandleHeap::grow):

(JSC::HandleHeap::markStrongHandles):

(JSC::HandleHeap::updateAfterMark):

(JSC::HandleHeap::clearWeakPointers):

(JSC::HandleHeap::writeBarrier):

* collector/handles/HandleHeap.h: Added.

(JSC::HandleHeap::heapFor):

(JSC::HandleHeap::toHandle):

(JSC::HandleHeap::toNode):

(JSC::HandleHeap::allocate):

(JSC::HandleHeap::deallocate):

(JSC::HandleHeap::makeWeak):

Convert a hard handle into weak handle that does not

protect the object it points to.

(JSC::HandleHeap::makeSelfDestroying):

Converts a handle to a weak handle that will be returned

to the free list when the referenced object dies.

(JSC::HandleHeap::Node::Node):

(JSC::HandleHeap::Node::slot):

(JSC::HandleHeap::Node::handleHeap):

(JSC::HandleHeap::Node::setFinalizer):

(JSC::HandleHeap::Node::makeWeak):

(JSC::HandleHeap::Node::isWeak):

(JSC::HandleHeap::Node::makeSelfDestroying):

(JSC::HandleHeap::Node::isSelfDestroying):

(JSC::HandleHeap::Node::finalizer):

(JSC::HandleHeap::Node::setPrev):

(JSC::HandleHeap::Node::prev):

(JSC::HandleHeap::Node::setNext):

(JSC::HandleHeap::Node::next):

* interpreter/Interpreter.cpp:

(JSC::Interpreter::Interpreter):

* interpreter/Interpreter.h:

* interpreter/RegisterFile.cpp:

(JSC::RegisterFile::globalObjectCollected):

* interpreter/RegisterFile.h:

(JSC::RegisterFile::RegisterFile):

* runtime/GCHandle.cpp: Removed.

* runtime/GCHandle.h: Removed.

* runtime/Heap.cpp:

(JSC::Heap::Heap):

(JSC::Heap::destroy):

(JSC::Heap::markRoots):

* runtime/Heap.h:

(JSC::Heap::allocateGlobalHandle):

(JSC::Heap::reportExtraMemoryCost):

* runtime/JSGlobalData.cpp:

(JSC::JSGlobalData::JSGlobalData):

* runtime/JSGlobalData.h:

(JSC::JSGlobalData::allocateGlobalHandle):

* runtime/JSGlobalObject.cpp:

(JSC::JSGlobalObject::~JSGlobalObject):

* runtime/JSPropertyNameIterator.cpp:

(JSC::JSPropertyNameIterator::create):

(JSC::JSPropertyNameIterator::~JSPropertyNameIterator):

* runtime/JSPropertyNameIterator.h:

(JSC::JSPropertyNameIterator::createStructure):

(JSC::Structure::setEnumerationCache):

(JSC::Structure::clearEnumerationCache):

* runtime/Protect.h:

* runtime/Structure.cpp:

(JSC::Structure::~Structure):

* runtime/Structure.h:

* runtime/WeakGCPtr.h:

(JSC::WeakGCPtrBase::get):

(JSC::WeakGCPtrBase::clear):

(JSC::WeakGCPtrBase::operator!):

(JSC::WeakGCPtrBase::operator UnspecifiedBoolType*):

(JSC::WeakGCPtrBase::~WeakGCPtrBase):

(JSC::WeakGCPtrBase::WeakGCPtrBase):

(JSC::WeakGCPtrBase::internalSet):

(JSC::LazyWeakGCPtr::LazyWeakGCPtr):

(JSC::LazyWeakGCPtr::set):

(JSC::WeakGCPtr::WeakGCPtr):

(JSC::WeakGCPtr::operator=):

* runtime/WriteBarrier.h:

* wtf/BlockStack.h: Added.

(WTF::::BlockStack):

(WTF::::~BlockStack):

(WTF::::blocks):

(WTF::::grow):

(WTF::::shrink):

* wtf/SentinelLinkedList.h: Added.

(WTF::::SentinelLinkedList):

(WTF::::begin):

(WTF::::end):

(WTF::::push):

(WTF::::remove):

* wtf/SinglyLinkedList.h: Added.

(WTF::::SinglyLinkedList):

(WTF::::isEmpty):

(WTF::::push):

(WTF::::pop):

2011-02-15 Pratik Solanki <psolanki@apple.com>

Move WTF_USE_CFNETWORK to Platform.h

https://bugs.webkit.org/show_bug.cgi?id=54168

Reviewed by Darin Adler.

* wtf/Platform.h: Define WTF_USE_CFNETWORK for Windows builds.

2011-02-15 Geoffrey Garen <ggaren@apple.com>

Reviewed by Darin Adler.

Moved MarkedBlock data members to the head of the block

https://bugs.webkit.org/show_bug.cgi?id=54482

This allows for a variable-sized tail, to accommodate oversized blocks.

SunSpider reports no change.

* runtime/JSCell.h:

(JSC::JSCell::MarkedBlock::allocate):

* runtime/MarkedBlock.cpp:

(JSC::MarkedBlock::destroy):

(JSC::MarkedBlock::MarkedBlock):

(JSC::MarkedBlock::sweep):

* runtime/MarkedBlock.h: Added missing element to the CELLS_PER_BLOCK

calculation. This kind of error is why we want to migrate to the system

described below.

(JSC::roundUpToMultipleOf):

(JSC::MarkedBlock::firstCell):

(JSC::MarkedBlock::cells):

(JSC::MarkedBlock::cellNumber): Use subtraction instead of masking to

calculate cell number. The mask is no longer correct because the first

cell is not at the head of the block.

(JSC::MarkedBlock::forEach): Replaced m_cells data member with a cells()

accessor. We want to use sizeof(MarkedBlock) to calculate the size of the

block header, so we can't have an explicit data member to represent the block tail.

Also replaced iteration from zero with iteration from startCell(), since

the first N cells are now occupied by the header.

* runtime/MarkedSpace.cpp:

(JSC::MarkedSpace::MarkedSpace):

(JSC::MarkedSpace::reset): Replaced iteration from zero as above.

2011-02-15 Chris Rogers <crogers@google.com>

Reviewed by Alexey Proskuryakov.

Fix Mutex::tryLock() on Windows to work properly with PlatformCondition::timedWait()

https://bugs.webkit.org/show_bug.cgi?id=54408

* wtf/ThreadingWin.cpp:

(WTF::PlatformCondition::timedWait):

2011-02-15 Xan Lopez <xlopez@igalia.com>

Reviewed by Martin Robinson.

Remove some dead code in ARMv7

https://bugs.webkit.org/show_bug.cgi?id=54461

* assembler/ARMv7Assembler.h: remove dead code.

2011-02-14 Geoffrey Garen <ggaren@apple.com>

Rubber-stamped by Gavin Barraclough.

Some MarkedBlock refactoring.

Made cells private.

Renamed cells => m_cells

marked => m_marks.

* runtime/JSCell.h:

(JSC::JSCell::MarkedBlock::allocate):

* runtime/MarkedBlock.cpp:

(JSC::MarkedBlock::destroy):

(JSC::MarkedBlock::MarkedBlock):

(JSC::MarkedBlock::sweep):

* runtime/MarkedBlock.h:

(JSC::MarkedBlock::isEmpty):

(JSC::MarkedBlock::clearMarks):

(JSC::MarkedBlock::markCount):

(JSC::MarkedBlock::isMarked):

(JSC::MarkedBlock::testAndSetMarked):

(JSC::MarkedBlock::setMarked):

(JSC::MarkedBlock::forEach):

2011-02-14 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

Add basic parser for Content Security Policy

https://bugs.webkit.org/show_bug.cgi?id=54379

Add a constructor for copying a Vector into a String. I suspect there

are a number of call sites that are doing this manually that would

benefit from being moved to this API.

* wtf/text/WTFString.h:

(WTF::String::String):

2011-02-14 Pavel Podivilov <podivilov@chromium.org>

Reviewed by Yury Semikhatsky.

Web Inspector: use call frame column to determine execution line in source frame.

https://bugs.webkit.org/show_bug.cgi?id=54001

* wtf/text/TextPosition.h:

(WTF::ZeroBasedNumber::convertAsOneBasedInt):

2011-02-13 Jeremy Moskovich <jeremy@chromium.org>

Reviewed by Adam Barth.

Add a compile-time option to completely disable WebArchive support.

https://bugs.webkit.org/show_bug.cgi?id=52712

Add an ENABLE(WEB_ARCHIVE) compile-time setting and use it for all WebArchive code.

Ports Affected:

WebArchive support is currently enabled for all ports that define PLATFORM(CF) apart from Qt.

This patch preserves this behavior except that it also disables support in the Chromium port.

* wtf/Platform.h: Add ENABLE_WEB_ARCHIVE definition and turn it off explicitly for Qt & Chromium ports.

2011-02-13 Cameron Zwarich <zwarich@apple.com>

Reviewed by Dan Bernstein.

Bug 53760 - JSC fails to build with TOT Clang

https://bugs.webkit.org/show_bug.cgi?id=53760

Fix -Woverloaded-virtual warnings. This is also a 6% speedup on the v8 raytrace

benchmark; it is nothing-to-noise on everything else.

* API/JSCallbackObject.h: Remove pointlessly overloaded method.

* API/JSCallbackObjectFunctions.h: Ditto.

* runtime/Arguments.cpp:

(JSC::Arguments::put): Change signature to match the base class. This implementation

was no longer being called by anyone. This wasn't noticed because it is merely an

optimization of the base class' implementation.

* runtime/Arguments.h: Ditto.

2011-02-12 Adam Barth <abarth@webkit.org>

Reviewed by Mark Rowe.

Use /dev/urandom as the OSRandomSource on OS(DARWIN)

https://bugs.webkit.org/show_bug.cgi?id=54279

I'm not sure it makes much of a difference whether we use arc4random or

/dev/urandom on Mac. However, there's some aesthetic benefit to using

the same underlying API on as many platforms as reasonable.

* config.h:

* wtf/OSRandomSource.cpp:

(WTF::cryptographicallyRandomValuesFromOS):

2011-02-12 Adam Barth <abarth@webkit.org>

Reviewed by Kenneth Russell.

Enable ArrayBuffers by default

https://bugs.webkit.org/show_bug.cgi?id=54310

Export the required functions.

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:

2011-02-11 Daniel Bates <dbates@rim.com>

Reviewed by Geoffrey Garen.

Remove explicit disable ENABLE_ASSEMBLER_WX_EXCLUSIVE on non-iOS ports

https://bugs.webkit.org/show_bug.cgi?id=54107

It is unnecessary to explicitly disable ENABLE_ASSEMBLER_WX_EXCLUSIVE

by the definition of ENABLE().

* wtf/Platform.h:

2011-02-11 Geoffrey Garen <ggaren@apple.com>

Not reviewed.

Randomly touch some build files in the hopes of fixing the Qt build.

* JavaScriptCore.gypi:

* JavaScriptCore.pri:

* JavaScriptCore.pro:

2011-02-11 Geoffrey Garen <ggaren@apple.com>

Reviewed by Sam Weinig.

Garbage collection timer cycles forever, even when nothing is happening

https://bugs.webkit.org/show_bug.cgi?id=54320

(Rolling back in r78386 with the build fixed.)

* runtime/GCActivityCallbackCF.cpp:

(JSC::DefaultGCActivityCallbackPlatformData::trigger): Be sure to make

our timer inert after forcing a GC, to avoid GC'ing repeatedly.

2011-02-11 Geoffrey Garen <ggaren@apple.com>

Not reviewed.

Used svn merge -r78386:78385 to roll out r78386 because it broke the build.

* runtime/GCActivityCallbackCF.cpp:

(JSC::DefaultGCActivityCallbackPlatformData::trigger):

2011-02-11 Geoffrey Garen <ggaren@apple.com>

Reviewed by Sam Weinig.

Garbage collection timer cycles forever, even when nothing is happening

https://bugs.webkit.org/show_bug.cgi?id=54320

* runtime/GCActivityCallbackCF.cpp:

(JSC::DefaultGCActivityCallbackPlatformData::trigger): Be sure to make

our timer inert after forcing a GC, to avoid GC'ing repeatedly.

2011-02-11 Geoffrey Garen <ggaren@apple.com>

Try to fix the Windows build: added an exported symbol.

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:

2011-02-11 Geoffrey Garen <ggaren@apple.com>

Reviewed by Oliver Hunt.

A little more encapsulation for the heap: Removed CollectorHeapIterator

https://bugs.webkit.org/show_bug.cgi?id=54298

CollectorHeapIterator is a God object that knows the internals of each

of the pieces of the heap. This undermines the encapsulation I'm trying

to achieve by splitting concepts into different classes.

As an alternative, I've given each class a forEach iteration function,

which takes a functor as an argument. Now, each class just needs to

know how to iterate the things it knows about.

* GNUmakefile.am:

* JavaScriptCore.exp:

* JavaScriptCore.gypi:

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Removed CollectorHeapIterator.

* debugger/Debugger.cpp:

(JSC::Recompiler::Recompiler):

(JSC::Recompiler::~Recompiler):

(JSC::Recompiler::operator()):

(JSC::Debugger::recompileAllJSFunctions): Updated to use forEach interface

instead of an iterator.

* runtime/CollectorHeapIterator.h: Removed.

* runtime/Heap.cpp:

(JSC::TypeCounter::TypeCounter):

(JSC::TypeCounter::typeName):

(JSC::TypeCounter::operator()):

(JSC::TypeCounter::take):

(JSC::Heap::protectedObjectTypeCounts):

(JSC::Heap::objectTypeCounts): Added forEach and removed iterator.

* runtime/Heap.h:

(JSC::Heap::forEach):

* runtime/JSGlobalData.cpp:

(JSC::Recompiler::operator()):

(JSC::JSGlobalData::recompileAllJSFunctions):

* runtime/MarkedBlock.h:

(JSC::MarkedBlock::forEach): Added forEach. Removed friend declaration

for CollectorHeapIterator. Now, we can make all our data private and

change it without breaking any other classes.

* runtime/MarkedSpace.cpp:

* runtime/MarkedSpace.h:

(JSC::MarkedSpace::forEach): Added forEach and removed iterator.

2011-02-11 Adam Barth <abarth@webkit.org>

Reviewed by Andreas Kling.

CryptographicRandomNumber has its threading ifdefs backwards

https://bugs.webkit.org/show_bug.cgi?id=54280

Turns out we want the mutex when thread is enabled. :)

* wtf/CryptographicallyRandomNumber.cpp:

(WTF::ARC4Stream::ARC4RandomNumberGenerator::randomNumber):

(WTF::ARC4Stream::ARC4RandomNumberGenerator::randomValues):

2011-02-10 Adam Barth <abarth@webkit.org>

Reviewed by Eric Seidel.

WebKit should have a cryptographic RNG

https://bugs.webkit.org/show_bug.cgi?id=22049

Teach JavaScriptCore how to export this function.

* JavaScriptCore.exp:

* JavaScriptCore.xcodeproj/project.pbxproj:

2011-02-10 Geoffrey Garen <ggaren@apple.com>

Reviewed by Sam Weinig.

A little more encapsulation for MarkedBlock: Made all constants private

so clients don't know whether allocations are fixed-sized or not

https://bugs.webkit.org/show_bug.cgi?id=54270

SunSpider reports no change.

* runtime/CollectorHeapIterator.h:

(JSC::CollectorHeapIterator::advance): Updated for removal of HeapConstants.

* runtime/Error.cpp: Switched to using ASSERT_CLASS_FITS_IN_CELL, like

all other classes.

* runtime/Heap.cpp:

(JSC::Heap::allocate): Updated for removal of HeapConstants.

(JSC::Heap::reset): Updated to use size(), instead of calculating size

on our own.

* runtime/Heap.h: Moved the ASSERT here to MarkedBlock, since it enforces

on special knowledge of fixed-sizery, which only MarkedBlock is supposed

to know about.

* runtime/JSCell.h:

(JSC::JSCell::MarkedBlock::allocate): Updated for removal of HeapConstants.

Also changed to reset nextCell to 0 at the end of a block, since that

seems more consistent.

* runtime/JSGlobalData.cpp:

(JSC::JSGlobalData::storeVPtrs): Changed to use a fixed array of char.

This hard-coded size is a little wonky, but the compiler will tell us

if it's ever wrong, so I think it's OK.

* runtime/MarkedBlock.cpp:

(JSC::MarkedBlock::destroy):

(JSC::MarkedBlock::MarkedBlock):

(JSC::MarkedBlock::sweep): Updated for removal of HeapConstants.

* runtime/MarkedBlock.h:

(JSC::MarkedBlock::isEmpty):

(JSC::MarkedBlock::clearMarks):

(JSC::MarkedBlock::size):

(JSC::MarkedBlock::capacity): Made constants private to this class.

Removed HeapConstants. Added size() and capacity() functions.

* runtime/MarkedSpace.cpp:

(JSC::MarkedSpace::allocate):

(JSC::MarkedSpace::objectCount):

(JSC::MarkedSpace::size):

(JSC::MarkedSpace::capacity):

* runtime/MarkedSpace.h: Use MarkedBlock helper functions instead of

direct knowledge of MarkedBlock internals.

2011-02-10 Geoffrey Garen <ggaren@apple.com>

Reviewed by Sam Weinig.

A little more encapsulation for MarkedBlock: Made mark bits private

https://bugs.webkit.org/show_bug.cgi?id=54264

SunSpider reports no change.

* runtime/Heap.cpp:

(JSC::Heap::markRoots):

(JSC::Heap::reset): Renamed clearMarkBits => clearMarks, since clients

don't need to know that marks are represented as bits. Renamed

markedCells => markCount, since clients don't need to know that blocks

are split into cells.

* runtime/MarkedBlock.h:

(JSC::MarkedBlock::isEmpty):

(JSC::MarkedBlock::clearMarks):

(JSC::MarkedBlock::markCount): New helper functions for encapsulating

the information clients actually need.

* runtime/MarkedSpace.cpp:

(JSC::MarkedSpace::destroy):

(JSC::MarkedSpace::shrink):

(JSC::MarkedSpace::clearMarks):

(JSC::MarkedSpace::markCount):

(JSC::MarkedSpace::objectCount):

* runtime/MarkedSpace.h: Use new helper functions instead of accessing

MarkedBlock data directly.

2011-02-10 Michael Saboff <msaboff@apple.com>

Reviewed by Geoffrey Garen.

Cached JavaScript Parser Data Being Left in Memory Cache

https://bugs.webkit.org/show_bug.cgi?id=54245

Added clear method which removes SourceProviderCache items.

Cleaned up extraneous whitespace.

* JavaScriptCore.exp:

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def:

* parser/SourceProviderCache.cpp:

(JSC::SourceProviderCache::~SourceProviderCache):

(JSC::SourceProviderCache::clear):

(JSC::SourceProviderCache::byteSize):

(JSC::SourceProviderCache::add):

* parser/SourceProviderCache.h:

2011-02-10 Joseph Pecoraro <joepeck@webkit.org>

Follow-up fix to r78291. I should pass (int) 0, not '0' to memset.

* runtime/GCActivityCallbackCF.cpp:

(JSC::DefaultGCActivityCallback::commonConstructor): fix mistaken post-review change.

2011-02-10 Joseph Pecoraro <joepeck@webkit.org>

Reviewed by Geoffrey Garen.

Make DefaultGCActivityCallback for PLATFORM(CF) Easier to Subclass

https://bugs.webkit.org/show_bug.cgi?id=54257

A subclass may want to specify the CFRunLoop that the Garbage Collection

will happen on. It was difficult to manipulate this in a subclass because

the current DefaultGCActivityCallback class does this in its constructor.

This patch generalizes things a bit more so that a specific run loop can

be passed in to the constructor. This makes it so all run loop management

can stay in DefaultGCActivityCallback and a subclass can specify any runloop.

* runtime/GCActivityCallback.h: expose a PLATFORM(CF) constructor that can

specify the runloop GC can be scheduled on.

* runtime/GCActivityCallbackCF.cpp:

(JSC::DefaultGCActivityCallback::DefaultGCActivityCallback): two constructors,

one that specifies the run loop and passes that on to commonConstructor and

the old constructor defaults to the current run loop.

(JSC::DefaultGCActivityCallback::commonConstructor): do normal construction,

adding the timer to the given runloop.

2011-02-10 Geoffrey Garen <ggaren@apple.com>

Reviewed by Oliver Hunt.

A little more encapsulation for MarkedBlock: Moved allocate() and sweep() into MarkedBlock

https://bugs.webkit.org/show_bug.cgi?id=54253

SunSpider reports no change.

* runtime/CollectorHeapIterator.h: Removed DeadObjectIterator, since it

is now unused.

* runtime/Heap.cpp:

(JSC::Heap::reset): Moved the call to shrink() here, since it seems a

little more clear for MarkedSpace's client to tell it explicitly when to

shrink.

* runtime/JSCell.h:

(JSC::JSCell::MarkedBlock::allocate): Split out from MarkedSpace::allocate.

* runtime/MarkedBlock.cpp:

(JSC::MarkedBlock::sweep): Split out from MarkedSpace::sweep, and

converted to more directly iterate a MarkedBlock based on knowing its

internal structure.

* runtime/MarkedBlock.h:

* runtime/MarkedSpace.cpp:

(JSC::MarkedSpace::allocate):

(JSC::MarkedSpace::sweep):

* runtime/MarkedSpace.h: Split out the code mentioned above.

2011-02-10 Patrick Gansterer <paroga@webkit.org>

Reviewed by Andreas Kling.

Fix compilation error on OpenBSD

https://bugs.webkit.org/show_bug.cgi?id=53766

Add a HAVE_STRNSTR in Platform.h and define it only on Darwin and FreeBSD.

* wtf/Platform.h:

* wtf/StringExtras.h: Use HAVE(STRNSTR) now.

2011-02-10 Adam Roben <aroben@apple.com>

Print locations of assertions and logs in a way that Visual Studio understands

With this change, double-clicking one of these locations in Visual Studio's Output Window

will focus that line of code in the editor.

Fixes <http://webkit.org/b/54208> Double-clicking locations of assertions/logs in Visual

Studio's Output Window does nothing

Reviewed by Alexey Proskuryakov.

* wtf/Assertions.cpp:

(printCallSite): Use OS(WINDOWS) instead of the meaningless OS(WIN). When we aren't using

the debug CRT (and thus can't call _CrtDbgReport), print the file and line number using the

same format that cl.exe uses when it prints compiler errors.

2011-02-10 Dan Bernstein <mitz@apple.com>

LLVM Compiler build fix.

* runtime/MarkedBlock.cpp:

(JSC::MarkedBlock::create):

2011-02-10 Peter Varga <pvarga@webkit.org>

Reviewed by Csaba Osztrogonác.

Remove PCRE source from trunk

https://bugs.webkit.org/show_bug.cgi?id=54188

* Android.mk:

* Android.v8.wtf.mk:

* CMakeLists.txt:

* DerivedSources.make:

* DerivedSources.pro:

* GNUmakefile.am:

* JavaScriptCore.gypi:

* JavaScriptCore.order:

* JavaScriptCore.pri:

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj:

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops:

* JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.make:

* JavaScriptCore.vcproj/WTF/WTFCommon.vsprops:

* JavaScriptCore.vcproj/jsc/jscCommon.vsprops:

* JavaScriptCore.xcodeproj/project.pbxproj:

* pcre/AUTHORS: Removed.

* pcre/COPYING: Removed.

* pcre/dftables: Removed.

* pcre/pcre.h: Removed.

* pcre/pcre.pri: Removed.

* pcre/pcre_compile.cpp: Removed.

* pcre/pcre_exec.cpp: Removed.

* pcre/pcre_internal.h: Removed.

* pcre/pcre_tables.cpp: Removed.

* pcre/pcre_ucp_searchfuncs.cpp: Removed.

* pcre/pcre_xclass.cpp: Removed.

* pcre/ucpinternal.h: Removed.

* pcre/ucptable.cpp: Removed.

* wscript:

2011-02-10 Patrick Gansterer <paroga@webkit.org>

Reviewed by Adam Barth.

Add randomValuesFromOS for OS(WINDOWS)

https://bugs.webkit.org/show_bug.cgi?id=54155

Use CryptGenRandom for generating cryptographically secure random numbers.

This will work on WinCE and MinGW too.

* config.h:

* wtf/OSRandomSource.cpp:

(WTF::randomValuesFromOS):

2011-02-10 Jarred Nicholls <jarred@sencha.com>

Reviewed by Adam Barth.

REGRESSION(r78149): Return value of read() shouldn't be ignored.

https://bugs.webkit.org/show_bug.cgi?id=54167

stdio read should have its return value handled. Build error in gcc 4.4.5.

* wtf/OSRandomSource.cpp:

(WTF::randomValuesFromOS):

2011-02-10 Patrick Gansterer <paroga@webkit.org>

Reviewed by Adam Barth.

Rename randomValuesFromOS to cryptographicallyRandomValuesFromOS

https://bugs.webkit.org/show_bug.cgi?id=54156

randomValuesFromOS generates random numbers of cryptographic quality.

Make this clear by adding "cryptographically" to the function name.

* wtf/CryptographicallyRandomNumber.cpp:

(WTF::ARC4Stream::ARC4RandomNumberGenerator::stir):

* wtf/OSRandomSource.cpp:

(WTF::cryptographicallyRandomValuesFromOS):

* wtf/OSRandomSource.h:

2011-02-09 Mark Rowe <mrowe@apple.com>

Reviewed by Sam Weinig.

<rdar://problem/8805364> Malloc zone enumeration code should be safe in the face of errors from the memory reader.

* wtf/FastMalloc.cpp:

(WTF::PageHeapAllocator::recordAdministrativeRegions): Use the new helper function to walk the linked list safely.

(WTF::TCMalloc_ThreadCache_FreeList::enumerateFreeObjects): Ditto.

(WTF::TCMalloc_Central_FreeList::enumerateFreeObjects): Ditto.

(WTF::TCMallocStats::PageMapFreeObjectFinder::visit): Bail out if the span could not be read.

(WTF::TCMallocStats::PageMapMemoryUsageRecorder::visit): Ditto.

* wtf/MallocZoneSupport.h:

(WTF::RemoteMemoryReader::operator()): Remove an assert that is not valid.

(WTF::RemoteMemoryReader::nextEntryInLinkedList): Add a helper function for retrieving the next entry in

a linked list. It maps a failed read of the remote memory in to a null pointer, which all callers can

handle gracefully.

2011-02-09 Gavin Barraclough <barraclough@apple.com>

Reviewed by Sam Weinig.

Bug 54164 - Optimize global_var accesses on JSVALUE64

Directly embed the pointer to d->registers, optimize out the load

from the variable object, as we do already in JSVALUE32_64.

This is a ~1.5% win on sunspidey.

* jit/JIT.cpp:

* jit/JIT.h:

* jit/JITOpcodes.cpp:

(JSC::JIT::emit_op_get_global_var):

(JSC::JIT::emit_op_put_global_var):

(JSC::JIT::emit_op_get_scoped_var):

(JSC::JIT::emit_op_put_scoped_var):

2011-02-09 Geoffrey Garen <ggaren@apple.com>

Reviewed by Oliver Hunt.

A little more encapsulation for MarkedBlock: Made MarkedBlock responsible

for its own initialization and destruction

https://bugs.webkit.org/show_bug.cgi?id=54137