blob: 156c25ed4af505ebd43c4a86368a713d13cf9b7e [file] [log] [blame]
/*
* Copyright (C) 2008, 2009, 2010, 2011, 2012, 2013 Apple Inc. All rights reserved.
* Copyright (C) 2008 Cameron Zwarich <cwzwarich@uwaterloo.ca>
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of Apple Computer, Inc. ("Apple") nor the names of
* its contributors may be used to endorse or promote products derived
* from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY APPLE AND ITS CONTRIBUTORS "AS IS" AND ANY
* EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
* WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL APPLE OR ITS CONTRIBUTORS BE LIABLE FOR ANY
* DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
* ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#ifndef CodeBlock_h
#define CodeBlock_h
#include "ArrayProfile.h"
#include "ByValInfo.h"
#include "BytecodeConventions.h"
#include "CallLinkInfo.h"
#include "CallReturnOffsetToBytecodeOffset.h"
#include "CodeBlockHash.h"
#include "CodeBlockSet.h"
#include "ConcurrentJITLock.h"
#include "CodeOrigin.h"
#include "CodeType.h"
#include "CompactJITCodeMap.h"
#include "DFGCommon.h"
#include "DFGCommonData.h"
#include "DFGExitProfile.h"
#include "DFGMinifiedGraph.h"
#include "DFGOSREntry.h"
#include "DFGOSRExit.h"
#include "DFGVariableEventStream.h"
#include "DeferredCompilationCallback.h"
#include "EvalCodeCache.h"
#include "ExecutionCounter.h"
#include "ExpressionRangeInfo.h"
#include "HandlerInfo.h"
#include "ObjectAllocationProfile.h"
#include "Options.h"
#include "Operations.h"
#include "PutPropertySlot.h"
#include "Instruction.h"
#include "JITCode.h"
#include "JITWriteBarrier.h"
#include "JSGlobalObject.h"
#include "JumpTable.h"
#include "LLIntCallLinkInfo.h"
#include "LazyOperandValueProfile.h"
#include "LineInfo.h"
#include "ProfilerCompilation.h"
#include "RegExpObject.h"
#include "StructureStubInfo.h"
#include "UnconditionalFinalizer.h"
#include "ValueProfile.h"
#include "VirtualRegister.h"
#include "Watchpoint.h"
#include <wtf/Bag.h>
#include <wtf/FastMalloc.h>
#include <wtf/PassOwnPtr.h>
#include <wtf/RefCountedArray.h>
#include <wtf/RefPtr.h>
#include <wtf/SegmentedVector.h>
#include <wtf/Vector.h>
#include <wtf/text/WTFString.h>
namespace JSC {
class ExecState;
class LLIntOffsetsExtractor;
class RepatchBuffer;
inline VirtualRegister unmodifiedArgumentsRegister(VirtualRegister argumentsRegister) { return VirtualRegister(argumentsRegister.offset() + 1); }
static ALWAYS_INLINE int missingThisObjectMarker() { return std::numeric_limits<int>::max(); }
enum ReoptimizationMode { DontCountReoptimization, CountReoptimization };
class CodeBlock : public ThreadSafeRefCounted<CodeBlock>, public UnconditionalFinalizer, public WeakReferenceHarvester {
WTF_MAKE_FAST_ALLOCATED;
friend class JIT;
friend class LLIntOffsetsExtractor;
public:
enum CopyParsedBlockTag { CopyParsedBlock };
protected:
CodeBlock(CopyParsedBlockTag, CodeBlock& other);
CodeBlock(ScriptExecutable* ownerExecutable, UnlinkedCodeBlock*, JSScope*, PassRefPtr<SourceProvider>, unsigned sourceOffset, unsigned firstLineColumnOffset);
WriteBarrier<JSGlobalObject> m_globalObject;
Heap* m_heap;
public:
JS_EXPORT_PRIVATE virtual ~CodeBlock();
UnlinkedCodeBlock* unlinkedCodeBlock() const { return m_unlinkedCode.get(); }
CString inferredName() const;
CodeBlockHash hash() const;
bool hasHash() const;
bool isSafeToComputeHash() const;
CString sourceCodeForTools() const; // Not quite the actual source we parsed; this will do things like prefix the source for a function with a reified signature.
CString sourceCodeOnOneLine() const; // As sourceCodeForTools(), but replaces all whitespace runs with a single space.
void dumpAssumingJITType(PrintStream&, JITCode::JITType) const;
void dump(PrintStream&) const;
int numParameters() const { return m_numParameters; }
void setNumParameters(int newValue);
int* addressOfNumParameters() { return &m_numParameters; }
static ptrdiff_t offsetOfNumParameters() { return OBJECT_OFFSETOF(CodeBlock, m_numParameters); }
CodeBlock* alternative() { return m_alternative.get(); }
PassRefPtr<CodeBlock> releaseAlternative() { return m_alternative.release(); }
void setAlternative(PassRefPtr<CodeBlock> alternative) { m_alternative = alternative; }
CodeSpecializationKind specializationKind() const
{
return specializationFromIsConstruct(m_isConstructor);
}
CodeBlock* baselineVersion();
void visitAggregate(SlotVisitor&);
static void dumpStatistics();
void dumpBytecode(PrintStream& = WTF::dataFile());
void dumpBytecode(PrintStream&, unsigned bytecodeOffset);
void printStructures(PrintStream&, const Instruction*);
void printStructure(PrintStream&, const char* name, const Instruction*, int operand);
bool isStrictMode() const { return m_isStrictMode; }
ECMAMode ecmaMode() const { return isStrictMode() ? StrictMode : NotStrictMode; }
inline bool isKnownNotImmediate(int index)
{
if (index == m_thisRegister.offset() && !m_isStrictMode)
return true;
if (isConstantRegisterIndex(index))
return getConstant(index).isCell();
return false;
}
ALWAYS_INLINE bool isTemporaryRegisterIndex(int index)
{
return index >= m_numVars;
}
HandlerInfo* handlerForBytecodeOffset(unsigned bytecodeOffset);
unsigned lineNumberForBytecodeOffset(unsigned bytecodeOffset);
unsigned columnNumberForBytecodeOffset(unsigned bytecodeOffset);
void expressionRangeForBytecodeOffset(unsigned bytecodeOffset, int& divot,
int& startOffset, int& endOffset, unsigned& line, unsigned& column);
#if ENABLE(JIT)
StructureStubInfo* addStubInfo();
Bag<StructureStubInfo>::iterator begin() { return m_stubInfos.begin(); }
Bag<StructureStubInfo>::iterator end() { return m_stubInfos.end(); }
void resetStub(StructureStubInfo&);
void getStubInfoMap(const ConcurrentJITLocker&, StubInfoMap& result);
ByValInfo& getByValInfo(unsigned bytecodeIndex)
{
return *(binarySearch<ByValInfo, unsigned>(m_byValInfos, m_byValInfos.size(), bytecodeIndex, getByValInfoBytecodeIndex));
}
CallLinkInfo& getCallLinkInfo(ReturnAddressPtr returnAddress)
{
return *(binarySearch<CallLinkInfo, void*>(m_callLinkInfos, m_callLinkInfos.size(), returnAddress.value(), getCallLinkInfoReturnLocation));
}
CallLinkInfo& getCallLinkInfo(unsigned bytecodeIndex)
{
ASSERT(!JITCode::isOptimizingJIT(jitType()));
return *(binarySearch<CallLinkInfo, unsigned>(m_callLinkInfos, m_callLinkInfos.size(), bytecodeIndex, getCallLinkInfoBytecodeIndex));
}
#endif // ENABLE(JIT)
void unlinkIncomingCalls();
#if ENABLE(JIT)
void unlinkCalls();
void linkIncomingCall(ExecState* callerFrame, CallLinkInfo*);
bool isIncomingCallAlreadyLinked(CallLinkInfo* incoming)
{
return m_incomingCalls.isOnList(incoming);
}
#endif // ENABLE(JIT)
#if ENABLE(LLINT)
void linkIncomingCall(ExecState* callerFrame, LLIntCallLinkInfo*);
#endif // ENABLE(LLINT)
void setJITCodeMap(PassOwnPtr<CompactJITCodeMap> jitCodeMap)
{
m_jitCodeMap = jitCodeMap;
}
CompactJITCodeMap* jitCodeMap()
{
return m_jitCodeMap.get();
}
unsigned bytecodeOffset(Instruction* returnAddress)
{
RELEASE_ASSERT(returnAddress >= instructions().begin() && returnAddress < instructions().end());
return static_cast<Instruction*>(returnAddress) - instructions().begin();
}
bool isNumericCompareFunction() { return m_unlinkedCode->isNumericCompareFunction(); }
unsigned numberOfInstructions() const { return m_instructions.size(); }
RefCountedArray<Instruction>& instructions() { return m_instructions; }
const RefCountedArray<Instruction>& instructions() const { return m_instructions; }
size_t predictedMachineCodeSize();
bool usesOpcode(OpcodeID);
unsigned instructionCount() { return m_instructions.size(); }
int argumentIndexAfterCapture(size_t argument);
bool hasSlowArguments();
const SlowArgument* machineSlowArguments();
// Exactly equivalent to codeBlock->ownerExecutable()->installCode(codeBlock);
void install();
// Exactly equivalent to codeBlock->ownerExecutable()->newReplacementCodeBlockFor(codeBlock->specializationKind())
PassRefPtr<CodeBlock> newReplacement();
void setJITCode(PassRefPtr<JITCode> code, MacroAssemblerCodePtr codeWithArityCheck)
{
ASSERT(m_heap->isDeferred());
m_heap->reportExtraMemoryCost(code->size());
ConcurrentJITLocker locker(m_lock);
WTF::storeStoreFence(); // This is probably not needed because the lock will also do something similar, but it's good to be paranoid.
m_jitCode = code;
m_jitCodeWithArityCheck = codeWithArityCheck;
}
PassRefPtr<JITCode> jitCode() { return m_jitCode; }
MacroAssemblerCodePtr jitCodeWithArityCheck() { return m_jitCodeWithArityCheck; }
JITCode::JITType jitType() const
{
JITCode* jitCode = m_jitCode.get();
WTF::loadLoadFence();
JITCode::JITType result = JITCode::jitTypeFor(jitCode);
WTF::loadLoadFence(); // This probably isn't needed. Oh well, paranoia is good.
return result;
}
#if ENABLE(JIT)
bool hasBaselineJITProfiling() const
{
return jitType() == JITCode::BaselineJIT;
}
void jettison(ReoptimizationMode = DontCountReoptimization);
virtual CodeBlock* replacement() = 0;
virtual DFG::CapabilityLevel capabilityLevelInternal() = 0;
DFG::CapabilityLevel capabilityLevel()
{
DFG::CapabilityLevel result = capabilityLevelInternal();
m_capabilityLevelState = result;
return result;
}
DFG::CapabilityLevel capabilityLevelState() { return m_capabilityLevelState; }
bool hasOptimizedReplacement(JITCode::JITType typeToReplace);
bool hasOptimizedReplacement(); // the typeToReplace is my JITType
#endif
ScriptExecutable* ownerExecutable() const { return m_ownerExecutable.get(); }
void setVM(VM* vm) { m_vm = vm; }
VM* vm() { return m_vm; }
void setThisRegister(VirtualRegister thisRegister) { m_thisRegister = thisRegister; }
VirtualRegister thisRegister() const { return m_thisRegister; }
bool needsFullScopeChain() const { return m_unlinkedCode->needsFullScopeChain(); }
bool usesEval() const { return m_unlinkedCode->usesEval(); }
void setArgumentsRegister(VirtualRegister argumentsRegister)
{
ASSERT(argumentsRegister.isValid());
m_argumentsRegister = argumentsRegister;
ASSERT(usesArguments());
}
VirtualRegister argumentsRegister() const
{
ASSERT(usesArguments());
return m_argumentsRegister;
}
VirtualRegister uncheckedArgumentsRegister()
{
if (!usesArguments())
return VirtualRegister();
return argumentsRegister();
}
void setActivationRegister(VirtualRegister activationRegister)
{
m_activationRegister = activationRegister;
}
VirtualRegister activationRegister() const
{
ASSERT(needsFullScopeChain());
return m_activationRegister;
}
VirtualRegister uncheckedActivationRegister()
{
if (!needsFullScopeChain())
return VirtualRegister();
return activationRegister();
}
bool usesArguments() const { return m_argumentsRegister.isValid(); }
bool needsActivation() const
{
return m_needsActivation;
}
bool isCaptured(VirtualRegister operand, InlineCallFrame* = 0) const;
int framePointerOffsetToGetActivationRegisters(int machineCaptureStart);
int framePointerOffsetToGetActivationRegisters();
CodeType codeType() const { return m_unlinkedCode->codeType(); }
PutPropertySlot::Context putByIdContext() const
{
if (codeType() == EvalCode)
return PutPropertySlot::PutByIdEval;
return PutPropertySlot::PutById;
}
SourceProvider* source() const { return m_source.get(); }
unsigned sourceOffset() const { return m_sourceOffset; }
unsigned firstLineColumnOffset() const { return m_firstLineColumnOffset; }
size_t numberOfJumpTargets() const { return m_unlinkedCode->numberOfJumpTargets(); }
unsigned jumpTarget(int index) const { return m_unlinkedCode->jumpTarget(index); }
void createActivation(CallFrame*);
void clearEvalCache();
String nameForRegister(VirtualRegister);
#if ENABLE(JIT)
void setNumberOfByValInfos(size_t size) { m_byValInfos.grow(size); }
size_t numberOfByValInfos() const { return m_byValInfos.size(); }
ByValInfo& byValInfo(size_t index) { return m_byValInfos[index]; }
void setNumberOfCallLinkInfos(size_t size) { m_callLinkInfos.grow(size); }
size_t numberOfCallLinkInfos() const { return m_callLinkInfos.size(); }
CallLinkInfo& callLinkInfo(int index) { return m_callLinkInfos[index]; }
#endif
#if ENABLE(VALUE_PROFILER)
unsigned numberOfArgumentValueProfiles()
{
ASSERT(m_numParameters >= 0);
ASSERT(m_argumentValueProfiles.size() == static_cast<unsigned>(m_numParameters));
return m_argumentValueProfiles.size();
}
ValueProfile* valueProfileForArgument(unsigned argumentIndex)
{
ValueProfile* result = &m_argumentValueProfiles[argumentIndex];
ASSERT(result->m_bytecodeOffset == -1);
return result;
}
unsigned numberOfValueProfiles() { return m_valueProfiles.size(); }
ValueProfile* valueProfile(int index) { return &m_valueProfiles[index]; }
ValueProfile* valueProfileForBytecodeOffset(int bytecodeOffset)
{
ValueProfile* result = binarySearch<ValueProfile, int>(
m_valueProfiles, m_valueProfiles.size(), bytecodeOffset,
getValueProfileBytecodeOffset<ValueProfile>);
ASSERT(result->m_bytecodeOffset != -1);
ASSERT(instructions()[bytecodeOffset + opcodeLength(
m_vm->interpreter->getOpcodeID(
instructions()[
bytecodeOffset].u.opcode)) - 1].u.profile == result);
return result;
}
SpeculatedType valueProfilePredictionForBytecodeOffset(const ConcurrentJITLocker& locker, int bytecodeOffset)
{
return valueProfileForBytecodeOffset(bytecodeOffset)->computeUpdatedPrediction(locker);
}
unsigned totalNumberOfValueProfiles()
{
return numberOfArgumentValueProfiles() + numberOfValueProfiles();
}
ValueProfile* getFromAllValueProfiles(unsigned index)
{
if (index < numberOfArgumentValueProfiles())
return valueProfileForArgument(index);
return valueProfile(index - numberOfArgumentValueProfiles());
}
RareCaseProfile* addRareCaseProfile(int bytecodeOffset)
{
m_rareCaseProfiles.append(RareCaseProfile(bytecodeOffset));
return &m_rareCaseProfiles.last();
}
unsigned numberOfRareCaseProfiles() { return m_rareCaseProfiles.size(); }
RareCaseProfile* rareCaseProfile(int index) { return &m_rareCaseProfiles[index]; }
RareCaseProfile* rareCaseProfileForBytecodeOffset(int bytecodeOffset)
{
return tryBinarySearch<RareCaseProfile, int>(
m_rareCaseProfiles, m_rareCaseProfiles.size(), bytecodeOffset,
getRareCaseProfileBytecodeOffset);
}
bool likelyToTakeSlowCase(int bytecodeOffset)
{
if (!hasBaselineJITProfiling())
return false;
unsigned value = rareCaseProfileForBytecodeOffset(bytecodeOffset)->m_counter;
return value >= Options::likelyToTakeSlowCaseMinimumCount();
}
bool couldTakeSlowCase(int bytecodeOffset)
{
if (!hasBaselineJITProfiling())
return false;
unsigned value = rareCaseProfileForBytecodeOffset(bytecodeOffset)->m_counter;
return value >= Options::couldTakeSlowCaseMinimumCount();
}
RareCaseProfile* addSpecialFastCaseProfile(int bytecodeOffset)
{
m_specialFastCaseProfiles.append(RareCaseProfile(bytecodeOffset));
return &m_specialFastCaseProfiles.last();
}
unsigned numberOfSpecialFastCaseProfiles() { return m_specialFastCaseProfiles.size(); }
RareCaseProfile* specialFastCaseProfile(int index) { return &m_specialFastCaseProfiles[index]; }
RareCaseProfile* specialFastCaseProfileForBytecodeOffset(int bytecodeOffset)
{
return tryBinarySearch<RareCaseProfile, int>(
m_specialFastCaseProfiles, m_specialFastCaseProfiles.size(), bytecodeOffset,
getRareCaseProfileBytecodeOffset);
}
bool likelyToTakeSpecialFastCase(int bytecodeOffset)
{
if (!hasBaselineJITProfiling())
return false;
unsigned specialFastCaseCount = specialFastCaseProfileForBytecodeOffset(bytecodeOffset)->m_counter;
return specialFastCaseCount >= Options::likelyToTakeSlowCaseMinimumCount();
}
bool couldTakeSpecialFastCase(int bytecodeOffset)
{
if (!hasBaselineJITProfiling())
return false;
unsigned specialFastCaseCount = specialFastCaseProfileForBytecodeOffset(bytecodeOffset)->m_counter;
return specialFastCaseCount >= Options::couldTakeSlowCaseMinimumCount();
}
bool likelyToTakeDeepestSlowCase(int bytecodeOffset)
{
if (!hasBaselineJITProfiling())
return false;
unsigned slowCaseCount = rareCaseProfileForBytecodeOffset(bytecodeOffset)->m_counter;
unsigned specialFastCaseCount = specialFastCaseProfileForBytecodeOffset(bytecodeOffset)->m_counter;
unsigned value = slowCaseCount - specialFastCaseCount;
return value >= Options::likelyToTakeSlowCaseMinimumCount();
}
bool likelyToTakeAnySlowCase(int bytecodeOffset)
{
if (!hasBaselineJITProfiling())
return false;
unsigned slowCaseCount = rareCaseProfileForBytecodeOffset(bytecodeOffset)->m_counter;
unsigned specialFastCaseCount = specialFastCaseProfileForBytecodeOffset(bytecodeOffset)->m_counter;
unsigned value = slowCaseCount + specialFastCaseCount;
return value >= Options::likelyToTakeSlowCaseMinimumCount();
}
unsigned numberOfArrayProfiles() const { return m_arrayProfiles.size(); }
const ArrayProfileVector& arrayProfiles() { return m_arrayProfiles; }
ArrayProfile* addArrayProfile(unsigned bytecodeOffset)
{
m_arrayProfiles.append(ArrayProfile(bytecodeOffset));
return &m_arrayProfiles.last();
}
ArrayProfile* getArrayProfile(unsigned bytecodeOffset);
ArrayProfile* getOrAddArrayProfile(unsigned bytecodeOffset);
#endif
// Exception handling support
size_t numberOfExceptionHandlers() const { return m_rareData ? m_rareData->m_exceptionHandlers.size() : 0; }
void allocateHandlers(const Vector<UnlinkedHandlerInfo>& unlinkedHandlers)
{
size_t count = unlinkedHandlers.size();
if (!count)
return;
createRareDataIfNecessary();
m_rareData->m_exceptionHandlers.resize(count);
for (size_t i = 0; i < count; ++i) {
m_rareData->m_exceptionHandlers[i].start = unlinkedHandlers[i].start;
m_rareData->m_exceptionHandlers[i].end = unlinkedHandlers[i].end;
m_rareData->m_exceptionHandlers[i].target = unlinkedHandlers[i].target;
m_rareData->m_exceptionHandlers[i].scopeDepth = unlinkedHandlers[i].scopeDepth;
}
}
HandlerInfo& exceptionHandler(int index) { RELEASE_ASSERT(m_rareData); return m_rareData->m_exceptionHandlers[index]; }
bool hasExpressionInfo() { return m_unlinkedCode->hasExpressionInfo(); }
#if ENABLE(DFG_JIT)
Vector<CodeOrigin, 0, UnsafeVectorOverflow>& codeOrigins()
{
return m_jitCode->dfgCommon()->codeOrigins;
}
// Having code origins implies that there has been some inlining.
bool hasCodeOrigins()
{
return JITCode::isOptimizingJIT(jitType());
}
bool canGetCodeOrigin(unsigned index)
{
if (!hasCodeOrigins())
return false;
return index < codeOrigins().size();
}
CodeOrigin codeOrigin(unsigned index)
{
return codeOrigins()[index];
}
bool addFrequentExitSite(const DFG::FrequentExitSite& site)
{
ASSERT(JITCode::isBaselineCode(jitType()));
ConcurrentJITLocker locker(m_lock);
return m_exitProfile.add(locker, site);
}
bool hasExitSite(const DFG::FrequentExitSite& site) const
{
ConcurrentJITLocker locker(m_lock);
return m_exitProfile.hasExitSite(locker, site);
}
DFG::ExitProfile& exitProfile() { return m_exitProfile; }
CompressedLazyOperandValueProfileHolder& lazyOperandValueProfiles()
{
return m_lazyOperandValueProfiles;
}
#else // ENABLE(DFG_JIT)
bool addFrequentExitSite(const DFG::FrequentExitSite&)
{
return false;
}
#endif // ENABLE(DFG_JIT)
// Constant Pool
#if ENABLE(DFG_JIT)
size_t numberOfIdentifiers() const { return m_unlinkedCode->numberOfIdentifiers() + numberOfDFGIdentifiers(); }
size_t numberOfDFGIdentifiers() const
{
if (!JITCode::isOptimizingJIT(jitType()))
return 0;
return m_jitCode->dfgCommon()->dfgIdentifiers.size();
}
const Identifier& identifier(int index) const
{
size_t unlinkedIdentifiers = m_unlinkedCode->numberOfIdentifiers();
if (static_cast<unsigned>(index) < unlinkedIdentifiers)
return m_unlinkedCode->identifier(index);
ASSERT(JITCode::isOptimizingJIT(jitType()));
return m_jitCode->dfgCommon()->dfgIdentifiers[index - unlinkedIdentifiers];
}
#else
size_t numberOfIdentifiers() const { return m_unlinkedCode->numberOfIdentifiers(); }
const Identifier& identifier(int index) const { return m_unlinkedCode->identifier(index); }
#endif
Vector<WriteBarrier<Unknown>>& constants() { return m_constantRegisters; }
size_t numberOfConstantRegisters() const { return m_constantRegisters.size(); }
unsigned addConstant(JSValue v)
{
unsigned result = m_constantRegisters.size();
m_constantRegisters.append(WriteBarrier<Unknown>());
m_constantRegisters.last().set(m_globalObject->vm(), m_ownerExecutable.get(), v);
return result;
}
unsigned addConstantLazily()
{
unsigned result = m_constantRegisters.size();
m_constantRegisters.append(WriteBarrier<Unknown>());
return result;
}
bool findConstant(JSValue, unsigned& result);
unsigned addOrFindConstant(JSValue);
WriteBarrier<Unknown>& constantRegister(int index) { return m_constantRegisters[index - FirstConstantRegisterIndex]; }
ALWAYS_INLINE bool isConstantRegisterIndex(int index) const { return index >= FirstConstantRegisterIndex; }
ALWAYS_INLINE JSValue getConstant(int index) const { return m_constantRegisters[index - FirstConstantRegisterIndex].get(); }
FunctionExecutable* functionDecl(int index) { return m_functionDecls[index].get(); }
int numberOfFunctionDecls() { return m_functionDecls.size(); }
FunctionExecutable* functionExpr(int index) { return m_functionExprs[index].get(); }
RegExp* regexp(int index) const { return m_unlinkedCode->regexp(index); }
unsigned numberOfConstantBuffers() const
{
if (!m_rareData)
return 0;
return m_rareData->m_constantBuffers.size();
}
unsigned addConstantBuffer(const Vector<JSValue>& buffer)
{
createRareDataIfNecessary();
unsigned size = m_rareData->m_constantBuffers.size();
m_rareData->m_constantBuffers.append(buffer);
return size;
}
Vector<JSValue>& constantBufferAsVector(unsigned index)
{
ASSERT(m_rareData);
return m_rareData->m_constantBuffers[index];
}
JSValue* constantBuffer(unsigned index)
{
return constantBufferAsVector(index).data();
}
JSGlobalObject* globalObject() { return m_globalObject.get(); }
JSGlobalObject* globalObjectFor(CodeOrigin);
// Jump Tables
size_t numberOfSwitchJumpTables() const { return m_rareData ? m_rareData->m_switchJumpTables.size() : 0; }
SimpleJumpTable& addSwitchJumpTable() { createRareDataIfNecessary(); m_rareData->m_switchJumpTables.append(SimpleJumpTable()); return m_rareData->m_switchJumpTables.last(); }
SimpleJumpTable& switchJumpTable(int tableIndex) { RELEASE_ASSERT(m_rareData); return m_rareData->m_switchJumpTables[tableIndex]; }
void clearSwitchJumpTables()
{
if (!m_rareData)
return;
m_rareData->m_switchJumpTables.clear();
}
size_t numberOfStringSwitchJumpTables() const { return m_rareData ? m_rareData->m_stringSwitchJumpTables.size() : 0; }
StringJumpTable& addStringSwitchJumpTable() { createRareDataIfNecessary(); m_rareData->m_stringSwitchJumpTables.append(StringJumpTable()); return m_rareData->m_stringSwitchJumpTables.last(); }
StringJumpTable& stringSwitchJumpTable(int tableIndex) { RELEASE_ASSERT(m_rareData); return m_rareData->m_stringSwitchJumpTables[tableIndex]; }
SharedSymbolTable* symbolTable() const { return m_unlinkedCode->symbolTable(); }
EvalCodeCache& evalCodeCache() { createRareDataIfNecessary(); return m_rareData->m_evalCodeCache; }
enum ShrinkMode {
// Shrink prior to generating machine code that may point directly into vectors.
EarlyShrink,
// Shrink after generating machine code, and after possibly creating new vectors
// and appending to others. At this time it is not safe to shrink certain vectors
// because we would have generated machine code that references them directly.
LateShrink
};
void shrinkToFit(ShrinkMode);
void copyPostParseDataFrom(CodeBlock* alternative);
void copyPostParseDataFromAlternative();
// Functions for controlling when JITting kicks in, in a mixed mode
// execution world.
bool checkIfJITThresholdReached()
{
return m_llintExecuteCounter.checkIfThresholdCrossedAndSet(this);
}
void dontJITAnytimeSoon()
{
m_llintExecuteCounter.deferIndefinitely();
}
void jitAfterWarmUp()
{
m_llintExecuteCounter.setNewThreshold(Options::thresholdForJITAfterWarmUp(), this);
}
void jitSoon()
{
m_llintExecuteCounter.setNewThreshold(Options::thresholdForJITSoon(), this);
}
const ExecutionCounter& llintExecuteCounter() const
{
return m_llintExecuteCounter;
}
// Functions for controlling when tiered compilation kicks in. This
// controls both when the optimizing compiler is invoked and when OSR
// entry happens. Two triggers exist: the loop trigger and the return
// trigger. In either case, when an addition to m_jitExecuteCounter
// causes it to become non-negative, the optimizing compiler is
// invoked. This includes a fast check to see if this CodeBlock has
// already been optimized (i.e. replacement() returns a CodeBlock
// that was optimized with a higher tier JIT than this one). In the
// case of the loop trigger, if the optimized compilation succeeds
// (or has already succeeded in the past) then OSR is attempted to
// redirect program flow into the optimized code.
// These functions are called from within the optimization triggers,
// and are used as a single point at which we define the heuristics
// for how much warm-up is mandated before the next optimization
// trigger files. All CodeBlocks start out with optimizeAfterWarmUp(),
// as this is called from the CodeBlock constructor.
// When we observe a lot of speculation failures, we trigger a
// reoptimization. But each time, we increase the optimization trigger
// to avoid thrashing.
unsigned reoptimizationRetryCounter() const;
void countReoptimization();
#if ENABLE(JIT)
unsigned numberOfDFGCompiles();
int32_t codeTypeThresholdMultiplier() const;
int32_t adjustedCounterValue(int32_t desiredThreshold);
int32_t* addressOfJITExecuteCounter()
{
return &m_jitExecuteCounter.m_counter;
}
static ptrdiff_t offsetOfJITExecuteCounter() { return OBJECT_OFFSETOF(CodeBlock, m_jitExecuteCounter) + OBJECT_OFFSETOF(ExecutionCounter, m_counter); }
static ptrdiff_t offsetOfJITExecutionActiveThreshold() { return OBJECT_OFFSETOF(CodeBlock, m_jitExecuteCounter) + OBJECT_OFFSETOF(ExecutionCounter, m_activeThreshold); }
static ptrdiff_t offsetOfJITExecutionTotalCount() { return OBJECT_OFFSETOF(CodeBlock, m_jitExecuteCounter) + OBJECT_OFFSETOF(ExecutionCounter, m_totalCount); }
const ExecutionCounter& jitExecuteCounter() const { return m_jitExecuteCounter; }
unsigned optimizationDelayCounter() const { return m_optimizationDelayCounter; }
// Check if the optimization threshold has been reached, and if not,
// adjust the heuristics accordingly. Returns true if the threshold has
// been reached.
bool checkIfOptimizationThresholdReached();
// Call this to force the next optimization trigger to fire. This is
// rarely wise, since optimization triggers are typically more
// expensive than executing baseline code.
void optimizeNextInvocation();
// Call this to prevent optimization from happening again. Note that
// optimization will still happen after roughly 2^29 invocations,
// so this is really meant to delay that as much as possible. This
// is called if optimization failed, and we expect it to fail in
// the future as well.
void dontOptimizeAnytimeSoon();
// Call this to reinitialize the counter to its starting state,
// forcing a warm-up to happen before the next optimization trigger
// fires. This is called in the CodeBlock constructor. It also
// makes sense to call this if an OSR exit occurred. Note that
// OSR exit code is code generated, so the value of the execute
// counter that this corresponds to is also available directly.
void optimizeAfterWarmUp();
// Call this to force an optimization trigger to fire only after
// a lot of warm-up.
void optimizeAfterLongWarmUp();
// Call this to cause an optimization trigger to fire soon, but
// not necessarily the next one. This makes sense if optimization
// succeeds. Successfuly optimization means that all calls are
// relinked to the optimized code, so this only affects call
// frames that are still executing this CodeBlock. The value here
// is tuned to strike a balance between the cost of OSR entry
// (which is too high to warrant making every loop back edge to
// trigger OSR immediately) and the cost of executing baseline
// code (which is high enough that we don't necessarily want to
// have a full warm-up). The intuition for calling this instead of
// optimizeNextInvocation() is for the case of recursive functions
// with loops. Consider that there may be N call frames of some
// recursive function, for a reasonably large value of N. The top
// one triggers optimization, and then returns, and then all of
// the others return. We don't want optimization to be triggered on
// each return, as that would be superfluous. It only makes sense
// to trigger optimization if one of those functions becomes hot
// in the baseline code.
void optimizeSoon();
void forceOptimizationSlowPathConcurrently();
void setOptimizationThresholdBasedOnCompilationResult(CompilationResult);
uint32_t osrExitCounter() const { return m_osrExitCounter; }
void countOSRExit() { m_osrExitCounter++; }
uint32_t* addressOfOSRExitCounter() { return &m_osrExitCounter; }
static ptrdiff_t offsetOfOSRExitCounter() { return OBJECT_OFFSETOF(CodeBlock, m_osrExitCounter); }
uint32_t adjustedExitCountThreshold(uint32_t desiredThreshold);
uint32_t exitCountThresholdForReoptimization();
uint32_t exitCountThresholdForReoptimizationFromLoop();
bool shouldReoptimizeNow();
bool shouldReoptimizeFromLoopNow();
#else // No JIT
void optimizeAfterWarmUp() { }
unsigned numberOfDFGCompiles() { return 0; }
#endif
#if ENABLE(VALUE_PROFILER)
bool shouldOptimizeNow();
void updateAllValueProfilePredictions();
void updateAllArrayPredictions();
void updateAllPredictions();
#else
bool updateAllPredictionsAndCheckIfShouldOptimizeNow() { return false; }
void updateAllValueProfilePredictions() { }
void updateAllArrayPredictions() { }
void updateAllPredictions() { }
#endif
#if ENABLE(VERBOSE_VALUE_PROFILE)
void dumpValueProfiles();
#endif
// FIXME: Make these remaining members private.
int m_numCalleeRegisters;
int m_numVars;
bool m_isConstructor;
// This is intentionally public; it's the responsibility of anyone doing any
// of the following to hold the lock:
//
// - Modifying any inline cache in this code block.
//
// - Quering any inline cache in this code block, from a thread other than
// the main thread.
//
// Additionally, it's only legal to modify the inline cache on the main
// thread. This means that the main thread can query the inline cache without
// locking. This is crucial since executing the inline cache is effectively
// "querying" it.
//
// Another exception to the rules is that the GC can do whatever it wants
// without holding any locks, because the GC is guaranteed to wait until any
// concurrent compilation threads finish what they're doing.
mutable ConcurrentJITLock m_lock;
bool m_shouldAlwaysBeInlined;
bool m_allTransitionsHaveBeenMarked; // Initialized and used on every GC.
bool m_didFailFTLCompilation;
protected:
virtual void visitWeakReferences(SlotVisitor&) OVERRIDE;
virtual void finalizeUnconditionally() OVERRIDE;
#if ENABLE(DFG_JIT)
void tallyFrequentExitSites();
#else
void tallyFrequentExitSites() { }
#endif
private:
friend class CodeBlockSet;
CodeBlock* specialOSREntryBlockOrNull();
void noticeIncomingCall(ExecState* callerFrame);
double optimizationThresholdScalingFactor();
#if ENABLE(JIT)
ClosureCallStubRoutine* findClosureCallForReturnPC(ReturnAddressPtr);
#endif
#if ENABLE(VALUE_PROFILER)
void updateAllPredictionsAndCountLiveness(unsigned& numberOfLiveNonArgumentValueProfiles, unsigned& numberOfSamplesInProfiles);
#endif
void setConstantRegisters(const Vector<WriteBarrier<Unknown>>& constants)
{
size_t count = constants.size();
m_constantRegisters.resize(count);
for (size_t i = 0; i < count; i++)
m_constantRegisters[i].set(*m_vm, ownerExecutable(), constants[i].get());
}
void dumpBytecode(PrintStream&, ExecState*, const Instruction* begin, const Instruction*&, const StubInfoMap& = StubInfoMap());
CString registerName(int r) const;
void printUnaryOp(PrintStream&, ExecState*, int location, const Instruction*&, const char* op);
void printBinaryOp(PrintStream&, ExecState*, int location, const Instruction*&, const char* op);
void printConditionalJump(PrintStream&, ExecState*, const Instruction*, const Instruction*&, int location, const char* op);
void printGetByIdOp(PrintStream&, ExecState*, int location, const Instruction*&);
void printGetByIdCacheStatus(PrintStream&, ExecState*, int location, const StubInfoMap&);
enum CacheDumpMode { DumpCaches, DontDumpCaches };
void printCallOp(PrintStream&, ExecState*, int location, const Instruction*&, const char* op, CacheDumpMode, bool& hasPrintedProfiling);
void printPutByIdOp(PrintStream&, ExecState*, int location, const Instruction*&, const char* op);
void printLocationAndOp(PrintStream& out, ExecState*, int location, const Instruction*&, const char* op)
{
out.printf("[%4d] %-17s ", location, op);
}
void printLocationOpAndRegisterOperand(PrintStream& out, ExecState* exec, int location, const Instruction*& it, const char* op, int operand)
{
printLocationAndOp(out, exec, location, it, op);
out.printf("%s", registerName(operand).data());
}
void beginDumpProfiling(PrintStream&, bool& hasPrintedProfiling);
void dumpValueProfiling(PrintStream&, const Instruction*&, bool& hasPrintedProfiling);
void dumpArrayProfiling(PrintStream&, const Instruction*&, bool& hasPrintedProfiling);
#if ENABLE(VALUE_PROFILER)
void dumpRareCaseProfile(PrintStream&, const char* name, RareCaseProfile*, bool& hasPrintedProfiling);
#endif
#if ENABLE(DFG_JIT)
bool shouldImmediatelyAssumeLivenessDuringScan()
{
// Interpreter and Baseline JIT CodeBlocks don't need to be jettisoned when
// their weak references go stale. So if a basline JIT CodeBlock gets
// scanned, we can assume that this means that it's live.
if (!JITCode::isOptimizingJIT(jitType()))
return true;
// For simplicity, we don't attempt to jettison code blocks during GC if
// they are executing. Instead we strongly mark their weak references to
// allow them to continue to execute soundly.
if (m_mayBeExecuting)
return true;
if (Options::forceDFGCodeBlockLiveness())
return true;
return false;
}
#else
bool shouldImmediatelyAssumeLivenessDuringScan() { return true; }
#endif
void propagateTransitions(SlotVisitor&);
void determineLiveness(SlotVisitor&);
void stronglyVisitStrongReferences(SlotVisitor&);
void stronglyVisitWeakReferences(SlotVisitor&);
void createRareDataIfNecessary()
{
if (!m_rareData)
m_rareData = adoptPtr(new RareData);
}
#if ENABLE(JIT)
void resetStubInternal(RepatchBuffer&, StructureStubInfo&);
void resetStubDuringGCInternal(RepatchBuffer&, StructureStubInfo&);
#endif
WriteBarrier<UnlinkedCodeBlock> m_unlinkedCode;
int m_numParameters;
WriteBarrier<ScriptExecutable> m_ownerExecutable;
VM* m_vm;
RefCountedArray<Instruction> m_instructions;
VirtualRegister m_thisRegister;
VirtualRegister m_argumentsRegister;
VirtualRegister m_activationRegister;
bool m_isStrictMode;
bool m_needsActivation;
bool m_mayBeExecuting;
uint8_t m_visitAggregateHasBeenCalled;
RefPtr<SourceProvider> m_source;
unsigned m_sourceOffset;
unsigned m_firstLineColumnOffset;
unsigned m_codeType;
#if ENABLE(LLINT)
Vector<LLIntCallLinkInfo> m_llintCallLinkInfos;
SentinelLinkedList<LLIntCallLinkInfo, BasicRawSentinelNode<LLIntCallLinkInfo>> m_incomingLLIntCalls;
#endif
RefPtr<JITCode> m_jitCode;
MacroAssemblerCodePtr m_jitCodeWithArityCheck;
#if ENABLE(JIT)
Bag<StructureStubInfo> m_stubInfos;
Vector<ByValInfo> m_byValInfos;
Vector<CallLinkInfo> m_callLinkInfos;
SentinelLinkedList<CallLinkInfo, BasicRawSentinelNode<CallLinkInfo>> m_incomingCalls;
#endif
OwnPtr<CompactJITCodeMap> m_jitCodeMap;
#if ENABLE(DFG_JIT)
// This is relevant to non-DFG code blocks that serve as the profiled code block
// for DFG code blocks.
DFG::ExitProfile m_exitProfile;
CompressedLazyOperandValueProfileHolder m_lazyOperandValueProfiles;
#endif
#if ENABLE(VALUE_PROFILER)
Vector<ValueProfile> m_argumentValueProfiles;
SegmentedVector<ValueProfile, 8> m_valueProfiles;
SegmentedVector<RareCaseProfile, 8> m_rareCaseProfiles;
SegmentedVector<RareCaseProfile, 8> m_specialFastCaseProfiles;
SegmentedVector<ArrayAllocationProfile, 8> m_arrayAllocationProfiles;
ArrayProfileVector m_arrayProfiles;
#endif
SegmentedVector<ObjectAllocationProfile, 8> m_objectAllocationProfiles;
// Constant Pool
Vector<Identifier> m_additionalIdentifiers;
COMPILE_ASSERT(sizeof(Register) == sizeof(WriteBarrier<Unknown>), Register_must_be_same_size_as_WriteBarrier_Unknown);
// TODO: This could just be a pointer to m_unlinkedCodeBlock's data, but the DFG mutates
// it, so we're stuck with it for now.
Vector<WriteBarrier<Unknown>> m_constantRegisters;
Vector<WriteBarrier<FunctionExecutable>> m_functionDecls;
Vector<WriteBarrier<FunctionExecutable>> m_functionExprs;
RefPtr<CodeBlock> m_alternative;
ExecutionCounter m_llintExecuteCounter;
ExecutionCounter m_jitExecuteCounter;
int32_t m_totalJITExecutions;
uint32_t m_osrExitCounter;
uint16_t m_optimizationDelayCounter;
uint16_t m_reoptimizationRetryCounter;
mutable CodeBlockHash m_hash;
struct RareData {
WTF_MAKE_FAST_ALLOCATED;
public:
Vector<HandlerInfo> m_exceptionHandlers;
// Buffers used for large array literals
Vector<Vector<JSValue>> m_constantBuffers;
// Jump Tables
Vector<SimpleJumpTable> m_switchJumpTables;
Vector<StringJumpTable> m_stringSwitchJumpTables;
EvalCodeCache m_evalCodeCache;
};
#if COMPILER(MSVC)
friend void WTF::deleteOwnedPtr<RareData>(RareData*);
#endif
OwnPtr<RareData> m_rareData;
#if ENABLE(JIT)
DFG::CapabilityLevel m_capabilityLevelState;
#endif
};
// Program code is not marked by any function, so we make the global object
// responsible for marking it.
class GlobalCodeBlock : public CodeBlock {
protected:
GlobalCodeBlock(CopyParsedBlockTag, GlobalCodeBlock& other)
: CodeBlock(CopyParsedBlock, other)
{
}
GlobalCodeBlock(ScriptExecutable* ownerExecutable, UnlinkedCodeBlock* unlinkedCodeBlock, JSScope* scope, PassRefPtr<SourceProvider> sourceProvider, unsigned sourceOffset, unsigned firstLineColumnOffset)
: CodeBlock(ownerExecutable, unlinkedCodeBlock, scope, sourceProvider, sourceOffset, firstLineColumnOffset)
{
}
};
class ProgramCodeBlock : public GlobalCodeBlock {
public:
ProgramCodeBlock(CopyParsedBlockTag, ProgramCodeBlock& other)
: GlobalCodeBlock(CopyParsedBlock, other)
{
}
ProgramCodeBlock(ProgramExecutable* ownerExecutable, UnlinkedProgramCodeBlock* unlinkedCodeBlock, JSScope* scope, PassRefPtr<SourceProvider> sourceProvider, unsigned firstLineColumnOffset)
: GlobalCodeBlock(ownerExecutable, unlinkedCodeBlock, scope, sourceProvider, 0, firstLineColumnOffset)
{
}
#if ENABLE(JIT)
protected:
virtual CodeBlock* replacement() OVERRIDE;
virtual DFG::CapabilityLevel capabilityLevelInternal() OVERRIDE;
#endif
};
class EvalCodeBlock : public GlobalCodeBlock {
public:
EvalCodeBlock(CopyParsedBlockTag, EvalCodeBlock& other)
: GlobalCodeBlock(CopyParsedBlock, other)
{
}
EvalCodeBlock(EvalExecutable* ownerExecutable, UnlinkedEvalCodeBlock* unlinkedCodeBlock, JSScope* scope, PassRefPtr<SourceProvider> sourceProvider)
: GlobalCodeBlock(ownerExecutable, unlinkedCodeBlock, scope, sourceProvider, 0, 1)
{
}
const Identifier& variable(unsigned index) { return unlinkedEvalCodeBlock()->variable(index); }
unsigned numVariables() { return unlinkedEvalCodeBlock()->numVariables(); }
#if ENABLE(JIT)
protected:
virtual CodeBlock* replacement() OVERRIDE;
virtual DFG::CapabilityLevel capabilityLevelInternal() OVERRIDE;
#endif
private:
UnlinkedEvalCodeBlock* unlinkedEvalCodeBlock() const { return jsCast<UnlinkedEvalCodeBlock*>(unlinkedCodeBlock()); }
};
class FunctionCodeBlock : public CodeBlock {
public:
FunctionCodeBlock(CopyParsedBlockTag, FunctionCodeBlock& other)
: CodeBlock(CopyParsedBlock, other)
{
}
FunctionCodeBlock(FunctionExecutable* ownerExecutable, UnlinkedFunctionCodeBlock* unlinkedCodeBlock, JSScope* scope, PassRefPtr<SourceProvider> sourceProvider, unsigned sourceOffset, unsigned firstLineColumnOffset)
: CodeBlock(ownerExecutable, unlinkedCodeBlock, scope, sourceProvider, sourceOffset, firstLineColumnOffset)
{
}
#if ENABLE(JIT)
protected:
virtual CodeBlock* replacement() OVERRIDE;
virtual DFG::CapabilityLevel capabilityLevelInternal() OVERRIDE;
#endif
};
inline CodeBlock* baselineCodeBlockForInlineCallFrame(InlineCallFrame* inlineCallFrame)
{
RELEASE_ASSERT(inlineCallFrame);
ExecutableBase* executable = inlineCallFrame->executable.get();
RELEASE_ASSERT(executable->structure()->classInfo() == FunctionExecutable::info());
return static_cast<FunctionExecutable*>(executable)->baselineCodeBlockFor(inlineCallFrame->isCall ? CodeForCall : CodeForConstruct);
}
inline CodeBlock* baselineCodeBlockForOriginAndBaselineCodeBlock(const CodeOrigin& codeOrigin, CodeBlock* baselineCodeBlock)
{
if (codeOrigin.inlineCallFrame)
return baselineCodeBlockForInlineCallFrame(codeOrigin.inlineCallFrame);
return baselineCodeBlock;
}
inline int CodeBlock::argumentIndexAfterCapture(size_t argument)
{
if (argument >= static_cast<size_t>(symbolTable()->parameterCount()))
return CallFrame::argumentOffset(argument);
const SlowArgument* slowArguments = symbolTable()->slowArguments();
if (!slowArguments || slowArguments[argument].status == SlowArgument::Normal)
return CallFrame::argumentOffset(argument);
ASSERT(slowArguments[argument].status == SlowArgument::Captured);
return slowArguments[argument].index;
}
inline bool CodeBlock::hasSlowArguments()
{
return !!symbolTable()->slowArguments();
}
inline Register& ExecState::r(int index)
{
CodeBlock* codeBlock = this->codeBlock();
if (codeBlock->isConstantRegisterIndex(index))
return *reinterpret_cast<Register*>(&codeBlock->constantRegister(index));
return this[index];
}
inline Register& ExecState::uncheckedR(int index)
{
RELEASE_ASSERT(index < FirstConstantRegisterIndex);
return this[index];
}
inline JSValue ExecState::argumentAfterCapture(size_t argument)
{
if (argument >= argumentCount())
return jsUndefined();
if (!codeBlock())
return this[argumentOffset(argument)].jsValue();
return this[codeBlock()->argumentIndexAfterCapture(argument)].jsValue();
}
inline void CodeBlockSet::mark(void* candidateCodeBlock)
{
// We have to check for 0 and -1 because those are used by the HashMap as markers.
uintptr_t value = reinterpret_cast<uintptr_t>(candidateCodeBlock);
// This checks for both of those nasty cases in one go.
// 0 + 1 = 1
// -1 + 1 = 0
if (value + 1 <= 1)
return;
HashSet<CodeBlock*>::iterator iter = m_set.find(static_cast<CodeBlock*>(candidateCodeBlock));
if (iter == m_set.end())
return;
(*iter)->m_mayBeExecuting = true;
}
} // namespace JSC
#endif // CodeBlock_h