2013-04-15 Roger Fong <roger_fong@apple.com> Build fix for Windows. <rdar://problem/13438603> * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: 2013-04-15 Lucas Forschler <lforschler@apple.com> Merge r141029 2013-01-28 Oliver Hunt <oliver@apple.com> Add more assertions to the property storage use in arrays https://bugs.webkit.org/show_bug.cgi?id=107728 Reviewed by Filip Pizlo. Add a bunch of assertions to array and object butterfly usage. This should make debugging somewhat easier. I also converted a couple of assertions to release asserts as they were so low cost it seemed a sensible thing to do. * runtime/JSArray.cpp: (JSC::JSArray::sortVector): (JSC::JSArray::compactForSorting): * runtime/JSObject.h: (JSC::JSObject::getHolyIndexQuickly): 2013-04-14 David Kilzer <ddkilzer@apple.com> Revert "Attempt to fix Windows build after r148370" This reverts commit r148375. * runtime/JSStringJoiner.h: Remove include of <wtf/CheckedArithmetic.h>. 2013-04-13 David Kilzer <ddkilzer@apple.com> Attempt to fix Windows build after r148370 Try to fix the following build failure: 4>JSStringJoiner.cpp 4>..\..\runtime\JSStringJoiner.cpp(112) : error C2653: 'CheckedState' : is not a class or namespace name 4>..\..\runtime\JSStringJoiner.cpp(112) : error C2065: 'DidOverflow' : undeclared identifier 4>..\..\runtime\JSStringJoiner.cpp(112) : warning C4805: '==' : unsafe mix of type 'bool' and type ''unknown-type'' in operation * runtime/JSStringJoiner.h: Include <wtf/CheckedArithmetic.h>. 2013-04-13 David Kilzer <ddkilzer@apple.com> Merge 145580, 145587, 145594, r145595, r145647, 145759. Part 2/2 for <rdar://problem/13438603>. 2013-03-11 Oliver Hunt <oliver@apple.com> Harden JSStringJoiner https://bugs.webkit.org/show_bug.cgi?id=112093 Reviewed by Filip Pizlo. Harden JSStringJoiner, make it use our CheckedArithmetic class to simplify everything. * runtime/JSStringJoiner.cpp: (JSC::JSStringJoiner::build): * runtime/JSStringJoiner.h: (JSStringJoiner): (JSC::JSStringJoiner::JSStringJoiner): (JSC::JSStringJoiner::append): 2013-03-12 Lucas Forschler <lforschler@apple.com> Merge r135794 2012-11-26 Daniel Bates <dbates@webkit.org> JavaScript fails to handle String.replace() with large replacement string https://bugs.webkit.org/show_bug.cgi?id=102956 <rdar://problem/12738012> Reviewed by Oliver Hunt. Fix an issue where we didn't check for overflow when computing the length of the result of String.replace() with a large replacement string. * runtime/StringPrototype.cpp: (JSC::jsSpliceSubstringsWithSeparators): 2013-03-12 Lucas Forschler <lforschler@apple.com> Merge r135454 2012-11-21 Daniel Bates <dbates@webkit.org> JavaScript fails to concatenate large strings <https://bugs.webkit.org/show_bug.cgi?id=102963> Reviewed by Michael Saboff. Fixes an issue where we inadvertently didn't check the length of a JavaScript string for overflow. * runtime/Operations.h: (JSC::jsString): (JSC::jsStringFromArguments): 2013-01-30 Lucas Forschler <lforschler@apple.com> Merge r138606 2013-01-01 Dan Bernstein <mitz@apple.com> <rdar://problem/12942239> Update copyright strings Reviewed by Sam Weinig. * Info.plist: 2012-11-16 Andy Estes <aestes@apple.com> Merge r130102. 2012-10-01 Filip Pizlo <fpizlo@apple.com> Address a FIXME in JSArray::sort https://bugs.webkit.org/show_bug.cgi?id=98080 <rdar://problem/12407844> Reviewed by Oliver Hunt. Get rid of fast sorting of sparse maps. I don't know that it's broken but I do know that we don't have coverage for it. Then also address the FIXME in JSArray::sort regarding side-effecting compare functions. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncSort): * runtime/JSArray.cpp: (JSC::JSArray::sortNumeric): (JSC::JSArray::sort): (JSC::JSArray::compactForSorting): * runtime/JSArray.h: (JSArray): * runtime/JSObject.h: (JSC::JSObject::hasSparseMap): (JSObject): 2012-11-14 Lucas Forschler <lforschler@apple.com> Merge fix for <rdar://problem/12691662>. 2012-11-14 Filip Pizlo <fpizlo@apple.com> Value recovery for arguments should be based on the argument-aware prediction. <rdar://problem/12046032> Reviewed by Gavin Barraclough, Oliver Hunt, and Filip Pizlo. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-11-12 Lucas Forschler <lforschler@apple.com> Merge fix for <rdar://problem/12486173>. 2012-11-12 Filip Pizlo <fpizlo@apple.com> Redundant phi elimination breaks CFA. <rdar://problem/11152598> Reviewed by Gavin Barraclough. * dfg/DFGDriver.cpp: (JSC::DFG::compile): 2012-11-12 Lucas Forschler <lforschler@apple.com> Merge fix for <rdar://problem/12486142>. 2012-08-13 Filip Pizlo <fpizlo@apple.com> Value recovery for arguments should be based on the argument-aware prediction. <rdar://problem/12046032> Reviewed by Gavin Barraclough. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-11-12 Lucas Forschler <lforschler@apple.com> Merge fix for <rdar://problem/12486142>. 2012-08-13 Filip Pizlo <fpizlo@apple.com> Value recovery for arguments should always just indicate that the argument is in the register file, and not attempt to claim anything about the tag. <rdar://problem/12046032> The code previously attempted to base the value recovery on the VariableAccessData prediction. This is wrong for two reasons: first, SetLocals speculate on the argument-aware prediction, which is different from the prediction, if the argument was used to store different types; and second, even if the argument-aware prediction indicates some unboxed type (int, bool, or cell) then it must be the case that the tag that was stored by the caller would be identical to the tag implied by the prediction - so OSR can just assume that the argument already holds a fully decorated JSValue. Of special note are double predictions: the DFG has never stored unboxed doubles into arguments, so those don't need special treatment here. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-07-24 Filip Pizlo <fpizlo@apple.com> DFG method checks should keep the receiver of the property access alive until all checks complete. <rdar://problem/11781190> Reviewed by Gavin Barraclough. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): 2012-11-12 Lucas Forschler <lforschler@apple.com> Merge r129440 2012-09-24 Filip Pizlo <fpizlo@apple.com> Nested try/finally should not confuse the finally unpopper in BytecodeGenerator::emitComplexJumpScopes https://bugs.webkit.org/show_bug.cgi?id=97508 <rdar://problem/12361132> Reviewed by Sam Weinig. We're reusing some vector for multiple iterations of a loop, but we were forgetting to clear its contents from one iteration to the next. Hence if you did multiple iterations of finally unpopping (like in a nested try/finally and a jump out of both of them) then you'd get a corrupted try context stack afterwards. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitComplexJumpScopes): 2012-11-12 Lucas Forschler <lforschler@apple.com> Merge r126718 2012-08-24 Filip Pizlo <fpizlo@apple.com> Finally inlining should correctly track the catch context https://bugs.webkit.org/show_bug.cgi?id=94986 <rdar://problem/11753784> Reviewed by Sam Weinig. This fixes two behaviors: 1) Throwing from a finally block. Previously, we would seem to reenter the finally block - though only once. 2) Executing a finally block from some nested context, for example due to a 'continue', 'break', or 'return' in the try. This would execute the finally block in the context of of the try block, which could lead to either scope depth mismatches or reexecutions of the finally block on throw, similarly to (1) but for different reasons. * bytecompiler/BytecodeGenerator.cpp: (JSC): (JSC::BytecodeGenerator::pushFinallyContext): (JSC::BytecodeGenerator::emitComplexJumpScopes): (JSC::BytecodeGenerator::pushTry): (JSC::BytecodeGenerator::popTryAndEmitCatch): * bytecompiler/BytecodeGenerator.h: (FinallyContext): (TryData): (JSC): (TryContext): (TryRange): (BytecodeGenerator): * bytecompiler/NodesCodegen.cpp: (JSC::TryNode::emitBytecode): 2012-11-12 Lucas Forschler <lforschler@apple.com> Merge r129577 2012-09-25 Filip Pizlo <fpizlo@apple.com> We shouldn't use the optimized versions of shift/unshift if the user is doing crazy things to the array https://bugs.webkit.org/show_bug.cgi?id=97603 <rdar://problem/12370864> Reviewed by Gavin Barraclough. You changed the length behind our backs? No optimizations for you then! * runtime/ArrayPrototype.cpp: (JSC::shift): (JSC::unshift): * runtime/JSArray.cpp: (JSC::JSArray::shiftCount): 2012-11-09 Lucas Forschler <lforschler@apple.com> Merge r126624 2012-08-24 Oliver Hunt <oliver@apple.com> Always null check cells before marking https://bugs.webkit.org/show_bug.cgi?id=94968 Reviewed by Geoffrey Garen. Originally we tried to minimise null checks by only null checking values that we knew could be null, however given that we can't ever guarantee when a GC will happen, we're better off just always assuming that a null check will be necessary. This results in a much less fragile code base as we can add GC allocations to object initialisers without having to subsequently worry about whether the object we are initialising will need to add a bunch of null checks in its visitChildren implementation. * heap/MarkStack.cpp: (JSC::MarkStack::internalAppend): * heap/MarkStackInlineMethods.h: (JSC::MarkStack::append): (JSC::MarkStack::appendUnbarrieredPointer): * runtime/Structure.h: (JSC::MarkStack::internalAppend): 2012-11-08 Lucas Forschler <lforschler@apple.com> Merge r125806 2012-08-16 Filip Pizlo <fpizlo@apple.com> All op_resolve_global instructions should end up in the list of global resolve instructions https://bugs.webkit.org/show_bug.cgi?id=94247 <rdar://problem/12103500> Reviewed by Mark Hahnenberg. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitResolveWithBase): 2012-09-20 Roger Fong <roger_fong@apple.com> Enable opportunistic garbage collecting on Windows. Reviewed by Mark Hahnenberg * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: 2012-08-14 Lucas Forschler <lforschler@apple.com> Merge r124272. 2012-07-31 Sam Weinig <sam@webkit.org> Fix the Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-08-14 Lucas Forschler <lforschler@apple.com> Merge r124268. 2012-07-31 Sam Weinig <sam@webkit.org> Stop masking 8 bits off of the visited link hash. We need all the bits! https://bugs.webkit.org/show_bug.cgi?id=92799 Reviewed by Anders Carlsson. * runtime/Identifier.cpp: (JSC::IdentifierCStringTranslator::hash): (JSC::IdentifierLCharFromUCharTranslator::hash): * runtime/Identifier.h: (JSC::IdentifierCharBufferTranslator::hash): Update for new function names. 2012-08-10 Lucas Forschler <lforschler@apple.com> Windows build fix after merging radar 12050720. Add missing symbols. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-08-10 Lucas Forschler <lforschler@apple.com> Windows build fix after merging radar 12050720. Part 2 - removing symbols. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-08-10 Lucas Forschler <lforschler@apple.com> Windows Build fix after merging radar 12050720. * runtime/JSLock.cpp: (JSC::JSLock::DropAllLocks::DropAllLocks): 2012-08-06 Lucas Forschler <lforschler@apple.com> Merge patch for <rdar://problem/11781190> 2012-07-24 Filip Pizlo <fpizlo@apple.com> DFG method checks should keep the receiver of the property access alive until all checks complete. <rdar://problem/11781190> Reviewed by Gavin Barraclough. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): 2012-08-02 Lucas Forschler <lforschler@apple.com> Merge 121307 2012-06-26 Filip Pizlo <fpizlo@apple.com> DFG PutByValAlias is too aggressive https://bugs.webkit.org/show_bug.cgi?id=90026 <rdar://problem/11751830> Reviewed by Gavin Barraclough. For CSE on normal arrays, we now treat PutByVal as impure. This does not appear to affect performance by much. For CSE on typed arrays, we fix PutByValAlias by making GetByVal speculate that the access is within bounds. This also has the effect of making our out-of-bounds handling consistent with WebCore. * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::performNodeCSE): * dfg/DFGGraph.h: (JSC::DFG::Graph::byValIsPure): (JSC::DFG::Graph::clobbersWorld): * dfg/DFGNodeType.h: (DFG): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileGetByValOnIntTypedArray): (JSC::DFG::SpeculativeJIT::compileGetByValOnFloatTypedArray): 2012-07-30 Lucas Forschler <lforschler@apple.com> Merge 121391 2012-06-27 Filip Pizlo <fpizlo@apple.com> Javascript SHA-512 gives wrong hash on second and subsequent runs unless Web Inspector Javascript Debugging is on https://bugs.webkit.org/show_bug.cgi?id=90053 <rdar://problem/11764613> Reviewed by Mark Hahnenberg. The problem is that the code was assuming that the recovery should be Undefined if the source of the SetLocal was !shouldGenerate(). But that's wrong, since the DFG optimizer may skip around a UInt32ToNumber node (hence making it !shouldGenerate()) and keep the source of that node alive. In that case we should base the recovery on the source of the UInt32ToNumber. The logic for this was already in place but the fast check for !shouldGenerate() broke it. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): 2012-06-06 Mark Rowe <mrowe@apple.com> Merge r118995. 2012-05-30 Oliver Hunt <oliver@apple.com> Really provide error information with the inspector disabled https://bugs.webkit.org/show_bug.cgi?id=87910 Reviewed by Filip Pizlo. Don't bother checking for anything other than pre-existing error info. In the absence of complete line number information you'll only get the line a function starts on, but at least it's something. * interpreter/Interpreter.cpp: (JSC::Interpreter::throwException): 2012-06-06 Mark Rowe <mrowe@apple.com> Merge r118992. 2012-05-30 Filip Pizlo <fpizlo@apple.com> LLInt broken on x86-32 with JIT turned off https://bugs.webkit.org/show_bug.cgi?id=87906 Reviewed by Geoffrey Garen. Fixed the code to not clobber registers that contain important things, like the call frame. * llint/LowLevelInterpreter32_64.asm: 2012-05-31 Ojan Vafai <ojan@chromium.org> add back the ability to disable flexbox https://bugs.webkit.org/show_bug.cgi?id=87147 Reviewed by Tony Chang. * Configurations/FeatureDefines.xcconfig: 2012-05-31 Tim Horton <timothy_horton@apple.com> Disable CSS3 flexbox <rdar://problem/11524921> Reviewed by John Sullivan. * Configurations/FeatureDefines.xcconfig: 2012-05-31 Tim Horton <timothy_horton@apple.com> Add feature defines for web-facing parts of CSS Regions and Exclusions https://bugs.webkit.org/show_bug.cgi?id=87442 <rdar://problem/10887709> Reviewed by Dan Bernstein. * Configurations/FeatureDefines.xcconfig: 2012-05-30 Lucas Forschler <lforschler@apple.com> Merge 118956 2012-05-30 Oliver Hunt <oliver@apple.com> DFG does not correctly handle exceptions caught in the LLInt https://bugs.webkit.org/show_bug.cgi?id=87885 Reviewed by Filip Pizlo. Make the DFG use genericThrow, rather than reimplementing a small portion of it. Also make the LLInt slow paths validate that their PC is correct. * dfg/DFGOperations.cpp: * llint/LLIntSlowPaths.cpp: (LLInt): 2012-05-30 Lucas Forschler <lforschler@apple.com> Merge 118810 2012-05-29 Mark Hahnenberg <mhahnenberg@apple.com> CopiedSpace::doneCopying could start another collection https://bugs.webkit.org/show_bug.cgi?id=86538 Reviewed by Geoffrey Garen. It's possible that if we don't have anything at the head of to-space after a collection and the BlockAllocator doesn't have any fresh blocks to give us right now we could start another collection while still in the middle of the first collection when we call CopiedSpace::addNewBlock(). One way to resolve this would be to have Heap::shouldCollect() check that m_operationInProgress is NoOperation. This would prevent the path in getFreshBlock() that starts the collection if we're already in the middle of one. I could not come up with a test case to reproduce this crash on ToT. * heap/Heap.h: (JSC::Heap::shouldCollect): We shouldn't collect if we're already in the middle of a collection, i.e. the current operation should be NoOperation. 2012-05-30 Lucas Forschler <lforschler@apple.com> Merge <rdar://problem/11519288> 2012-05-23 Lucas Forschler <lforschler@apple.com> Merge 117860 2012-05-21 Michael Saboff <msaboff@apple.com> Cleanup of Calls to operationStrCat and operationNewArray and Use Constructor after r117729 https://bugs.webkit.org/show_bug.cgi?id=87027 Reviewed by Oliver Hunt. Change calls to operationStrCat and operationNewArray to provide the pointer to the EncodedJSValue* data buffer instead of the ScratchBuffer that contains it. Added a ScratchBuffer::create() function. This is a clean-up to r117729. * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/JSGlobalData.h: (JSC::ScratchBuffer::create): (JSC::ScratchBuffer::dataBuffer): (JSC::JSGlobalData::scratchBufferForSize): 2012-05-23 Lucas Forschler <lforschler@apple.com> Merge 117729 2012-05-20 Michael Saboff <msaboff@apple.com> JSGlobalData ScratchBuffers Are Not Visited During Garbage Collection https://bugs.webkit.org/show_bug.cgi?id=86553 Reviewed by Gavin Barraclough. Scratch buffers can contain the only reference to live objects. Therefore visit scratch buffer contents as conservative roots. Changed the scratch buffers to be a struct with an "active" length and the actual buffer. The users of the scratch buffer emit code where needed to set and clear the active length as appropriate. During marking, the active count is used for conservative marking. * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::debugCall): * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGThunks.cpp: (JSC::DFG::osrExitGenerationThunkGenerator): * heap/Heap.cpp: (JSC::Heap::markRoots): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::gatherConservativeRoots): * runtime/JSGlobalData.h: (JSC::ScratchBuffer::ScratchBuffer): (ScratchBuffer): (JSC::ScratchBuffer::allocationSize): (JSC::ScratchBuffer::setActiveLength): (JSC::ScratchBuffer::activeLength): (JSC::ScratchBuffer::activeLengthPtr): (JSC::ScratchBuffer::dataBuffer): (JSGlobalData): (JSC::JSGlobalData::scratchBufferForSize): 2012-05-21 Lucas Forschler <lforschler@apple.com> Merge 117523 2012-05-17 Filip Pizlo <fpizlo@apple.com> Setting array index -1 and looping over array causes bad behavior https://bugs.webkit.org/show_bug.cgi?id=86733 <rdar://problem/11477670> Reviewed by Oliver Hunt. * dfg/DFGOperations.cpp: 2012-05-21 Lucas Forschler <lforschler@apple.com> Merge 117193 2012-05-15 Oliver Hunt <oliver@apple.com> Make error information available even if all we have is line number information. https://bugs.webkit.org/show_bug.cgi?id=86547 Reviewed by Filip Pizlo. We don't need expression information to generate useful line, file, and stack information, so only require that we have line number info available. * interpreter/Interpreter.cpp: (JSC::Interpreter::throwException): * runtime/Executable.h: (JSC): 2012-05-21 Lucas Forschler <lforschler@apple.com> Merge 117201 2012-05-15 Mark Hahnenberg <mhahnenberg@apple.com> Block freeing thread should not free blocks when we are actively requesting them https://bugs.webkit.org/show_bug.cgi?id=86519 Reviewed by Geoff Garen. * heap/BlockAllocator.h: (JSC::BlockAllocator::allocate): Reordering the setting of the flag so its done while we hold the lock to ensure proper locking. 2012-05-21 Lucas Forschler <lforschler@apple.com> Merge 117183 2012-05-15 Mark Hahnenberg <mhahnenberg@apple.com> Block freeing thread should not free blocks when we are actively requesting them https://bugs.webkit.org/show_bug.cgi?id=86519 Reviewed by Geoffrey Garen. The block freeing thread shoots us in the foot if it decides to run while we're actively requesting blocks and returning them. This situation can arise when there is a lot of copying collection going on in steady state. We allocate a large swath of pages to copy into, then we return all the newly free old pages to the BlockAllocator. In this state, if the block freeing thread wakes up in between collections (which is more likely than it waking up during a collection) and frees half of these pages, they will be needed almost immediately during the next collection, causing a storm of VM allocations which we know are going to be very slow. What we'd like is for when things have quieted down the block freeing thread can then return memory to the OS. Usually this will be when a page has fully loaded and has a low allocation rate. In this situation, our opportunistic collections will only be running at least every few seconds, thus the extra time spent doing VM allocations won't matter nearly as much as, say, while a page is loading. * heap/BlockAllocator.cpp: (JSC::BlockAllocator::BlockAllocator): Initialize our new field. (JSC::BlockAllocator::blockFreeingThreadMain): We check if we've seen any block requests recently. If so, reset our flag and go back to sleep. We also don't bother with locking here. If we miss out on an update, we'll see it when we wake up again. * heap/BlockAllocator.h: Add new field to track whether or not we've received recent block requests. (BlockAllocator): (JSC::BlockAllocator::allocate): If we receive a request for a block, set our field that tracks that to true. We don't bother locking since we assume that writing to a bool is atomic. 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116785 2012-05-11 Sam Weinig <sam@webkit.org> Fix crash seen when running with libgmalloc <rdar://problem/11435411> https://bugs.webkit.org/show_bug.cgi?id=86232 Reviewed by Gavin Barraclough. * heap/MarkStack.cpp: (JSC::MarkStackThreadSharedData::markingThreadMain): Don't delete the SlotVisitor before the ParallelModeEnabler has had a chance to run its destructor. 2012-05-15 Sam Weinig <sam@webkit.org> <rdar://problem/11401642> ENABLE_IFRAME_SEAMLESS should be turned off on the branch Reviewed by Andy Estes. * Configurations/FeatureDefines.xcconfig: Disable ENABLE_IFRAME_SEAMLESS. 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116925 2012-05-13 Filip Pizlo <fpizlo@apple.com> DFG performs incorrect constant folding on double-to-uint32 conversion in Uint32Array PutByVal https://bugs.webkit.org/show_bug.cgi?id=86330 Reviewed by Darin Adler. static_cast<int>(d) is wrong, since JS semantics require us to use toInt32(d). In particular, C++ casts on typical hardware (like x86 and similar) will return 0x80000000 for double values that are out of range of the int32 domain (i.e. less than -2^31 or greater than or equal to 2^31). But JS semantics call for wrap-around; for example the double value 4294967297 ought to become the int32 value 1, not 0x80000000. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116809 2012-05-11 Geoffrey Garen <ggaren@apple.com> Clarified JSGlobalData (JavaScript VM) lifetime https://bugs.webkit.org/show_bug.cgi?id=85142 Reviewed by Alexey Proskuryakov. (Follow-up fix.) * API/JSContextRef.cpp: (JSGlobalContextCreate): Restored some code I removed because I misread an #ifdef. (We don't need to test BUILDING_ON_LEOPARD, but we still need the linked-on test, because apps might have been linked on older OS's.) 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116813 2012-05-11 Filip Pizlo <fpizlo@apple.com> JIT memory allocator is not returning memory to the OS on Darwin https://bugs.webkit.org/show_bug.cgi?id=86047 Reviewed by Geoff Garen. * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::FixedVMPoolExecutableAllocator::notifyPageIsFree): 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116593 2012-05-09 Filip Pizlo <fpizlo@apple.com> JIT memory allocator is not returning memory to the OS on Darwin https://bugs.webkit.org/show_bug.cgi?id=86047 <rdar://problem/11414948> Reviewed by Geoff Garen. Work around the problem by using a different madvise() flag, but only for the JIT memory allocator. Also put in ASSERTs that the call is actually working. * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::FixedVMPoolExecutableAllocator::notifyNeedPage): (JSC::FixedVMPoolExecutableAllocator::notifyPageIsFree): 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116565 2012-05-09 Mark Hahnenberg <mhahnenberg@apple.com> CopiedSpace does not add pinned blocks back to the to-space filter https://bugs.webkit.org/show_bug.cgi?id=86011 Reviewed by Geoffrey Garen. After a collection has finished, we go through the blocks in from-space and move any of them that are pinned into to-space. At the beginning of collection, we reset the to-space block filter that is used during conservative scanning and add back the blocks that are filled during the collection. However, we neglect to add back those blocks that are moved from from-space to to-space, which can cause the conservative scan to think that some pinned items are not actually in CopiedSpace. * heap/CopiedSpace.cpp: (JSC::CopiedSpace::doneCopying): Add the pinned blocks back to the to-space filter. Also added a comment and assert for future readers that indicates that it's okay that we don't also add the block to the to-space block set since it was never removed. 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116484 2012-05-08 Mark Hahnenberg <mhahnenberg@apple.com> Heap should not continually allocate new pages in steady state https://bugs.webkit.org/show_bug.cgi?id=85936 Reviewed by Geoff Garen. Currently, in steady state (i.e. a constant amount of live GC memory with a constant rate of allocation) assuming we've just finished a collection with X live blocks in CopiedSpace, we increase our working set by X blocks in CopiedSpace with each collection we perform. This is due to the fact that we allocate until we run out of free blocks to use in the Heap before we consider whether we should run a collection. In the longer term, this issue will be mostly resolved by implementing quick release for the CopiedSpace. In the shorter term, we should change our policy to check whether we should allocate before trying to use a free block from the Heap. We can change our policy to something more appropriate once we have implemented quick release. This change should also have the convenient side effect of reducing the variance in GC-heavy tests (e.g. v8-splay) due to fact that we are doing less VM allocation during copying collection. Overall, this patch is performance neutral across the benchmarks we track. * heap/CopiedSpace.cpp: (JSC::CopiedSpace::getFreshBlock): Shuffle the request from the BlockAllocator around so that we only do it if the block request must succeed i.e. after we've already checked whether we should do a collection. * heap/MarkedAllocator.cpp: (JSC::MarkedAllocator::allocateSlowCase): Ditto. (JSC::MarkedAllocator::allocateBlock): We no longer have a failure mode in this function because by the time we've called it, we've already checked whether we should run a collection so there's no point in returning null. * heap/MarkedAllocator.h: Removing old arguments from function declaration. (MarkedAllocator): 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116372 2012-05-07 Oliver Hunt <oliver@apple.com> Rolling out r110287 RS=Filip Pizlo r110287 was meant to be refactoring only, but changed behavior enough to break some websites, including qq.com. 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116363 2012-05-07 Oliver Hunt <oliver@apple.com> Fix release build. * llint/LLIntSlowPaths.cpp: (JSC::LLInt::LLINT_SLOW_PATH_DECL): 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116361 2012-05-07 Oliver Hunt <oliver@apple.com> LLInt doesn't check for Ropes when performing a character switch https://bugs.webkit.org/show_bug.cgi?id=85837 Reviewed by Filip Pizlo. Make LLint check if the scrutinee of a char switch is a rope, and if so fall back to a slow case. * llint/LLIntSlowPaths.cpp: (JSC::LLInt::LLINT_SLOW_PATH_DECL): (LLInt): * llint/LowLevelInterpreter32_64.asm: * llint/LowLevelInterpreter64.asm: 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116367 2012-05-07 Andy Estes <aestes@apple.com> ENABLE_IFRAME_SEAMLESS should be part of FEATURE_DEFINES. * Configurations/FeatureDefines.xcconfig: 2012-05-15 Lucas Forschler <lforschler@apple.com> Merge 116356 2012-05-07 Eric Seidel <eric@webkit.org> Add ENABLE_IFRAME_SEAMLESS so Apple can turn off SEAMLESS if needed https://bugs.webkit.org/show_bug.cgi?id=85822 Reviewed by Adam Barth. * Configurations/FeatureDefines.xcconfig: 2012-05-04 Filip Pizlo <fpizlo@apple.com> DFG should not Flush GetLocal's https://bugs.webkit.org/show_bug.cgi?id=85663 <rdar://problem/11373600> Reviewed by Oliver Hunt. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::flushArgument): (JSC::DFG::ByteCodeParser::handleCall): 2012-05-04 Allan Sandfeld Jensen <allan.jensen@nokia.com> Doesn't build with ENABLE_JIT=0 https://bugs.webkit.org/show_bug.cgi?id=85042 Reviewed by Gavin Barraclough. * bytecode/Operands.h: 2012-05-03 Oliver Hunt <oliver@apple.com> Regression(r114702): Clobbering the caller frame register before we've stored it. https://bugs.webkit.org/show_bug.cgi?id=85564 Reviewed by Filip Pizlo. Don't use t0 as a temporary, when we're about to use the value in t0. * llint/LowLevelInterpreter32_64.asm: 2012-05-03 Mark Hahnenberg <mhahnenberg@apple.com> Removing remainder of accidental printfs. * heap/Heap.cpp: (JSC::Heap::collect): 2012-05-03 Andy Estes <aestes@apple.com> If you add printf()s to your garbage collector, the layout tests are gonna have a bad time. * runtime/GCActivityCallbackCF.cpp: (JSC::DefaultGCActivityCallbackPlatformData::timerDidFire): 2012-05-03 Mark Hahnenberg <mhahnenberg@apple.com> Heap::reportAbandonedObjectGraph should not hasten an allocation-triggered collection https://bugs.webkit.org/show_bug.cgi?id=85543 Reviewed by Filip Pizlo. Currently reportAbandonedObjectGraph causes the Heap to think it is closer to its allocation limit for the current cycle, thus hastening an allocation-triggered collection. In reality, it should just affect the opportunistic GC timer. We should track the bytes we think have been abandoned and the bytes that have been allocated separately. * heap/Heap.cpp: Added a new field m_abandonedBytes to Heap to keep track of how much we think we've abandoned. (JSC::Heap::Heap): (JSC::Heap::reportAbandonedObjectGraph): (JSC): (JSC::Heap::didAbandon): Added this function for reportAbandonedObjectGraph to call rather than didAllocate. Works the same as didAllocate, but modifies bytes abandoned rather than bytes allocated. Also notifies the timer, summing the two values together. (JSC::Heap::collect): (JSC::Heap::didAllocate): Now adds the bytes allocated and bytes abandoned when reporting to GCActivityCallback. * heap/Heap.h: (Heap): 2012-05-02 Eric Seidel <eric@webkit.org> Sort ENABLE_ defines in FeatureDefines.xcconfig files to make them easier to compare with one another (and easier to autogenerate) https://bugs.webkit.org/show_bug.cgi?id=85433 Reviewed by Adam Barth. I have a script which can autogenerate these xcconfig files as well as the vsprops files (and soon the Chromium, cmake, gnumake and qmake) feature lists from a central feature list file. In preparation for posting such a tool, I'm re-sorting these xcconfig files to be alphabetically ordered (currently they're close, but not quite). There is also at least one inconsistency between these files (CSS_LEGACY_PREFIXES) which I will fix in a second pass. I will also sort the FEATURE_DEFINES = line in a follow-up patch. * Configurations/FeatureDefines.xcconfig: 2012-05-02 Hojong Han <hojong.han@samsung.com> ARM_TRADITIONAL build fix https://bugs.webkit.org/show_bug.cgi?id=85358 Reviewed by Gavin Barraclough. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::lshift32): (MacroAssemblerARM): (JSC::MacroAssemblerARM::or32): (JSC::MacroAssemblerARM::urshift32): (JSC::MacroAssemblerARM::xor32): (JSC::MacroAssemblerARM::branchSub32): 2012-05-02 Mark Hahnenberg <mhahnenberg@apple.com> Opportunistic GC should give up if the Heap is paged out https://bugs.webkit.org/show_bug.cgi?id=85411 Reviewed by Filip Pizlo. Opportunistic GC is punishing us severely in limited memory situations because its assumptions about how much time a collection will take are way out of whack when the Heap has been paged out by the OS. We should add a simple detection function to the Heap that detects if its is paged out. It will do this by iterating each block of both the MarkedSpace and CopiedSpace. If that operation takes longer than a fixed amount of time (e.g. 100ms), the function returns true. This function will only be run prior to an opportunistic collection (i.e. it will not run during our normal allocation-triggered collections). In my tests, steady state was drastically improved in high memory pressure situations (i.e. the browser was still usable, significant reduction in SPODs). Occasionally, a normal GC would be triggered due to pages doing things in the background, which would cause a significant pause. As we close pages we now cause normal collections rather than full collections, which prevents us from collecting all of the dead memory immediately. One nice way to deal with this issue might be to do incremental sweeping. * heap/CopiedSpace.cpp: (JSC::isBlockListPagedOut): Helper function to reduce code duplication when iterating over to-space, from-space, and the oversize blocks. (JSC): (JSC::CopiedSpace::isPagedOut): Tries to determine whether or not CopiedSpace is paged out by iterating all of the blocks. * heap/CopiedSpace.h: (CopiedSpace): * heap/Heap.cpp: (JSC::Heap::isPagedOut): Tries to determine whether the Heap is paged out by asking the MarkedSpace and CopiedSpace if they are paged out. (JSC): * heap/Heap.h: (Heap): (JSC::Heap::increaseLastGCLength): Added this so that the GC timer can linearly back off each time it determines that the Heap is paged out. * heap/MarkedAllocator.cpp: (JSC::MarkedAllocator::isPagedOut): Tries to determine if this particular MarkedAllocator's list of blocks are paged out. (JSC): * heap/MarkedAllocator.h: (MarkedAllocator): * heap/MarkedSpace.cpp: (JSC::MarkedSpace::isPagedOut): For each MarkedAllocator, check to see if they're paged out. * heap/MarkedSpace.h: (MarkedSpace): * runtime/GCActivityCallback.cpp: (JSC::DefaultGCActivityCallback::cancel): (JSC): * runtime/GCActivityCallback.h: (JSC::GCActivityCallback::cancel): (DefaultGCActivityCallback): * runtime/GCActivityCallbackCF.cpp: Added a constant of 100ms for the timeout in determining whether the Heap is paged out or not. (JSC): (JSC::DefaultGCActivityCallbackPlatformData::timerDidFire): Added the check to see if we should attempt a collection based on whether or not we can iterate the blocks of the Heap in 100ms. If we can't, we cancel the timer and tell the Heap we just wasted 100ms more trying to do a collection. This gives us a nice linear backoff so we're not constantly re-trying in steady state paged-out-ness. (JSC::DefaultGCActivityCallback::cancel): Added this function which, while currently doing exactly the same thing as willCollect, is more obvious as to what it's doing when we call it in timerDidFire. 2012-05-02 Yong Li <yoli@rim.com> Fix GCC X86 build error https://bugs.webkit.org/show_bug.cgi?id=85379 Reviewed by Rob Buis. Always explicitly claim ".text" to make sure functions defined with inline assembly will be created in the correct section. * dfg/DFGOperations.cpp: (JSC): 2012-05-02 Oliver Hunt <oliver@apple.com> Unreviewed, rolling out r115388. http://trac.webkit.org/changeset/115388 https://bugs.webkit.org/show_bug.cgi?id=85011 This caused many weird performance problems, and needs to be landed in pieces. * dfg/DFGOperations.cpp: * heap/Heap.cpp: (JSC::Heap::getConservativeRegisterRoots): (JSC::Heap::markRoots): * interpreter/CallFrame.cpp: (JSC::CallFrame::dumpCaller): (JSC): * interpreter/CallFrame.h: (JSC::ExecState::init): (ExecState): * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): (JSC::Interpreter::executeCall): (JSC::Interpreter::executeConstruct): (JSC::Interpreter::prepareForRepeatCall): (JSC::Interpreter::privateExecute): * interpreter/Interpreter.h: (JSC::Interpreter::execute): * interpreter/RegisterFile.cpp: (JSC::RegisterFile::growSlowCase): (JSC::RegisterFile::gatherConservativeRoots): * interpreter/RegisterFile.h: (JSC::RegisterFile::end): (JSC::RegisterFile::size): (JSC::RegisterFile::addressOfEnd): (RegisterFile): (JSC::RegisterFile::RegisterFile): (JSC::RegisterFile::shrink): (JSC::RegisterFile::grow): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): (JSC::jitCompileFor): (JSC::lazyLinkFor): * llint/LLIntSlowPaths.cpp: (JSC::LLInt::LLINT_SLOW_PATH_DECL): (JSC::LLInt::handleHostCall): * llint/LowLevelInterpreter.asm: * runtime/CommonSlowPaths.h: (JSC::CommonSlowPaths::arityCheckFor): 2012-05-01 Oliver Hunt <oliver@apple.com> Physijs demo crashes due to DFG not updating topCallFrame correctly. https://bugs.webkit.org/show_bug.cgi?id=85311 Reviewed by Filip Pizlo. A few of the dfg operations failed to correctly set the topCallFrame, and so everything goes wrong. This patch corrects the effected operations, and makes debug builds poison topCallFrame before calling a dfg operation. * dfg/DFGOperations.cpp: (JSC::DFG::putByVal): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::callOperation): (SpeculativeJIT): (JSC::DFG::SpeculativeJIT::prepareForExternalCall): (JSC::DFG::SpeculativeJIT::appendCallWithExceptionCheck): (JSC::DFG::SpeculativeJIT::appendCallSetResult): 2012-04-30 Gavin Barraclough <barraclough@apple.com> Should be able to use YARR JIT without the JS language JIT https://bugs.webkit.org/show_bug.cgi?id=85252 Reviewed by Geoff Garen. Need to split canUseRegExpJIT out of canUseJIT. * runtime/JSGlobalData.cpp: (JSC): (JSC::useJIT): (JSC::JSGlobalData::JSGlobalData): - replace m_canUseJIT with m_canUseAssembler * runtime/JSGlobalData.h: (JSGlobalData): (JSC::JSGlobalData::canUseRegExpJIT): - Added canUseRegExpJIT, distinct from canUseJIT. * runtime/RegExp.cpp: (JSC::RegExp::compile): (JSC::RegExp::compileMatchOnly): - Call canUseRegExpJIT instead of canUseJIT. 2012-04-30 Gavin Barraclough <barraclough@apple.com> Should be able to build YARR JIT without the JS language JIT https://bugs.webkit.org/show_bug.cgi?id=85242 Reviewed by Michael Saboff. Some build macros are wrong. * assembler/RepatchBuffer.h: * jit/ExecutableAllocator.h: (JSC): * jit/JITExceptions.cpp: * runtime/InitializeThreading.cpp: (JSC::initializeThreadingOnce): 2012-04-26 Gavin Barraclough <barraclough@apple.com> Arguments object resets attributes on redefinition of a parameter https://bugs.webkit.org/show_bug.cgi?id=84994 Rubber stamped by Oliver Hunt. There is a bug that we always re-add the original property before redefinition, doing so in a way that will reset the attributes without checking configurability. * runtime/Arguments.cpp: (JSC::Arguments::defineOwnProperty): - Only instantiate the property once - do not re-add if it has already been added, or if it has been deleted. 2012-04-30 Ryosuke Niwa <rniwa@webkit.org> Remove an erroneous assertion after r115655. * runtime/NumberPrototype.cpp: (JSC::toUStringWithRadix): 2012-04-30 Myles Maxfield <mmaxfield@google.com> End of Interpreter::tryCacheGetByID can trigger the garbage collector https://bugs.webkit.org/show_bug.cgi?id=84927 Reviewed by Oliver Hunt. * interpreter/Interpreter.cpp: (JSC::Interpreter::tryCacheGetByID): 2012-04-30 Benjamin Poulain <benjamin@webkit.org> jsSingleCharacterString and jsSingleCharacterSubstring are not inlined https://bugs.webkit.org/show_bug.cgi?id=85147 Reviewed by Darin Adler. The functions jsSingleCharacterString() and jsSingleCharacterSubstring() were not inlined by the compiler. This annihilate the gains of using SmallStrings. On stringProtoFuncCharAt(), this patch improves the performance by 11%. * runtime/JSString.h: (JSC::jsSingleCharacterString): (JSC::jsSingleCharacterSubstring): 2012-04-30 Benjamin Poulain <bpoulain@apple.com> Add fast patch for radix == 10 on numberProtoFuncToString https://bugs.webkit.org/show_bug.cgi?id=85120 Reviewed by Darin Adler. When radix, we use to turn the doubleValue into a JSValue just to convert it to a String. The problem is that was using the slow path for conversion and for the toString() operation. This patch shortcuts the creation of a JSValue and uses NumericStrings directly. The conversion is split between Integer and Double to ensure the fastest conversion for the common case of integer arguments. Converting number with radix 10 becomes 5% faster. Due to the simpler conversion of number to string for integer, converting integers that do not fall in the two previous optimizations get 32% faster. * runtime/NumberPrototype.cpp: (JSC::extractRadixFromArgs): (JSC::integerValueToString): (JSC::numberProtoFuncToString): 2012-04-30 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck. * GNUmakefile.list.am: Add missing header. 2012-04-28 Geoffrey Garen <ggaren@apple.com> Factored threaded block allocation into a separate object https://bugs.webkit.org/show_bug.cgi?id=85148 Reviewed by Sam Weinig. 99% of this patch just moves duplicated block allocation and deallocation code into a new object named BlockAllocator, with these exceptions: * heap/BlockAllocator.h: Added. (BlockAllocator::BlockAllocator): The order of declarations here now guards us against an unlikely race condition during startup. * heap/BlockAllocator.cpp: JSC::BlockAllocator::blockFreeingThreadMain): Added a FIXME to highlight a lack of clarity we have in our block deallocation routines. 2012-04-28 Sam Weinig <sam@webkit.org> Try to fix the Qt build. * heap/Heap.cpp: (JSC::Heap::lastChanceToFinalize): 2012-04-28 Geoffrey Garen <ggaren@apple.com> Try to fix the Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-04-28 Geoffrey Garen <ggaren@apple.com> Clarified JSGlobalData (JavaScript VM) lifetime https://bugs.webkit.org/show_bug.cgi?id=85142 Reviewed by Anders Carlsson. This was so confusing that I didn't feel like I could reason about memory lifetime in the heap without fixing it. The rules are: (1) JSGlobalData owns the virtual machine and all memory in it. (2) Deleting a JSGlobalData frees the virtual machine and all memory in it. (Caveat emptor: if you delete the virtual machine while you're running JIT code or accessing GC objects, you're gonna have a bad time.) (I opted not to make arbitrary sub-objects keep the virtual machine alive automatically because: (a) doing that right would be complex and slow; (b) in the case of an exiting thread or process, there's no clear way to give the garbage collector a chance to try again later; (c) continuing to run the garbage collector after we've been asked to shut down the virtual machine seems rude; (d) we've never really supported that feature, anyway.) (3) Normal ref-counting will do. No need to call a battery of specialty functions to tear down a JSGlobalData. Its foibles notwithstanding, C++ does in fact know how to execute destructors in order. * API/JSContextRef.cpp: (JSGlobalContextCreate): Removed compatibility shim for older operating systems because it's no longer used. (JSGlobalContextRelease): Now that we can rely on JSGlobalData to "do the right thing", this code is much simpler. We still have one special case to notify the garbage collector if we're removing the last reference to the global object, since this can improve memory behavior. * heap/CopiedSpace.cpp: (JSC::CopiedSpace::freeAllBlocks): * heap/CopiedSpace.h: (CopiedSpace): Renamed "destroy" => "freeAllBlocks" because true destruction-time behaviors should be limited to our C++ destructor. * heap/Heap.cpp: (JSC::Heap::~Heap): (JSC): (JSC::Heap::lastChanceToFinalize): * heap/Heap.h: (Heap): (JSC::Heap::heap): Renamed "destroy" => "lastChanceToFinalize" because true destruction-time behaviors should be limited to our C++ destructor. Reorganized the code, putting code that must run before any objects get torn down into lastChanceToFinalize, and code that just tears down objects into our destructor. * heap/Local.h: (JSC::LocalStack::LocalStack): (JSC::LocalStack::push): (LocalStack): See rule (2). * jsc.cpp: (functionQuit): (main): (printUsageStatement): (parseArguments): (jscmain): * testRegExp.cpp: (main): (printUsageStatement): (parseArguments): (realMain): See rule (3). I removed the feature of ensuring orderly tear-down when calling quit() or running in --help mode because it didn't seem very useful and making it work with Windows structured exception handling and NO_RETURN didn't seem like a fun way to spend a Saturday. * runtime/JSGlobalData.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): Moved heap to be the first data member in JSGlobalData to ensure that it's destructed last, so other objects that reference it destruct without crashing. This allowed me to remove clearBuiltinStructures() altogether, and helped guarantee rule (3). (JSC::JSGlobalData::~JSGlobalData): Explicitly call lastChanceToFinalize() at the head of our destructor to ensure that all pending finalizers run while the virtual machine is still in a valid state. Trying to resurrect (re-ref) the virtual machine at this point is not valid, but all other operations are. Changed a null to a 0xbbadbeef to clarify just how bad this beef is. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::init): * runtime/JSGlobalObject.h: (JSGlobalObject): (JSC::JSGlobalObject::globalData): See rule (3). 2012-04-27 Geoffrey Garen <ggaren@apple.com> Try to fix the Windows build. * heap/WeakBlock.h: (WeakBlock): 2012-04-27 Geoffrey Garen <ggaren@apple.com> Made WeakSet::allocate() static and removed its JSGlobalData argument https://bugs.webkit.org/show_bug.cgi?id=85128 Reviewed by Anders Carlsson. This is a step toward faster finalization. WeakSet::allocate() now deduces which WeakSet to allocate from based on its JSCell* argument. (Currently, there's only one WeakSet, but soon there will be many.) This was a global replace of "globalData.heap.weakSet()->allocate" with "WeakSet::allocate", plus by-hand removal of the JSGlobalData argument. * heap/WeakSetInlines.h: Copied from Source/JavaScriptCore/heap/WeakSet.h. I had to split out WeakSet::allocate() in to a separate header to avoid a cycle. (JSC::WeakSet::allocate): We can mask the pointer we're passed to figure out where to allocate our WeakImpl. (Soon, we'll use this to associate the WeakImpl with the GC block it references.) 2012-04-27 Geoffrey Garen <ggaren@apple.com> Stop using aligned allocation for WeakBlock https://bugs.webkit.org/show_bug.cgi?id=85124 Reviewed by Anders Carlsson. We don't actually use the alignment for anything. * heap/WeakBlock.cpp: (JSC::WeakBlock::create): (JSC::WeakBlock::WeakBlock): Switched from aligned allocation to regular allocation. * heap/WeakBlock.h: (WeakBlock): Don't use HeapBlock because HeapBlock requires aligned allocation. This change required me to add some declarations that we used to inherit from HeapBlock. (WeakBlock::blockFor): Removed. This function relied on aligned allocation but didn't do anything for us. (WeakBlock::deallocate): Removed. WeakBlock doesn't own any of the deallocation logic, so it shouldn't own the function. * heap/WeakSet.cpp: (JSC::WeakSet::~WeakSet): (JSC::WeakSet::finalizeAll): (JSC::WeakSet::visitLiveWeakImpls): (JSC::WeakSet::visitDeadWeakImpls): (JSC::WeakSet::sweep): (JSC::WeakSet::shrink): (JSC::WeakSet::resetAllocator): (JSC::WeakSet::tryFindAllocator): * heap/WeakSet.h: (WeakSet): Updated declarations to reflect WeakBlock not inheriting from HeapBlock. This allowed me to remove some casts, which was nice. (JSC::WeakSet::deallocate): Directly set the deallocated flag instead of asking WeakBlock to do it for us. We don't need to have a WeakBlock pointer to set the flag, so stop asking for one. 2012-04-27 Kentaro Hara <haraken@chromium.org> [JSC] Implement a helper method createNotEnoughArgumentsError() https://bugs.webkit.org/show_bug.cgi?id=85102 Reviewed by Geoffrey Garen. In bug 84787, kbr@ requested to avoid hard-coding createTypeError(exec, "Not enough arguments") here and there. This patch implements createNotEnoughArgumentsError(exec) and uses it in JSC bindings. c.f. a corresponding bug for V8 bindings is bug 85097. * runtime/Error.cpp: (JSC::createNotEnoughArgumentsError): (JSC): * runtime/Error.h: (JSC): 2012-04-27 Geoffrey Garen <ggaren@apple.com> Only allow non-null pointers in the WeakSet https://bugs.webkit.org/show_bug.cgi?id=85119 Reviewed by Darin Adler. This is a step toward more efficient finalization. No clients put non-pointers (JSValues) into Weak<T> and PassWeak<T>. Some clients put null pointers into Weak<T> and PassWeak<T>, but this is more efficient and straight-forward to model with a null in the Weak<T> or PassWeak<T> instead of allocating a WeakImpl just to hold null. * heap/PassWeak.h: (JSC): Removed the Unknown (JSValue) type of weak pointer because it's unused now. (PassWeak): Don't provide a default initializer for our JSCell* argument. This feature was only used in one place, and it was a bug. (JSC::::get): Don't check for a null stored inside our WeakImpl: that's not allowed anymore. (JSC::PassWeak::PassWeak): Handle null as a null WeakImpl instead of allocating a WeakImpl and storing null into it. * heap/Weak.h: (Weak): (JSC::::Weak): Same changes as in PassWeak<T>. * heap/WeakBlock.cpp: (JSC::WeakBlock::visitLiveWeakImpls): (JSC::WeakBlock::visitDeadWeakImpls): Only non-null cells are valid in the WeakSet now, so no need to check for non-cells and null cell pointers. * heap/WeakImpl.h: (JSC::WeakImpl::WeakImpl): Only non-null cells are valid in the WeakSet now, so ASSERT that. 2012-04-27 Gavin Barraclough <barraclough@apple.com> <rdar://problem/7909395> Math in JavaScript is inaccurate on iOS By defalut IEEE754 denormal support is disabled on iOS; turn it on. Reviewed by Filip Pizlo. * jsc.cpp: (main): - clear the appropriate bit in the fpscr. 2012-04-27 Michael Saboff <msaboff@apple.com> Memory wasted in JSString for non-rope strings https://bugs.webkit.org/show_bug.cgi?id=84907 Reviewed by Geoffrey Garen. Split JSString into two classes, JSString as a base class that does not include the fibers of a Rope, and a subclass JSRopeString that has the rope functionality. Both classes "share" the same ClassInfo. Added a bool to JSString to indicate that the string was allocated as a JSRopeString to properly handle visiting the fiber children when the rope is resolved and the JSRopeString appears as a JSString. Didn't change the interface of JSString to require any JIT changes. As part of this change, removed "cellSize" from ClassInfo since both classes share the same ClassInfo, but have different sizes. The only use I could find for cellSize was an ASSERT in allocateCell(). This appears to be neutral on performance tests. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Changed JSString::resolveRope to JSRopeString::resolveRope * runtime/ClassInfo.h: (JSC): (ClassInfo): * runtime/JSCell.h: (JSC::allocateCell): * runtime/JSString.cpp: (JSC::JSRopeString::RopeBuilder::expand): (JSC::JSString::visitChildren): (JSC): (JSC::JSRopeString::visitFibers): (JSC::JSRopeString::resolveRope): (JSC::JSRopeString::resolveRopeSlowCase8): (JSC::JSRopeString::resolveRopeSlowCase): (JSC::JSRopeString::outOfMemory): (JSC::JSRopeString::getIndexSlowCase): * runtime/JSString.h: (JSC): (JSString): (JSC::JSString::finishCreation): (JSC::JSString::create): (JSC::JSString::isRope): (JSC::JSString::is8Bit): (JSRopeString): (RopeBuilder): (JSC::JSRopeString::RopeBuilder::RopeBuilder): (JSC::JSRopeString::RopeBuilder::append): (JSC::JSRopeString::RopeBuilder::release): (JSC::JSRopeString::RopeBuilder::length): (JSC::JSRopeString::JSRopeString): (JSC::JSRopeString::finishCreation): (JSC::JSRopeString::createNull): (JSC::JSRopeString::create): (JSC::JSString::value): (JSC::JSString::tryGetValue): (JSC::JSString::getIndex): (JSC::jsStringBuilder): * runtime/Operations.h: (JSC::jsString): (JSC::jsStringFromArguments): 2012-04-27 Oliver Hunt <oliver@apple.com> Correct assertion. * interpreter/Interpreter.cpp: (JSC::Interpreter::throwException): 2012-04-27 Oliver Hunt <oliver@apple.com> Lazy link phase of baseline jit fails to propagate exception https://bugs.webkit.org/show_bug.cgi?id=85092 Reviewed by Filip Pizlo. Very simple patch, when linking produces an error we need to actually store the exception prior to throwing it. I can't find any other examples of this, but as we're already in the slow path when throwing an exception I've hardened exception throwing against null exceptions. * interpreter/Interpreter.cpp: (JSC::Interpreter::throwException): * jit/JITStubs.cpp: (JSC::lazyLinkFor): 2012-04-27 Benjamin Poulain <benjamin@webkit.org> Generalize the single character optimization of numberProtoFuncToString https://bugs.webkit.org/show_bug.cgi?id=85027 Reviewed by Geoffrey Garen. The function numberProtoFuncToString() has an optimization to use SmallStrings::singleCharacterString() when the radix is 36. This patch generalize the optimization for any radix. Any positive number smaller than its radix can be represented by a single character of radixDigits. This makes numberProtoFuncToString() about twice as fast for this case of single digit conversion. * runtime/NumberPrototype.cpp: (JSC::numberProtoFuncToString): 2012-04-27 Gavin Peters <gavinp@chromium.org> Add new ENABLE_LINK_PRERENDER define to control the Prerendering API https://bugs.webkit.org/show_bug.cgi?id=84871 Reviewed by Adam Barth. Prerendering is currently covered by the ENABLE_LINK_PREFETCH macro, but the new Prerendering API separates it from prefetching. Having separate include guards lets ports enable prefetching, a relatively easy change, without needing to build the infrastructure for prerendering, which is considerably more complicated. * Configurations/FeatureDefines.xcconfig: 2012-04-26 Oliver Hunt <oliver@apple.com> Allocating WeakImpl should not trigger GC, as that makes the world very tricksy. https://bugs.webkit.org/show_bug.cgi?id=85020 Reviewed by Gavin Barraclough. Now in the event that we are unable to find an allocator for a new handle, just add a new allocator rather than trying to recover "dead" handles through a GC. Find allocator is now much simpler, and addAllocator directly reports the increased memory usage to the heap without causing any GC to happen immediately. * heap/WeakSet.cpp: (JSC::WeakSet::findAllocator): (JSC::WeakSet::addAllocator): 2012-04-26 Oliver Hunt <oliver@apple.com> Remove RegisterFile::end()/m_end https://bugs.webkit.org/show_bug.cgi?id=85011 Reviewed by Gavin Barraclough. Get rid of end() and m_end from RegisterFile. From now on we only care about the end of the committed region when calling code. When re-entering the VM we now plant the new CallFrame immediately after whatever the current topCallFrame is. This required adding a routine to CallFrame to determine exactly what we should be doing (in the absence of an existing CallFrame, we can't reason about the frameExtent() so we check for that). This also now means that the GC only marks the portion of the RegisterFile that is actually in use, and that VM re-entry doesn't exhaust the RegisterFile as rapidly. * dfg/DFGOperations.cpp: * heap/Heap.cpp: (JSC::Heap::getConservativeRegisterRoots): (JSC::Heap::markRoots): * interpreter/CallFrame.h: (JSC::ExecState::init): (JSC::ExecState::startOfReusableRegisterFile): (ExecState): * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): (JSC::Interpreter::executeCall): (JSC::Interpreter::executeConstruct): (JSC::Interpreter::prepareForRepeatCall): (JSC::Interpreter::privateExecute): * interpreter/Interpreter.h: (JSC::Interpreter::execute): * interpreter/RegisterFile.cpp: (JSC::RegisterFile::growSlowCase): (JSC::RegisterFile::gatherConservativeRoots): * interpreter/RegisterFile.h: (JSC::RegisterFile::commitEnd): (JSC::RegisterFile::addressOfEnd): (RegisterFile): (JSC::RegisterFile::RegisterFile): (JSC::RegisterFile::shrink): (JSC::RegisterFile::grow): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): (JSC::jitCompileFor): (JSC::lazyLinkFor): * llint/LLIntSlowPaths.cpp: (JSC::LLInt::LLINT_SLOW_PATH_DECL): (JSC::LLInt::handleHostCall): * llint/LowLevelInterpreter.asm: * runtime/CommonSlowPaths.h: (JSC::CommonSlowPaths::arityCheckFor): 2012-04-26 Filip Pizlo <fpizlo@apple.com> DFG ARMv7 backend should optimize Float32 arrays https://bugs.webkit.org/show_bug.cgi?id=85000 <rdar://problem/10652827> Reviewed by Gavin Barraclough. * assembler/ARMv7Assembler.h: (ARMv7Assembler): (JSC::ARMv7Assembler::flds): (JSC::ARMv7Assembler::fsts): (JSC::ARMv7Assembler::vcvtds): (JSC::ARMv7Assembler::vcvtsd): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::loadFloat): (MacroAssemblerARMv7): (JSC::MacroAssemblerARMv7::storeFloat): (JSC::MacroAssemblerARMv7::convertFloatToDouble): (JSC::MacroAssemblerARMv7::convertDoubleToFloat): * bytecode/PredictedType.h: (JSC::isActionableFloatMutableArrayPrediction): * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateFloat32Array): 2012-04-25 Benjamin Poulain <benjamin@webkit.org> Add a version of StringImpl::find() without offset https://bugs.webkit.org/show_bug.cgi?id=83968 Reviewed by Sam Weinig. Add support for the new StringImpl::find() to UString. Change stringProtoFuncIndexOf() to specifically take advatage of the feature. This gives a 12% gains on a distribution of strings between 30 and 100 characters. * runtime/StringPrototype.cpp: (JSC::substituteBackreferences): (JSC::stringProtoFuncIndexOf): * runtime/UString.h: (UString): (JSC::UString::find): 2012-04-25 Mark Hahnenberg <mhahnenberg@apple.com> WebCore shouldn't call collectAllGarbage directly https://bugs.webkit.org/show_bug.cgi?id=84897 Reviewed by Geoffrey Garen. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Exported symbol for reportAbanondedObjectGraph so WebCore can use it. * heap/Heap.h: Ditto. 2012-04-25 Oliver Hunt <oliver@apple.com> Biolab disaster crashes on ToT https://bugs.webkit.org/show_bug.cgi?id=84898 Reviewed by Filip Pizlo. Whoops, committed without saving reviewer requested change. * dfg/DFGVirtualRegisterAllocationPhase.cpp: (JSC::DFG::VirtualRegisterAllocationPhase::run): 2012-04-25 Oliver Hunt <oliver@apple.com> Biolab disaster crashes on ToT https://bugs.webkit.org/show_bug.cgi?id=84898 Reviewed by Filip Pizlo. I recently added an assertion to the Interpreter to catch incorrect updates of topCallFrame. This caused a bunch of sites (including biolab disaster) to crash as we were not correctly handling callee registers of inlined functions, leading to a mismatch. I could not actually make this trigger directly, although it does trigger already on some of the GTK and QT bots. * dfg/DFGVirtualRegisterAllocationPhase.cpp: (JSC::DFG::VirtualRegisterAllocationPhase::run): 2012-04-25 Kenneth Russell <kbr@google.com> Delete CanvasPixelArray, ByteArray, JSByteArray and JSC code once unreferenced https://bugs.webkit.org/show_bug.cgi?id=83655 Reviewed by Oliver Hunt. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.order: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * bytecode/PredictedType.cpp: (JSC::predictionToString): (JSC::predictionToAbbreviatedString): (JSC::predictionFromClassInfo): * bytecode/PredictedType.h: (JSC): (JSC::isActionableIntMutableArrayPrediction): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::initialize): (JSC::DFG::AbstractState::execute): * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::performNodeCSE): * dfg/DFGFixupPhase.cpp: (JSC::DFG::FixupPhase::fixupNode): * dfg/DFGNode.h: * dfg/DFGNodeType.h: (DFG): * dfg/DFGOperations.cpp: (JSC::DFG::putByVal): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::compileGetIndexedPropertyStorage): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::ValueSource::forPrediction): (SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jit/JITStubs.h: * llint/LLIntSlowPaths.cpp: (JSC::LLInt::getByVal): (JSC::LLInt::LLINT_SLOW_PATH_DECL): * runtime/JSByteArray.cpp: Removed. * runtime/JSByteArray.h: Removed. * runtime/JSGlobalData.cpp: 2012-04-25 Filip Pizlo <fpizlo@apple.com> http://bellard.org/jslinux/ triggers an assertion failure in the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=84815 <rdar://problem/11319514> Reviewed by Gavin Barraclough. * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::forwardSpeculationCheck): 2012-04-25 Michael Saboff <msaboff@apple.com> Closure in try {} with catch captures all locals from the enclosing function https://bugs.webkit.org/show_bug.cgi?id=84804 Reviewed by Oliver Hunt. Changed the capturing of local variables from capturing when eval is used, within a "with" or within a "catch" to be just when an eval is used. Renamed the function returning that we should capture from getCapturedVariables() to usesEval(), since that what it noew returns. Needed to fix the "with" code to only range check when the activation has actually been torn off. Added m_isTornOff to JSActivation to track this. * parser/Parser.h: (JSC::Scope::usesEval): (JSC::Scope::getCapturedVariables): * runtime/JSActivation.cpp: (JSC::JSActivation::JSActivation): (JSC::JSActivation::symbolTableGet): (JSC::JSActivation::symbolTablePut): * runtime/JSActivation.h: (JSActivation): (JSC::JSActivation::tearOff): 2012-04-24 Mark Hahnenberg <mhahnenberg@apple.com> GC Activity Callback timer should be based on how much has been allocated since the last collection https://bugs.webkit.org/show_bug.cgi?id=84763 Reviewed by Geoffrey Garen. The desired behavior for the GC timer is to collect at some point in the future, regardless of how little we've allocated. A secondary goal, which is almost if not as important, is for the timer to collect sooner if there is the potential to collect a greater amount of memory. Conversely, as we allocate more memory we'd like to reduce the delay to the next collection. If we're allocating quickly enough, the timer should be preempted in favor of a normal allocation-triggered collection. If allocation were to slow or stop, we'd like the timer to be able to opportunistically run a collection without us having to allocate to the hard limit set by the Heap. This type of policy can be described in terms of the amount of CPU we are willing to dedicate to reclaim a single MB of memory. For example, we might be willing to dedicate 1% of our CPU to reclaim 1 MB. We base our CPU usage off of the length of the last collection, e.g. if our last collection took 1ms, we would want to wait about 100ms before running another collection to reclaim 1 MB. These constants should be tune-able, e.g. 0.1% CPU = 1 MB vs. 1% CPU = 1 MB vs. 10% CPU = 1 MB. * API/JSBase.cpp: Use the new reportAbandonedObjectGraph. (JSGarbageCollect): * API/JSContextRef.cpp: Ditto. * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::reportAbandonedObjectGraph): Similar to reportExtraMemoryCost. Clients call this function to notify the Heap that some unknown number of JSC objects might have just been abandoned and are now garbage. The Heap might schedule a new collection timer based on this notification. (JSC): (JSC::Heap::collect): Renamed m_lastFullGCSize to the less confusing m_sizeAfterLastCollect. * heap/Heap.h: (Heap): * heap/MarkedAllocator.h: (JSC::MarkedAllocator::zapFreeList): Fixed a bug in zapFreeList that failed to nullify the current allocator's FreeList once zapping was complete. * runtime/GCActivityCallback.cpp: Removed didAbandonObjectGraph because it was replaced by Heap::reportAbandonedObjectGraph. (JSC): * runtime/GCActivityCallback.h: (JSC::GCActivityCallback::willCollect): (DefaultGCActivityCallback): * runtime/GCActivityCallbackCF.cpp: Refactored the GC timer code so that we now schedule the timer based on how much we have allocated since the last collection up to a certain amount. We use the length of the previous GC to try to keep our total cost of opportunistic timer-triggered collections around 1% of the CPU per MB of garbage we expect to reclaim up to a maximum of 5 MB. (DefaultGCActivityCallbackPlatformData): (JSC): (JSC::DefaultGCActivityCallback::~DefaultGCActivityCallback): (JSC::DefaultGCActivityCallback::commonConstructor): (JSC::scheduleTimer): (JSC::cancelTimer): (JSC::DefaultGCActivityCallback::didAllocate): 2012-04-24 Michael Saboff <msaboff@apple.com> objectProtoFuncToString creates new string every invocation https://bugs.webkit.org/show_bug.cgi?id=84781 Reviewed by Geoffrey Garen. Cache the results of object toString() in the attached Structure. * runtime/ObjectPrototype.cpp: (JSC::objectProtoFuncToString): * runtime/Structure.cpp: (JSC::Structure::visitChildren): visit new m_hasObjectToStringValue. * runtime/Structure.h: Added new member m_hasObjectToStringValue (JSC): (JSC::Structure::objectToStringValue): (Structure): (JSC::Structure::setObjectToStringValue): 2012-04-24 Thouraya ANDOLSI <thouraya.andolsi@st.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=84727. Fix build when ENABLE_JIT_CONSTANT_BLINDING enabled. * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::or32): (JSC::MacroAssemblerSH4::and32): (JSC::MacroAssemblerSH4::lshift32): (JSC::MacroAssemblerSH4::xor32): (JSC::MacroAssemblerSH4::branchSub32): (JSC::MacroAssemblerSH4::urshift32): 2012-04-24 Gavin Barraclough <barraclough@apple.com> Add explicit patchableBranchPtrWithPatch/patchableJump methods https://bugs.webkit.org/show_bug.cgi?id=84498 Reviewed by Filip Pizlo. Don't rely on inUninterruptedSequence to distinguish which jumps we need to be able to repatch. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::PatchableJump::PatchableJump): (PatchableJump): (JSC::AbstractMacroAssembler::PatchableJump::operator Jump&): (AbstractMacroAssembler): (JSC::AbstractMacroAssembler::AbstractMacroAssembler): - Added PatchableJump type, removed inUninterruptedSequence. * assembler/LinkBuffer.h: (LinkBuffer): (JSC::LinkBuffer::locationOf): - Only allow the location to be taken of patchable branches * assembler/MacroAssembler.h: (MacroAssembler): (JSC::MacroAssembler::patchableBranchPtrWithPatch): (JSC::MacroAssembler::patchableJump): (JSC::MacroAssembler::shouldBlind): - Added default implementation of patchableBranchPtrWithPatch, patchableJump. * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::MacroAssemblerARMv7): (MacroAssemblerARMv7): (JSC::MacroAssemblerARMv7::patchableBranchPtrWithPatch): (JSC::MacroAssemblerARMv7::patchableJump): (JSC::MacroAssemblerARMv7::jump): (JSC::MacroAssemblerARMv7::makeBranch): - Added ARMv7 implementation of patchableBranchPtrWithPatch, patchableJump. * dfg/DFGCorrectableJumpPoint.h: (DFG): (JSC::DFG::CorrectableJumpPoint::switchToLateJump): - Late jumps are PatchableJumps. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::linkOSRExits): - replace use of inUninterruptedSequence * dfg/DFGJITCompiler.h: (JSC::DFG::PropertyAccessRecord::PropertyAccessRecord): (PropertyAccessRecord): - replace use of inUninterruptedSequence * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::cachedGetById): (JSC::DFG::SpeculativeJIT::cachedPutById): - replace use of inUninterruptedSequence * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::cachedGetById): (JSC::DFG::SpeculativeJIT::cachedPutById): - replace use of inUninterruptedSequence * jit/JIT.h: (PropertyStubCompilationInfo): - replace use of inUninterruptedSequence * jit/JITInlineMethods.h: (JSC::JIT::beginUninterruptedSequence): (JSC::JIT::endUninterruptedSequence): - replace use of inUninterruptedSequence * jit/JITPropertyAccess.cpp: (JSC::JIT::compileGetByIdHotPath): - replace use of inUninterruptedSequence * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::compileGetByIdHotPath): - replace use of inUninterruptedSequence 2012-04-24 Benjamin Poulain <bpoulain@apple.com> Generalize the single character optimization of r114072 https://bugs.webkit.org/show_bug.cgi?id=83961 Reviewed by Eric Seidel. Use the regular String::find(StringImpl*) in all cases now that it has been made faster. * runtime/StringPrototype.cpp: (JSC::replaceUsingStringSearch): 2012-04-24 Filip Pizlo <fpizlo@apple.com> Unreviewed, 32-bit build fix. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-04-24 Filip Pizlo <fpizlo@apple.com> DFG performs incorrect DCE on (some?) intrinsics https://bugs.webkit.org/show_bug.cgi?id=84746 <rdar://problem/11310772> Reviewed by Oliver Hunt. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (ByteCodeParser): (JSC::DFG::ByteCodeParser::setIntrinsicResult): (JSC::DFG::ByteCodeParser::handleMinMax): (JSC::DFG::ByteCodeParser::handleIntrinsic): * dfg/DFGNodeType.h: (DFG): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-04-24 Mark Hahnenberg <mhahnenberg@apple.com> Failure to allocate ArrayStorage in emit_op_new_array leads to poisonous JSArray https://bugs.webkit.org/show_bug.cgi?id=84648 Reviewed by Geoffrey Garen. When emit_op_new_array successfully allocates a new JSArray but fails to allocate the corresponding ArrayStorage for it, it falls back to the out-of-line stub call to constructArray, which constructs and entirely new JSArray/ArrayStorage pair. This leaves us with a JSArray hanging around on the stack or in a register that did not go through its own constructor, thus giving it uninitialized memory in the two fields that are checked in JSArray::visitChildren. * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateJSArray): We try to allocate the ArrayStorage first, so that if we fail we haven't generated the poisonous JSArray that can cause a GC crash. * jit/JITOpcodes.cpp: (JSC::JIT::emitSlow_op_new_array): 2012-04-23 Filip Pizlo <fpizlo@apple.com> DFG on ARMv7 should not OSR exit on every integer division https://bugs.webkit.org/show_bug.cgi?id=84661 Reviewed by Oliver Hunt. On ARMv7, ArithDiv no longer has to know whether or not to speculate integer (since that was broken with the introduction of Int32ToDouble) nor does it have to know whether or not to convert its result to integer. This is now taken care of for free with the addition of the DoubleAsInt32 node, which represents a double-is-really-int speculation. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::performNodeCSE): * dfg/DFGFixupPhase.cpp: (JSC::DFG::FixupPhase::fixupNode): * dfg/DFGNodeType.h: (DFG): * dfg/DFGOSRExit.cpp: (JSC::DFG::OSRExit::OSRExit): (JSC::DFG::OSRExit::considerAddingAsFrequentExitSiteSlow): * dfg/DFGOSRExit.h: (OSRExit): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): (JSC::DFG::SpeculativeJIT::compileDoubleAsInt32): (DFG): * dfg/DFGSpeculativeJIT.h: (SpeculativeJIT): (JSC::DFG::SpeculativeJIT::speculationCheck): (JSC::DFG::SpeculativeJIT::forwardSpeculationCheck): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-04-24 Geoffrey Garen <ggaren@apple.com> "GlobalHandle" HandleHeap (now WeakSet) allocations grow but do not shrink https://bugs.webkit.org/show_bug.cgi?id=84740 <rdar://problem/9917638> Reviewed by Gavin Barraclough. Shrink! * heap/Heap.cpp: (JSC::Heap::destroy): Be more specific about what's shrinking, since we can also shrink the WeakSet, but we don't do so here. (JSC::Heap::collect): If we're going to shrink the heap, shrink the WeakSet too. Otherwise, its footprint is permanent. * heap/Heap.h: (Heap): Removed shrink() as a public interface, since it's vague about which parts of the heap it affects, and it's really an internal detail. * heap/WeakSet.cpp: (JSC::WeakSet::shrink): Nix any free blocks. We assume that sweep() has already taken place, since that's the convention for shrink() in the heap. * heap/WeakSet.h: (WeakSet): New function! 2012-04-24 Adam Klein <adamk@chromium.org> Fix includes in StrongInlines.h and ScriptValue.h https://bugs.webkit.org/show_bug.cgi?id=84659 Reviewed by Geoffrey Garen. * heap/StrongInlines.h: Include JSGlobalData.h, since JSGlobalData's definiition is required here. 2012-04-23 Filip Pizlo <fpizlo@apple.com> DFG OSR exit should ensure that all variables have been initialized https://bugs.webkit.org/show_bug.cgi?id=84653 <rdar://problem/11258183> Reviewed by Gavin Barraclough. Initialize all uncaptured dead variables to undefined on OSR exit. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::ValueSource::dump): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * dfg/DFGSpeculativeJIT.h: 2012-04-23 Oliver Hunt <oliver@apple.com> Call instruction for the baseline JIT stores origin info in wrong callframe https://bugs.webkit.org/show_bug.cgi?id=84645 Reviewed by Gavin Barraclough. The baseline JIT was updating the wrong callframe when making a call. If the call failed during dispatch (unable to perform codegen, calling a non-object) we would attempt to use this information, but it would be completely wrong. * jit/JITCall.cpp: (JSC::JIT::compileOpCall): * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCall): 2012-04-23 Filip Pizlo <fpizlo@apple.com> DFG must keep alive values that it will perform speculations on https://bugs.webkit.org/show_bug.cgi?id=84638 <rdar://problem/11258183> Reviewed by Oliver Hunt. * dfg/DFGNodeType.h: (DFG): 2012-04-23 Oliver Hunt <oliver@apple.com> Fix non-LLInt builds by temporarily removing an over-enthusiastic assertion * interpreter/Interpreter.cpp: (JSC::Interpreter::executeCall): 2012-04-22 Jon Lee <jonlee@apple.com> Remove notifications support on Mac Lion. https://bugs.webkit.org/show_bug.cgi?id=84554 <rdar://problem/11297128> Reviewed by Sam Weinig. * Configurations/FeatureDefines.xcconfig: 2012-04-21 Darin Adler <darin@apple.com> Change JavaScript lexer to use 0 instead of -1 for sentinel, eliminating the need to put characters into ints https://bugs.webkit.org/show_bug.cgi?id=84523 Reviewed by Oliver Hunt. Profiles showed that checks against -1 were costly, and I saw they could be eliminated. Streamlined this code to use standard character types and 0 rather than -1. One benefit of this is that there's no widening and narrowing. Another is that there are many cases where we already have the correct behavior for 0, so can eliminate a branch that was used to test for -1 before. Also eliminates typecasts in the code. * parser/Lexer.cpp: (JSC::Lexer::invalidCharacterMessage): Updated use of String::format since m_current is now a character type, not an int. (JSC::Lexer::setCode): Use 0 rather than -1 when past the end. (JSC::Lexer::shift): Ditto. Also spruced up the comment a bit. (JSC::Lexer::atEnd): Added. New function that distinguishes an actual 0 character from the end of the code. This can be used places we used to cheeck for -1. (JSC::Lexer::peek): Updated to use -1 instead of 0. Removed meaningless comment. (JSC::Lexer::parseFourDigitUnicodeHex): Changed to use character types instead of int. (JSC::Lexer::shiftLineTerminator): Removed now-unneeded type casts. Changed local variable that had a data-member-style name. (JSC::Lexer::parseIdentifier): Removed now-unneeded explicit checks for -1, since the isIdentPart function already returns false for the 0 character. Updated types in a couple other places. Used the atEnd function where needed. (JSC::Lexer::parseIdentifierSlowCase): More of the same. (JSC::characterRequiresParseStringSlowCase): Added overloaded helper function for parseString. (JSC::Lexer::parseString): Ditto. (JSC::Lexer::parseStringSlowCase): Ditto. (JSC::Lexer::parseMultilineComment): Ditto. (JSC::Lexer::lex): More of the same. Also changed code to set the startOffset directly in the tokenInfo instead of putting it in a local variable first, saving some memory access. (JSC::Lexer::scanRegExp): Ditto. (JSC::Lexer::skipRegExp): Ditto. * parser/Lexer.h: Changed return type of the peek function and type of m_current from int to the character type. Added atEnd function. (JSC::Lexer::setOffset): Used 0 instead of -1 and removed an overzealous attempt to optimize. (JSC::Lexer::lexExpectIdentifier): Used 0 instead of -1. 2012-04-21 Darin Adler <darin@apple.com> Change JavaScript lexer to use 0 instead of -1 for sentinel, eliminating the need to put characters into ints https://bugs.webkit.org/show_bug.cgi?id=84523 Reviewed by Oliver Hunt. Separate preparation step of copyright dates, renaming, and other small tweaks. * parser/Lexer.cpp: (JSC::Lexer::invalidCharacterMessage): Removed "get" from name to match WebKit naming conventions. (JSC::Lexer::peek): Removed meaningless comment. (JSC::Lexer::parseFourDigitUnicodeHex): Renamed from getUnicodeCharacter to be more precise about what this function does. (JSC::Lexer::shiftLineTerminator): Renamed local variable that had a data-member-style name. (JSC::Lexer::parseStringSlowCase): Updated for new name of parseFourDigitUnicodeHex. (JSC::Lexer::lex): Updated for new name of invalidCharacterMessage. * parser/Lexer.h: Removed an unneeded forward declaration of the RegExp class. Renamed getInvalidCharMessage to invalidCharacterMessage and made it const. Renamed getUnicodeCharacter to parseFourDigitUnicodeHex. 2012-04-20 Filip Pizlo <fpizlo@apple.com> DFG should optimize int8 and int16 arrays on ARMv7 https://bugs.webkit.org/show_bug.cgi?id=84503 Reviewed by Oliver Hunt. * assembler/ARMv7Assembler.h: (ARMv7Assembler): (JSC::ARMv7Assembler::ldrsb): (JSC::ARMv7Assembler::ldrsh): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::load16Signed): (JSC::MacroAssemblerARMv7::load8Signed): * bytecode/PredictedType.h: (JSC::isActionableIntMutableArrayPrediction): * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateInt8Array): (JSC::DFG::Node::shouldSpeculateInt16Array): 2012-04-20 Oliver Hunt <oliver@apple.com> Add an ability to find the extent of a callframe https://bugs.webkit.org/show_bug.cgi?id=84513 Reviewed by Filip Pizlo. Add a function to get the extent of a callframe and use that function for a new assertion to make sure the RegisterFile makes sense using that information. * interpreter/CallFrame.cpp: (JSC::CallFrame::frameExtentInternal): (JSC): * interpreter/CallFrame.h: (JSC::ExecState::frameExtent): (ExecState): * interpreter/Interpreter.cpp: (JSC::Interpreter::executeCall): 2012-04-20 Benjamin Poulain <bpoulain@apple.com> Inline the JSArray constructor https://bugs.webkit.org/show_bug.cgi?id=84416 Reviewed by Geoffrey Garen. The constructor is trivial, no reason to jump for it. This makes the creation of array ~5% faster (on non-trivial cases, no empty arrays). * runtime/JSArray.cpp: (JSC): * runtime/JSArray.h: (JSC::JSArray::JSArray): * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-04-20 Mark Hahnenberg <mhahnenberg@apple.com> Heap should cancel GC timer at the start of the collection https://bugs.webkit.org/show_bug.cgi?id=84477 Reviewed by Geoffrey Garen. Currently the Heap cancels the GC timer at the conclusion of a collection. We should change this to be at the beginning because something (e.g. a finalizer) could call didAbandonObjectGraph(), which will schedule the timer, but then we'll immediately unschedule the timer at the conclusion of the collection, thus potentially preventing large swaths of memory from being reclaimed in a timely manner. * API/JSBase.cpp: (JSGarbageCollect): Remove outdated fix-me and remove check for whether the Heap is busy or not, since we're just scheduling a timer to run a GC in the future. * heap/Heap.cpp: (JSC::Heap::collect): Rename didCollect to willCollect and move the call to the top of Heap::collect. * runtime/GCActivityCallback.cpp: Renamed didCollect to willCollect. (JSC::DefaultGCActivityCallback::willCollect): * runtime/GCActivityCallback.h: Ditto. (JSC::GCActivityCallback::willCollect): (DefaultGCActivityCallback): * runtime/GCActivityCallbackCF.cpp: Ditto. (JSC::DefaultGCActivityCallback::willCollect): 2012-04-20 Mark Hahnenberg <mhahnenberg@apple.com> JSGarbageCollect should not call collectAllGarbage() https://bugs.webkit.org/show_bug.cgi?id=84476 Reviewed by Geoffrey Garen. * API/JSBase.cpp: (JSGarbageCollect): Notify the Heap's GCActivityCallback using didAbandonObjectGraph. 2012-04-19 Oliver Hunt <oliver@apple.com> Exception stack traces aren't complete when the exception starts in native code https://bugs.webkit.org/show_bug.cgi?id=84073 Reviewed by Filip Pizlo. Refactored building the stack trace to so that we can construct it earlier, and don't rely on any prior work performed in the exception handling machinery. Also updated LLInt and the DFG to completely initialise the callframes of host function calls. Also fixed a few LLInt paths that failed to correctly update the topCallFrame. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * dfg/DFGJITCompiler.h: * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): * interpreter/Interpreter.cpp: (JSC::eval): (JSC::Interpreter::getStackTrace): (JSC::Interpreter::addStackTraceIfNecessary): (JSC): (JSC::Interpreter::throwException): * interpreter/Interpreter.h: (Interpreter): * jit/JITCall.cpp: (JSC::JIT::compileOpCall): * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCall): * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTINativeCall): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTINativeCall): * jsc.cpp: (functionJSCStack): * llint/LLIntExceptions.cpp: (JSC::LLInt::interpreterThrowInCaller): (JSC::LLInt::returnToThrow): (JSC::LLInt::callToThrow): * llint/LLIntSlowPaths.cpp: (JSC::LLInt::handleHostCall): * llint/LowLevelInterpreter32_64.asm: * llint/LowLevelInterpreter64.asm: * parser/Parser.h: (JSC::::parse): * runtime/Error.cpp: (JSC::addErrorInfo): (JSC::throwError): * runtime/Error.h: (JSC): 2012-04-19 Mark Hahnenberg <mhahnenberg@apple.com> We're collecting pathologically due to small allocations https://bugs.webkit.org/show_bug.cgi?id=84404 Reviewed by Geoffrey Garen. No change in performance on run-jsc-benchmarks. * dfg/DFGSpeculativeJIT.h: Replacing m_firstFreeCell with m_freeList. (JSC::DFG::SpeculativeJIT::emitAllocateBasicJSObject): * heap/CopiedSpace.cpp: Getting rid of any water mark related stuff, since it's no longer useful. (JSC::CopiedSpace::CopiedSpace): (JSC::CopiedSpace::tryAllocateSlowCase): We now only call didAllocate here rather than carrying out a somewhat complicated accounting job for our old water mark throughout CopiedSpace. (JSC::CopiedSpace::tryAllocateOversize): Call the new didAllocate to notify the Heap of newly allocated stuff. (JSC::CopiedSpace::tryReallocateOversize): (JSC::CopiedSpace::doneFillingBlock): (JSC::CopiedSpace::doneCopying): (JSC::CopiedSpace::destroy): * heap/CopiedSpace.h: (CopiedSpace): * heap/CopiedSpaceInlineMethods.h: (JSC::CopiedSpace::startedCopying): * heap/Heap.cpp: Removed water mark related stuff, replaced with new bytesAllocated and bytesAllocatedLimit to track how much memory has been allocated since the last collection. (JSC::Heap::Heap): (JSC::Heap::reportExtraMemoryCostSlowCase): (JSC::Heap::collect): We now set the new limit of bytes that we can allocate before triggering a collection to be the size of the Heap after the previous collection. Thus, we still have our 2x allocation amount. (JSC::Heap::didAllocate): Notifies the GC activity timer of how many bytes have been allocated thus far and then adds the new number of bytes to the current total. (JSC): * heap/Heap.h: Removed water mark related stuff. (JSC::Heap::notifyIsSafeToCollect): (Heap): (JSC::Heap::shouldCollect): (JSC): * heap/MarkedAllocator.cpp: (JSC::MarkedAllocator::tryAllocateHelper): Refactored to use MarkedBlock's new FreeList struct. (JSC::MarkedAllocator::allocateSlowCase): (JSC::MarkedAllocator::addBlock): * heap/MarkedAllocator.h: (MarkedAllocator): (JSC::MarkedAllocator::MarkedAllocator): (JSC::MarkedAllocator::allocate): (JSC::MarkedAllocator::zapFreeList): Refactored to take in a FreeList instead of a FreeCell. * heap/MarkedBlock.cpp: (JSC::MarkedBlock::specializedSweep): (JSC::MarkedBlock::sweep): (JSC::MarkedBlock::sweepHelper): (JSC::MarkedBlock::zapFreeList): * heap/MarkedBlock.h: (FreeList): Added a new struct that keeps track of the current MarkedAllocator's free list including the number of bytes of stuff in the free list so that when the free list is exhausted, the correct amount can be reported to Heap. (MarkedBlock): (JSC::MarkedBlock::FreeList::FreeList): (JSC): * heap/MarkedSpace.cpp: Removing all water mark related stuff. (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::resetAllocators): * heap/MarkedSpace.h: (MarkedSpace): (JSC): * heap/WeakSet.cpp: (JSC::WeakSet::findAllocator): Refactored to use the didAllocate interface with the Heap. This function still needs work though now that the Heap knows how many bytes have been allocated since the last collection. * jit/JITInlineMethods.h: Refactored to use MarkedBlock's new FreeList struct. (JSC::JIT::emitAllocateBasicJSObject): Ditto. * llint/LowLevelInterpreter.asm: Ditto. * runtime/GCActivityCallback.cpp: (JSC::DefaultGCActivityCallback::didAllocate): * runtime/GCActivityCallback.h: (JSC::GCActivityCallback::didAllocate): Renamed willAllocate to didAllocate to indicate that the allocation that is being reported has already taken place. (DefaultGCActivityCallback): * runtime/GCActivityCallbackCF.cpp: (JSC): (JSC::DefaultGCActivityCallback::didAllocate): Refactored to return early if the amount of allocation since the last collection is not above a threshold (initially arbitrarily chosen to be 128KB). 2012-04-19 Filip Pizlo <fpizlo@apple.com> MacroAssemblerARMv7::branchTruncateDoubleToUint32 should obey the overflow signal https://bugs.webkit.org/show_bug.cgi?id=84401 Reviewed by Gavin Barraclough. * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::branchTruncateDoubleToUint32): 2012-04-19 Don Olmstead <don.olmstead@am.sony.com> KeywordLookupGenerator.py should take an output file as an argument https://bugs.webkit.org/show_bug.cgi?id=84292 Reviewed by Eric Seidel. Extended KeywordLookupGenerator to accept an additional argument specifying an output file. If this argument is found stdout is redirected to a file for the duration of the script. * KeywordLookupGenerator.py: 2012-04-19 Filip Pizlo <fpizlo@apple.com> It should be possible to perform debugCall on ARMv7 https://bugs.webkit.org/show_bug.cgi?id=84381 Reviewed by Oliver Hunt. debugCall() was clobbering the argument to the call it was making, leading to a corrupt ExecState*. This change fixes that issue by using a scratch register that does not clobber arguments, and it also introduces more assertions that we have a valid call frame. * dfg/DFGAssemblyHelpers.cpp: (DFG): (JSC::DFG::AssemblyHelpers::jitAssertHasValidCallFrame): * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::selectScratchGPR): (AssemblyHelpers): (JSC::DFG::AssemblyHelpers::debugCall): (JSC::DFG::AssemblyHelpers::jitAssertHasValidCallFrame): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::linkOSRExits): * dfg/DFGOSRExitCompiler.cpp: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::selectScratchGPR): 2012-04-19 Filip Pizlo <fpizlo@apple.com> LLInt no-JIT fallback native call trampoline's exception handler incorrectly assumes that the PB/PC has been preserved https://bugs.webkit.org/show_bug.cgi?id=84367 Reviewed by Oliver Hunt. * llint/LowLevelInterpreter32_64.asm: * llint/LowLevelInterpreter64.asm: 2012-04-19 Filip Pizlo <fpizlo@apple.com> It should be possible to load from Float64 arrays on ARMv7 without crashing https://bugs.webkit.org/show_bug.cgi?id=84361 Reviewed by Oliver Hunt. * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::loadDouble): (JSC::MacroAssemblerARMv7::storeDouble): 2012-04-19 Dominik Röttsches <dominik.rottsches@linux.intel.com> [CMake] Build fix after r114575 https://bugs.webkit.org/show_bug.cgi?id=84322 Reviewed by Simon Hausmann. Build fix, adding WTF when linking jsc shell. * shell/CMakeLists.txt: 2012-04-18 Filip Pizlo <fpizlo@apple.com> JSC testing should have complete coverage over typed array types https://bugs.webkit.org/show_bug.cgi?id=84302 Reviewed by Geoff Garen. Added Uint8ClampedArray to the set of typed arrays that are supported by jsc command-line. * JSCTypedArrayStubs.h: (JSC): * jsc.cpp: (GlobalObject::finishCreation): 2012-04-18 Filip Pizlo <fpizlo@apple.com> jsc command line should support typed arrays by default https://bugs.webkit.org/show_bug.cgi?id=84298 Rubber stamped by Gavin Barraclough. * JSCTypedArrayStubs.h: (JSC): * jsc.cpp: (GlobalObject::finishCreation): 2012-04-18 Filip Pizlo <fpizlo@apple.com> JSVALUE32_64 should be able to perform division on ARM without crashing, and variables forced double should not be scrambled when performing OSR entry https://bugs.webkit.org/show_bug.cgi?id=84272 Reviewed by Geoff Garen. * dfg/DFGFixupPhase.cpp: (JSC::DFG::FixupPhase::fixupNode): * dfg/DFGOSREntry.cpp: (JSC::DFG::prepareOSREntry): 2012-04-18 Don Olmstead <don.olmstead@am.sony.com> JavaScriptCore.gypi not current https://bugs.webkit.org/show_bug.cgi?id=84224 Reviewed by Eric Seidel. Updated JavaScriptCore.gypi to contain the latest sources. Removed os-win32 as it wasn't used. Also removed references to ICU files in the gypi file as ICU is most likely specified by the port itself. Private and public header files were determined by looking at copy-files.cmd within Apple's Visual Studio directory. * JavaScriptCore.gypi: 2012-04-18 Benjamin Poulain <bpoulain@apple.com> Remove m_subclassData from JSArray, move the attribute to subclass as needed https://bugs.webkit.org/show_bug.cgi?id=84249 Reviewed by Geoffrey Garen. JSArray's m_subclassData is only used by WebCore's RuntimeArray. This patch moves the attribute to RuntimeArray to avoid allocating memory for the pointer in the common case. This gives ~1% improvement in JSArray creation microbenchmark thanks to fewer allocations of CopiedSpace. * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateJSArray): * runtime/JSArray.cpp: (JSC::JSArray::JSArray): * runtime/JSArray.h: 2012-04-18 Benjamin Poulain <bpoulain@apple.com> replaceUsingStringSearch: delay the creation of the replace string until needed https://bugs.webkit.org/show_bug.cgi?id=83841 Reviewed by Geoffrey Garen. We do not need to obtain the replaceValue until we have a match. By moving the intialization of replaceValue when needed, we save a few instructions when there is no match. * runtime/StringPrototype.cpp: (JSC::replaceUsingRegExpSearch): (JSC::replaceUsingStringSearch): (JSC::stringProtoFuncReplace): 2012-04-18 Mark Hahnenberg <mhahnenberg@apple.com> GC activity timer should be tied to allocation, not collection https://bugs.webkit.org/show_bug.cgi?id=83919 Reviewed by Geoffrey Garen. * API/JSContextRef.cpp: Used the new didAbandonObjectGraph callback to indicate that now that we've released a global object, we're abandoning a potentially large number of objects that JSC might want to collect. * heap/CopiedSpace.cpp: (JSC::CopiedSpace::tryAllocateSlowCase): Added the call to timer's willAllocate function to indicate that we've hit a slow path and are allocating now, so schedule the timer. * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::collectAllGarbage): Removed the call to discardAllCompiledCode because it was causing us to throw away too much code during our benchmarks (especially vp8, which is very large and thus has large amounts of compiled code). (JSC::Heap::collect): Added the new call to didCollect at the conclusion of a collection so that we can cancel the timer if we no longer need to run a collection. Also added a check at the beginning of a collection to see if we should throw away our compiled code. Currently this is set to happen about once every minute. * heap/Heap.h: Added field to keep track of the last time we threw away our compiled code. * heap/MarkedAllocator.cpp: (JSC::MarkedAllocator::allocateSlowCase): Added call to willAllocate on the allocation slow path, just like in CopiedSpace. * runtime/GCActivityCallback.cpp: Added default stubs for non-CF platforms. (JSC::DefaultGCActivityCallback::willAllocate): (JSC): (JSC::DefaultGCActivityCallback::didCollect): (JSC::DefaultGCActivityCallback::didAbandonObjectGraph): * runtime/GCActivityCallback.h: Added new functions to make JSC's GC timer less arcane. This includes replacing the operator () with willAllocate() and adding an explicit didCollect() to cancel the timer after a collection occurs rather than relying on the way the timer is invoked to cancel itself. Also added a callback for when somebody else (e.g. WebCore or the JSC API) to notify JSC that they have just abandoned an entire graph of objects and that JSC might want to clean them up. (JSC::GCActivityCallback::~GCActivityCallback): (JSC::GCActivityCallback::willAllocate): (JSC::GCActivityCallback::didCollect): (JSC::GCActivityCallback::didAbandonObjectGraph): (JSC::GCActivityCallback::synchronize): (DefaultGCActivityCallback): * runtime/GCActivityCallbackCF.cpp: Re-wired all the run loop stuff to implement the aforementioned functions. We added a flag to check whether the timer was active because the call to CFRunLoopTimerSetNextFireDate actually turned out to be quite expensive (although Instruments couldn't tell us this). (DefaultGCActivityCallbackPlatformData): (JSC): (JSC::DefaultGCActivityCallbackPlatformData::timerDidFire): (JSC::DefaultGCActivityCallback::commonConstructor): (JSC::scheduleTimer): (JSC::cancelTimer): (JSC::DefaultGCActivityCallback::willAllocate): (JSC::DefaultGCActivityCallback::didCollect): (JSC::DefaultGCActivityCallback::didAbandonObjectGraph): 2012-04-17 Filip Pizlo <fpizlo@apple.com> DFG should not attempt to get rare case counts for op_mod on ARM https://bugs.webkit.org/show_bug.cgi?id=84218 Reviewed by Geoff Garen. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::makeSafe): * dfg/DFGCommon.h: (JSC::DFG::isX86): (DFG): 2012-04-17 Myles Maxfield <mmaxfield@google.com> BumpPointerAllocator assumes page size is less than MINIMUM_BUMP_POOL_SIZE https://bugs.webkit.org/show_bug.cgi?id=80912 Reviewed by Hajime Morita. * wtf/BumpPointerAllocator.h: (WTF::BumpPointerPool::create): 2012-04-17 Filip Pizlo <fpizlo@apple.com> Attempt to fix Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-04-17 Filip Pizlo <fpizlo@apple.com> It should be possible to create an inheritorID for the global this object without crashing https://bugs.webkit.org/show_bug.cgi?id=84200 <rdar://problem/11251082> Reviewed by Oliver Hunt. * runtime/JSGlobalThis.cpp: (JSC::JSGlobalThis::setUnwrappedObject): * runtime/JSGlobalThis.h: (JSC::JSGlobalThis::unwrappedObject): (JSGlobalThis): * runtime/JSObject.cpp: (JSC::JSObject::createInheritorID): * runtime/JSObject.h: (JSObject): (JSC::JSObject::resetInheritorID): 2012-04-17 Filip Pizlo <fpizlo@apple.com> DFG and LLInt should not clobber the frame pointer on ARMv7 https://bugs.webkit.org/show_bug.cgi?id=84185 <rdar://problem/10767252> Reviewed by Gavin Barraclough. Changed LLInt to use a different register. Changed DFG to use one fewer registers. We should revisit this and switch the DFG to use a different register instead of r7, but we can do that in a subsequent step since the performance effect is tiny. * dfg/DFGGPRInfo.h: (GPRInfo): (JSC::DFG::GPRInfo::toRegister): (JSC::DFG::GPRInfo::toIndex): * offlineasm/armv7.rb: 2012-04-17 Filip Pizlo <fpizlo@apple.com> use after free in JSC::DFG::Node::op / JSC::DFG::ByteCodeParser::flushArgument https://bugs.webkit.org/show_bug.cgi?id=83942 <rdar://problem/11247370> Reviewed by Gavin Barraclough. Don't use references to the graph after resizing the graph. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::flushArgument): 2012-04-16 Gavin Barraclough <barraclough@apple.com> Array.prototype.toString should be generic https://bugs.webkit.org/show_bug.cgi?id=81588 Reviewed by Sam Weinig. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncToString): - check for join function, use fast case if base object is array & join is present & default. * runtime/CommonIdentifiers.h: - added 'join'. 2012-04-16 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck issues. * GNUmakefile.list.am: Add missing files. 2012-04-16 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r114309. http://trac.webkit.org/changeset/114309 https://bugs.webkit.org/show_bug.cgi?id=84097 it broke everything (Requested by olliej on #webkit). * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * bytecode/CodeBlock.h: * dfg/DFGOperations.cpp: * interpreter/Interpreter.cpp: (JSC::Interpreter::getStackTrace): (JSC::Interpreter::throwException): * interpreter/Interpreter.h: (Interpreter): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jsc.cpp: (functionJSCStack): * llint/LLIntSlowPaths.cpp: (JSC::LLInt::handleHostCall): * parser/Parser.h: (JSC::::parse): * runtime/Error.cpp: (JSC::addErrorInfo): (JSC::throwError): * runtime/Error.h: (JSC): 2012-04-16 Oliver Hunt <oliver@apple.com> Exception stack traces aren't complete when the exception starts in native code https://bugs.webkit.org/show_bug.cgi?id=84073 Reviewed by Gavin Barraclough. Refactored building the stack trace to so that we can construct it earlier, and don't rely on any prior work performed in the exception handling machinery. Also updated LLInt and the DFG to completely initialise the callframes of host function calls. * bytecode/CodeBlock.h: (JSC::CodeBlock::codeOriginIndexForReturn): (CodeBlock): * dfg/DFGOperations.cpp: * interpreter/Interpreter.cpp: (JSC::Interpreter::getStackTrace): (JSC::Interpreter::addStackTraceIfNecessary): (JSC): (JSC::Interpreter::throwException): * interpreter/Interpreter.h: (Interpreter): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jsc.cpp: (functionJSCStack): * llint/LLIntSlowPaths.cpp: (JSC::LLInt::handleHostCall): * parser/Parser.h: (JSC::::parse): * runtime/Error.cpp: (JSC::addErrorInfo): (JSC::throwError): * runtime/Error.h: (JSC): 2012-04-16 Oliver Hunt <oliver@apple.com> Fix COMMANDLINE_TYPEDARRAYS build https://bugs.webkit.org/show_bug.cgi?id=84051 Reviewed by Gavin Barraclough. Update for new putByIndex API and wtf changes. * JSCTypedArrayStubs.h: (JSC): 2012-04-16 Mark Hahnenberg <mhahnenberg@apple.com> GC in the middle of JSObject::allocatePropertyStorage can cause badness https://bugs.webkit.org/show_bug.cgi?id=83839 Reviewed by Geoffrey Garen. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * jit/JITStubs.cpp: Making changes to use the new return value of growPropertyStorage. (JSC::DEFINE_STUB_FUNCTION): * runtime/JSObject.cpp: (JSC::JSObject::growPropertyStorage): Renamed to more accurately reflect that we're growing our already-existing PropertyStorage. * runtime/JSObject.h: (JSObject): (JSC::JSObject::setPropertyStorage): "Atomically" sets the new property storage and the new structure so that we can be sure a GC never occurs when our Structure info is out of sync with our PropertyStorage. (JSC): (JSC::JSObject::putDirectInternal): Moved the check to see if we should allocate more backing store before the actual property insertion into the structure. (JSC::JSObject::putDirectWithoutTransition): Ditto. (JSC::JSObject::transitionTo): Ditto. * runtime/Structure.cpp: (JSC::Structure::suggestedNewPropertyStorageSize): Added to keep the resize policy for property backing stores contained within the Structure class. (JSC): * runtime/Structure.h: (JSC::Structure::shouldGrowPropertyStorage): Lets clients know if another insertion into the Structure would require resizing the property backing store so that they can preallocate the required storage. (Structure): 2012-04-13 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r114185. http://trac.webkit.org/changeset/114185 https://bugs.webkit.org/show_bug.cgi?id=83967 Broke a bunch of JavaScript related tests (Requested by andersca on #webkit). * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncToString): (JSC::arrayProtoFuncToLocaleString): * runtime/CommonIdentifiers.h: * tests/mozilla/ecma/Array/15.4.4.2.js: (getTestCases): 2012-04-13 Gavin Barraclough <barraclough@apple.com> Don't rely on fixed offsets to patch calls https://bugs.webkit.org/show_bug.cgi?id=83966 Rubber stamped by Oliver Hunt. These aren't being used anywhere! * jit/JIT.h: * jit/JITCall.cpp: (JSC::JIT::compileOpCall): * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCall): 2012-04-13 Hojong Han <hojong.han@samsung.com> Array.prototype.toString and Array.prototype.toLocaleString should be generic https://bugs.webkit.org/show_bug.cgi?id=81588 Reviewed by Gavin Barraclough. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncToString): (JSC::arrayProtoFuncToLocaleString): * runtime/CommonIdentifiers.h: * tests/mozilla/ecma/Array/15.4.4.2.js: (getTestCases.array.item.new.TestCase): (getTestCases): 2012-04-13 Gavin Barraclough <barraclough@apple.com> Don't rely on fixed offsets to patch method checks https://bugs.webkit.org/show_bug.cgi?id=83958 Reviewed by Oliver Hunt. * bytecode/StructureStubInfo.h: - Add fields for the method check info. * jit/JIT.cpp: (JSC::PropertyStubCompilationInfo::copyToStubInfo): - Store the offsets on the stub info, instead of asserting. * jit/JIT.h: - Delete all the method check related offsets. * jit/JITPropertyAccess.cpp: (JSC::JIT::patchMethodCallProto): - Use the offset from the stubInfo. * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): - Pass the stubInfo to patchMethodCallProto. 2012-04-13 Gavin Barraclough <barraclough@apple.com> Don't rely on fixed offsets to patch get_by_id/put_by_id https://bugs.webkit.org/show_bug.cgi?id=83924 Reviewed by Oliver Hunt. Store offsets in the structure stub info, as we do for the DFG JIT. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::differenceBetween): - this method can be static (now used from PropertyStubCompilationInfo::copyToStubInfo, will be removed soon!) * bytecode/StructureStubInfo.h: - added new fields for baseline JIT offsets. * jit/JIT.cpp: (JSC::PropertyStubCompilationInfo::copyToStubInfo): - moved out from JIT::privateCompile. (JSC::JIT::privateCompile): - moved out code to PropertyStubCompilationInfo::copyToStubInfo. * jit/JIT.h: (PropertyStubCompilationInfo): - added helper functions to initializae PropertyStubCompilationInfo, state to store more offset info. - removed many offsets. * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_method_check): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::compileGetByIdSlowCase): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::emitSlow_op_put_by_id): (JSC::JIT::patchGetByIdSelf): (JSC::JIT::patchPutByIdReplace): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): (JSC::JIT::resetPatchGetById): (JSC::JIT::resetPatchPutById): - changed code generation to use new interface to store info on PropertyStubCompilationInfo. - changed repatch functions to read offsets from the structure stub info. * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_method_check): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::compileGetByIdSlowCase): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::emitSlow_op_put_by_id): (JSC::JIT::patchGetByIdSelf): (JSC::JIT::patchPutByIdReplace): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): (JSC::JIT::resetPatchGetById): (JSC::JIT::resetPatchPutById): - changed code generation to use new interface to store info on PropertyStubCompilationInfo. - changed repatch functions to read offsets from the structure stub info. 2012-04-13 Rob Buis <rbuis@rim.com> Fix some compiler warnings (miscellaneous) https://bugs.webkit.org/show_bug.cgi?id=80790 Reviewed by Antonio Gomes. Fix signed/unsigned comparison warning. * parser/Lexer.cpp: (JSC::::record16): 2012-04-12 Benjamin Poulain <bpoulain@apple.com> Improve replaceUsingStringSearch() for case of a single character searchValue https://bugs.webkit.org/show_bug.cgi?id=83738 Reviewed by Geoffrey Garen. This patch improves replaceUsingStringSearch() with the following: -Add a special case for single character search, taking advantage of the faster WTF::find(). -Inline replaceUsingStringSearch(). -Use StringImpl::create() instead of UString::substringSharingImpl() since we know we are in the bounds by definition. This gives less than 1% improvement for the multicharacter replace. The single character search show about 9% improvement. * runtime/StringPrototype.cpp: (JSC::replaceUsingStringSearch): 2012-04-12 Michael Saboff <msaboff@apple.com> StructureStubInfo::reset() causes leaks of PolymorphicAccessStructureList and ExecutableMemoryHandle objects https://bugs.webkit.org/show_bug.cgi?id=83823 Reviewed by Gavin Barraclough. Put the clearing of the accessType to after the call to deref() so that deref() can use the accessType to delete referenced objects as needed. * bytecode/StructureStubInfo.h: (JSC::StructureStubInfo::reset): 2012-04-12 Balazs Kelemen <kbalazs@webkit.org> [Qt] Fix WebKit1 build with V8 https://bugs.webkit.org/show_bug.cgi?id=83322 Reviewed by Adam Barth. * yarr/yarr.pri: 2012-04-12 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=83821 Move dfg repatching properties of structure stub info into a union Reviewed by Oliver Hunt. We want to be able to have similar properties for the baseline JIT, some restructuring to prepare for this. * bytecode/StructureStubInfo.h: (StructureStubInfo): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGRepatch.cpp: (JSC::DFG::dfgRepatchByIdSelfAccess): (JSC::DFG::linkRestoreScratch): (JSC::DFG::generateProtoChainAccessStub): (JSC::DFG::tryCacheGetByID): (JSC::DFG::tryBuildGetByIDList): (JSC::DFG::tryBuildGetByIDProtoList): (JSC::DFG::emitPutReplaceStub): (JSC::DFG::emitPutTransitionStub): (JSC::DFG::tryCachePutByID): (JSC::DFG::tryBuildPutByIdList): (JSC::DFG::dfgResetGetByID): (JSC::DFG::dfgResetPutByID): 2012-04-12 Gavin Barraclough <barraclough@apple.com> Delete a bunch of unused, copy & pasted values in JIT.h https://bugs.webkit.org/show_bug.cgi?id=83822 Reviewed by Oliver Hunt. The only architecture we support the JSVALUE64 JIT on is x86-64, all the patch offsets for other architectures are just nonsense. * jit/JIT.h: (JIT): 2012-04-12 Csaba Osztrogonác <ossy@webkit.org> [Qt][ARM] Buildfix after r113934. Reviewed by Zoltan Herczeg. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::compare8): (MacroAssemblerARM): 2012-04-11 Filip Pizlo <fpizlo@apple.com> It is incorrect to short-circuit Branch(LogicalNot(@a)) if boolean speculations on @a may fail https://bugs.webkit.org/show_bug.cgi?id=83744 <rdar://problem/11206946> Reviewed by Andy Estes. This does the conservative thing: it only short-circuits Branch(LogicalNot(@a)) if @a is a node that is statically known to return boolean results. * dfg/DFGFixupPhase.cpp: (JSC::DFG::FixupPhase::fixupNode): 2012-04-11 Michael Saboff <msaboff@apple.com> Invalid Union Reference in StructureStubInfo.{cpp.h} https://bugs.webkit.org/show_bug.cgi?id=83735 Reviewed by Filip Pizlo. Changed the references to u.getByIdProtoList and u.getByIdSelfList to be consistent. * bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::visitWeakReferences): * bytecode/StructureStubInfo.h: (JSC::StructureStubInfo::initGetByIdSelfList): 2012-04-11 Filip Pizlo <fpizlo@apple.com> Unreviewed attempting to make Qt's eccentric hardware work. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::compare8): (MacroAssemblerARM): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::compare8): (MacroAssemblerMIPS): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::compare8): (MacroAssemblerSH4): 2012-04-11 Filip Pizlo <fpizlo@apple.com> op_is_foo should be optimized https://bugs.webkit.org/show_bug.cgi?id=83666 Reviewed by Gavin Barraclough. This implements inlining of op_is_undefined, op_is_string, op_is_number, and op_is_boolean in LLInt and the baseline JIT. op_is_object and op_is_function are not inlined because they are quite a bit more complex. This also implements all of the op_is_foo opcodes in the DFG, but it does not do any type profiling based optimizations, yet. * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::compare8): (MacroAssemblerARMv7): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::compare8): (MacroAssemblerX86Common): * assembler/MacroAssemblerX86_64.h: (MacroAssemblerX86_64): (JSC::MacroAssemblerX86_64::testPtr): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCCallHelpers.h: (JSC::DFG::CCallHelpers::setupArguments): (CCallHelpers): * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::performNodeCSE): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGNodeType.h: (DFG): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::callOperation): (JSC::DFG::SpeculativeJIT::appendCallSetResult): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): * jit/JIT.h: (JIT): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_is_undefined): (JSC): (JSC::JIT::emit_op_is_boolean): (JSC::JIT::emit_op_is_number): (JSC::JIT::emit_op_is_string): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_is_undefined): (JSC): (JSC::JIT::emit_op_is_boolean): (JSC::JIT::emit_op_is_number): (JSC::JIT::emit_op_is_string): * jit/JITStubs.cpp: (JSC): * llint/LLIntSlowPaths.cpp: (LLInt): * llint/LLIntSlowPaths.h: (LLInt): * llint/LowLevelInterpreter.asm: * llint/LowLevelInterpreter32_64.asm: * llint/LowLevelInterpreter64.asm: * offlineasm/armv7.rb: * offlineasm/instructions.rb: * offlineasm/x86.rb: 2012-04-11 Filip Pizlo <fpizlo@apple.com> If you use an IntegerOperand and want to return it with integerResult, you need to zero extend to get rid of the box https://bugs.webkit.org/show_bug.cgi?id=83734 <rdar://problem/11232296> Reviewed by Oliver Hunt. * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillInteger): (JSC::DFG::SpeculativeJIT::nonSpeculativeValueToInt32): 2012-04-11 Filip Pizlo <fpizlo@apple.com> SpeculativeJIT::fillStorage() should work with all the states that a cell may be in https://bugs.webkit.org/show_bug.cgi?id=83722 Reviewed by Gavin Barraclough. It's now possible to do StorageOperand on a cell, in the case that the storage is inline. But this means that fillStorage() must be able to handle all of the states that a cell might be in. Previously it didn't. With this change, it now does handle all of the states, and moreover, it does so by preserving the DataFormat of cells and performing all of the cell speculations that should be performed if you're using a cell as storage. But if you use this on something that is known to be storage already then it behaves as it did before. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillStorage): 2012-04-11 Filip Pizlo <fpizlo@apple.com> Global variable predictions should not be coalesced unnecessarily https://bugs.webkit.org/show_bug.cgi?id=83678 Reviewed by Geoff Garen. Removed the PredictionTracker and everyone who used it. Converted GetGlobalVar to have a heapPrediction like a civilized DFG opcode ought to. No performance effect. * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.h: * bytecode/PredictionTracker.h: Removed. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGenerationInfo.h: * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (Graph): * dfg/DFGNode.h: (JSC::DFG::Node::hasHeapPrediction): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): 2012-04-11 Benjamin Poulain <bpoulain@apple.com> Optimize String.split() for 1 character separator https://bugs.webkit.org/show_bug.cgi?id=83546 Reviewed by Gavin Barraclough. This patch adds a serie of optimizations to make stringProtoFuncSplit() faster in the common case where the separator is a single character. The two main gains are: -Use of the find() function with a single character instead of doing a full string matching. -Use of WTF::find() instead of UString::find() to avoid branching on is8Bit() and have a simpler inline function. The code is also changed to avoid making unnecessary allocations by converting the 8bit string to 16bits. This makes String.split() faster by about 13% in that particular case. * runtime/StringPrototype.cpp: (JSC): (JSC::splitStringByOneCharacterImpl): (JSC::stringProtoFuncSplit): 2012-04-10 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck issues. * GNUmakefile.list.am: Ad missing files. 2012-04-10 Mark Rowe <mrowe@apple.com> Attempt to fix the Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-04-10 Patrick Gansterer <paroga@webkit.org> Cleanup wtf/Platform.h and config.h files https://bugs.webkit.org/show_bug.cgi?id=83431 Reviewed by Eric Seidel. The ENABLE() and USE() macros take care about the case when the flag isn't defined. So there is no need to define anything with 0. Also move duplicated code from the config.h files to Platform.h and merge a few preprocessor commands to make the file more readable. * config.h: 2012-04-10 Filip Pizlo <fpizlo@apple.com> DFG should flush SetLocals to arguments https://bugs.webkit.org/show_bug.cgi?id=83554 Reviewed by Gavin Barraclough. This is necessary to match baseline JIT argument capture behavior. But to make this work right we need to have a story for arguments into which we store values of different formats. This patch introduces the notion of an ArgumentPosition - i.e. an argument in a particular inline call frame - and forces unification of all data pertinent to selecting the argument's data format. Also fixed an amusing bug in the handling of OSR on SetLocals if there was any insertion/deletion of nodes in the basic block. This is benign for now but won't be eventually since the DFG is getting smarter. So better fix it now. Also fixed an amusing bug in the handling of OSR on SetLocals if they are immediately followed by a Flush. I think this bug might have always been there but now it'll happen more commonly, and it's covered by the run-javascriptcore-tests. * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGArgumentPosition.h: Added. (DFG): (ArgumentPosition): (JSC::DFG::ArgumentPosition::ArgumentPosition): (JSC::DFG::ArgumentPosition::addVariable): (JSC::DFG::ArgumentPosition::mergeArgumentAwareness): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::setLocal): (JSC::DFG::ByteCodeParser::setArgument): (InlineStackEntry): (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): * dfg/DFGDoubleFormatState.h: Added. (DFG): (JSC::DFG::mergeDoubleFormatStates): (JSC::DFG::mergeDoubleFormatState): (JSC::DFG::doubleFormatStateToString): * dfg/DFGGraph.h: (Graph): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::doRoundOfDoubleVoting): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGVariableAccessData.h: (JSC::DFG::VariableAccessData::VariableAccessData): (JSC::DFG::VariableAccessData::predict): (JSC::DFG::VariableAccessData::argumentAwarePrediction): (VariableAccessData): (JSC::DFG::VariableAccessData::mergeArgumentAwarePrediction): (JSC::DFG::VariableAccessData::doubleFormatState): (JSC::DFG::VariableAccessData::shouldUseDoubleFormat): (JSC::DFG::VariableAccessData::tallyVotesForShouldUseDoubleFormat): (JSC::DFG::VariableAccessData::mergeDoubleFormatState): (JSC::DFG::VariableAccessData::makePredictionForDoubleFormat): 2012-04-10 Adam Klein <adamk@chromium.org> Remove unused NonNullPassRefPtr from WTF https://bugs.webkit.org/show_bug.cgi?id=82389 Reviewed by Kentaro Hara. * JavaScriptCore.order: Remove nonexistent symbols referencing NonNullPassRefPtr. 2012-04-10 Darin Adler <darin@apple.com> Remove unused data member from Lexer class https://bugs.webkit.org/show_bug.cgi?id=83429 Reviewed by Kentaro Hara. I noticed that m_delimited was "write-only", so I deleted it. * parser/Lexer.cpp: (JSC::Lexer::setCode): Removed code to set m_delimited. (JSC::Lexer::parseIdentifier): Ditto. (JSC::Lexer::parseIdentifierSlowCase): Ditto. (JSC::Lexer::lex): Ditto. * parser/Lexer.h: Deleted m_delimited. 2012-04-10 Patrick Gansterer <paroga@webkit.org> [CMake] Enable USE_FOLDERS property https://bugs.webkit.org/show_bug.cgi?id=83571 Reviewed by Daniel Bates. Setting the FOLDER property on targets gives more structure to the generated Visual Studio solutions. This does not affect other CMake generators. * CMakeLists.txt: * shell/CMakeLists.txt: 2012-04-10 Filip Pizlo <fpizlo@apple.com> It should be possible to see why a code block was not compiled by the DFG https://bugs.webkit.org/show_bug.cgi?id=83553 Reviewed by Geoff Garen. If DFG_ENABLE(DEBUG_VERBOSE) and a code block is rejected, then print the opcode that caused the rejection. * dfg/DFGCapabilities.cpp: (JSC::DFG::debugFail): (DFG): (JSC::DFG::canHandleOpcodes): 2012-04-09 Gavin Barraclough <barraclough@apple.com> If a callback constructor returns a C++ null, throw a type error. https://bugs.webkit.org/show_bug.cgi?id=83537 Rubber Stamped by Geoff Garen. * API/JSCallbackConstructor.cpp: (JSC::constructJSCallback): - If a callback constructor returns a C++ null, throw a type error. * API/tests/testapi.c: (Base_returnHardNull): * API/tests/testapi.js: - Add a test case for callback constructors that return a C++ null. 2012-04-09 Gavin Barraclough <barraclough@apple.com> If a callback function returns a C++ null, convert to undefined. https://bugs.webkit.org/show_bug.cgi?id=83534 Reviewed by Geoff Garen. * API/JSCallbackFunction.cpp: - If a callback function returns a C++ null, convert to undefined. (JSC::JSCallbackFunction::call): * API/tests/testapi.c: (Base_returnHardNull): * API/tests/testapi.js: - Add a test case for callback functions that return a C++ null. 2012-04-09 Filip Pizlo <fpizlo@apple.com> Classic interpreter's GC hooks shouldn't attempt to scan instructions for code blocks that are currently being generated https://bugs.webkit.org/show_bug.cgi?id=83531 <rdar://problem/11215200> Reviewed by Gavin Barraclough. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::stronglyVisitStrongReferences): 2012-04-09 Filip Pizlo <fpizlo@apple.com> Unreviewed, modernize and clean up uses of ARM assembly mnemonics in inline asm blocks. * dfg/DFGOperations.cpp: (JSC): * offlineasm/armv7.rb: 2012-04-09 Patrick Gansterer <paroga@webkit.org> Remove HAVE_STDINT_H https://bugs.webkit.org/show_bug.cgi?id=83434 Reviewed by Kentaro Hara. HAVE_STDINT_H is defined with 1 all the time and we us stdint.h without HAVE(STDINT_H) already. * config.h: 2012-04-08 Filip Pizlo <fpizlo@apple.com> DFG should not load the property storage if it is inline. https://bugs.webkit.org/show_bug.cgi?id=83455 Reviewed by Gavin Barraclough. We had previously decided to have all property storage accesses go through the property storage pointer even if they don't "really" have to, because we were thinking this would help GC barriers somehow. Well, we never ended up doing anything with that. Hence, doing these wasted loads of the property storage pointer when the storage is inline is just a waste of CPU cycles. This change makes the DFG's inline property accesses (GetByOffset and PutByOffset) go directly to the inline property storage if the structure(s) tell us that it's OK. This looks like an across-the-board 1% win. * bytecode/StructureSet.h: (JSC): (JSC::StructureSet::allAreUsingInlinePropertyStorage): (StructureSet): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillStorage): 2012-04-08 Filip Pizlo <fpizlo@apple.com> Command-line jsc's exception handling should be rationalized https://bugs.webkit.org/show_bug.cgi?id=83437 Reviewed by Dan Bernstein. - If an exception is thrown during run() execution, it is now propagated, so that it will terminate program execution unless it is caught. - If program execution terminates with an exception, the exception is now always printed. - When printing the exception, the backtrace is now also printed if one is available. It will only not be available if you use something akin to my favorite line of code, 'throw "error"', since primitives don't have properties and hence we cannot attach a "stack" property to them. * jsc.cpp: (functionRun): (runWithScripts): 2012-04-04 Filip Pizlo <fpizlo@apple.com> Forced OSR exits should lead to recompilation based on count, not rate https://bugs.webkit.org/show_bug.cgi?id=83247 <rdar://problem/10720925> Reviewed by Geoff Garen. Track which OSR exits happen because of inadequate coverage. Count them separately. If the count reaches a threshold, immediately trigger reoptimization. This is in contrast to the recompilation trigger for all other OSR exits. Normally recomp is triggered when the exit rate exceeds a certain ratio. Looks like a slight V8 speedup (sub 1%). * bytecode/CodeBlock.cpp: (JSC::CodeBlock::CodeBlock): * bytecode/CodeBlock.h: (JSC::CodeBlock::forcedOSRExitCounter): (JSC::CodeBlock::addressOfForcedOSRExitCounter): (JSC::CodeBlock::offsetOfForcedOSRExitCounter): (JSC::CodeBlock::shouldReoptimizeNow): (JSC::CodeBlock::shouldReoptimizeFromLoopNow): (CodeBlock): * bytecode/DFGExitProfile.h: (JSC::DFG::exitKindToString): * dfg/DFGOSRExitCompiler.cpp: (JSC::DFG::OSRExitCompiler::handleExitCounts): (DFG): * dfg/DFGOSRExitCompiler.h: (OSRExitCompiler): * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileGetIndexedPropertyStorage): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/Options.cpp: (Options): (JSC::Options::initializeOptions): * runtime/Options.h: (Options): 2012-04-06 Benjamin Poulain <bpoulain@apple.com> Do not abuse ArrayStorage's m_length for testing array consistency https://bugs.webkit.org/show_bug.cgi?id=83403 Reviewed by Geoffrey Garen. Array creation from a list of values is a 3 steps process: -JSArray::tryCreateUninitialized() -JSArray::initializeIndex() for each values -JSArray::completeInitialization() Previously, the attribute m_length was not set to the final size JSArray::tryCreateUninitialized() because it was used to test the array consistency JSArray::initializeIndex(). This caused the initialization loop using JSArray::initializeIndex() maintain two counters: -index of the loop -storage->m_length++ This patch fixes this by using the index of the initialization loop for the indinces of JSArray::initializeIndex(). For testing consistency, the variable m_initializationIndex is introduced if CHECK_ARRAY_CONSISTENCY is defined. The patch also fixes minor unrelated build issue when CHECK_ARRAY_CONSISTENCY is defined. This improves the performance of JSArray creation from literals by 8%. * runtime/JSArray.cpp: (JSC::JSArray::tryFinishCreationUninitialized): (JSC::JSArray::checkConsistency): * runtime/JSArray.h: (ArrayStorage): (JSC::JSArray::initializeIndex): (JSC::JSArray::completeInitialization): 2012-04-06 Jon Lee <jonlee@apple.com> Build fix for Windows bots. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: export missing symbol. 2012-04-06 Geoffrey Garen <ggaren@apple.com> Renamed WeakHeap => WeakSet HandleHeap => HandleSet Reviewed by Sam Weinig. These sets do have internal allocators, but it's confusing to call them heaps because they're sub-objects of an object called "heap". * heap/HandleHeap.cpp: Removed. * heap/HandleHeap.h: Removed. * heap/HandleSet.cpp: Copied from JavaScriptCore/heap/HandleHeap.cpp. * heap/WeakHeap.cpp: Removed. * heap/WeakHeap.h: Removed. * heap/WeakSet.cpp: Copied from JavaScriptCore/heap/WeakHeap.cpp. * heap/WeakSet.h: Copied from JavaScriptCore/heap/WeakHeap.h. Plus global rename using grep. 2012-04-06 Dan Bernstein <mitz@apple.com> <rdar://problem/10912476> HiDPI: Have canvas use a hidpi backing store, but downsample upon access Reviewed by Sam Weinig. * Configurations/FeatureDefines.xcconfig: Added ENABLE_HIGH_DPI_CANVAS. 2012-04-06 Rob Buis <rbuis@rim.com> Fix cast-align warnings in JSC https://bugs.webkit.org/show_bug.cgi?id=80790 Reviewed by George Staikos. * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::computeJumpType): (JSC::ARMv7Assembler::link): * assembler/LinkBuffer.h: (JSC::LinkBuffer::linkCode): * heap/MarkStack.cpp: (JSC::SlotVisitor::copyAndAppend): * runtime/JSArray.cpp: (JSC::JSArray::visitChildren): * wtf/RefCountedArray.h: (WTF::RefCountedArray::Header::payload): 2012-04-06 Darin Adler <darin@apple.com> Streamline strtod and fix some related problems https://bugs.webkit.org/show_bug.cgi?id=82857 Reviewed by Geoffrey Garen. * parser/Lexer.cpp: (JSC::Lexer<>::lex): Use parseDouble. Since we have already scanned the number and we know it has only correct characters, leading spaces, trailing junk, and trailing spaces are not a possibility. No need to add a trailing null character. * runtime/JSGlobalObjectFunctions.cpp: (JSC::parseInt): Changed overflow based 10 case to use parseDouble. No need to allow trailing junk since the code above already allows only numeric digits in the string. This code path is used only in unusual cases, so it's not optimized for 8-bit strings, but easily could be. (JSC::jsStrDecimalLiteral): Removed the allow trailing junk argument to this function template because all the callers are OK with trailing junk. Use the parseDouble function. No need to copy the data into a byte buffer, because parseDouble handles that. (JSC::toDouble): Got rid of the DisallowTrailingJunk argument to the jsStrDecimalLiteral function template. That's OK because this function already checks for trailing junk and handles it appropriately. The old code path was doing it twice. (JSC::parseFloat): Got rid of the AllowTrailingJunk argument to the jsStrDecimalLiteral function template; the template allows junk unconditionally. * runtime/LiteralParser.cpp: (JSC::::Lexer::lexNumber): Use parseDouble. Since we have already scanned the number and we know it has only correct characters, leading spaces, trailing junk, and trailing spaces are not a possibility. No need to add a trailing null character. No need to copy the data into a byte buffer, because parseDouble handles that. We could optimize the UChar case even more because we know all the characters are ASCII, but not doing that at this time. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Updated. 2012-04-06 Patrick Gansterer <paroga@webkit.org> Remove JSC dependency from GregorianDateTime https://bugs.webkit.org/show_bug.cgi?id=83290 Reviewed by Geoffrey Garen. This allows us to move it to WTF later. * runtime/DateConstructor.cpp: (JSC::callDate): * runtime/JSDateMath.h: 2012-04-05 Michael Saboff <msaboff@apple.com> Call Heap::discardAllCompiledCode() in low memory situations https://bugs.webkit.org/show_bug.cgi?id=83335 Reviewed by Geoffrey Garen. Restructured Heap::discardAllCompiledCode() to do the "Is JavaScriptRunning?" check inline so that it can be called directly without this check. * heap/Heap.cpp: (JSC::Heap::discardAllCompiledCode): (JSC::Heap::collectAllGarbage): * heap/Heap.h: Added JS_EXPORT_PRIVATE to discardAllCompiledCode() so it can be called from WebCore. (Heap): * runtime/JSGlobalData.h: Removed unused " void discardAllCompiledCode()" declaration. (JSGlobalData): 2012-04-05 Benjamin Poulain <bpoulain@apple.com> Speed up the conversion from JSValue to String for bulk operations https://bugs.webkit.org/show_bug.cgi?id=83243 Reviewed by Geoffrey Garen. When making operations on primitive types, we loose some time converting values to JSString in order to extract the string. This patch speeds up some basic Array operations by avoiding the creation of intermediary JSString when possible. For the cases where we need to convert a lot of JSValue in a tight loop, an inline conversion is used. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncToString): (JSC::arrayProtoFuncToLocaleString): (JSC::arrayProtoFuncJoin): (JSC::arrayProtoFuncPush): (JSC::arrayProtoFuncSort): * runtime/CommonIdentifiers.h: * runtime/JSArray.cpp: (JSC::JSArray::sort): * runtime/JSString.h: (JSC::JSValue::toUString): (JSC): (JSC::inlineJSValueNotStringtoUString): (JSC::JSValue::toUStringInline): * runtime/JSValue.cpp: (JSC::JSValue::toUStringSlowCase): (JSC): * runtime/JSValue.h: (JSValue): 2012-04-05 Benjamin Poulain <bpoulain@apple.com> Use QuickSort when sorting primitive values by string representation https://bugs.webkit.org/show_bug.cgi?id=83312 Reviewed by Gavin Barraclough. When the value we are sorting are all primitive values, we do not need to ensure a stable sort as two values with equal string representation are indistinguishable from JavaScript. This gives about 16% performance increase when sorting primitive values. * runtime/JSArray.cpp: (JSC::JSArray::sort): 2012-04-05 Oliver Hunt <oliver@apple.com> SIGILL in JavaScriptCore on a Geode processor https://bugs.webkit.org/show_bug.cgi?id=82496 Reviewed by Gavin Barraclough. Don't attempt to use the DFG when SSE2 is not available. * dfg/DFGCapabilities.cpp: (JSC::DFG::canCompileOpcodes): 2012-04-05 Oliver Hunt <oliver@apple.com> Fix 32-bit build. * API/APICast.h: (toJS): 2012-04-05 Oliver Hunt <oliver@apple.com> Replace static_cast with jsCast when casting JSCell subclasses in JSC https://bugs.webkit.org/show_bug.cgi?id=83307 Reviewed by Gavin Barraclough. Replace all usage of static_cast<JSCell subtype*> with jsCast<> in JavaScriptCore. This results in assertions when unsafe casts are performed, but simply leaves a static_cast<> in release builds. * API/APICast.h: (toJS): * API/JSCallbackConstructor.cpp: (JSC::constructJSCallback): * API/JSCallbackFunction.cpp: (JSC::JSCallbackFunction::call): * API/JSCallbackObjectFunctions.h: (JSC::::asCallbackObject): (JSC::::finishCreation): (JSC::::construct): (JSC::::call): * API/JSObjectRef.cpp: (JSObjectGetPrivate): (JSObjectSetPrivate): (JSObjectGetPrivateProperty): (JSObjectSetPrivateProperty): (JSObjectDeletePrivateProperty): * API/JSValueRef.cpp: (JSValueIsObjectOfClass): * API/JSWeakObjectMapRefPrivate.cpp: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::resolve): (JSC::BytecodeGenerator::resolveConstDecl): * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::finishCreation): * dfg/DFGOperations.cpp: * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Executable.h: (JSC::isHostFunction): * runtime/JSActivation.h: (JSC::asActivation): * runtime/JSArray.cpp: (JSC::JSArray::defineOwnProperty): * runtime/JSArray.h: (JSC::asArray): * runtime/JSBoundFunction.cpp: (JSC::boundFunctionCall): (JSC::boundFunctionConstruct): * runtime/JSByteArray.h: (JSC::asByteArray): * runtime/JSCell.cpp: (JSC::JSCell::toObject): * runtime/JSCell.h: (JSC::jsCast): * runtime/JSGlobalObject.h: (JSC::asGlobalObject): * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncEval): * runtime/JSObject.cpp: (JSC::JSObject::setPrototypeWithCycleCheck): (JSC::JSObject::allowsAccessFrom): (JSC::JSObject::toThisObject): (JSC::JSObject::unwrappedObject): * runtime/JSObject.h: (JSC::asObject): * runtime/JSPropertyNameIterator.h: (JSC::Register::propertyNameIterator): * runtime/JSString.h: (JSC::asString): (JSC::JSValue::toString): * runtime/StringPrototype.cpp: (JSC::stringProtoFuncSubstr): 2012-04-05 Benjamin Poulain <bpoulain@apple.com> Make something faster than JSStringBuilder for joining an array of JSValue https://bugs.webkit.org/show_bug.cgi?id=83180 Reviewed by Geoffrey Garen. This patch add the class JSStringJoiner optimized for join() operations. This class makes stricter constraints than JSStringBuilder in order avoid memory allocations. In the best case, the class allocate memory only twice: -Allocate an array to keep a list of UString to join. -Allocate the final string. We also avoid the conversion from 8bits strings to 16bits strings since they are costly and unlikly to help for subsequent calls. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncToLocaleString): (JSC::arrayProtoFuncJoin): * runtime/JSStringJoiner.cpp: Added. (JSC): (JSC::appendStringToData): (JSC::joinStrings): (JSC::JSStringJoiner::build): * runtime/JSStringJoiner.h: Added. (JSC): (JSStringJoiner): (JSC::JSStringJoiner::JSStringJoiner): (JSC::JSStringJoiner::append): 2012-04-05 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=77293 [Un]Reserve 'let' Rubber stamped by Oliver Hunt. Revert r106198. This does break the web - e.g. https://bvi.bnc.ca/index/bnc/indexen.html If we're going to reserve let, we're going to have to do so in a more circumspect fashion. * parser/Keywords.table: 2012-04-05 Michael Saboff <msaboff@apple.com> Rolling out http://trac.webkit.org/changeset/113262. Original code was fine. Rubber-stamped by Oliver Hunt. * assembler/MacroAssembler.h: (JSC::MacroAssembler::additionBlindedConstant): 2012-04-05 Patrick Gansterer <paroga@webkit.org> [WinCE] Remove unnecessary function decleration https://bugs.webkit.org/show_bug.cgi?id=83155 Reviewed by Kentaro Hara. * runtime/JSDateMath.cpp: 2012-04-04 Patrick Gansterer <paroga@webkit.org> Add WTF::getCurrentLocalTime() https://bugs.webkit.org/show_bug.cgi?id=83164 Reviewed by Alexey Proskuryakov. Replace the calls to WTF::getLocalTime() with time(0) with the new function. This allows us to use Win32 API on windows to get the same result in a next step. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/DateConstructor.cpp: (JSC::callDate): 2012-04-04 Oliver Hunt <oliver@apple.com> Parser fails to revert some state after parsing expression and object literals. https://bugs.webkit.org/show_bug.cgi?id=83236 Reviewed by Gavin Barraclough. Reset left hand side counter after parsing the literals. * parser/Parser.cpp: (JSC::::parseObjectLiteral): (JSC::::parseStrictObjectLiteral): (JSC::::parseArrayLiteral): 2012-04-04 Filip Pizlo <fpizlo@apple.com> DFG InstanceOf should not uselessly speculate cell https://bugs.webkit.org/show_bug.cgi?id=83234 Reviewed by Oliver Hunt. If InstanceOf is the only user of its child then don't speculate cell, since the not-cell case is super easy to handle. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileInstanceOf): 2012-04-04 Michael Saboff <msaboff@apple.com> Fixed minor error: "& 3" should be "& 2". Rubber-stamped by Oliver Hunt. * assembler/MacroAssembler.h: (JSC::MacroAssembler::additionBlindedConstant): 2012-04-04 Michael Saboff <msaboff@apple.com> Constant Blinding for add/sub immediate crashes in ArmV7 when dest is SP https://bugs.webkit.org/show_bug.cgi?id=83191 Reviewed by Oliver Hunt. Make are that blinded constant pairs are similarly aligned to the original immediate values so that instructions that expect that alignment work correctly. One example is ARMv7 add/sub imm to SP. * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::add): Added ASSERT that immediate is word aligned. (JSC::ARMv7Assembler::sub): Added ASSERT that immediate is word aligned. (JSC::ARMv7Assembler::sub_S): Added ASSERT that immediate is word aligned. * assembler/MacroAssembler.h: (JSC::MacroAssembler::additionBlindedConstant): 2012-04-04 Filip Pizlo <fpizlo@apple.com> DFG should short-circuit Branch(LogicalNot(...)) https://bugs.webkit.org/show_bug.cgi?id=83181 Reviewed by Geoff Garen. Slight (sub 1%) speed-up on V8. * dfg/DFGFixupPhase.cpp: (JSC::DFG::FixupPhase::fixupNode): 2012-04-04 Geoffrey Garen <ggaren@apple.com> [Qt] REGRESSION(r113141): All tests assert on 32 bit debug mode https://bugs.webkit.org/show_bug.cgi?id=83139 Reviewed by Sam Weinig. * heap/PassWeak.h: (JSC::::get): 32-bit JSValue treats JSValue(nullptr).asCell() as an error, so work around that here. (Long-term, we should make 32-bit and 64-bit agree on the right behavior.) 2012-04-03 Geoffrey Garen <ggaren@apple.com> Updated JSC expected test results to reflect recent bug fixes <disapproving look>. Reviewed by Sam Weinig. * tests/mozilla/expected.html: 2012-03-29 Geoffrey Garen <ggaren@apple.com> First step toward incremental Weak<T> finalization https://bugs.webkit.org/show_bug.cgi?id=82670 Reviewed by Filip Pizlo. This patch implements a Weak<T> heap that is compatible with incremental finalization, while making as few behavior changes as possible. The behavior changes it makes are: (*) Weak<T>'s raw JSValue no longer reverts to JSValue() automatically -- instead, a separate flag indicates that the JSValue is no longer valid. (This is required so that the JSValue can be preserved for later finalization.) Objects dealing with WeakImpls directly must change to check the flag. (*) Weak<T> is no longer a subclass of Handle<T>. (*) DOM GC performance is different -- 9% faster in the geometric mean, but 15% slower in one specific case: gc-dom1.html: 6% faster gc-dom2.html: 23% faster gc-dom3.html: 17% faster gc-dom4.html: 15% *slower* The key features of this new heap are: (*) Each block knows its own state, independent of any other blocks. (*) Each block caches its own sweep result. (*) The heap visits dead Weak<T>s at the end of GC. (It doesn't mark them yet, since that would be a behavior change.) * API/JSCallbackObject.cpp: (JSC::JSCallbackObjectData::finalize): * API/JSCallbackObjectFunctions.h: (JSC::::init): Updated to use the new WeakHeap API. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: Paid the build system tax since I added some new files. * heap/Handle.h: Made WeakBlock a friend and exposed slot() as public, so we can keep passing a Handle<T> to finalizers, to avoid more surface area change in this patch. A follow-up patch should change the type we pass to finalizers. * heap/HandleHeap.cpp: (JSC): (JSC::HandleHeap::writeBarrier): (JSC::HandleHeap::isLiveNode): * heap/HandleHeap.h: (JSC): (HandleHeap): (Node): (JSC::HandleHeap::Node::Node): Removed all code related to Weak<T>, since we have a separate WeakHeap now. * heap/Heap.cpp: (JSC::Heap::Heap): Removed m_extraCost because extra cost is accounted for through our watermark now. Removed m_waterMark because it was unused. (JSC::Heap::destroy): Updated for addition of WeakHeap. (JSC::Heap::reportExtraMemoryCostSlowCase): Changed from using its own variable to participating in the watermark strategy. I wanted to standardize WeakHeap and all other Heap clients on this strategy, to make sure it's accurate. (JSC::Heap::markRoots): Updated for addition of WeakHeap. Added WeakHeap dead visit pass, as explained above. (JSC::Heap::collect): (JSC::Heap::resetAllocators): Updated for addition of WeakHeap. (JSC::Heap::addFinalizer): (JSC::Heap::FinalizerOwner::finalize): Updated for new Weak<T> API. * heap/Heap.h: (JSC::Heap::weakHeap): (Heap): (JSC::Heap::addToWaterMark): Added a way to participate in the watermarking strategy, since this is the best way for WeakHeap to report its memory cost. (I plan to update this in a follow-up patch to make it more accurate, but for now it is not less accurate than it used to be.) * heap/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::resetAllocators): * heap/MarkedSpace.h: (MarkedSpace): (JSC::MarkedSpace::addToWaterMark): (JSC::MarkedSpace::didConsumeFreeList): Removed m_nurseryWaterMark because it was unused, and I didn't want to update WeakHeap to keep an usused variable working. Added API for above. * heap/PassWeak.h: (JSC): (WeakImplAccessor): (PassWeak): (JSC::::operator): (JSC::::get): (JSC::::was): (JSC::::PassWeak): (JSC::::~PassWeak): (JSC::UnspecifiedBoolType): (JSC::::leakImpl): (JSC::adoptWeak): * heap/Strong.h: (JSC::Strong::operator!): (Strong): (JSC::Strong::operator UnspecifiedBoolType*): (JSC::Strong::get): * heap/Weak.h: (Weak): (JSC::::Weak): (JSC): (JSC::::isHashTableDeletedValue): (JSC::::~Weak): (JSC::::swap): (JSC::=): (JSC::::operator): (JSC::UnspecifiedBoolType): (JSC::::release): (JSC::::clear): (JSC::::hashTableDeletedValue): Lots of code changes here, but they boil down to two things: (*) Allocate WeakImpls from the WeakHeap instead of Handles from the HandleHeap. (*) Explicitly check WeakImpl::state() for non-liveness before returning a value (explained above). These files implement the new Weak<T> heap behavior described above: * heap/WeakBlock.cpp: Added. * heap/WeakBlock.h: Added. * heap/WeakHandleOwner.cpp: Added. * heap/WeakHandleOwner.h: Added. * heap/WeakHeap.cpp: Added. * heap/WeakHeap.h: Added. * heap/WeakImpl.h: Added. One interesting difference from the old heap is that we don't allow clients to overwrite a WeakImpl after allocating it, and we don't recycle WeakImpls prior to garbage collection. This is required for lazy finalization, but it will also help us esablish a useful invariant in the future: allocating a WeakImpl will be a binding contract to run a finalizer at some point in the future, even if the WeakImpl is later deallocated. * jit/JITStubs.cpp: (JSC::JITThunks::hostFunctionStub): Check the Weak<T> for ! instead of its JSValue, since that's our API contract now, and the JSValue might be stale. * runtime/JSCell.h: (JSC::jsCast): Allow casting NULL pointers because it's useful and harmless. * runtime/Structure.cpp: (JSC::StructureTransitionTable::add): I can't remember why I did this. * runtime/StructureTransitionTable.h: * runtime/WeakGCMap.h: I had to update these classes because they allocate and deallocate weak pointers manually. They should probably stop doing that. 2012-04-03 Keishi Hattori <keishi@webkit.org> Disable ENABLE_DATALIST for now https://bugs.webkit.org/show_bug.cgi?id=82871 Reviewed by Kent Tamura. * Configurations/FeatureDefines.xcconfig: Disabled ENABLE_DATALIST. 2012-04-02 Filip Pizlo <fpizlo@apple.com> jsr/sret should be removed https://bugs.webkit.org/show_bug.cgi?id=82986 <rdar://problem/11017015> Reviewed by Sam Weinig and Geoff Garen. Replaces jsr/sret with finally block inlining. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): * bytecode/Opcode.h: (JSC): (JSC::padOpcodeName): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::pushFinallyContext): (JSC::BytecodeGenerator::emitComplexJumpScopes): (JSC): * bytecompiler/BytecodeGenerator.h: (FinallyContext): (BytecodeGenerator): * bytecompiler/NodesCodegen.cpp: (JSC::TryNode::emitBytecode): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompile): * jit/JIT.h: (JIT): * jit/JITOpcodes.cpp: (JSC): * jit/JITOpcodes32_64.cpp: (JSC): * llint/LowLevelInterpreter32_64.asm: * llint/LowLevelInterpreter64.asm: 2012-04-03 Mark Rowe <mrowe@apple.com> Make it possible to install the JavaScriptCore test tools. Part of <rdar://problem/11158607>. Reviewed by Filip Pizlo. * JavaScriptCore.xcodeproj/project.pbxproj: Introduce an aggregate target named Test Tools that builds testapi, minidom and testRegExp. Switch All from depending on those targets individually to depending on the new aggregate target. 2012-04-03 Filip Pizlo <fpizlo@apple.com> Offlineasm ARM backend has a very convoluted way of saying it wants to emit a three-operand multiply instruction https://bugs.webkit.org/show_bug.cgi?id=83100 Reviewed by Darin Adler. Changed the "muli"/"mulp" case to call emitArmV7() since that helper method was already smart enough to do the Right Thing for multiply. * offlineasm/armv7.rb: 2012-04-03 Filip Pizlo <fpizlo@apple.com> Offlineasm ARM backend uses the wrong mnemonic for multiply https://bugs.webkit.org/show_bug.cgi?id=83098 <rdar://problem/11168744> Reviewed by Gavin Barraclough. Use "mul" instead of "muls" since we're passing three operands, not two. * offlineasm/armv7.rb: 2012-04-03 Gavin Barraclough <barraclough@apple.com> Linux crashes during boot https://bugs.webkit.org/show_bug.cgi?id=83096 Reviewed by Filip Pizlo. The bug here is that we add empty JSValues to the sparse map, and then set them - but a GC may occur before doing so (due to a call to reportExtraMemory cost). We may want to consider making it safe to mark empty JSValues, but the simple & contained fix to this specific bug is to just initialize these values to something other than JSValue(). * runtime/JSArray.cpp: (JSC::SparseArrayValueMap::add): - Initialize sparse map entries. 2012-04-02 Oliver Hunt <oliver@apple.com> Incorrect liveness information when inlining https://bugs.webkit.org/show_bug.cgi?id=82985 Reviewed by Filip Pizlo. Don't remap register numbers that have already been remapped. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleInlining): 2012-04-02 Filip Pizlo <fpizlo@apple.com> Activation tear-off neglects to copy the callee and scope chain, leading to crashes if we try to create an arguments object from the activation https://bugs.webkit.org/show_bug.cgi?id=82947 <rdar://problem/11058598> Reviewed by Gavin Barraclough. We now copy the entire call frame header just to be sure. This is mostly perf-netural, except for a 3.7% slow-down in V8/earley. * runtime/JSActivation.cpp: (JSC::JSActivation::visitChildren): * runtime/JSActivation.h: (JSC::JSActivation::tearOff): 2012-04-02 Daniel Bates <dbates@webkit.org> Remove Source/JavaScriptCore/wtf and its empty subdirectories Rubber-stamped by Eric Seidel. Following the move of WTF from Source/JavaScriptCore/wtf to Source/WTF (https://bugs.webkit.org/show_bug.cgi?id=75673), remove directory Source/JavaScriptCore/wtf and its empty subdirectories. * wtf: Removed. * wtf/android: Removed. * wtf/blackberry: Removed. * wtf/chromium: Removed. * wtf/dtoa: Removed. * wtf/efl: Removed. * wtf/gobject: Removed. * wtf/gtk: Removed. * wtf/mac: Removed. * wtf/qt: Removed. * wtf/qt/compat: Removed. * wtf/tests: Removed. * wtf/text: Removed. * wtf/threads: Removed. * wtf/threads/win: Removed. * wtf/unicode: Removed. * wtf/unicode/glib: Removed. * wtf/unicode/icu: Removed. * wtf/unicode/qt4: Removed. * wtf/unicode/wince: Removed. * wtf/url: Removed. * wtf/url/api: Removed. * wtf/url/src: Removed. * wtf/win: Removed. * wtf/wince: Removed. * wtf/wx: Removed. 2012-04-02 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck issues. * GNUmakefile.list.am: Add missing file. 2012-04-01 Darin Adler <darin@apple.com> Fix incorrect path for libWTF.a in Mac project file. * JavaScriptCore.xcodeproj/project.pbxproj: Removed the "../Release" prefix that would cause other configurations to try to link with the "Release" version of libWTF.a instead of the correct version. 2012-03-29 Filip Pizlo <fpizlo@apple.com> DFG should optimize a==b for a being an object and b being either an object or null/undefined, and vice versa https://bugs.webkit.org/show_bug.cgi?id=82656 Reviewed by Oliver Hunt. Implements additional object equality optimizations for the case that one operand is predicted to be an easily speculated object (like FinalObject or Array) and the other is either an easily speculated object or Other, i.e. Null or Undefined. 2-5% speed-up on V8/raytrace, leading to a sub-1% progression on V8. I also took the opportunity to clean up the control flow for the speculation decisions in the various Compare opcodes. And to fix a build bug in SamplingTool. And to remove debug cruft I stupidly committed in my last patch. * bytecode/SamplingTool.h: (SamplingRegion): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleBranch): (JSC::DFG::SpeculativeJIT::compare): * dfg/DFGSpeculativeJIT.h: (SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranch): (JSC::DFG::SpeculativeJIT::compileObjectToObjectOrOtherEquality): (DFG): (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectToObjectOrOtherEquality): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranch): (JSC::DFG::SpeculativeJIT::compileObjectToObjectOrOtherEquality): (DFG): (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectToObjectOrOtherEquality): 2012-03-30 David Barr <davidbarr@chromium.org> Split up top-level .gitignore and .gitattributes https://bugs.webkit.org/show_bug.cgi?id=82687 Reviewed by Tor Arne Vestbø. * JavaScriptCore.gyp/.gitignore: Added. 2012-03-30 Steve Falkenburg <sfalken@apple.com> Windows (make based) build fix. * JavaScriptCore.vcproj/JavaScriptCore.make: Copy WTF header files into a place where JavaScriptCore build can see them. 2012-03-30 Keishi Hattori <keishi@webkit.org> Change ENABLE_INPUT_COLOR to ENABLE_INPUT_TYPE_COLOR and enable it for chromium https://bugs.webkit.org/show_bug.cgi?id=80972 Reviewed by Kent Tamura. * Configurations/FeatureDefines.xcconfig: 2012-03-29 Mark Hahnenberg <mhahnenberg@apple.com> Refactor recompileAllJSFunctions() to be less expensive https://bugs.webkit.org/show_bug.cgi?id=80330 Reviewed by Filip Pizlo. This change is performance neutral on the JS benchmarks we track. It's mostly to improve page load performance, which currently does at least a couple full GCs per navigation. * heap/Heap.cpp: (JSC::Heap::discardAllCompiledCode): Rename recompileAllJSFunctions to discardAllCompiledCode because the function doesn't actually recompile anything (and never did); it simply throws code away for it to be recompiled later if we determine we should do so. (JSC): (JSC::Heap::collectAllGarbage): (JSC::Heap::addFunctionExecutable): Adds a newly created FunctionExecutable to the Heap's list. (JSC::Heap::removeFunctionExecutable): Removes the specified FunctionExecutable from the Heap's list. * heap/Heap.h: (JSC): (Heap): * runtime/Executable.cpp: Added next and prev fields to FunctionExecutables so that they can be used in DoublyLinkedLists. (JSC::FunctionExecutable::FunctionExecutable): (JSC::FunctionExecutable::finalize): Removes the FunctionExecutable from the Heap's list. * runtime/Executable.h: (FunctionExecutable): (JSC::FunctionExecutable::create): Adds the FunctionExecutable to the Heap's list. * runtime/JSGlobalData.cpp: Remove recompileAllJSFunctions, as it's the Heap's job to own and manage the list of FunctionExecutables. * runtime/JSGlobalData.h: (JSGlobalData): * runtime/JSGlobalObject.cpp: (JSC::DynamicGlobalObjectScope::DynamicGlobalObjectScope): Use the new discardAllCompiledCode. 2012-03-29 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix for non-x86 platforms. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileSoftModulo): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::callOperation): * jit/JITArithmetic32_64.cpp: (JSC::JIT::emitSlow_op_mod): 2012-03-29 Gavin Barraclough <barraclough@apple.com> Windows build fix p2. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-03-29 Gavin Barraclough <barraclough@apple.com> Windows build fix p1. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-03-29 Gavin Barraclough <barraclough@apple.com> Template the Yarr::Interpreter on the character type https://bugs.webkit.org/show_bug.cgi?id=82637 Reviewed by Sam Weinig. We should be able to call to the interpreter after having already checked the character type, without having to re-package the character pointer back up into a string! * runtime/RegExp.cpp: (JSC::RegExp::match): (JSC::RegExp::matchCompareWithInterpreter): - Don't pass length. * yarr/Yarr.h: - moved function declarations to YarrInterpreter.h. * yarr/YarrInterpreter.cpp: (Yarr): (Interpreter): (JSC::Yarr::Interpreter::InputStream::InputStream): (InputStream): (JSC::Yarr::Interpreter::Interpreter): (JSC::Yarr::interpret): - templated Interpreter class on CharType. * yarr/YarrInterpreter.h: (Yarr): - added function declarations. 2012-03-29 David Kilzer <ddkilzer@apple.com> Don't use a flattened framework path when building on OS X Reviewed by Mark Rowe. * Configurations/ToolExecutable.xcconfig: Use REAL_PLATFORM_NAME to select different INSTALL_PATH values. 2012-03-29 Kevin Ollivier <kevino@theolliviers.com> [wx] Unreviewed build fix, add Win-specific sources the wx port needs after WTF move. * wscript: 2012-03-29 Andy Estes <aestes@apple.com> Remove an unused variable that breaks the build with newer versions of clang. Rubber stamped by Gavin Barraclough. * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::backtrackCharacterClassNonGreedy): 2012-03-29 Caio Marcelo de Oliveira Filho <caio.oliveira@openbossa.org> HashMap<>::add should return a more descriptive object https://bugs.webkit.org/show_bug.cgi?id=71063 Reviewed by Ryosuke Niwa. Update code to use AddResult instead of a pair. Note that since WeakGCMap wraps the iterator type, there's a need for its own AddResult type -- instantiated from HashTableAddResult template class. * API/JSCallbackObject.h: (JSC::JSCallbackObjectData::JSPrivatePropertyMap::setPrivateProperty): * API/JSClassRef.cpp: (OpaqueJSClass::contextData): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::addVar): (JSC::BytecodeGenerator::addGlobalVar): (JSC::BytecodeGenerator::addConstant): (JSC::BytecodeGenerator::addConstantValue): (JSC::BytecodeGenerator::emitLoad): (JSC::BytecodeGenerator::addStringConstant): (JSC::BytecodeGenerator::emitLazyNewFunction): * bytecompiler/NodesCodegen.cpp: (JSC::PropertyListNode::emitBytecode): * debugger/Debugger.cpp: * dfg/DFGAssemblyHelpers.cpp: (JSC::DFG::AssemblyHelpers::decodedCodeMapFor): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::cellConstant): (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): * jit/JITStubs.cpp: (JSC::JITThunks::ctiStub): (JSC::JITThunks::hostFunctionStub): * parser/Parser.cpp: (JSC::::parseStrictObjectLiteral): * parser/Parser.h: (JSC::Scope::declareParameter): * runtime/Identifier.cpp: (JSC::Identifier::add): (JSC::Identifier::add8): (JSC::Identifier::addSlowCase): * runtime/Identifier.h: (JSC::Identifier::add): (JSC::IdentifierTable::add): * runtime/JSArray.cpp: (JSC::SparseArrayValueMap::add): (JSC::SparseArrayValueMap::put): (JSC::SparseArrayValueMap::putDirect): (JSC::JSArray::enterDictionaryMode): (JSC::JSArray::defineOwnNumericProperty): * runtime/JSArray.h: (SparseArrayValueMap): * runtime/PropertyNameArray.cpp: (JSC::PropertyNameArray::add): * runtime/StringRecursionChecker.h: (JSC::StringRecursionChecker::performCheck): * runtime/Structure.cpp: (JSC::StructureTransitionTable::add): * runtime/WeakGCMap.h: (WeakGCMap): (JSC::WeakGCMap::add): (JSC::WeakGCMap::set): * tools/ProfileTreeNode.h: (JSC::ProfileTreeNode::sampleChild): 2012-03-29 Patrick Gansterer <paroga@webkit.org> Build fix for !ENABLE(YARR_JIT) after r112454. * runtime/RegExp.cpp: (JSC::RegExp::invalidateCode): 2012-03-28 Filip Pizlo <fpizlo@apple.com> DFG object equality speculations should be simplified https://bugs.webkit.org/show_bug.cgi?id=82557 Reviewed by Gavin Barraclough. * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateFinalObject): (JSC::DFG::Node::shouldSpeculateArray): 2012-03-28 David Kilzer <ddkilzer@apple.com> minidom configurations should be based on ToolExecutable.xcconfig <http://webkit.org/b/82513> Reviewed by Mark Rowe. Note that this patch changes minidom from being installed in /usr/local/bin to JavaScriptCore.framework/Resources. * Configurations/ToolExecutable.xcconfig: Add semi-colon. * JavaScriptCore.xcodeproj/project.pbxproj: Base minidom configurations on ToolExecutable.xcconfig. Remove redundant PRODUCT_NAME and SKIP_INSTALL variables. 2012-03-28 Gavin Barraclough <barraclough@apple.com> Build fix - some compiles generating NORETURN related warnings. * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::setSubpatternStart): (JSC::Yarr::YarrGenerator::setSubpatternEnd): (JSC::Yarr::YarrGenerator::clearSubpatternStart): 2012-03-28 Kevin Ollivier <kevino@theolliviers.com> [wx] Unreviewed. Build fix, move WTF back into JSCore target until issues with JSCore not linking in all WTF symbols are resolved. * wscript: 2012-03-28 Gavin Barraclough <barraclough@apple.com> Yarr: if we're not using the output array, don't populate it! https://bugs.webkit.org/show_bug.cgi?id=82519 Reviewed by Sam Weinig. * runtime/RegExp.cpp: (JSC): - Missed review comment! - didn't fully remove RegExpRepresentation. 2012-03-28 Gavin Barraclough <barraclough@apple.com> Yarr: if we're not using the output array, don't populate it! https://bugs.webkit.org/show_bug.cgi?id=82519 Reviewed by Sam Weinig. Add a new variant of the match method to RegExp that returns a MatchResult, and modify YarrJIT to be able to compile code that doesn't use an output vector. This is a 3% progression on v8-regexp. * JavaScriptCore.xcodeproj/project.pbxproj: - Moved MatchResult into its own header. * assembler/AbstractMacroAssembler.h: - Added missing include. * runtime/MatchResult.h: Added. (MatchResult::MatchResult): (MatchResult): (MatchResult::failed): (MatchResult::operator bool): (MatchResult::empty): - Moved MatchResult into its own header. * runtime/RegExp.cpp: (JSC::RegExp::compile): (JSC::RegExp::compileIfNecessary): (JSC::RegExp::match): - Changed due to execute & representation changes. (JSC::RegExp::compileMatchOnly): (JSC::RegExp::compileIfNecessaryMatchOnly): - Added helper to compile MatchOnly code. (JSC::RegExp::invalidateCode): (JSC::RegExp::matchCompareWithInterpreter): (JSC::RegExp::printTraceData): - Changed due representation changes. * runtime/RegExp.h: (RegExp): (JSC::RegExp::hasCode): - Made YarrCodeBlock a member. * runtime/RegExpConstructor.h: (RegExpConstructor): (JSC::RegExpConstructor::performMatch): - Added no-ovector form. * runtime/RegExpMatchesArray.cpp: (JSC::RegExpMatchesArray::reifyAllProperties): - Match now takes a reference to ovector, not a pointer. * runtime/RegExpObject.h: (JSC): - Moved MatchResult into its own header. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncSplit): - Match now takes a reference to ovector, not a pointer. * testRegExp.cpp: (testOneRegExp): - Match now takes a reference to ovector, not a pointer. * yarr/YarrJIT.cpp: (Yarr): (YarrGenerator): (JSC::Yarr::YarrGenerator::initCallFrame): (JSC::Yarr::YarrGenerator::removeCallFrame): (JSC::Yarr::YarrGenerator::setSubpatternStart): (JSC::Yarr::YarrGenerator::setSubpatternEnd): (JSC::Yarr::YarrGenerator::clearSubpatternStart): (JSC::Yarr::YarrGenerator::setMatchStart): (JSC::Yarr::YarrGenerator::getMatchStart): - Added helper functions to intermediate access to output. (JSC::Yarr::YarrGenerator::generateDotStarEnclosure): (JSC::Yarr::YarrGenerator::generate): (JSC::Yarr::YarrGenerator::backtrack): (JSC::Yarr::YarrGenerator::generateEnter): (JSC::Yarr::YarrGenerator::compile): - Changed to use the new helpers, only generate subpatterns if IncludeSubpatterns. (JSC::Yarr::jitCompile): - Needs to template of MatchOnly or IncludeSubpatterns. * yarr/YarrJIT.h: (YarrCodeBlock): (JSC::Yarr::YarrCodeBlock::set8BitCode): (JSC::Yarr::YarrCodeBlock::set16BitCode): (JSC::Yarr::YarrCodeBlock::has8BitCodeMatchOnly): (JSC::Yarr::YarrCodeBlock::has16BitCodeMatchOnly): (JSC::Yarr::YarrCodeBlock::set8BitCodeMatchOnly): (JSC::Yarr::YarrCodeBlock::set16BitCodeMatchOnly): (JSC::Yarr::YarrCodeBlock::execute): (JSC::Yarr::YarrCodeBlock::clear): - Added a second set of CodeRefs, so that we can compile RexExps with/without subpattern matching. 2012-03-27 Filip Pizlo <fpizlo@apple.com> DFG OSR exit should not generate an exit for variables of inlinees if the inlinees are not in scope https://bugs.webkit.org/show_bug.cgi?id=82312 Reviewed by Oliver Hunt. * bytecode/CodeBlock.h: (JSC::baselineCodeBlockForInlineCallFrame): (JSC): (JSC::baselineCodeBlockForOriginAndBaselineCodeBlock): * dfg/DFGOSRExit.cpp: (JSC::DFG::computeNumVariablesForCodeOrigin): (DFG): (JSC::DFG::OSRExit::OSRExit): 2012-03-27 Matt Lilek <mrl@apple.com> Stop compiling Interpreter.cpp with -fno-var-tracking https://bugs.webkit.org/show_bug.cgi?id=82299 Reviewed by Anders Carlsson. * JavaScriptCore.xcodeproj/project.pbxproj: 2012-03-27 Pratik Solanki <psolanki@apple.com> Compiler warning when JIT is not enabled https://bugs.webkit.org/show_bug.cgi?id=82352 Reviewed by Filip Pizlo. * runtime/JSFunction.cpp: (JSC::JSFunction::create): 2012-03-26 Thouraya ANDOLSI <thouraya.andolsi@st.com> Unaligned userspace access for SH4 platforms https://bugs.webkit.org/show_bug.cgi?id=79104 Reviewed by Gavin Barraclough. * assembler/AbstractMacroAssembler.h: (Jump): (JSC::AbstractMacroAssembler::Jump::Jump): (JSC::AbstractMacroAssembler::Jump::link): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::load16Unaligned): (JSC::MacroAssemblerSH4::load32WithUnalignedHalfWords): (JSC::MacroAssemblerSH4::branchDouble): (JSC::MacroAssemblerSH4::branchTrue): (JSC::MacroAssemblerSH4::branchFalse): * assembler/SH4Assembler.h: (JSC::SH4Assembler::extraInstrForBranch): (SH4Assembler): (JSC::SH4Assembler::bra): (JSC::SH4Assembler::linkJump): * jit/JIT.h: (JIT): * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::generatePatternCharacterOnce): 2012-03-26 Ryosuke Niwa <rniwa@webkit.org> cssText should use shorthand notations https://bugs.webkit.org/show_bug.cgi?id=81737 Reviewed by Enrica Casucci. Export symbols of BitVector on Windows. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-03-26 Filip Pizlo <fpizlo@apple.com> DFG should assert that argument value recoveries can only be AlreadyInRegisterFile or Constant https://bugs.webkit.org/show_bug.cgi?id=82249 Reviewed by Michael Saboff. Made the assertions that the DFG makes for argument value recoveries match what Arguments expects. * bytecode/ValueRecovery.h: (JSC::ValueRecovery::isConstant): (ValueRecovery): (JSC::ValueRecovery::isAlreadyInRegisterFile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-03-26 Dan Bernstein <mitz@apple.com> Tried to fix the Windows build. * yarr/YarrPattern.cpp: (JSC::Yarr::CharacterClassConstructor::putRange): 2012-03-26 Gavin Barraclough <barraclough@apple.com> Unreviewed - speculative Windows build fix. * yarr/YarrCanonicalizeUCS2.h: (JSC::Yarr::getCanonicalPair): 2012-03-26 Dan Bernstein <mitz@apple.com> Fixed builds with assertions disabled. * yarr/YarrCanonicalizeUCS2.h: (JSC::Yarr::areCanonicallyEquivalent): 2012-03-26 Gavin Barraclough <barraclough@apple.com> Unreviewed - errk! - accidentally the whole pbxproj. * JavaScriptCore.xcodeproj/project.pbxproj: 2012-03-25 Gavin Barraclough <barraclough@apple.com> Greek sigma is handled wrong in case independent regexp. https://bugs.webkit.org/show_bug.cgi?id=82063 Reviewed by Oliver Hunt. The bug here is that we assume that any given codepoint has at most one additional value it should match under a case insensitive match, and that the pair of codepoints that match (if a codepoint does not only match itself) can be determined by calling toUpper/toLower on the given codepoint). Life is not that simple. Instead, pre-calculate a set of tables mapping from a UCS2 codepoint to the set of characters it may match, under the ES5.1 case-insensitive matching rules. Since unicode is fairly regular we can pack this table quite nicely, and get it down to 364 entries. This means we can use a simple binary search to find an entry in typically eight compares. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * yarr/yarr.pri: - Added new files to build systems. * yarr/YarrCanonicalizeUCS2.cpp: Added. - New - autogenerated, UCS2 canonicalized comparison tables. * yarr/YarrCanonicalizeUCS2.h: Added. (JSC::Yarr::rangeInfoFor): - Look up the canonicalization info for a UCS2 character. (JSC::Yarr::getCanonicalPair): - For a UCS2 character with a single equivalent value, look it up. (JSC::Yarr::isCanonicallyUnique): - Returns true if no other UCS2 code points are canonically equal. (JSC::Yarr::areCanonicallyEquivalent): - Compare two values, under canonicalization rules. * yarr/YarrCanonicalizeUCS2.js: Added. - script used to generate YarrCanonicalizeUCS2.cpp. * yarr/YarrInterpreter.cpp: (JSC::Yarr::Interpreter::tryConsumeBackReference): - Use isCanonicallyUnique, rather than Unicode toUpper/toLower. * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::jumpIfCharNotEquals): (JSC::Yarr::YarrGenerator::generatePatternCharacterOnce): (JSC::Yarr::YarrGenerator::generatePatternCharacterFixed): - Use isCanonicallyUnique, rather than Unicode toUpper/toLower. * yarr/YarrPattern.cpp: (JSC::Yarr::CharacterClassConstructor::putChar): - Updated to determine canonical equivalents correctly. (JSC::Yarr::CharacterClassConstructor::putUnicodeIgnoreCase): - Added, used to put a non-ascii, non-unique character in a case-insensitive match. (JSC::Yarr::CharacterClassConstructor::putRange): - Updated to determine canonical equivalents correctly. (JSC::Yarr::YarrPatternConstructor::atomPatternCharacter): - Changed to call putUnicodeIgnoreCase, instead of putChar, avoid a double lookup of rangeInfo. 2012-03-26 Kevin Ollivier <kevino@theolliviers.com> [wx] Unreviewed build fix. Add the build outputs dir to the list of build dirs, so we make sure it finds the API headers on all platforms. * wscript: 2012-03-26 Patrick Gansterer <paroga@webkit.org> Build fix for WinCE after r112039. * interpreter/Register.h: (Register): Removed inline keyword from decleration since there is an ALWAYS_INLINE at the definition anyway. 2012-03-26 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck. * GNUmakefile.list.am: Add missing files. 2012-03-25 Kevin Ollivier <kevino@theolliviers.com> [wx] Unreviewed build fix. Move WTF to its own static lib build. * wscript: 2012-03-25 Filip Pizlo <fpizlo@apple.com> DFG int-to-double conversion should be revealed to CSE https://bugs.webkit.org/show_bug.cgi?id=82135 Reviewed by Oliver Hunt. This introduces the notion of an Int32ToDouble node, which is injected into the graph anytime we know that we have a double use of a node that was predicted integer. The Int32ToDouble simplifies double speculation on integers by skipping the path that would unbox doubles, if we know that the value is already proven to be an integer. It allows integer to double conversions to be subjected to common subexpression elimination (CSE) by allowing the CSE phase to see where these conversions are occurring. Finally, it allows us to see when a constant is being used as both a double and an integer. This is a bit odd, since it means that sometimes a double use of a constant will not refer directly to the constant. This should not cause problems, for now, but it may require some canonizalization in the future if we want to support strength reductions of double operations based on constants. To allow injection of nodes into the graph, this change introduces the DFG::InsertionSet, which is a way of lazily inserting elements into a list. This allows the FixupPhase to remain O(N) despite performing multiple injections in a single basic block. Without the InsertionSet, each injection would require performing an insertion into a vector, which is O(N), leading to O(N^2) performance overall. With the InsertionSet, each injection simply records what insertion would have been performed, and all insertions are performed at once (via InsertionSet::execute) after processing of a basic block is completed. * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/PredictedType.h: (JSC::isActionableIntMutableArrayPrediction): (JSC): (JSC::isActionableFloatMutableArrayPrediction): (JSC::isActionableTypedMutableArrayPrediction): (JSC::isActionableMutableArrayPrediction): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::performNodeCSE): * dfg/DFGCommon.h: (JSC::DFG::useKindToString): (DFG): * dfg/DFGFixupPhase.cpp: (JSC::DFG::FixupPhase::run): (JSC::DFG::FixupPhase::fixupBlock): (FixupPhase): (JSC::DFG::FixupPhase::fixupNode): (JSC::DFG::FixupPhase::fixDoubleEdge): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGInsertionSet.h: Added. (DFG): (Insertion): (JSC::DFG::Insertion::Insertion): (JSC::DFG::Insertion::index): (JSC::DFG::Insertion::element): (InsertionSet): (JSC::DFG::InsertionSet::InsertionSet): (JSC::DFG::InsertionSet::append): (JSC::DFG::InsertionSet::execute): * dfg/DFGNodeType.h: (DFG): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): (JSC::DFG::SpeculativeJIT::compileValueToInt32): (JSC::DFG::SpeculativeJIT::compileInt32ToDouble): (DFG): * dfg/DFGSpeculativeJIT.h: (SpeculativeJIT): (JSC::DFG::IntegerOperand::IntegerOperand): (JSC::DFG::DoubleOperand::DoubleOperand): (JSC::DFG::JSValueOperand::JSValueOperand): (JSC::DFG::StorageOperand::StorageOperand): (JSC::DFG::SpeculateIntegerOperand::SpeculateIntegerOperand): (JSC::DFG::SpeculateStrictInt32Operand::SpeculateStrictInt32Operand): (JSC::DFG::SpeculateDoubleOperand::SpeculateDoubleOperand): (JSC::DFG::SpeculateCellOperand::SpeculateCellOperand): (JSC::DFG::SpeculateBooleanOperand::SpeculateBooleanOperand): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-03-25 Filip Pizlo <fpizlo@apple.com> DFGOperands should be moved out of the DFG and into bytecode https://bugs.webkit.org/show_bug.cgi?id=82151 Reviewed by Dan Bernstein. * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/Operands.h: Copied from Source/JavaScriptCore/dfg/DFGOperands.h. * dfg/DFGBasicBlock.h: * dfg/DFGNode.h: * dfg/DFGOSREntry.h: * dfg/DFGOSRExit.h: * dfg/DFGOperands.h: Removed. * dfg/DFGVariableAccessData.h: 2012-03-24 Filip Pizlo <fpizlo@apple.com> DFG 64-bit Branch implementation should not be creating a JSValueOperand that it isn't going to use https://bugs.webkit.org/show_bug.cgi?id=82136 Reviewed by Geoff Garen. * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::emitBranch): 2012-03-24 Kevin Ollivier <kevino@theolliviers.com> [wx] Unreviewed. Fix the build after WTF move. * wscript: 2012-03-23 Filip Pizlo <fpizlo@apple.com> DFG double voting may be overzealous in the case of variables that end up being used as integers https://bugs.webkit.org/show_bug.cgi?id=82008 Reviewed by Oliver Hunt. Cleaned up propagation, making the intent more explicit in most places. Back-propagate NodeUsedAsInt for cases where a node was used in a context that is known to strongly prefer integers. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dumpCodeOrigin): (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (Graph): * dfg/DFGNodeFlags.cpp: (JSC::DFG::nodeFlagsAsString): * dfg/DFGNodeFlags.h: (DFG): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::run): (JSC::DFG::PredictionPropagationPhase::propagate): (PredictionPropagationPhase): (JSC::DFG::PredictionPropagationPhase::mergeDefaultFlags): (JSC::DFG::PredictionPropagationPhase::vote): (JSC::DFG::PredictionPropagationPhase::doRoundOfDoubleVoting): (JSC::DFG::PredictionPropagationPhase::fixupNode): * dfg/DFGVariableAccessData.h: (JSC::DFG::VariableAccessData::shouldUseDoubleFormatAccordingToVote): 2012-03-24 Filip Pizlo <fpizlo@apple.com> DFG::Node::shouldNotSpeculateInteger() should be eliminated https://bugs.webkit.org/show_bug.cgi?id=82123 Reviewed by Geoff Garen. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGNode.h: (Node): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): 2012-03-24 Yong Li <yoli@rim.com> Increase getByIdSlowCase ConstantSpace/InstructionSpace for CPU(ARM_TRADITIONAL) https://bugs.webkit.org/show_bug.cgi?id=81521 Increase sequenceGetByIdSlowCaseConstantSpace and sequenceGetByIdSlowCaseInstructionSpace for CPU(ARM_TRADITIONAL) to fit actual need. Reviewed by Oliver Hunt. * jit/JIT.h: (JIT): 2012-03-23 Filip Pizlo <fpizlo@apple.com> DFG Fixup should be able to short-circuit trivial ValueToInt32's https://bugs.webkit.org/show_bug.cgi?id=82030 Reviewed by Michael Saboff. Takes the fixup() method of the prediction propagation phase and makes it into its own phase. Adds the ability to short-circuit trivial ValueToInt32 nodes, and mark pure ValueToInt32's as such. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::makeSafe): (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCommon.h: * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGFixupPhase.cpp: Added. (DFG): (FixupPhase): (JSC::DFG::FixupPhase::FixupPhase): (JSC::DFG::FixupPhase::run): (JSC::DFG::FixupPhase::fixupNode): (JSC::DFG::FixupPhase::fixIntEdge): (JSC::DFG::performFixup): * dfg/DFGFixupPhase.h: Added. (DFG): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::run): (PredictionPropagationPhase): 2012-03-23 Mark Hahnenberg <mhahnenberg@apple.com> tryReallocate could break the zero-ed memory invariant of CopiedBlocks https://bugs.webkit.org/show_bug.cgi?id=82087 Reviewed by Filip Pizlo. Removing this optimization turned out to be ~1% regression on kraken, so I simply undid the modification to the current block if we fail. * heap/CopiedSpace.cpp: (JSC::CopiedSpace::tryReallocate): Undid the reset in the CopiedAllocator if we fail to reallocate from the current block. 2012-03-23 Alexey Proskuryakov <ap@apple.com> [Mac] No need for platform-specific ENABLE_BLOB values https://bugs.webkit.org/show_bug.cgi?id=82102 Reviewed by David Kilzer. * Configurations/FeatureDefines.xcconfig: 2012-03-23 Michael Saboff <msaboff@apple.com> DFG::compileValueToInt32 Sometime Generates GPR to FPR reg back to GPR https://bugs.webkit.org/show_bug.cgi?id=81805 Reviewed by Filip Pizlo. Added SpeculativeJIT::checkGeneratedType() to determine the current format of an operand. Used that information in SpeculativeJIT::compileValueToInt32 to generate code that will use integer and JSValue types in integer format directly without a conversion to double. * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::checkGeneratedType): (DFG): (JSC::DFG::SpeculativeJIT::compileValueToInt32): * dfg/DFGSpeculativeJIT.h: (DFG): (SpeculativeJIT): 2012-03-23 Steve Falkenburg <sfalken@apple.com> Update Apple Windows build files for WTF move https://bugs.webkit.org/show_bug.cgi?id=82069 Reviewed by Jessie Berlin. * JavaScriptCore.vcproj/JavaScriptCoreSubmit.sln: Removed WTF and WTFGenerated. 2012-03-23 Dean Jackson <dino@apple.com> Disable CSS_SHADERS in Apple builds https://bugs.webkit.org/show_bug.cgi?id=81996 Reviewed by Simon Fraser. Remove ENABLE_CSS_SHADERS from FeatureDefines. It's now in Platform.h. * Configurations/FeatureDefines.xcconfig: 2012-03-23 Gavin Barraclough <barraclough@apple.com> RexExp constructor last match properties should not rely on previous ovector https://bugs.webkit.org/show_bug.cgi?id=82077 Reviewed by Oliver Hunt. This change simplifies matching, and will enable subpattern results to be fully lazily generated in the future. This patch changes the scheme used to lazily generate the last match properties of the RegExp object. Instead of relying on the results in the ovector, we can instead lazily generate the subpatters using a RegExpMatchesArray. To do so we just need to store the input, the regexp matched, and the match location (the MatchResult). When the match is accessed or the input is set, we reify results. We use a special value of setting the saved result to MatchResult::failed() to indicated that we're in a reified state. This means that next time a match is performed, the store of the result will automatically blow away the reified value. * JavaScriptCore.xcodeproj/project.pbxproj: - Added new files. * runtime/RegExp.cpp: (JSC::RegExpFunctionalTestCollector::outputOneTest): - changed 'subPattern' -> 'subpattern' (there was a mix in JSC, 'subpattern' was more common). * runtime/RegExpCachedResult.cpp: Added. (JSC::RegExpCachedResult::visitChildren): (JSC::RegExpCachedResult::lastResult): (JSC::RegExpCachedResult::setInput): - New methods, mark GC objects, lazily create the matches array, and record a user provided input (via assignment to RegExp.inupt). * runtime/RegExpCachedResult.h: Added. (RegExpCachedResult): - Added new class. (JSC::RegExpCachedResult::RegExpCachedResult): (JSC::RegExpCachedResult::record): (JSC::RegExpCachedResult::input): - Initialize the object, record the result of a RegExp match, access the stored input property. * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::RegExpConstructor): - Initialize m_result/m_multiline properties. (JSC::RegExpConstructor::visitChildren): - Make sure the cached results (or lazy source for them) are marked. (JSC::RegExpConstructor::getBackref): (JSC::RegExpConstructor::getLastParen): (JSC::RegExpConstructor::getLeftContext): (JSC::RegExpConstructor::getRightContext): - Moved from RegExpConstructor, moved to RegExpCachedResult, and using new caching scheme. (JSC::regExpConstructorInput): (JSC::setRegExpConstructorInput): - Changed to use RegExpCachedResult. * runtime/RegExpConstructor.h: (JSC::RegExpConstructor::create): (RegExpConstructor): (JSC::RegExpConstructor::setMultiline): (JSC::RegExpConstructor::multiline): - Move multiline property onto the constructor object; it is not affected by the last match. (JSC::RegExpConstructor::setInput): (JSC::RegExpConstructor::input): - These defer to RegExpCachedResult. (JSC::RegExpConstructor::performMatch): * runtime/RegExpMatchesArray.cpp: Added. (JSC::RegExpMatchesArray::visitChildren): - Eeeep! added missing visitChildren! (JSC::RegExpMatchesArray::finishCreation): (JSC::RegExpMatchesArray::reifyAllProperties): (JSC::RegExpMatchesArray::reifyMatchProperty): - Moved from RegExpConstructor.cpp. (JSC::RegExpMatchesArray::leftContext): (JSC::RegExpMatchesArray::rightContext): - Since the match start/ * runtime/RegExpMatchesArray.h: (RegExpMatchesArray): - Declare new methods & structure flags. * runtime/RegExpObject.cpp: (JSC::RegExpObject::match): - performMatch now requires the JSString input, to cache. * runtime/StringPrototype.cpp: (JSC::removeUsingRegExpSearch): (JSC::replaceUsingRegExpSearch): (JSC::stringProtoFuncMatch): (JSC::stringProtoFuncSearch): - performMatch now requires the JSString input, to cache. 2012-03-23 Tony Chang <tony@chromium.org> [chromium] rename newwtf target back to wtf https://bugs.webkit.org/show_bug.cgi?id=82064 Reviewed by Adam Barth. * JavaScriptCore.gyp/JavaScriptCore.gyp: 2012-03-23 Mark Hahnenberg <mhahnenberg@apple.com> Simplify memory usage tracking in CopiedSpace https://bugs.webkit.org/show_bug.cgi?id=80705 Reviewed by Filip Pizlo. * heap/CopiedAllocator.h: (CopiedAllocator): Rename currentUtilization to currentSize. (JSC::CopiedAllocator::currentCapacity): * heap/CopiedBlock.h: (CopiedBlock): (JSC::CopiedBlock::payload): Move the implementation of payload() out of the class declaration. (JSC): (JSC::CopiedBlock::size): Add new function to calculate the block's size. (JSC::CopiedBlock::capacity): Ditto for capacity. * heap/CopiedSpace.cpp: (JSC::CopiedSpace::CopiedSpace): Remove old bogus memory stats fields and add a new field for the water mark. (JSC::CopiedSpace::init): (JSC::CopiedSpace::tryAllocateSlowCase): When we fail to allocate from the current block, we need to update our current water mark with the size of the block. (JSC::CopiedSpace::tryAllocateOversize): When we allocate a new oversize block, we need to update our current water mark with the size of the used portion of the block. (JSC::CopiedSpace::tryReallocate): We don't need to update the water mark when reallocating because it will either get accounted for when we fill up the block later in the case of being able to reallocate in the current block or it will get picked up immediately because we'll have to get a new block. (JSC::CopiedSpace::tryReallocateOversize): We do, however, need to update in when realloc-ing an oversize block because we deallocate the old block and allocate a brand new one. (JSC::CopiedSpace::doneFillingBlock): Update the water mark as blocks are returned to the CopiedSpace by the SlotVisitors. (JSC::CopiedSpace::doneCopying): Add in any pinned blocks to the water mark. (JSC::CopiedSpace::getFreshBlock): We use the Heap's new function to tell us whether or not we should collect now instead of doing the calculation ourself. (JSC::CopiedSpace::destroy): (JSC): (JSC::CopiedSpace::size): Manually calculate the size of the CopiedSpace, similar to how MarkedSpace does. (JSC::CopiedSpace::capacity): Ditto for capacity. * heap/CopiedSpace.h: (JSC::CopiedSpace::waterMark): (CopiedSpace): * heap/CopiedSpaceInlineMethods.h: (JSC::CopiedSpace::startedCopying): Reset water mark to 0 when we start copying during a collection. (JSC::CopiedSpace::allocateNewBlock): (JSC::CopiedSpace::fitsInBlock): (JSC::CopiedSpace::allocateFromBlock): * heap/Heap.cpp: (JSC::Heap::size): Incorporate size of CopiedSpace into the total size of the Heap. (JSC::Heap::capacity): Ditto for capacity. (JSC::Heap::collect): * heap/Heap.h: (Heap): (JSC::Heap::shouldCollect): New function for other sub-parts of the Heap to use to determine whether they should initiate a collection or continue to allocate new blocks. (JSC): (JSC::Heap::waterMark): Now is the sum of the water marks of the two sub-parts of the Heap (MarkedSpace and CopiedSpace). * heap/MarkedAllocator.cpp: (JSC::MarkedAllocator::allocateSlowCase): Changed to use the Heap's new shouldCollect() function. 2012-03-23 Ryosuke Niwa <rniwa@webkit.org> BitVector::resizeOutOfLine doesn't memset when converting an inline buffer https://bugs.webkit.org/show_bug.cgi?id=82012 Reviewed by Filip Pizlo. Initialize out-of-line buffers while extending an inline buffer. Also export symbols to be used in WebCore. * wtf/BitVector.cpp: (WTF::BitVector::resizeOutOfLine): * wtf/BitVector.h: (BitVector): (OutOfLineBits): 2012-03-22 Michael Saboff <msaboff@apple.com> ExecutableAllocator::memoryPressureMultiplier() might can return NaN https://bugs.webkit.org/show_bug.cgi?id=82002 Reviewed by Filip Pizlo. Guard against divide by zero and then make sure the return value is >= 1.0. * jit/ExecutableAllocator.cpp: (JSC::ExecutableAllocator::memoryPressureMultiplier): * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::ExecutableAllocator::memoryPressureMultiplier): 2012-03-22 Jessie Berlin <jberlin@apple.com> Windows build fix after r111778. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: Don't include and try to build files owned by WTF. Also, let VS have its way with the vcproj in terms of file ordering. 2012-03-22 Raphael Kubo da Costa <rakuco@FreeBSD.org> [CMake] Unreviewed build fix after r111778. * CMakeLists.txt: Move ${WTF_DIR} after ${JAVASCRIPTCORE_DIR} in the include paths so that the right config.h is used. 2012-03-22 Tony Chang <tony@chromium.org> Unreviewed, fix chromium build after wtf move. Remove old wtf_config and wtf targets. * JavaScriptCore.gyp/JavaScriptCore.gyp: 2012-03-22 Martin Robinson <mrobinson@igalia.com> Fixed the GTK+ WTF/JavaScriptCore build after r111778. * GNUmakefile.list.am: Removed an extra trailing backslash. 2012-03-22 Mark Rowe <mrowe@apple.com> Fix the build. * Configurations/JavaScriptCore.xcconfig: Tell the linker to pull in all members from static libraries rather than only those that contain symbols that JavaScriptCore itself uses. * JavaScriptCore.xcodeproj/project.pbxproj: Remove some bogus settings that crept in to the Xcode project. 2012-03-22 Filip Pizlo <fpizlo@apple.com> DFG NodeFlags has some duplicate code and naming issues https://bugs.webkit.org/show_bug.cgi?id=81975 Reviewed by Gavin Barraclough. Removed most references to "ArithNodeFlags" since those are now just part of the node flags. Fixed some renaming goofs (EdgedAsNum is once again NodeUsedAsNum). Got rid of setArithNodeFlags() and mergeArithNodeFlags() because the former was never called and the latter did the same things as mergeFlags(). * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::makeSafe): (JSC::DFG::ByteCodeParser::makeDivSafe): (JSC::DFG::ByteCodeParser::handleIntrinsic): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGNode.h: (JSC::DFG::Node::arithNodeFlags): (Node): * dfg/DFGNodeFlags.cpp: (JSC::DFG::nodeFlagsAsString): * dfg/DFGNodeFlags.h: (DFG): (JSC::DFG::nodeUsedAsNumber): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): (JSC::DFG::PredictionPropagationPhase::mergeDefaultArithFlags): 2012-03-22 Eric Seidel <eric@webkit.org> Actually move WTF files to their new home https://bugs.webkit.org/show_bug.cgi?id=81844 Unreviewed. The details of the port-specific changes have been seen by contributors from those ports, but the whole 5MB change isn't very reviewable as-is. * GNUmakefile.am: * GNUmakefile.list.am: * JSCTypedArrayStubs.h: * JavaScriptCore.gypi: * JavaScriptCore.xcodeproj/project.pbxproj: * jsc.cpp: 2012-03-22 Kevin Ollivier <kevino@theolliviers.com> [wx] Unreviewed. Adding Source/WTF to the build. * wscript: 2012-03-22 Gavin Barraclough <barraclough@apple.com> Add JSValue::isFunction https://bugs.webkit.org/show_bug.cgi?id=81935 Reviewed by Geoff Garen. This would be useful in the WebCore bindings code. Also, remove asFunction, replace with jsCast<JSFunction*>. * API/JSContextRef.cpp: * debugger/Debugger.cpp: * debugger/DebuggerCallFrame.cpp: (JSC::DebuggerCallFrame::functionName): * dfg/DFGGraph.h: (JSC::DFG::Graph::valueOfFunctionConstant): * dfg/DFGOperations.cpp: * interpreter/CallFrame.cpp: (JSC::CallFrame::isInlineCallFrameSlow): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): (JSC::jitCompileFor): (JSC::lazyLinkFor): * llint/LLIntSlowPaths.cpp: (JSC::LLInt::traceFunctionPrologue): (JSC::LLInt::LLINT_SLOW_PATH_DECL): (JSC::LLInt::setUpCall): * runtime/Arguments.h: (JSC::Arguments::finishCreation): * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncFilter): (JSC::arrayProtoFuncMap): (JSC::arrayProtoFuncEvery): (JSC::arrayProtoFuncForEach): (JSC::arrayProtoFuncSome): (JSC::arrayProtoFuncReduce): (JSC::arrayProtoFuncReduceRight): * runtime/CommonSlowPaths.h: (JSC::CommonSlowPaths::arityCheckFor): * runtime/Executable.h: (JSC::FunctionExecutable::compileFor): (JSC::FunctionExecutable::compileOptimizedFor): * runtime/FunctionPrototype.cpp: (JSC::functionProtoFuncToString): * runtime/JSArray.cpp: (JSC::JSArray::sort): * runtime/JSFunction.cpp: (JSC::JSFunction::argumentsGetter): (JSC::JSFunction::callerGetter): (JSC::JSFunction::lengthGetter): * runtime/JSFunction.h: (JSC): (JSC::asJSFunction): (JSC::JSValue::isFunction): * runtime/JSGlobalData.cpp: (WTF::Recompiler::operator()): (JSC::JSGlobalData::releaseExecutableMemory): * runtime/JSValue.h: * runtime/StringPrototype.cpp: (JSC::replaceUsingRegExpSearch): 2012-03-21 Filip Pizlo <fpizlo@apple.com> DFG speculation on booleans should be rationalized https://bugs.webkit.org/show_bug.cgi?id=81840 Reviewed by Gavin Barraclough. This removes isKnownBoolean() and replaces it with AbstractState-based optimization, and cleans up the control flow in code gen methods for Branch and LogicalNot. Also fixes a goof in Node::shouldSpeculateNumber, and removes isKnownNotBoolean() since that method appeared to be a helper used solely by 32_64's speculateBooleanOperation(). This is performance-neutral. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateNumber): * dfg/DFGSpeculativeJIT.cpp: (DFG): * dfg/DFGSpeculativeJIT.h: (SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): 2012-03-21 Mark Rowe <mrowe@apple.com> Fix the build. * wtf/MetaAllocator.h: (MetaAllocator): Export the destructor. 2012-03-21 Eric Seidel <eric@webkit.org> Fix remaining WTF includes in JavaScriptCore in preparation for moving WTF headers out of JavaScriptCore https://bugs.webkit.org/show_bug.cgi?id=81834 Reviewed by Adam Barth. * jsc.cpp: * os-win32/WinMain.cpp: * runtime/JSDateMath.cpp: * runtime/TimeoutChecker.cpp: * testRegExp.cpp: * tools/CodeProfiling.cpp: 2012-03-21 Eric Seidel <eric@webkit.org> WTF::MetaAllocator has a weak vtable (discovered when building wtf as a static library) https://bugs.webkit.org/show_bug.cgi?id=81838 Reviewed by Geoffrey Garen. My understanding is that weak vtables happen when the compiler/linker cannot determine which compilation unit should constain the vtable. In this case because there were only pure virtual functions as well as an "inline" virtual destructor (thus the virtual destructor was defined in many compilation units). Since you can't actually "inline" a virtual function (it still has to bounce through the vtable), the "inline" on this virutal destructor doesn't actually help performance, and is only serving to confuse the compiler here. I've moved the destructor implementation to the .cpp file, thus making it clear to the compiler where the vtable should be stored, and solving the error. * wtf/MetaAllocator.cpp: (WTF::MetaAllocator::~MetaAllocator): (WTF): * wtf/MetaAllocator.h: 2012-03-20 Gavin Barraclough <barraclough@apple.com> RegExpMatchesArray should not copy the ovector https://bugs.webkit.org/show_bug.cgi?id=81742 Reviewed by Michael Saboff. Currently, all RegExpMatchesArray object contain Vector<int, 32>, used to hold any sub-pattern results. This makes allocation/construction/destruction of these objects more expensive. Instead, just store the main match, and recreate the sub-pattern ranges only if necessary (these are often only used for grouping, and the results never accessed). If the main match (index 0) of the RegExpMatchesArray is accessed, reify that value alone. * dfg/DFGOperations.cpp: - RegExpObject match renamed back to test (test returns a bool). * runtime/RegExpConstructor.cpp: (JSC): - Removed RegExpResult, RegExpMatchesArray constructor, destroy method. (JSC::RegExpMatchesArray::finishCreation): - Removed RegExpConstructorPrivate parameter. (JSC::RegExpMatchesArray::reifyAllProperties): - (Was fillArrayInstance) Reify all properties of the RegExpMatchesArray. If there are sub-pattern properties, the RegExp is re-run to generate their values. (JSC::RegExpMatchesArray::reifyMatchProperty): - Reify just the match (index 0) property of the RegExpMatchesArray. * runtime/RegExpConstructor.h: (RegExpConstructor): (JSC::RegExpConstructor::performMatch): - performMatch now returns a MatchResult, rather than using out-parameters. * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::RegExpMatchesArray): - Moved from .cpp, stores the input/regExp/result to use when lazily reifying properties. (RegExpMatchesArray): (JSC::RegExpMatchesArray::create): - Now passed the input string matched against, the RegExp, and the MatchResult. (JSC::RegExpMatchesArray::reifyAllPropertiesIfNecessary): (JSC::RegExpMatchesArray::reifyMatchPropertyIfNecessary): - Helpers to conditionally reify properties. (JSC::RegExpMatchesArray::getOwnPropertySlot): (JSC::RegExpMatchesArray::getOwnPropertySlotByIndex): (JSC::RegExpMatchesArray::getOwnPropertyDescriptor): (JSC::RegExpMatchesArray::put): (JSC::RegExpMatchesArray::putByIndex): (JSC::RegExpMatchesArray::deleteProperty): (JSC::RegExpMatchesArray::deletePropertyByIndex): (JSC::RegExpMatchesArray::getOwnPropertyNames): (JSC::RegExpMatchesArray::defineOwnProperty): - Changed to use reifyAllPropertiesIfNecessary/reifyMatchPropertyIfNecessary (getOwnPropertySlotByIndex calls reifyMatchPropertyIfNecessary if index is 0). * runtime/RegExpObject.cpp: (JSC::RegExpObject::exec): (JSC::RegExpObject::match): - match now returns a MatchResult. * runtime/RegExpObject.h: (JSC::MatchResult::MatchResult): - Added the result of a match is a start & end tuple. (JSC::MatchResult::failed): - A failure is indicated by (notFound, 0). (JSC::MatchResult::operator bool): - Evaluates to false if the match failed. (JSC::MatchResult::empty): - Evaluates to true if the match succeeded with length 0. (JSC::RegExpObject::test): - Now returns a bool. * runtime/RegExpPrototype.cpp: (JSC::regExpProtoFuncTest): - RegExpObject match renamed back to test (test returns a bool). * runtime/StringPrototype.cpp: (JSC::removeUsingRegExpSearch): (JSC::replaceUsingRegExpSearch): (JSC::stringProtoFuncMatch): (JSC::stringProtoFuncSearch): - performMatch now returns a MatchResult, rather than using out-parameters. 2012-03-21 Hojong Han <hojong.han@samsung.com> Fix out of memory by allowing overcommit https://bugs.webkit.org/show_bug.cgi?id=81743 Reviewed by Geoffrey Garen. Garbage collection is not triggered and new blocks are added because overcommit is allowed by MAP_NORESERVE flag when high water mark is big enough. * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveAndCommit): 2012-03-21 Jessie Berlin <jberlin@apple.com> More Windows build fixing. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: Fix the order of the include directories to look in include/private first before looking in include/private/JavaScriptCore. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleasePGO.vsprops: Look in the Production output directory (where the wtf headers will be). This is the same thing that is done for jsc and testRegExp in ReleasePGO. 2012-03-21 Jessie Berlin <jberlin@apple.com> WTF headers should be in $(ConfigurationBuildDir)\include\private\wtf, not $(ConfigurationBuildDir)\include\private\JavaScriptCore\wtf. https://bugs.webkit.org/show_bug.cgi?id=81739 Reviewed by Dan Bernstein. * JavaScriptCore.vcproj/jsc/jsc.vcproj: Look for AtomicString.cpp, StringBuilder.cpp, StringImpl.cpp, and WTFString.cpp in the wtf subdirectory of the build output, not the JavaScriptCore/wtf subdirectory. * JavaScriptCore.vcproj/testRegExp/testRegExp.vcproj: Ditto. * JavaScriptCore.vcproj/testRegExp/testRegExpReleasePGO.vsprops: Get the headers for those 4 files from the wtf subdirectory of the build output, not the JavaScriptCore/wtf subdirectory. * JavaScriptCore.vcproj/jsc/jscReleasePGO.vsprops: Ditto. 2012-03-20 Eric Seidel <eric@webkit.org> Move wtf/Platform.h from JavaScriptCore to Source/WTF/wtf https://bugs.webkit.org/show_bug.cgi?id=80911 Reviewed by Adam Barth. Update the various build systems to depend on Source/WTF headers as well as remove references to Platform.h (since it's now moved). * CMakeLists.txt: * JavaScriptCore.pri: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/CMakeLists.txt: 2012-03-20 Filip Pizlo <fpizlo@apple.com> op_mod fails on many interesting corner cases https://bugs.webkit.org/show_bug.cgi?id=81648 Reviewed by Oliver Hunt. Removed most strength reduction for op_mod, and fixed the integer handling to do the right thing for corner cases. Oddly, this revealed bugs in OSR, which this patch also fixes. This patch is performance neutral on all of the major benchmarks we track. * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT.cpp: (DFG): (JSC::DFG::SpeculativeJIT::compileSoftModulo): (JSC::DFG::SpeculativeJIT::compileArithMod): * jit/JIT.h: (JIT): * jit/JITArithmetic.cpp: (JSC): (JSC::JIT::emit_op_mod): (JSC::JIT::emitSlow_op_mod): * jit/JITArithmetic32_64.cpp: (JSC::JIT::emit_op_mod): (JSC::JIT::emitSlow_op_mod): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC): * jit/JITStubs.h: (TrampolineStructure): (JSC::JITThunks::ctiNativeConstruct): * llint/LowLevelInterpreter64.asm: * wtf/Platform.h: * wtf/SimpleStats.h: (WTF::SimpleStats::variance): 2012-03-20 Steve Falkenburg <sfalken@apple.com> Windows (make based) build fix. <rdar://problem/11069015> * JavaScriptCore.vcproj/JavaScriptCore.make: devenv /rebuild doesn't work with JavaScriptCore.vcproj. Use /clean and /build instead. 2012-03-20 Steve Falkenburg <sfalken@apple.com> Move WTF-related Windows project files out of JavaScriptCore https://bugs.webkit.org/show_bug.cgi?id=80680 This change only moves the vcproj and related files from JavaScriptCore/JavaScriptCore.vcproj/WTF. It does not move any source code. This is in preparation for the WTF source move out of JavaScriptCore. Reviewed by Jessie Berlin. * JavaScriptCore.vcproj/JavaScriptCore.sln: * JavaScriptCore.vcproj/JavaScriptCoreSubmit.sln: * JavaScriptCore.vcproj/WTF: Removed. * JavaScriptCore.vcproj/WTF/WTF.vcproj: Removed. * JavaScriptCore.vcproj/WTF/WTFCommon.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFDebug.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFDebugAll.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFDebugCairoCFLite.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFGenerated.make: Removed. * JavaScriptCore.vcproj/WTF/WTFGenerated.vcproj: Removed. * JavaScriptCore.vcproj/WTF/WTFGeneratedCommon.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFGeneratedDebug.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFGeneratedDebugAll.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFGeneratedDebugCairoCFLite.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFGeneratedProduction.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFGeneratedRelease.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFGeneratedReleaseCairoCFLite.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFPostBuild.cmd: Removed. * JavaScriptCore.vcproj/WTF/WTFPreBuild.cmd: Removed. * JavaScriptCore.vcproj/WTF/WTFProduction.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFRelease.vsprops: Removed. * JavaScriptCore.vcproj/WTF/WTFReleaseCairoCFLite.vsprops: Removed. * JavaScriptCore.vcproj/WTF/build-generated-files.sh: Removed. * JavaScriptCore.vcproj/WTF/copy-files.cmd: Removed. * JavaScriptCore.vcproj/WTF/work-around-vs-dependency-tracking-bugs.py: Removed. 2012-03-20 Benjamin Poulain <bpoulain@apple.com> Cache the type string of JavaScript object https://bugs.webkit.org/show_bug.cgi?id=81446 Reviewed by Geoffrey Garen. Instead of creating the JSString every time, we create lazily the strings in JSGlobalData. This avoid the construction of the StringImpl and of the JSString, which gives some performance improvements. * runtime/CommonIdentifiers.h: * runtime/JSValue.cpp: (JSC::JSValue::toStringSlowCase): * runtime/Operations.cpp: (JSC::jsTypeStringForValue): * runtime/SmallStrings.cpp: (JSC::SmallStrings::SmallStrings): (JSC::SmallStrings::finalizeSmallStrings): (JSC::SmallStrings::initialize): (JSC): * runtime/SmallStrings.h: (SmallStrings): 2012-03-20 Oliver Hunt <oliver@apple.com> Allow LLINT to work even when executable allocation fails. https://bugs.webkit.org/show_bug.cgi?id=81693 Reviewed by Gavin Barraclough. Don't crash if executable allocation fails if we can fall back on LLINT * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::FixedVMPoolExecutableAllocator::FixedVMPoolExecutableAllocator): * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveAndCommit): 2012-03-20 Csaba Osztrogonác <ossy@webkit.org> Division optimizations fail to infer cases of truncated division and mishandle -2147483648/-1 https://bugs.webkit.org/show_bug.cgi?id=81428 32 bit buildfix after r111355. 2147483648 (2^31) isn't valid int literal in ISO C90, because 2147483647 (2^31-1) is the biggest int. The smallest int is -2147483648 (-2^31) == -2147483647 - 1 == -INT32_MAX-1 == INT32_MIN (stdint.h). Reviewed by Zoltan Herczeg. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileIntegerArithDivForX86): 2012-03-19 Jochen Eisinger <jochen@chromium.org> Split WTFReportBacktrace into WTFReportBacktrace and WTFPrintBacktrace https://bugs.webkit.org/show_bug.cgi?id=80983 Reviewed by Darin Adler. This allows printing a backtrace acquired by an earlier WTFGetBacktrace call which is useful for local debugging. * wtf/Assertions.cpp: * wtf/Assertions.h: 2012-03-19 Benjamin Poulain <benjamin@webkit.org> Do not copy the script source in the SourceProvider, just reference the existing string https://bugs.webkit.org/show_bug.cgi?id=81466 Reviewed by Geoffrey Garen. * parser/SourceCode.h: Remove the unused, and incorrect, function data(). * parser/SourceProvider.h: Add OVERRIDE for clarity. 2012-03-19 Filip Pizlo <fpizlo@apple.com> Division optimizations fail to infer cases of truncated division and mishandle -2147483648/-1 https://bugs.webkit.org/show_bug.cgi?id=81428 <rdar://problem/11067382> Reviewed by Oliver Hunt. If you're a division over integers and you're only used as an integer, then you're an integer division and remainder checks become unnecessary. If you're dividing -2147483648 by -1, don't crash. * assembler/MacroAssemblerX86Common.h: (MacroAssemblerX86Common): (JSC::MacroAssemblerX86Common::add32): * dfg/DFGSpeculativeJIT.cpp: (DFG): (JSC::DFG::SpeculativeJIT::compileIntegerArithDivForX86): * dfg/DFGSpeculativeJIT.h: (SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * llint/LowLevelInterpreter64.asm: 2012-03-19 Benjamin Poulain <bpoulain@apple.com> Simplify SmallStrings https://bugs.webkit.org/show_bug.cgi?id=81445 Reviewed by Gavin Barraclough. SmallStrings had two methods that should not be public: count() and clear(). The method clear() is effectively replaced by finalizeSmallStrings(). The body of the method was moved to the constructor since the code is obvious. The method count() is unused. * runtime/SmallStrings.cpp: (JSC::SmallStrings::SmallStrings): * runtime/SmallStrings.h: (SmallStrings): 2012-03-19 Filip Pizlo <fpizlo@apple.com> DFG can no longer compile V8-v4/regexp in debug mode https://bugs.webkit.org/show_bug.cgi?id=81592 Reviewed by Gavin Barraclough. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-03-19 Filip Pizlo <fpizlo@apple.com> Prediction propagation for UInt32ToNumber incorrectly assumes that outs outcome does not change throughout the fixpoint https://bugs.webkit.org/show_bug.cgi?id=81583 Reviewed by Michael Saboff. * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): 2012-03-19 Filip Pizlo <fpizlo@apple.com> GC should not attempt to clear LLInt instruction inline caches for code blocks that are in the process of being generated https://bugs.webkit.org/show_bug.cgi?id=81565 Reviewed by Oliver Hunt. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::finalizeUnconditionally): 2012-03-19 Eric Seidel <eric@webkit.org> Fix WTF header include discipline in Chromium WebKit https://bugs.webkit.org/show_bug.cgi?id=81281 Reviewed by James Robinson. * JavaScriptCore.gyp/JavaScriptCore.gyp: * wtf/unicode/icu/CollatorICU.cpp: 2012-03-19 Filip Pizlo <fpizlo@apple.com> DFG NodeUse should be called Edge and NodeReferenceBlob should be called AdjacencyList https://bugs.webkit.org/show_bug.cgi?id=81556 Rubber stamped by Gavin Barraclough. * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGAbstractState.h: (JSC::DFG::AbstractState::forNode): * dfg/DFGAdjacencyList.h: Copied from Source/JavaScriptCore/dfg/DFGNodeReferenceBlob.h. (JSC::DFG::AdjacencyList::AdjacencyList): (JSC::DFG::AdjacencyList::child): (JSC::DFG::AdjacencyList::setChild): (JSC::DFG::AdjacencyList::child1): (JSC::DFG::AdjacencyList::child2): (JSC::DFG::AdjacencyList::child3): (JSC::DFG::AdjacencyList::setChild1): (JSC::DFG::AdjacencyList::setChild2): (JSC::DFG::AdjacencyList::setChild3): (JSC::DFG::AdjacencyList::child1Unchecked): (JSC::DFG::AdjacencyList::initialize): (AdjacencyList): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::addVarArgChild): (JSC::DFG::ByteCodeParser::processPhiStack): * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::canonicalize): (JSC::DFG::CSEPhase::performSubstitution): * dfg/DFGEdge.h: Copied from Source/JavaScriptCore/dfg/DFGNodeUse.h. (DFG): (JSC::DFG::Edge::Edge): (JSC::DFG::Edge::operator==): (JSC::DFG::Edge::operator!=): (Edge): (JSC::DFG::operator==): (JSC::DFG::operator!=): * dfg/DFGGraph.h: (JSC::DFG::Graph::operator[]): (JSC::DFG::Graph::at): (JSC::DFG::Graph::ref): (JSC::DFG::Graph::deref): (JSC::DFG::Graph::clearAndDerefChild1): (JSC::DFG::Graph::clearAndDerefChild2): (JSC::DFG::Graph::clearAndDerefChild3): (Graph): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::getPrediction): * dfg/DFGNode.h: (JSC::DFG::Node::Node): (JSC::DFG::Node::child1): (JSC::DFG::Node::child1Unchecked): (JSC::DFG::Node::child2): (JSC::DFG::Node::child3): (Node): * dfg/DFGNodeFlags.cpp: (JSC::DFG::arithNodeFlagsAsString): * dfg/DFGNodeFlags.h: (DFG): (JSC::DFG::nodeUsedAsNumber): * dfg/DFGNodeReferenceBlob.h: Removed. * dfg/DFGNodeUse.h: Removed. * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): (JSC::DFG::PredictionPropagationPhase::mergeDefaultArithFlags): (JSC::DFG::PredictionPropagationPhase::vote): (JSC::DFG::PredictionPropagationPhase::fixupNode): * dfg/DFGScoreBoard.h: (JSC::DFG::ScoreBoard::use): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::useChildren): (JSC::DFG::SpeculativeJIT::writeBarrier): (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): (JSC::DFG::SpeculativeJIT::compilePutByValForFloatTypedArray): (JSC::DFG::SpeculativeJIT::compileStrictEqForConstant): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::at): (JSC::DFG::SpeculativeJIT::canReuse): (JSC::DFG::SpeculativeJIT::use): (SpeculativeJIT): (JSC::DFG::SpeculativeJIT::speculationCheck): (JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution): (JSC::DFG::IntegerOperand::IntegerOperand): (JSC::DFG::DoubleOperand::DoubleOperand): (JSC::DFG::JSValueOperand::JSValueOperand): (JSC::DFG::StorageOperand::StorageOperand): (JSC::DFG::SpeculateIntegerOperand::SpeculateIntegerOperand): (JSC::DFG::SpeculateStrictInt32Operand::SpeculateStrictInt32Operand): (JSC::DFG::SpeculateDoubleOperand::SpeculateDoubleOperand): (JSC::DFG::SpeculateCellOperand::SpeculateCellOperand): (JSC::DFG::SpeculateBooleanOperand::SpeculateBooleanOperand): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::cachedPutById): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompareNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull): (JSC::DFG::SpeculativeJIT::nonSpeculativeCompareNull): (JSC::DFG::SpeculativeJIT::emitCall): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::cachedPutById): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompareNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull): (JSC::DFG::SpeculativeJIT::nonSpeculativeCompareNull): (JSC::DFG::SpeculativeJIT::emitCall): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): 2012-03-19 Gavin Barraclough <barraclough@apple.com> Object.freeze broken on latest Nightly https://bugs.webkit.org/show_bug.cgi?id=80577 Reviewed by Oliver Hunt. * runtime/Arguments.cpp: (JSC::Arguments::defineOwnProperty): - defineOwnProperty was checking for correct behaviour, provided that length/callee hadn't been overrridden. instead, just reify length/callee & rely on JSObject::defineOwnProperty. * runtime/JSFunction.cpp: (JSC::JSFunction::defineOwnProperty): - for arguments/caller/length properties, defineOwnProperty was incorrectly asserting that the object must be extensible; this is incorrect since these properties should already exist on the object. In addition, it was asserting that the arguments/caller values must match the corresponding magic data properties, but for strict mode function this is incorrect. Instead, just reify the arguments/caller accessor & defer to JSObject::defineOwnProperty. 2012-03-19 Filip Pizlo <fpizlo@apple.com> LLInt get_by_pname slow path incorrectly assumes that the operands are not constants https://bugs.webkit.org/show_bug.cgi?id=81559 Reviewed by Michael Saboff. * llint/LLIntSlowPaths.cpp: (JSC::LLInt::LLINT_SLOW_PATH_DECL): 2012-03-19 Yong Li <yoli@rim.com> [BlackBerry] Implement OSAllocator::commit/decommit in the correct way https://bugs.webkit.org/show_bug.cgi?id=77013 We should use mmap(PROT_NONE, MAP_LAZY) instead of posix_madvise() to implement memory decommitting for QNX. Reviewed by Rob Buis. * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveUncommitted): (WTF::OSAllocator::commit): (WTF::OSAllocator::decommit): 2012-03-19 Gavin Barraclough <barraclough@apple.com> Unreviewed - revent a couple of files accidentally committed. * runtime/Arguments.cpp: (JSC::Arguments::defineOwnProperty): * runtime/JSFunction.cpp: (JSC::JSFunction::defineOwnProperty): 2012-03-19 Jessie Berlin <jberlin@apple.com> Another Windows build fix after r111129. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-03-19 Raphael Kubo da Costa <rakuco@FreeBSD.org> Cross-platform processor core counter: fix build on FreeBSD. https://bugs.webkit.org/show_bug.cgi?id=81482 Reviewed by Zoltan Herczeg. The documentation of sysctl(3) shows that <sys/types.h> should be included before <sys/sysctl.h> (sys/types.h tends to be the first included header in general). This should fix the build on FreeBSD and other systems where sysctl.h really depends on types defined in types.h. * wtf/NumberOfCores.cpp: 2012-03-19 Jessie Berlin <jberlin@apple.com> Windows build fix after r111129. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-03-19 Gavin Barraclough <barraclough@apple.com> JSCallbackFunction::toStringCallback/valueOfCallback do not handle 0 return value from convertToType https://bugs.webkit.org/show_bug.cgi?id=81468 <rdar://problem/11034745> Reviewed by Oliver Hunt. The API specifies that convertToType may opt not to handle a conversion: "@result The objects's converted value, or NULL if the object was not converted." In which case, it would propagate first up the JSClass hierarchy, calling its superclass's conversion functions, and failing that call the JSObject::defaultValue function. Unfortunately this behaviour was removed in bug#69677/bug#69858, and instead we now rely on the toStringCallback/valueOfCallback function introduced in bug#69156. Even after a fix in bug#73368, these will return the result from the first convertToType they find, regardless of whether this result is null, and if no convertToType method is found in the api class hierarchy (possible if toStringCallback/valueOfCallback was accessed off the prototype chain), they will also return a null pointer. This is unsafe. It would be easy to make the approach based around toStringCallback/valueOfCallback continue to walk the api class hierarchy, but making the fallback to defaultValue would be problematic (since defaultValue calls toStringCallback/valueOfCallback, this would infinitely recurse). Making the fallback work with toString/valueOf methods attached to api objects is probably not the right thing to do – instead, we should just implement the defaultValue trap for api objects. In addition, this bug highlights that fact that JSCallbackFunction::call will allow a hard null to be returned from C to JavaScript - this is not okay. Handle with an exception. * API/JSCallbackFunction.cpp: (JSC::JSCallbackFunction::call): - Should be null checking the return value. (JSC): - Remove toStringCallback/valueOfCallback. * API/JSCallbackFunction.h: (JSCallbackFunction): - Remove toStringCallback/valueOfCallback. * API/JSCallbackObject.h: (JSCallbackObject): - Add defaultValue mthods to JSCallbackObject. * API/JSCallbackObjectFunctions.h: (JSC::::defaultValue): - Add defaultValue mthods to JSCallbackObject. * API/JSClassRef.cpp: (OpaqueJSClass::prototype): - Remove toStringCallback/valueOfCallback. * API/tests/testapi.js: - Revert this test, now we no longer artificially introduce a toString method onto the api object. 2012-03-18 Raphael Kubo da Costa <rakuco@FreeBSD.org> [EFL] Include ICU_INCLUDE_DIRS when building. https://bugs.webkit.org/show_bug.cgi?id=81483 Reviewed by Daniel Bates. So far, only the ICU libraries were being included when building JavaScriptCore, however the include path is also needed, otherwise the build will fail when ICU is installed into a non-standard location. * PlatformEfl.cmake: Include ${ICU_INCLUDE_DIRS}. 2012-03-17 Gavin Barraclough <barraclough@apple.com> Strength reduction, RegExp.exec -> RegExp.test https://bugs.webkit.org/show_bug.cgi?id=81459 Reviewed by Sam Weinig. RegExp.prototype.exec & RegExp.prototype.test can both be used to test a regular expression for a match against a string - however exec is more expensive, since it allocates a matches array object. In cases where the result is consumed in a boolean context the allocation of the matches array can be trivially elided. For example: function f() { for (i =0; i < 10000000; ++i) if(!/a/.exec("a")) err = true; } This is a 2.5x speedup on this example microbenchmark loop. In a more advanced form of this optimization, we may be able to avoid allocating the array where access to the array can be observed. * create_hash_table: * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleIntrinsic): * dfg/DFGNode.h: (JSC::DFG::Node::hasHeapPrediction): * dfg/DFGNodeType.h: (DFG): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileRegExpExec): (DFG): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::callOperation): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * jsc.cpp: (GlobalObject::addConstructableFunction): * runtime/Intrinsic.h: * runtime/JSFunction.cpp: (JSC::JSFunction::create): (JSC): * runtime/JSFunction.h: (JSFunction): * runtime/Lookup.cpp: (JSC::setUpStaticFunctionSlot): * runtime/RegExpObject.cpp: (JSC::RegExpObject::exec): (JSC::RegExpObject::match): * runtime/RegExpObject.h: (RegExpObject): * runtime/RegExpPrototype.cpp: (JSC::regExpProtoFuncTest): (JSC::regExpProtoFuncExec): 2012-03-16 Michael Saboff <msaboff@apple.com> Improve diagnostic benefit of JSGlobalData::m_isInitializingObject https://bugs.webkit.org/show_bug.cgi?id=81244 Rubber stamped by Filip Pizlo. Changed type and name of JSGlobalData::m_isInitializingObject to ClassInfo* and m_initializingObjectClass. Changed JSGlobalData::setInitializingObject to JSGlobalData::setInitializingObjectClass. This pointer can be used within the debugger to determine what type of object is being initialized. * runtime/JSCell.h: (JSC::JSCell::finishCreation): (JSC::allocateCell): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: (JSGlobalData): (JSC::JSGlobalData::isInitializingObject): (JSC::JSGlobalData::setInitializingObjectClass): * runtime/Structure.h: (JSC::JSCell::finishCreation): 2012-03-16 Mark Rowe <mrowe@apple.com> Build fix. Do not preserve owner and group information when installing the WTF headers. * JavaScriptCore.xcodeproj/project.pbxproj: 2012-03-15 David Dorwin <ddorwin@chromium.org> Make the array pointer parameters in the Typed Array create() methods const. https://bugs.webkit.org/show_bug.cgi?id=81147 Reviewed by Kenneth Russell. This allows const arrays to be passed to these methods. They use PassRefPtr<Subclass> create(), which already has a const parameter. * wtf/Int16Array.h: (Int16Array): (WTF::Int16Array::create): * wtf/Int32Array.h: (Int32Array): (WTF::Int32Array::create): * wtf/Int8Array.h: (Int8Array): (WTF::Int8Array::create): * wtf/Uint16Array.h: (Uint16Array): (WTF::Uint16Array::create): * wtf/Uint32Array.h: (Uint32Array): (WTF::Uint32Array::create): * wtf/Uint8Array.h: (Uint8Array): (WTF::Uint8Array::create): * wtf/Uint8ClampedArray.h: (Uint8ClampedArray): (WTF::Uint8ClampedArray::create): 2012-03-15 Myles Maxfield <mmaxfield@google.com> CopiedSpace::tryAllocateOversize assumes system page size https://bugs.webkit.org/show_bug.cgi?id=80615 Reviewed by Geoffrey Garen. * heap/CopiedSpace.cpp: (JSC::CopiedSpace::tryAllocateOversize): * heap/CopiedSpace.h: (CopiedSpace): * heap/CopiedSpaceInlineMethods.h: (JSC::CopiedSpace::oversizeBlockFor): * wtf/BumpPointerAllocator.h: (WTF::BumpPointerPool::create): * wtf/StdLibExtras.h: (WTF::roundUpToMultipleOf): 2012-03-15 Mark Hahnenberg <mhahnenberg@apple.com> Fixing Windows build breakage * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-03-15 Patrick Gansterer <paroga@webkit.org> [EFL] Make zlib a general build requirement https://bugs.webkit.org/show_bug.cgi?id=80153 Reviewed by Hajime Morita. After r109538 WebSocket module needs zlib to support deflate-frame extension. * wtf/Platform.h: 2012-03-15 Benjamin Poulain <bpoulain@apple.com> NumericStrings should be inlined https://bugs.webkit.org/show_bug.cgi?id=81183 Reviewed by Gavin Barraclough. NumericStrings is not always inlined. When it is not, the class is not faster than using UString::number() directly. * runtime/NumericStrings.h: (JSC::NumericStrings::add): (JSC::NumericStrings::lookupSmallString): 2012-03-15 Andras Becsi <andras.becsi@nokia.com> Fix ARM build after r110792. Unreviewed build fix. * jit/ExecutableAllocator.h: (JSC::ExecutableAllocator::cacheFlush): Remove superfluous curly brackets. 2012-03-15 Gavin Barraclough <barraclough@apple.com> ARMv7: prefer vmov(gpr,gpr->double) over vmov(gpr->single) https://bugs.webkit.org/show_bug.cgi?id=81256 Reviewed by Oliver Hunt. This is a 0.5% sunspider progression. * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::convertInt32ToDouble): - switch which form of vmov we use. 2012-03-15 YoungTaeck Song <youngtaeck.song@samsung.com> [EFL] Add OwnPtr specialization for Ecore_Timer. https://bugs.webkit.org/show_bug.cgi?id=80119 Reviewed by Hajime Morita. Add an overload for deleteOwnedPtr(Ecore_Timer*) on EFL port. * wtf/OwnPtrCommon.h: (WTF): * wtf/efl/OwnPtrEfl.cpp: (WTF::deleteOwnedPtr): (WTF): 2012-03-15 Hojong Han <hojong.han@samsung.com> Linux has madvise enough to support OSAllocator::commit/decommit https://bugs.webkit.org/show_bug.cgi?id=80505 Reviewed by Geoffrey Garen. * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveUncommitted): (WTF::OSAllocator::commit): (WTF::OSAllocator::decommit): 2012-03-15 Steve Falkenburg <sfalken@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleasePGO.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleasePGOOptimize.vsprops: * JavaScriptCore.vcproj/WTF/copy-files.cmd: * JavaScriptCore.vcproj/jsc/jscReleasePGO.vsprops: 2012-03-15 Steve Falkenburg <sfalken@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.vcproj: 2012-03-15 Kevin Ollivier <kevino@theolliviers.com> Move wx port to using export macros https://bugs.webkit.org/show_bug.cgi?id=77279 Reviewed by Hajime Morita. * wscript: * wtf/Platform.h: 2012-03-14 Benjamin Poulain <bpoulain@apple.com> Avoid StringImpl::getData16SlowCase() when sorting array https://bugs.webkit.org/show_bug.cgi?id=81070 Reviewed by Geoffrey Garen. The function codePointCompare() is used intensively when sorting strings. This patch improves its performance by: -Avoiding character conversion. -Inlining the function. This makes Peacekeeper's arrayCombined test 30% faster. * wtf/text/StringImpl.cpp: * wtf/text/StringImpl.h: (WTF): (WTF::codePointCompare): (WTF::codePointCompare8): (WTF::codePointCompare16): (WTF::codePointCompare8To16): 2012-03-14 Hojong Han <hojong.han@samsung.com> Fix memory allocation failed by fastmalloc https://bugs.webkit.org/show_bug.cgi?id=79614 Reviewed by Geoffrey Garen. Memory allocation failed even if the heap grows successfully. It is wrong to get the span only from the large list after the heap grows, because new span could be added in the normal list. * wtf/FastMalloc.cpp: (WTF::TCMalloc_PageHeap::New): 2012-03-14 Hojong Han <hojong.han@samsung.com> Run cacheFlush page by page to assure of flushing all the requested ranges https://bugs.webkit.org/show_bug.cgi?id=77712 Reviewed by Geoffrey Garen. Current MetaAllocator concept, always coalesces adjacent free spaces, doesn't meet memory management of Linux kernel. In a certain case Linux kernel doesn't regard contiguous virtual memory areas as one but two. Therefore cacheFlush page by page guarantees a flush-requested range. * jit/ExecutableAllocator.h: (JSC::ExecutableAllocator::cacheFlush): 2012-03-14 Oliver Hunt <oliver@apple.com> Make ARMv7 work again https://bugs.webkit.org/show_bug.cgi?id=81157 Reviewed by Geoffrey Garen. We were trying to use the ARMv7 dataRegister as a scratch register in a scenario where we the ARMv7MacroAssembler would also try to use dataRegister for its own nefarious purposes. * assembler/MacroAssembler.h: (JSC::MacroAssembler::store32): * assembler/MacroAssemblerARMv7.h: (MacroAssemblerARMv7): 2012-03-14 Mark Hahnenberg <mhahnenberg@apple.com> Heap::destroy leaks CopiedSpace https://bugs.webkit.org/show_bug.cgi?id=81055 Reviewed by Geoffrey Garen. Added a destroy() function to CopiedSpace that moves all normal size CopiedBlocks from the CopiedSpace to the Heap's list of free blocks as well as deallocates all of the oversize blocks in the CopiedSpace. This function is now called in Heap::destroy(). * heap/CopiedSpace.cpp: (JSC::CopiedSpace::destroy): (JSC): * heap/CopiedSpace.h: (CopiedSpace): * heap/Heap.cpp: (JSC::Heap::destroy): 2012-03-14 Andrew Lo <anlo@rim.com> [BlackBerry] Implement REQUEST_ANIMATION_FRAME_DISPLAY_MONITOR using AnimationFrameRateController https://bugs.webkit.org/show_bug.cgi?id=81000 Enable WTF_USE_REQUEST_ANIMATION_FRAME_DISPLAY_MONITOR for BlackBerry. Reviewed by Antonio Gomes. * wtf/Platform.h: 2012-03-13 Filip Pizlo <fpizlo@apple.com> ValueToInt32 speculation will cause OSR exits even when it does not have to https://bugs.webkit.org/show_bug.cgi?id=81068 <rdar://problem/11043926> Reviewed by Anders Carlsson. Two related changes: 1) ValueToInt32 will now always just defer to the non-speculative path, instead of exiting, if it doesn't know what speculations to perform. 2) ValueToInt32 will speculate boolean if it sees this to be profitable. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateBoolean): (Node): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileValueToInt32): 2012-03-13 Mark Hahnenberg <mhahnenberg@apple.com> More Windows build fixing * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-03-13 Mark Hahnenberg <mhahnenberg@apple.com> Windows build fix * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-03-13 Mark Hahnenberg <mhahnenberg@apple.com> Type conversion of exponential part failed https://bugs.webkit.org/show_bug.cgi?id=80673 Reviewed by Geoffrey Garen. * parser/Lexer.cpp: (JSC::::lex): * runtime/JSGlobalObjectFunctions.cpp: (JSC::parseInt): (JSC): (JSC::jsStrDecimalLiteral): Added another template argument that exposes whether or not we accept trailing junk to clients of jsStrDecimalLiteral. Also added additional template parameter for strtod to allow trailing spaces. (JSC::toDouble): (JSC::parseFloat): Accept trailing junk, as per the ECMA 262 spec (15.1.2.3). * runtime/LiteralParser.cpp: (JSC::::Lexer::lexNumber): * tests/mozilla/expected.html: Update the expected page for run-javascriptcore-tests so that we will run ecma/TypeConversion/9.3.1-3.js as a regression test now. * wtf/dtoa.cpp: (WTF): (WTF::strtod): We also needed to sometimes accept trailing spaces to pass a few other tests that were broken by changing the default allowance of trailing junk in jsStrDecimalLiteral. * wtf/dtoa.h: * wtf/dtoa/double-conversion.cc: When the AdvanceToNonspace function was lifted out of the Chromium codebase, the person porting it only thought to check for spaces when skipping whitespace. A few of our JSC tests check for other types of trailing whitespace, so I've added checks for those here to cover those cases (horizontal tab, vertical tab, carriage return, form feed, and line feed). * wtf/text/WTFString.cpp: (WTF::toDoubleType): Disallow trailing spaces, as this breaks form input verification stuff. 2012-03-13 Filip Pizlo <fpizlo@apple.com> Unreviewed, build fix since is_pod<> includes some header that I didn't know about. Removing the assert for now. * dfg/DFGOperations.h: * llint/LLIntSlowPaths.h: 2012-03-13 Filip Pizlo <fpizlo@apple.com> Functions with C linkage should return POD types https://bugs.webkit.org/show_bug.cgi?id=81061 Reviewed by Mark Rowe. * dfg/DFGOperations.h: * llint/LLIntSlowPaths.h: (LLInt): (SlowPathReturnType): (JSC::LLInt::encodeResult): 2012-03-13 Filip Pizlo <fpizlo@apple.com> Loads from UInt32Arrays should not result in a double up-convert if it isn't necessary https://bugs.webkit.org/show_bug.cgi?id=80979 <rdar://problem/11036848> Reviewed by Oliver Hunt. Also improved DFG IR dumping to include type information in a somewhat more intuitive way. * bytecode/PredictedType.cpp: (JSC::predictionToAbbreviatedString): (JSC): * bytecode/PredictedType.h: (JSC): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileUInt32ToNumber): (JSC::DFG::SpeculativeJIT::compileGetByValOnIntTypedArray): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::forwardSpeculationCheck): 2012-03-13 George Staikos <staikos@webkit.org> The callback is only used if SA_RESTART is defined. Compile it out otherwise to avoid a warning. https://bugs.webkit.org/show_bug.cgi?id=80926 Reviewed by Alexey Proskuryakov. * heap/MachineStackMarker.cpp: (JSC): 2012-03-13 Hojong Han <hojong.han@samsung.com> Dump the generated code for ARM_TRADITIONAL https://bugs.webkit.org/show_bug.cgi?id=80975 Reviewed by Gavin Barraclough. * assembler/LinkBuffer.h: (JSC::LinkBuffer::dumpCode): 2012-03-13 Adam Barth <abarth@webkit.org> && Benjamin Poulain <bpoulain@apple.com> Always enable ENABLE(CLIENT_BASED_GEOLOCATION) https://bugs.webkit.org/show_bug.cgi?id=78853 Reviewed by Adam Barth. * Configurations/FeatureDefines.xcconfig: * wtf/Platform.h: 2012-03-13 Kwonjin Jeong <gram@company100.net> Remove SlotVisitor::copy() method. https://bugs.webkit.org/show_bug.cgi?id=80973 Reviewed by Geoffrey Garen. SlotVisitor::copy() method isn't called anywhere. * heap/MarkStack.cpp: Remove definition of SlotVisitor::copy() method. * heap/SlotVisitor.h: Remove declaration of SlotVisitor::copy() method. 2012-03-12 Hojong Han <hojong.han@samsung.com> Fix test cases for RegExp multiline https://bugs.webkit.org/show_bug.cgi?id=80822 Reviewed by Gavin Barraclough. * tests/mozilla/js1_2/regexp/RegExp_multiline.js: * tests/mozilla/js1_2/regexp/RegExp_multiline_as_array.js: * tests/mozilla/js1_2/regexp/beginLine.js: * tests/mozilla/js1_2/regexp/endLine.js: 2012-03-12 Filip Pizlo <fpizlo@apple.com> Arithmetic use inference should be procedure-global and should run in tandem with type propagation https://bugs.webkit.org/show_bug.cgi?id=80819 <rdar://problem/11034006> Reviewed by Gavin Barraclough. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * dfg/DFGArithNodeFlagsInferencePhase.cpp: Removed. * dfg/DFGArithNodeFlagsInferencePhase.h: Removed. * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::isNotNegZero): (PredictionPropagationPhase): (JSC::DFG::PredictionPropagationPhase::isNotZero): (JSC::DFG::PredictionPropagationPhase::propagate): (JSC::DFG::PredictionPropagationPhase::mergeDefaultArithFlags): * dfg/DFGVariableAccessData.h: (JSC::DFG::VariableAccessData::VariableAccessData): (JSC::DFG::VariableAccessData::flags): (VariableAccessData): (JSC::DFG::VariableAccessData::mergeFlags): 2012-03-12 Filip Pizlo <fpizlo@apple.com> Node::op and Node::flags should be private https://bugs.webkit.org/show_bug.cgi?id=80824 <rdar://problem/11033435> Reviewed by Gavin Barraclough. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::initialize): (JSC::DFG::AbstractState::execute): (JSC::DFG::AbstractState::mergeStateAtTail): (JSC::DFG::AbstractState::mergeToSuccessors): * dfg/DFGArithNodeFlagsInferencePhase.cpp: (JSC::DFG::ArithNodeFlagsInferencePhase::propagate): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::injectLazyOperandPrediction): (JSC::DFG::ByteCodeParser::getLocal): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::flushArgument): (JSC::DFG::ByteCodeParser::toInt32): (JSC::DFG::ByteCodeParser::isJSConstant): (JSC::DFG::ByteCodeParser::makeSafe): (JSC::DFG::ByteCodeParser::makeDivSafe): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::processPhiStack): (JSC::DFG::ByteCodeParser::linkBlock): * dfg/DFGCFAPhase.cpp: (JSC::DFG::CFAPhase::performBlockCFA): * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::canonicalize): (JSC::DFG::CSEPhase::endIndexForPureCSE): (JSC::DFG::CSEPhase::pureCSE): (JSC::DFG::CSEPhase::byValIsPure): (JSC::DFG::CSEPhase::clobbersWorld): (JSC::DFG::CSEPhase::impureCSE): (JSC::DFG::CSEPhase::globalVarLoadElimination): (JSC::DFG::CSEPhase::getByValLoadElimination): (JSC::DFG::CSEPhase::checkFunctionElimination): (JSC::DFG::CSEPhase::checkStructureLoadElimination): (JSC::DFG::CSEPhase::getByOffsetLoadElimination): (JSC::DFG::CSEPhase::getPropertyStorageLoadElimination): (JSC::DFG::CSEPhase::getIndexedPropertyStorageLoadElimination): (JSC::DFG::CSEPhase::getScopeChainLoadElimination): (JSC::DFG::CSEPhase::performNodeCSE): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): (DFG): * dfg/DFGGraph.h: (JSC::DFG::Graph::addShouldSpeculateInteger): (JSC::DFG::Graph::negateShouldSpeculateInteger): (JSC::DFG::Graph::methodOfGettingAValueProfileFor): * dfg/DFGNode.cpp: Removed. * dfg/DFGNode.h: (DFG): (JSC::DFG::Node::Node): (Node): (JSC::DFG::Node::op): (JSC::DFG::Node::flags): (JSC::DFG::Node::setOp): (JSC::DFG::Node::setFlags): (JSC::DFG::Node::mergeFlags): (JSC::DFG::Node::filterFlags): (JSC::DFG::Node::clearFlags): (JSC::DFG::Node::setOpAndDefaultFlags): (JSC::DFG::Node::mustGenerate): (JSC::DFG::Node::isConstant): (JSC::DFG::Node::isWeakConstant): (JSC::DFG::Node::valueOfJSConstant): (JSC::DFG::Node::hasVariableAccessData): (JSC::DFG::Node::hasIdentifier): (JSC::DFG::Node::resolveGlobalDataIndex): (JSC::DFG::Node::hasArithNodeFlags): (JSC::DFG::Node::arithNodeFlags): (JSC::DFG::Node::setArithNodeFlag): (JSC::DFG::Node::mergeArithNodeFlags): (JSC::DFG::Node::hasConstantBuffer): (JSC::DFG::Node::hasRegexpIndex): (JSC::DFG::Node::hasVarNumber): (JSC::DFG::Node::hasScopeChainDepth): (JSC::DFG::Node::hasResult): (JSC::DFG::Node::hasInt32Result): (JSC::DFG::Node::hasNumberResult): (JSC::DFG::Node::hasJSResult): (JSC::DFG::Node::hasBooleanResult): (JSC::DFG::Node::isJump): (JSC::DFG::Node::isBranch): (JSC::DFG::Node::isTerminal): (JSC::DFG::Node::hasHeapPrediction): (JSC::DFG::Node::hasFunctionCheckData): (JSC::DFG::Node::hasStructureTransitionData): (JSC::DFG::Node::hasStructureSet): (JSC::DFG::Node::hasStorageAccessData): (JSC::DFG::Node::hasFunctionDeclIndex): (JSC::DFG::Node::hasFunctionExprIndex): (JSC::DFG::Node::child1): (JSC::DFG::Node::child2): (JSC::DFG::Node::child3): (JSC::DFG::Node::firstChild): (JSC::DFG::Node::numChildren): * dfg/DFGNodeFlags.cpp: Copied from Source/JavaScriptCore/dfg/DFGNode.cpp. * dfg/DFGNodeFlags.h: Added. (DFG): (JSC::DFG::nodeUsedAsNumber): (JSC::DFG::nodeCanTruncateInteger): (JSC::DFG::nodeCanIgnoreNegativeZero): (JSC::DFG::nodeMayOverflow): (JSC::DFG::nodeCanSpeculateInteger): * dfg/DFGNodeType.h: Added. (DFG): (JSC::DFG::defaultFlags): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): (JSC::DFG::PredictionPropagationPhase::vote): (JSC::DFG::PredictionPropagationPhase::doRoundOfDoubleVoting): (JSC::DFG::PredictionPropagationPhase::fixupNode): * dfg/DFGRedundantPhiEliminationPhase.cpp: (JSC::DFG::RedundantPhiEliminationPhase::run): (JSC::DFG::RedundantPhiEliminationPhase::replacePhiChild): (JSC::DFG::RedundantPhiEliminationPhase::updateBlockVariableInformation): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::useChildren): (JSC::DFG::SpeculativeJIT::compilePeepHoleBranch): (JSC::DFG::SpeculativeJIT::compileMovHint): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): (JSC::DFG::SpeculativeJIT::compileUInt32ToNumber): (JSC::DFG::SpeculativeJIT::compileAdd): (JSC::DFG::SpeculativeJIT::compare): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::detectPeepHoleBranch): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGVirtualRegisterAllocationPhase.cpp: (JSC::DFG::VirtualRegisterAllocationPhase::run): 2012-03-12 Laszlo Gombos <laszlo.1.gombos@nokia.com> Minor DataLog fixes https://bugs.webkit.org/show_bug.cgi?id=80826 Reviewed by Andreas Kling. * bytecode/ExecutionCounter.cpp: Do not include DataLog.h, it is not used. * jit/ExecutableAllocator.cpp: Ditto. * wtf/DataLog.cpp: (WTF::initializeLogFileOnce): Add missing semi-colon to the code path where DATA_LOG_FILENAME is defined. * wtf/HashTable.cpp: Include DataLog as it is used. 2012-03-12 SangGyu Lee <sg5.lee@samsung.com> Integer overflow check code in arithmetic operation in classic interpreter https://bugs.webkit.org/show_bug.cgi?id=80465 Reviewed by Gavin Barraclough. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): 2012-03-12 Zeno Albisser <zeno@webkit.org> [Qt][Mac] Build fails after enabling LLINT when JIT is disabled (r109863) https://bugs.webkit.org/show_bug.cgi?id=80827 Qt on Mac uses OS(DARWIN) as well, but we do not want to enable LLINT. Reviewed by Simon Hausmann. * wtf/Platform.h: 2012-03-12 Simon Hausmann <simon.hausmann@nokia.com> Unreviewed prospective Qt/Mac build fix * runtime/JSGlobalData.cpp: use #USE(CF) instead of PLATFORM(MAC) to determine whether to include CoreFoundation headers, used for JIT configuration in JSGlobalData constructor. 2012-03-12 Filip Pizlo <fpizlo@apple.com> All DFG nodes should have a mutable set of flags https://bugs.webkit.org/show_bug.cgi?id=80779 <rdar://problem/11026218> Reviewed by Gavin Barraclough. Got rid of NodeId, and placed all of the flags that distinguished NodeId from NodeType into a separate Node::flags field. Combined what was previously ArithNodeFlags into Node::flags. In the process of debugging, I found that the debug support in the virtual register allocator was lacking, so I improved it. I also realized that the virtual register allocator was assuming that the nodes in a basic block were contiguous, which is no longer the case. So I fixed that. The fix also made it natural to have more extreme assertions, so I added them. I suspect this will make it easier to catch virtual register allocation bugs in the future. This is mostly performance neutral; if anything it looks like a slight speed-up. This patch does leave some work for future refactorings; for example, Node::op is unencapsulated. This was already the case, though now it feels even more like it should be. I avoided doing that because this patch has already grown way bigger than I wanted. Finally, this patch creates a DFGNode.cpp file and makes a slight effort to move some unnecessarily inline stuff out of DFGNode.h. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * dfg/DFGArithNodeFlagsInferencePhase.cpp: (JSC::DFG::ArithNodeFlagsInferencePhase::propagate): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::addToGraph): (JSC::DFG::ByteCodeParser::makeSafe): (JSC::DFG::ByteCodeParser::makeDivSafe): (JSC::DFG::ByteCodeParser::handleMinMax): (JSC::DFG::ByteCodeParser::handleIntrinsic): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCFAPhase.cpp: (JSC::DFG::CFAPhase::performBlockCFA): * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::endIndexForPureCSE): (JSC::DFG::CSEPhase::pureCSE): (JSC::DFG::CSEPhase::clobbersWorld): (JSC::DFG::CSEPhase::impureCSE): (JSC::DFG::CSEPhase::setReplacement): (JSC::DFG::CSEPhase::eliminate): (JSC::DFG::CSEPhase::performNodeCSE): (JSC::DFG::CSEPhase::performBlockCSE): (CSEPhase): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::opName): (JSC::DFG::Graph::dump): (DFG): * dfg/DFGNode.cpp: Added. (DFG): (JSC::DFG::arithNodeFlagsAsString): * dfg/DFGNode.h: (DFG): (JSC::DFG::nodeUsedAsNumber): (JSC::DFG::nodeCanTruncateInteger): (JSC::DFG::nodeCanIgnoreNegativeZero): (JSC::DFG::nodeMayOverflow): (JSC::DFG::nodeCanSpeculateInteger): (JSC::DFG::defaultFlags): (JSC::DFG::Node::Node): (Node): (JSC::DFG::Node::setOpAndDefaultFlags): (JSC::DFG::Node::mustGenerate): (JSC::DFG::Node::arithNodeFlags): (JSC::DFG::Node::setArithNodeFlag): (JSC::DFG::Node::mergeArithNodeFlags): (JSC::DFG::Node::hasResult): (JSC::DFG::Node::hasInt32Result): (JSC::DFG::Node::hasNumberResult): (JSC::DFG::Node::hasJSResult): (JSC::DFG::Node::hasBooleanResult): (JSC::DFG::Node::isJump): (JSC::DFG::Node::isBranch): (JSC::DFG::Node::isTerminal): (JSC::DFG::Node::child1): (JSC::DFG::Node::child2): (JSC::DFG::Node::child3): (JSC::DFG::Node::firstChild): (JSC::DFG::Node::numChildren): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): (JSC::DFG::PredictionPropagationPhase::vote): (JSC::DFG::PredictionPropagationPhase::fixupNode): * dfg/DFGScoreBoard.h: (ScoreBoard): (JSC::DFG::ScoreBoard::~ScoreBoard): (JSC::DFG::ScoreBoard::assertClear): (JSC::DFG::ScoreBoard::use): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::useChildren): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGVirtualRegisterAllocationPhase.cpp: (JSC::DFG::VirtualRegisterAllocationPhase::run): 2012-03-10 Filip Pizlo <fpizlo@apple.com> LLInt should support JSVALUE64 https://bugs.webkit.org/show_bug.cgi?id=79609 <rdar://problem/10063437> Reviewed by Gavin Barraclough and Oliver Hunt. Ported the LLInt, which previously only worked on 32-bit, to 64-bit. This patch moves a fair bit of code from LowLevelInterpreter32_64.asm to the common file, LowLevelInterpreter.asm. About 1/3 of the LLInt did not have to be specialized for value representation. Also made some minor changes to offlineasm and the slow-paths. * llint/LLIntData.cpp: (JSC::LLInt::Data::performAssertions): * llint/LLIntEntrypoints.cpp: * llint/LLIntSlowPaths.cpp: (LLInt): (JSC::LLInt::llint_trace_value): (JSC::LLInt::LLINT_SLOW_PATH_DECL): (JSC::LLInt::jitCompileAndSetHeuristics): * llint/LLIntSlowPaths.h: (LLInt): (SlowPathReturnType): (JSC::LLInt::SlowPathReturnType::SlowPathReturnType): (JSC::LLInt::encodeResult): * llint/LLIntThunks.cpp: * llint/LowLevelInterpreter.asm: * llint/LowLevelInterpreter32_64.asm: * llint/LowLevelInterpreter64.asm: * offlineasm/armv7.rb: * offlineasm/asm.rb: * offlineasm/ast.rb: * offlineasm/backends.rb: * offlineasm/instructions.rb: * offlineasm/parser.rb: * offlineasm/registers.rb: * offlineasm/transform.rb: * offlineasm/x86.rb: * wtf/Platform.h: 2012-03-10 Yong Li <yoli@rim.com> Web Worker crashes with WX_EXCLUSIVE https://bugs.webkit.org/show_bug.cgi?id=80532 Let each JS global object own a meta allocator for WX_EXCLUSIVE to avoid conflicts from Web Worker. Also fix a mutex leak in MetaAllocator's dtor. Reviewed by Filip Pizlo. * jit/ExecutableAllocator.cpp: (JSC::DemandExecutableAllocator::DemandExecutableAllocator): (JSC::DemandExecutableAllocator::~DemandExecutableAllocator): (JSC::DemandExecutableAllocator::bytesAllocatedByAllAllocators): (DemandExecutableAllocator): (JSC::DemandExecutableAllocator::bytesCommittedByAllocactors): (JSC::DemandExecutableAllocator::dumpProfileFromAllAllocators): (JSC::DemandExecutableAllocator::allocateNewSpace): (JSC::DemandExecutableAllocator::allocators): (JSC::DemandExecutableAllocator::allocatorsMutex): (JSC): (JSC::ExecutableAllocator::initializeAllocator): (JSC::ExecutableAllocator::ExecutableAllocator): (JSC::ExecutableAllocator::underMemoryPressure): (JSC::ExecutableAllocator::memoryPressureMultiplier): (JSC::ExecutableAllocator::allocate): (JSC::ExecutableAllocator::committedByteCount): (JSC::ExecutableAllocator::dumpProfile): * jit/ExecutableAllocator.h: (JSC): (ExecutableAllocator): (JSC::ExecutableAllocator::allocator): * wtf/MetaAllocator.h: (WTF::MetaAllocator::~MetaAllocator): Finalize the spin lock. * wtf/TCSpinLock.h: (TCMalloc_SpinLock::Finalize): Add empty Finalize() to some implementations. 2012-03-09 Gavin Barraclough <barraclough@apple.com> Object.freeze broken on latest Nightly https://bugs.webkit.org/show_bug.cgi?id=80577 Reviewed by Oliver Hunt. The problem here is that deleteProperty rejects deletion of prototype. This is correct in most cases, however defineOwnPropery is presently implemented internally to ensure the attributes change by deleting the old property, and creating a new one. * runtime/JSFunction.cpp: (JSC::JSFunction::deleteProperty): - If deletePropery is called via defineOwnPropery, allow old prototype to be removed. 2012-03-09 Gavin Barraclough <barraclough@apple.com> Array.prototype.toLocaleString visits elements in wrong order under certain conditions https://bugs.webkit.org/show_bug.cgi?id=80663 Reviewed by Michael Saboff. The bug here is actually that we're continuing to process the array after an exception has been thrown, and that the second value throw is overriding the first. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncToLocaleString): 2012-03-09 Ryosuke Niwa <rniwa@webkit.org> WebKit compiled by gcc (Xcode 3.2.6) hangs while running DOM/Accessors.html https://bugs.webkit.org/show_bug.cgi?id=80080 Reviewed by Filip Pizlo. * bytecode/SamplingTool.cpp: (JSC::SamplingRegion::Locker::Locker): (JSC::SamplingRegion::Locker::~Locker): * bytecode/SamplingTool.h: (JSC::SamplingRegion::exchangeCurrent): * wtf/Atomics.h: (WTF): (WTF::weakCompareAndSwap): (WTF::weakCompareAndSwapUIntPtr): 2012-03-09 Gavin Barraclough <barraclough@apple.com> REGRESSION: Date.parse("Tue Nov 23 20:40:05 2010 GMT") returns NaN https://bugs.webkit.org/show_bug.cgi?id=49989 Reviewed by Oliver Hunt. Patch originally by chris reiss <christopher.reiss@nokia.com>, allow the year to appear before the timezone in date strings. * wtf/DateMath.cpp: (WTF::parseDateFromNullTerminatedCharacters): 2012-03-09 Mark Rowe <mrowe@apple.com> Ensure that the WTF headers are copied at installhdrs time. Reviewed by Dan Bernstein and Jessie Berlin. * Configurations/JavaScriptCore.xcconfig: Set INSTALLHDRS_SCRIPT_PHASE = YES so that our script phases are invoked at installhdrs time. The only one that does any useful work at that time is the one that installs WTF headers. 2012-03-09 Jon Lee <jonlee@apple.com> Add support for ENABLE(LEGACY_NOTIFICATIONS) https://bugs.webkit.org/show_bug.cgi?id=80497 Reviewed by Adam Barth. Prep for b80472: Update API for Web Notifications * Configurations/FeatureDefines.xcconfig: 2012-03-09 Ashod Nakashian <ashodnakashian@yahoo.com> Bash scripts should support LF endings only https://bugs.webkit.org/show_bug.cgi?id=79509 Reviewed by David Kilzer. * gyp/generate-derived-sources.sh: Added property svn:eol-style. * gyp/run-if-exists.sh: Added property svn:eol-style. * gyp/update-info-plist.sh: Added property svn:eol-style. 2012-03-09 Jessie Berlin <jberlin@apple.com> Windows debug build fix. * assembler/MacroAssembler.h: (JSC::MacroAssembler::shouldBlind): Fix unreachable code warnings (which we treat as errors). 2012-03-09 Thouraya ANDOLSI <thouraya.andolsi@st.com> Reviewed by Zoltan Herczeg. [Qt] Fix the SH4 build after r109834 https://bugs.webkit.org/show_bug.cgi?id=80492 * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::branchAdd32): (JSC::MacroAssemblerSH4::branchSub32): 2012-03-09 Andy Wingo <wingo@igalia.com> Refactor code feature analysis in the parser https://bugs.webkit.org/show_bug.cgi?id=79112 Reviewed by Geoffrey Garen. This commit refactors the parser to more uniformly propagate flag bits down and up the parse process, as the parser descends and returns into nested blocks. Some flags get passed town to subscopes, some apply to specific scopes only, and some get unioned up after parsing subscopes. The goal is to eventually be very precise with scoping information, once we have block scopes: one block scope might use `eval', which would require the emission of a symbol table within that block and containing blocks, whereas another block in the same function might not, allowing us to not emit a symbol table. * parser/Nodes.h: (JSC::ScopeFlags): Rename from CodeFeatures. (JSC::ScopeNode::addScopeFlags): (JSC::ScopeNode::scopeFlags): New accessors for m_scopeFlags. (JSC::ScopeNode::isStrictMode): (JSC::ScopeNode::usesEval): (JSC::ScopeNode::usesArguments): (JSC::ScopeNode::setUsesArguments): (JSC::ScopeNode::usesThis): (JSC::ScopeNode::needsActivationForMoreThanVariables): (JSC::ScopeNode::needsActivation): Refactor these accessors to operate on the m_scopeFlags member. (JSC::ScopeNode::source): (JSC::ScopeNode::sourceURL): (JSC::ScopeNode::sourceID): Shuffle these definitions around; no semantic change. (JSC::ScopeNode::ScopeNode) (JSC::ProgramNode::ProgramNode) (JSC::EvalNode::EvalNode) (JSC::FunctionBodyNode::FunctionBodyNode): Have these constructors take a ScopeFlags as an argument, instead of a bool inStrictContext. * parser/Nodes.cpp: (JSC::ScopeNode::ScopeNode): (JSC::ProgramNode::ProgramNode): (JSC::ProgramNode::create): (JSC::EvalNode::EvalNode): (JSC::EvalNode::create): (JSC::FunctionBodyNode::FunctionBodyNode): (JSC::FunctionBodyNode::create): Adapt constructors to change. * parser/ASTBuilder.h: (JSC::ASTBuilder::ASTBuilder): (JSC::ASTBuilder::thisExpr): (JSC::ASTBuilder::createResolve): (JSC::ASTBuilder::createFunctionBody): (JSC::ASTBuilder::createFuncDeclStatement): (JSC::ASTBuilder::createTryStatement): (JSC::ASTBuilder::createWithStatement): (JSC::ASTBuilder::addVar): (JSC::ASTBuilder::Scope::Scope): (Scope): (ASTBuilder): (JSC::ASTBuilder::makeFunctionCallNode): Don't track scope features here. Instead rely on the base Parser mechanism to track features. * parser/NodeInfo.h (NodeInfo, NodeDeclarationInfo): "ScopeFlags". * parser/Parser.h: (JSC::Scope::Scope): Manage scope through flags, not bit-booleans. This lets us uniformly propagate them up and down. (JSC::Scope::declareWrite): (JSC::Scope::declareParameter): (JSC::Scope::useVariable): (JSC::Scope::collectFreeVariables): (JSC::Scope::getCapturedVariables): (JSC::Scope::saveFunctionInfo): (JSC::Scope::restoreFunctionInfo): (JSC::Parser::pushScope): Adapt to use scope flags and their accessors instead of bit-booleans. * parser/Parser.cpp: (JSC::::Parser): (JSC::::parseInner): (JSC::::didFinishParsing): (JSC::::parseSourceElements): (JSC::::parseVarDeclarationList): (JSC::::parseConstDeclarationList): (JSC::::parseWithStatement): (JSC::::parseTryStatement): (JSC::::parseFunctionBody): (JSC::::parseFunctionInfo): (JSC::::parseFunctionDeclaration): (JSC::::parsePrimaryExpression): Hoist some of the flag handling out of the "context" (ASTBuilder or SyntaxChecker) and to here. Does not seem to have a performance impact. * parser/SourceProviderCacheItem.h (SourceProviderCacheItem): Cache the scopeflags. * parser/SyntaxChecker.h: Remove evalCount() decl. * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::produceCodeBlockFor): * runtime/Executable.h: (JSC::ScriptExecutable::ScriptExecutable): (JSC::ScriptExecutable::usesEval): (JSC::ScriptExecutable::usesArguments): (JSC::ScriptExecutable::needsActivation): (JSC::ScriptExecutable::isStrictMode): (JSC::ScriptExecutable::recordParse): (ScriptExecutable): ScopeFlags, not features. 2012-03-08 Benjamin Poulain <bpoulain@apple.com> Build fix for MSVC after r110266 Unreviewed. A #ifdef for MSVC was left over in r110266. * runtime/RegExpObject.h: (RegExpObject): 2012-03-08 Benjamin Poulain <bpoulain@apple.com> Allocate the RegExpObject's data with the Cell https://bugs.webkit.org/show_bug.cgi?id=80654 Reviewed by Gavin Barraclough. This patch removes the creation of RegExpObject's data to avoid the overhead create by the allocation and destruction. We RegExp are created repeatedly, this provides some performance improvment. The PeaceKeeper test stringDetectBrowser improves by 10%. * runtime/RegExpObject.cpp: (JSC::RegExpObject::RegExpObject): (JSC::RegExpObject::visitChildren): (JSC::RegExpObject::getOwnPropertyDescriptor): (JSC::RegExpObject::defineOwnProperty): (JSC::RegExpObject::match): * runtime/RegExpObject.h: (JSC::RegExpObject::setRegExp): (JSC::RegExpObject::regExp): (JSC::RegExpObject::setLastIndex): (JSC::RegExpObject::getLastIndex): (RegExpObject): 2012-03-08 Steve Falkenburg <sfalken@apple.com> Separate WTF parts of JavaScriptCoreGenerated into WTFGenerated for Windows build https://bugs.webkit.org/show_bug.cgi?id=80657 Preparation for WTF separation from JavaScriptCore. The "Generated" vcproj files on Windows are necessary so Visual Studio can calculate correct dependencies for generated files. This also removes the PGO build targets from the WTF code, since we can't build instrumentation/optimization versions of the WTF code independent of the JavaScriptCore code. Reviewed by Jessie Berlin. * JavaScriptCore.vcproj/JavaScriptCore.sln: Add WTFGenerated, update dependent projects. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.make: Removed WTF specific parts. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.vcproj: Removed WTF specific parts. * JavaScriptCore.vcproj/JavaScriptCore/build-generated-files.sh: Removed WTF specific parts. * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: Removed WTF specific parts. * JavaScriptCore.vcproj/JavaScriptCore/work-around-vs-dependency-tracking-bugs.py: Removed. * JavaScriptCore.vcproj/JavaScriptCoreSubmit.sln: Add WTFGenerated, update dependent projects. * JavaScriptCore.vcproj/WTF/WTF.vcproj: Remove PGO targets from WTF. * JavaScriptCore.vcproj/WTF/WTFGenerated.make: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.make. * JavaScriptCore.vcproj/WTF/WTFGenerated.vcproj: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.vcproj. * JavaScriptCore.vcproj/WTF/WTFGeneratedCommon.vsprops: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedCommon.vsprops. * JavaScriptCore.vcproj/WTF/WTFGeneratedDebug.vsprops: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedDebug.vsprops. * JavaScriptCore.vcproj/WTF/WTFGeneratedDebugAll.vsprops: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedDebugAll.vsprops. * JavaScriptCore.vcproj/WTF/WTFGeneratedDebugCairoCFLite.vsprops: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedDebugCairoCFLite.vsprops. * JavaScriptCore.vcproj/WTF/WTFGeneratedProduction.vsprops: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedProduction.vsprops. * JavaScriptCore.vcproj/WTF/WTFGeneratedRelease.vsprops: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedRelease.vsprops. * JavaScriptCore.vcproj/WTF/WTFGeneratedReleaseCairoCFLite.vsprops: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedReleaseCairoCFLite.vsprops. * JavaScriptCore.vcproj/WTF/WTFReleasePGO.vsprops: Removed. * JavaScriptCore.vcproj/WTF/build-generated-files.sh: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/build-generated-files.sh. * JavaScriptCore.vcproj/WTF/copy-files.cmd: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd. * JavaScriptCore.vcproj/WTF/work-around-vs-dependency-tracking-bugs.py: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/work-around-vs-dependency-tracking-bugs.py. 2012-03-08 Benjamin Poulain <benjamin@webkit.org> Fix the build of WebKit with WTFURL following the removal of ForwardingHeaders/wtf https://bugs.webkit.org/show_bug.cgi?id=80652 Reviewed by Eric Seidel. Fix the header, URLSegments.h is not part of the API. * wtf/url/api/ParsedURL.h: 2012-03-08 Ryosuke Niwa <rniwa@webkit.org> Mac build fix for micro data API. * Configurations/FeatureDefines.xcconfig: 2012-03-08 Gavin Barraclough <barraclough@apple.com> String.prototype.match and replace do not clear global regexp lastIndex per ES5.1 15.5.4.10 https://bugs.webkit.org/show_bug.cgi?id=26890 Reviewed by Oliver Hunt. Per 15.10.6.2 step 9.a.1 called via the action of the last iteration of 15.5.4.10 8.f.i. * runtime/StringPrototype.cpp: (JSC::replaceUsingRegExpSearch): (JSC::stringProtoFuncMatch): - added calls to setLastIndex. 2012-03-08 Matt Lilek <mrl@apple.com> Don't enable VIDEO_TRACK on all OS X platforms https://bugs.webkit.org/show_bug.cgi?id=80635 Reviewed by Eric Carlson. * Configurations/FeatureDefines.xcconfig: 2012-03-08 Oliver Hunt <oliver@apple.com> Build fix. That day is not today. * assembler/MacroAssembler.h: (JSC::MacroAssembler::shouldBlind): * assembler/MacroAssemblerX86Common.h: (MacroAssemblerX86Common): (JSC::MacroAssemblerX86Common::shouldBlindForSpecificArch): 2012-03-08 Oliver Hunt <oliver@apple.com> Build fix. One of these days I'll manage to commit something that works everywhere. * assembler/AbstractMacroAssembler.h: (AbstractMacroAssembler): * assembler/MacroAssemblerARMv7.h: (MacroAssemblerARMv7): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::shouldBlindForSpecificArch): (MacroAssemblerX86Common): 2012-03-08 Chao-ying Fu <fu@mips.com> Update MIPS patchOffsetGetByIdSlowCaseCall https://bugs.webkit.org/show_bug.cgi?id=80302 Reviewed by Oliver Hunt. * jit/JIT.h: (JIT): 2012-03-08 Oliver Hunt <oliver@apple.com> Missing some places where we should be blinding 64bit values (and blinding something we shouldn't) https://bugs.webkit.org/show_bug.cgi?id=80633 Reviewed by Gavin Barraclough. Add 64-bit trap for shouldBlindForSpecificArch, so that we always blind if there isn't a machine specific implementation (otherwise the 64bit value got truncated and 32bit checks were used -- leaving 32bits untested). Also add a bit of logic to ensure that we don't try to blind a few common constants that go through the ImmPtr paths -- encoded numeric JSValues and unencoded doubles with common "safe" values. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::shouldBlindForSpecificArch): * assembler/MacroAssembler.h: (JSC::MacroAssembler::shouldBlindDouble): (MacroAssembler): (JSC::MacroAssembler::shouldBlind): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::shouldBlindForSpecificArch): 2012-03-08 Mark Rowe <mrowe@apple.com> <rdar://problem/11012572> Ensure that the staged frameworks path is in the search path for JavaScriptCore Reviewed by Dan Bernstein. * Configurations/Base.xcconfig: 2012-03-08 Steve Falkenburg <sfalken@apple.com> Fix line endings for copy-files.cmd. If a cmd file doesn't have Windows line endings, it doesn't work properly. In this case, the label :clean wasn't found, breaking the clean build. Reviewed by Jessie Berlin. * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: 2012-03-07 Filip Pizlo <fpizlo@apple.com> DFG CFA incorrectly handles ValueToInt32 https://bugs.webkit.org/show_bug.cgi?id=80568 Reviewed by Gavin Barraclough. Changed it match exactly the decision pattern used in DFG::SpeculativeJIT::compileValueToInt32 * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): 2012-03-08 Viatcheslav Ostapenko <ostapenko.viatcheslav@nokia.com> [Qt] [WK2] Webkit fails to link when compiled with force_static_libs_as_shared https://bugs.webkit.org/show_bug.cgi?id=80524 Reviewed by Simon Hausmann. Move IdentifierTable methods defintion to WTFThreadData.cpp to fix linking of WTF library. * runtime/Identifier.cpp: * wtf/WTFThreadData.cpp: (JSC): (JSC::IdentifierTable::~IdentifierTable): (JSC::IdentifierTable::add): 2012-03-08 Filip Pizlo <fpizlo@apple.com> DFG instruction count threshold should be lifted to 10000 https://bugs.webkit.org/show_bug.cgi?id=80579 Reviewed by Gavin Barraclough. * runtime/Options.cpp: (JSC::Options::initializeOptions): 2012-03-07 Filip Pizlo <fpizlo@apple.com> Incorrect tracking of abstract values of variables forced double https://bugs.webkit.org/show_bug.cgi?id=80566 <rdar://problem/11001442> Reviewed by Gavin Barraclough. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::mergeStateAtTail): 2012-03-07 Chao-yng Fu <fu@mips.com> [Qt] Fix the MIPS/SH4 build after r109834 https://bugs.webkit.org/show_bug.cgi?id=80492 Reviewed by Oliver Hunt. Implement three-argument branch(Add,Sub)32. * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::add32): (MacroAssemblerMIPS): (JSC::MacroAssemblerMIPS::sub32): (JSC::MacroAssemblerMIPS::branchAdd32): (JSC::MacroAssemblerMIPS::branchSub32): 2012-03-07 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r110127. http://trac.webkit.org/changeset/110127 https://bugs.webkit.org/show_bug.cgi?id=80562 compile failed on AppleWin (Requested by ukai on #webkit). * heap/Heap.cpp: (JSC::Heap::collectAllGarbage): * heap/Heap.h: (JSC): (Heap): * runtime/Executable.cpp: (JSC::FunctionExecutable::FunctionExecutable): (JSC::FunctionExecutable::finalize): * runtime/Executable.h: (FunctionExecutable): (JSC::FunctionExecutable::create): * runtime/JSGlobalData.cpp: (WTF): (Recompiler): (WTF::Recompiler::operator()): (JSC::JSGlobalData::recompileAllJSFunctions): (JSC): * runtime/JSGlobalData.h: (JSGlobalData): * runtime/JSGlobalObject.cpp: (JSC::DynamicGlobalObjectScope::DynamicGlobalObjectScope): 2012-03-07 Hojong Han <hojong.han@samsung.com> The end atom of the marked block considered to filter invalid cells https://bugs.webkit.org/show_bug.cgi?id=79191 Reviewed by Geoffrey Garen. Register file could have stale pointers beyond the end atom of marked block. Those pointers can weasel out of filtering in-middle-of-cell pointer. * heap/MarkedBlock.h: (JSC::MarkedBlock::isLiveCell): 2012-03-07 Jessie Berlin <jberlin@apple.com> Clean Windows build fails after r110033 https://bugs.webkit.org/show_bug.cgi?id=80553 Rubber-stamped by Jon Honeycutt and Eric Seidel. * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: Place the implementation files next to their header files in the wtf/text subdirectory. Use echo -F to tell xcopy that these are files (since there is apparently no flag). * JavaScriptCore.vcproj/jsc/jsc.vcproj: Update the path to those implementation files. * JavaScriptCore.vcproj/testRegExp/testRegExp.vcproj: Ditto. 2012-03-07 Yuqiang Xian <yuqiang.xian@intel.com> Eliminate redundant Phis in DFG https://bugs.webkit.org/show_bug.cgi?id=80415 Reviewed by Filip Pizlo. Although this may not have any advantage at current stage, this is towards minimal SSA to make more high level optimizations (like bug 76770) easier. We have the choices either to build minimal SSA from scratch or to keep current simple Phi insertion mechanism and remove the redundancy in another phase. Currently we choose the latter because the change could be smaller. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGRedundantPhiEliminationPhase.cpp: Added. (DFG): (RedundantPhiEliminationPhase): (JSC::DFG::RedundantPhiEliminationPhase::RedundantPhiEliminationPhase): (JSC::DFG::RedundantPhiEliminationPhase::run): (JSC::DFG::RedundantPhiEliminationPhase::getRedundantReplacement): (JSC::DFG::RedundantPhiEliminationPhase::replacePhiChild): (JSC::DFG::RedundantPhiEliminationPhase::fixupPhis): (JSC::DFG::RedundantPhiEliminationPhase::updateBlockVariableInformation): (JSC::DFG::performRedundantPhiElimination): * dfg/DFGRedundantPhiEliminationPhase.h: Added. (DFG): 2012-03-07 Mark Hahnenberg <mhahnenberg@apple.com> Refactor recompileAllJSFunctions() to be less expensive https://bugs.webkit.org/show_bug.cgi?id=80330 Reviewed by Geoffrey Garen. This change is performance neutral on the JS benchmarks we track. It's mostly to improve page load performance, which currently does at least a couple full GCs per navigation. * heap/Heap.cpp: (JSC::Heap::discardAllCompiledCode): Rename recompileAllJSFunctions to discardAllCompiledCode because the function doesn't actually recompile anything (and never did); it simply throws code away for it to be recompiled later if we determine we should do so. (JSC): (JSC::Heap::collectAllGarbage): (JSC::Heap::addFunctionExecutable): Adds a newly created FunctionExecutable to the Heap's list. (JSC::Heap::removeFunctionExecutable): Removes the specified FunctionExecutable from the Heap's list. * heap/Heap.h: (JSC): (Heap): * runtime/Executable.cpp: Added next and prev fields to FunctionExecutables so that they can be used in DoublyLinkedLists. (JSC::FunctionExecutable::FunctionExecutable): (JSC::FunctionExecutable::finalize): Removes the FunctionExecutable from the Heap's list. * runtime/Executable.h: (FunctionExecutable): (JSC::FunctionExecutable::create): Adds the FunctionExecutable to the Heap's list. * runtime/JSGlobalData.cpp: Remove recompileAllJSFunctions, as it's the Heap's job to own and manage the list of FunctionExecutables. * runtime/JSGlobalData.h: (JSGlobalData): * runtime/JSGlobalObject.cpp: (JSC::DynamicGlobalObjectScope::DynamicGlobalObjectScope): Use the new discardAllCompiledCode. 2012-03-06 Oliver Hunt <oliver@apple.com> Further harden 64-bit JIT https://bugs.webkit.org/show_bug.cgi?id=80457 Reviewed by Filip Pizlo. This patch implements blinding for ImmPtr. Rather than xor based blinding we perform randomised pointer rotations in order to avoid the significant cost in executable memory that would otherwise be necessary (and to avoid the need for an additional scratch register in some cases). As with the prior blinding patch there's a moderate amount of noise as we correct the use of ImmPtr vs. TrustedImmPtr. * assembler/AbstractMacroAssembler.h: (ImmPtr): (JSC::AbstractMacroAssembler::ImmPtr::asTrustedImmPtr): * assembler/MacroAssembler.h: (MacroAssembler): (JSC::MacroAssembler::storePtr): (JSC::MacroAssembler::branchPtr): (JSC::MacroAssembler::shouldBlind): (JSC::MacroAssembler::RotatedImmPtr::RotatedImmPtr): (RotatedImmPtr): (JSC::MacroAssembler::rotationBlindConstant): (JSC::MacroAssembler::loadRotationBlindedConstant): (JSC::MacroAssembler::convertInt32ToDouble): (JSC::MacroAssembler::move): (JSC::MacroAssembler::poke): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::storeDouble): (JSC::MacroAssemblerARMv7::branchAdd32): * assembler/MacroAssemblerX86_64.h: (MacroAssemblerX86_64): (JSC::MacroAssemblerX86_64::rotateRightPtr): (JSC::MacroAssemblerX86_64::xorPtr): * assembler/X86Assembler.h: (X86Assembler): (JSC::X86Assembler::xorq_rm): (JSC::X86Assembler::rorq_i8r): * dfg/DFGCCallHelpers.h: (CCallHelpers): (JSC::DFG::CCallHelpers::setupArgumentsWithExecState): * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::createOSREntries): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::silentFillGPR): (JSC::DFG::SpeculativeJIT::callOperation): (JSC::DFG::SpeculativeJIT::emitEdgeCode): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillInteger): (JSC::DFG::SpeculativeJIT::fillDouble): (JSC::DFG::SpeculativeJIT::fillJSValue): (JSC::DFG::SpeculativeJIT::emitCall): (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitBranch): * jit/JIT.cpp: (JSC::JIT::emitOptimizationCheck): * jit/JITArithmetic32_64.cpp: (JSC::JIT::emitSlow_op_post_inc): * jit/JITInlineMethods.h: (JSC::JIT::emitValueProfilingSite): (JSC::JIT::emitGetVirtualRegister): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_mov): (JSC::JIT::emit_op_new_object): (JSC::JIT::emit_op_strcat): (JSC::JIT::emit_op_ensure_property_exists): (JSC::JIT::emit_op_resolve_skip): (JSC::JIT::emitSlow_op_resolve_global): (JSC::JIT::emit_op_resolve_with_base): (JSC::JIT::emit_op_resolve_with_this): (JSC::JIT::emit_op_jmp_scopes): (JSC::JIT::emit_op_switch_imm): (JSC::JIT::emit_op_switch_char): (JSC::JIT::emit_op_switch_string): (JSC::JIT::emit_op_throw_reference_error): (JSC::JIT::emit_op_debug): (JSC::JIT::emitSlow_op_resolve_global_dynamic): (JSC::JIT::emit_op_new_array): (JSC::JIT::emitSlow_op_new_array): (JSC::JIT::emit_op_new_array_buffer): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_new_object): (JSC::JIT::emit_op_strcat): (JSC::JIT::emit_op_ensure_property_exists): (JSC::JIT::emit_op_resolve_skip): (JSC::JIT::emitSlow_op_resolve_global): (JSC::JIT::emit_op_resolve_with_base): (JSC::JIT::emit_op_resolve_with_this): (JSC::JIT::emit_op_jmp_scopes): (JSC::JIT::emit_op_switch_imm): (JSC::JIT::emit_op_switch_char): (JSC::JIT::emit_op_switch_string): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_put_by_index): * jit/JITStubCall.h: (JITStubCall): (JSC::JITStubCall::addArgument): 2012-03-07 Simon Hausmann <simon.hausmann@nokia.com> ARM build fix. Reviewed by Zoltan Herczeg. Implement three-argument branch(Add,Sub)32. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::add32): (MacroAssemblerARM): (JSC::MacroAssemblerARM::sub32): (JSC::MacroAssemblerARM::branchAdd32): (JSC::MacroAssemblerARM::branchSub32): 2012-03-07 Andy Wingo <wingo@igalia.com> Parser: Inline ScopeNodeData into ScopeNode https://bugs.webkit.org/show_bug.cgi?id=79776 Reviewed by Geoffrey Garen. It used to be that some ScopeNode members were kept in a separate structure because sometimes they wouldn't be needed, and allocating a ParserArena was expensive. This patch makes ParserArena lazily allocate its IdentifierArena, allowing the members to be included directly, which is simpler and easier to reason about. * parser/ParserArena.cpp: (JSC::ParserArena::ParserArena): (JSC::ParserArena::reset): (JSC::ParserArena::isEmpty): * parser/ParserArena.h: (JSC::ParserArena::identifierArena): Lazily allocate the IdentifierArena. * parser/Nodes.cpp: (JSC::ScopeNode::ScopeNode): (JSC::ScopeNode::singleStatement): (JSC::ProgramNode::create): (JSC::EvalNode::create): (JSC::FunctionBodyNode::create): * parser/Nodes.h: (JSC::ScopeNode::destroyData): (JSC::ScopeNode::needsActivationForMoreThanVariables): (JSC::ScopeNode::needsActivation): (JSC::ScopeNode::hasCapturedVariables): (JSC::ScopeNode::capturedVariableCount): (JSC::ScopeNode::captures): (JSC::ScopeNode::varStack): (JSC::ScopeNode::functionStack): (JSC::ScopeNode::neededConstants): (ScopeNode): * bytecompiler/NodesCodegen.cpp: (JSC::ScopeNode::emitStatementsBytecode): Inline ScopeNodeData into ScopeNode. Adapt accessors. 2012-03-06 Eric Seidel <eric@webkit.org> Make WTF public headers use fully-qualified include paths and remove ForwardingHeaders/wtf https://bugs.webkit.org/show_bug.cgi?id=80363 Reviewed by Mark Rowe. Historically WTF has been part of JavaScriptCore, and on Mac and Windows its headers have appeared as part of the "private" headers exported by JavaScriptCore. All of the WTF headers there are "flattened" into a single private headers directory, and WebCore, WebKit and WebKit2 have used "ForwardingHeaders" to re-map fully-qualified <wtf/text/Foo.h> includes to simple <JavaScriptCore/Foo.h> includes. However, very soon, we are moving the WTF source code out of JavaScriptCore into its own directory and project. As part of such, the WTF headers will no longer be part of the JavaScriptCore private interfaces. In preparation for that, this change makes both the Mac and Win builds export WTF headers in a non-flattened manner. On Mac, that means into usr/local/include/wtf (and subdirectories), on Windows for now that means JavaScriptCore/wtf (and subdirectories). There are 5 parts to this change. 1. Updates the JavaScriptCore XCode and VCProj files to actually install these headers (and header directories) into the appropriate places in the build directory. 2. Updates JavaScriptCore.xcodeproj to look for these WTF headers in this install location (WebCore, WebKit, etc. had already been taught to look in previous patches). 3. Fixes all JavaScriptCore source files, and WTF headers to include WTF headers using fully qualified paths. 4. Stops the Mac and Win builds from installing these WTF headers in their old "flattened" location. 5. Removes WebCore and WebKit ForwardingHeaders/wtf directories now that the flattened headers no longer exist. Unfortunately we see no way to do this change in smaller parts, since all of these steps are interdependant. It is possible there are internal Apple projects which depend on JavaScriptCore/Foo.h working for WTF headers, those will have to be updated to use <wtf/Foo.h> after this change. I've discussed this proposed change at length with Mark Rowe, and my understanding is they are ready for (and interested in) this change happening. * API/tests/JSNode.c: * API/tests/JSNodeList.c: * Configurations/Base.xcconfig: * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: * JavaScriptCore.xcodeproj/project.pbxproj: * assembler/MacroAssemblerCodeRef.h: * bytecompiler/BytecodeGenerator.h: * dfg/DFGOperations.cpp: * heap/GCAssertions.h: * heap/HandleHeap.h: * heap/HandleStack.h: * heap/MarkedSpace.h: * heap/PassWeak.h: * heap/Strong.h: * heap/Weak.h: * jit/HostCallReturnValue.cpp: * jit/JIT.cpp: * jit/JITStubs.cpp: * jit/ThunkGenerators.cpp: * parser/Lexer.cpp: * runtime/Completion.cpp: * runtime/Executable.cpp: * runtime/Identifier.h: * runtime/InitializeThreading.cpp: * runtime/JSDateMath.cpp: * runtime/JSGlobalObjectFunctions.cpp: * runtime/JSStringBuilder.h: * runtime/JSVariableObject.h: * runtime/NumberPrototype.cpp: * runtime/WriteBarrier.h: * tools/CodeProfile.cpp: * tools/TieredMMapArray.h: * wtf/AVLTree.h: * wtf/Alignment.h: * wtf/AlwaysInline.h: * wtf/ArrayBufferView.h: * wtf/Assertions.h: * wtf/Atomics.h: * wtf/Bitmap.h: * wtf/BoundsCheckedPointer.h: * wtf/CheckedArithmetic.h: * wtf/Deque.h: * wtf/ExportMacros.h: * wtf/FastAllocBase.h: * wtf/FastMalloc.h: * wtf/Float32Array.h: * wtf/Float64Array.h: * wtf/Functional.h: * wtf/HashCountedSet.h: * wtf/HashFunctions.h: * wtf/HashMap.h: * wtf/HashSet.h: * wtf/HashTable.h: * wtf/HashTraits.h: * wtf/Int16Array.h: * wtf/Int32Array.h: * wtf/Int8Array.h: * wtf/IntegralTypedArrayBase.h: * wtf/ListHashSet.h: * wtf/MainThread.h: * wtf/MetaAllocator.h: * wtf/Noncopyable.h: * wtf/OwnArrayPtr.h: * wtf/OwnPtr.h: * wtf/PackedIntVector.h: * wtf/ParallelJobs.h: * wtf/PassOwnArrayPtr.h: * wtf/PassOwnPtr.h: * wtf/PassRefPtr.h: * wtf/PassTraits.h: * wtf/Platform.h: * wtf/PossiblyNull.h: * wtf/RefCounted.h: * wtf/RefCountedLeakCounter.h: * wtf/RefPtr.h: * wtf/RetainPtr.h: * wtf/SimpleStats.h: * wtf/Spectrum.h: * wtf/StdLibExtras.h: * wtf/TCPageMap.h: * wtf/TemporaryChange.h: * wtf/ThreadSafeRefCounted.h: * wtf/Threading.h: * wtf/ThreadingPrimitives.h: * wtf/TypeTraits.h: * wtf/TypedArrayBase.h: * wtf/Uint16Array.h: * wtf/Uint32Array.h: * wtf/Uint8Array.h: * wtf/Uint8ClampedArray.h: * wtf/UnusedParam.h: * wtf/Vector.h: * wtf/VectorTraits.h: * wtf/dtoa/double-conversion.h: * wtf/dtoa/utils.h: * wtf/gobject/GRefPtr.h: * wtf/gobject/GlibUtilities.h: * wtf/text/AtomicString.h: * wtf/text/AtomicStringImpl.h: * wtf/text/CString.h: * wtf/text/StringConcatenate.h: * wtf/text/StringHash.h: * wtf/text/WTFString.h: * wtf/unicode/CharacterNames.h: * wtf/unicode/UTF8.h: * wtf/unicode/glib/UnicodeGLib.h: * wtf/unicode/qt4/UnicodeQt4.h: * wtf/unicode/wince/UnicodeWinCE.h: * wtf/url/api/ParsedURL.h: * wtf/url/api/URLString.h: * wtf/wince/FastMallocWinCE.h: * yarr/YarrJIT.cpp: 2012-03-06 Gavin Barraclough <barraclough@apple.com> Array.prototype functions should throw if delete fails https://bugs.webkit.org/show_bug.cgi?id=80467 Reviewed by Oliver Hunt. All calls to [[Delete]] from Array.prototype are specified to pass 'true' as the value of Throw. In the case of shift/unshift, these are also missing a throw from the 'put' in the implementations in JSArray.cpp. There are effectively three copies of each of the generic shift/unshift routines, one in splice, one in ArrayPrototype's shift/unshift methods, and one in JSArray's shift/unshift routines, for handling arrays with holes. These three copies should be unified. * runtime/ArrayPrototype.cpp: (JSC::shift): (JSC::unshift): - Added - shared copies of the shift/unshift functionality. (JSC::arrayProtoFuncPop): - should throw if the delete fails. (JSC::arrayProtoFuncReverse): - should throw if the delete fails. (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): - use shift/unshift. * runtime/JSArray.cpp: (JSC::JSArray::shiftCount): (JSC::JSArray::unshiftCount): - Don't try to handle arrays with holes; return a value indicating the generic routine should be used instead. * runtime/JSArray.h: - declaration for shiftCount/unshiftCount changed. * tests/mozilla/js1_6/Array/regress-304828.js: - this was asserting incorrect behaviour. 2012-03-06 Raphael Kubo da Costa <kubo@profusion.mobi> [CMake] Make the removal of transitive library dependencies work with CMake < 2.8.7. https://bugs.webkit.org/show_bug.cgi?id=80469 Reviewed by Antonio Gomes. * CMakeLists.txt: Manually set the LINK_INTERFACE_LIBRARIES target property on the library being created. 2012-03-06 Yuqiang Xian <yuqiang.xian@intel.com> DFG BasicBlock should group the Phi nodes together and separate them from the other nodes https://bugs.webkit.org/show_bug.cgi?id=80361 Reviewed by Filip Pizlo. This would make it more efficient to remove the redundant Phi nodes or insert new Phi nodes for SSA, besides providing a cleaner BasicBlock structure. This is performance neutral on SunSpider, V8 and Kraken. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::clobberStructures): (JSC::DFG::AbstractState::dump): * dfg/DFGBasicBlock.h: (JSC::DFG::BasicBlock::BasicBlock): (BasicBlock): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::addToGraph): (JSC::DFG::ByteCodeParser::insertPhiNode): * dfg/DFGCFAPhase.cpp: (JSC::DFG::CFAPhase::performBlockCFA): * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::pureCSE): (JSC::DFG::CSEPhase::impureCSE): (JSC::DFG::CSEPhase::globalVarLoadElimination): (JSC::DFG::CSEPhase::getByValLoadElimination): (JSC::DFG::CSEPhase::checkFunctionElimination): (JSC::DFG::CSEPhase::checkStructureLoadElimination): (JSC::DFG::CSEPhase::getByOffsetLoadElimination): (JSC::DFG::CSEPhase::getPropertyStorageLoadElimination): (JSC::DFG::CSEPhase::getIndexedPropertyStorageLoadElimination): (JSC::DFG::CSEPhase::getScopeChainLoadElimination): (JSC::DFG::CSEPhase::performBlockCSE): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-03-06 Mark Hahnenberg <mhahnenberg@apple.com> GCActivityCallback timer should vary with the length of the previous GC https://bugs.webkit.org/show_bug.cgi?id=80344 Reviewed by Geoffrey Garen. * heap/Heap.cpp: Gave Heap the ability to keep track of the length of its last GC length so that the GC Activity Callback can use it. (JSC::Heap::Heap): (JSC::Heap::collect): * heap/Heap.h: (JSC::Heap::lastGCLength): (Heap): * runtime/GCActivityCallbackCF.cpp: (JSC): (JSC::DefaultGCActivityCallback::operator()): Use the length of the Heap's last GC to determine the length of our timer trigger (currently set at 100x the duration of the last GC). 2012-03-06 Rob Buis <rbuis@rim.com> BlackBerry] Fix cast-align gcc warnings when compiling JSC https://bugs.webkit.org/show_bug.cgi?id=80420 Reviewed by Gavin Barraclough. Fix warnings given in Blackberry build. * heap/CopiedBlock.h: (JSC::CopiedBlock::CopiedBlock): * wtf/RefCountedArray.h: (WTF::RefCountedArray::Header::fromPayload): 2012-03-06 Gavin Barraclough <barraclough@apple.com> writable/configurable not respected for some properties of Function/String/Arguments https://bugs.webkit.org/show_bug.cgi?id=80436 Reviewed by Oliver Hunt. Special properties should behave like regular properties. * runtime/Arguments.cpp: (JSC::Arguments::defineOwnProperty): - Mis-nested logic for making read-only properties non-live. * runtime/JSFunction.cpp: (JSC::JSFunction::put): - arguments/length/caller are non-writable, non-configurable - reject appropriately. (JSC::JSFunction::deleteProperty): - Attempting to delete prototype/caller should fail. (JSC::JSFunction::defineOwnProperty): - Ensure prototype is reified on attempt to reify it. - arguments/length/caller are non-writable, non-configurable - reject appropriately. * runtime/JSFunction.h: - added declaration for defineOwnProperty. (JSFunction): * runtime/StringObject.cpp: (JSC::StringObject::put): - length is non-writable, non-configurable - reject appropriately. 2012-03-06 Ulan Degenbaev <ulan@chromium.org> TypedArray subarray call for subarray does not clamp the end index parameter properly https://bugs.webkit.org/show_bug.cgi?id=80285 Reviewed by Kenneth Russell. * wtf/ArrayBufferView.h: (WTF::ArrayBufferView::calculateOffsetAndLength): 2012-03-06 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r109837. http://trac.webkit.org/changeset/109837 https://bugs.webkit.org/show_bug.cgi?id=80399 breaks Mac Productions builds, too late to try and fix it tonight (Requested by eseidel on #webkit). * API/tests/JSNode.c: * API/tests/JSNodeList.c: * Configurations/Base.xcconfig: * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: * JavaScriptCore.xcodeproj/project.pbxproj: * assembler/MacroAssemblerCodeRef.h: * bytecompiler/BytecodeGenerator.h: * dfg/DFGOperations.cpp: * heap/GCAssertions.h: * heap/HandleHeap.h: * heap/HandleStack.h: * heap/MarkedSpace.h: * heap/PassWeak.h: * heap/Strong.h: * heap/Weak.h: * jit/HostCallReturnValue.cpp: * jit/JIT.cpp: * jit/JITStubs.cpp: * jit/ThunkGenerators.cpp: * parser/Lexer.cpp: * runtime/Completion.cpp: * runtime/Executable.cpp: * runtime/Identifier.h: * runtime/InitializeThreading.cpp: * runtime/JSDateMath.cpp: * runtime/JSGlobalObjectFunctions.cpp: * runtime/JSStringBuilder.h: * runtime/JSVariableObject.h: * runtime/NumberPrototype.cpp: * runtime/WriteBarrier.h: * tools/CodeProfile.cpp: * tools/TieredMMapArray.h: * yarr/YarrJIT.cpp: 2012-03-06 Zoltan Herczeg <zherczeg@webkit.org> [Qt][ARM] Speculative buildfix after r109834. Reviewed by Csaba Osztrogonác. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::and32): (MacroAssemblerARM): 2012-03-05 Gavin Barraclough <barraclough@apple.com> Unreviewed windows build fix pt 2. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-03-05 Gavin Barraclough <barraclough@apple.com> Unreviewed windows build fix pt 1. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-03-05 Gavin Barraclough <barraclough@apple.com> putByIndex should throw in strict mode https://bugs.webkit.org/show_bug.cgi?id=80335 Reviewed by Filip Pizlo. Make the MethodTable PutByIndex trap take a boolean 'shouldThrow' parameter. This is a largely mechanical change, simply adding an extra parameter to a number of functions. Some call sites need perform additional exception checks, and operationPutByValBeyondArrayBounds needs to know whether it is strict or not. This patch doesn't fix a missing throw from some cases of shift/unshift (this is an existing bug), I'll follow up with a third patch to handle that. * API/JSObjectRef.cpp: (JSObjectSetPropertyAtIndex): * JSCTypedArrayStubs.h: (JSC): * dfg/DFGOperations.cpp: (JSC::DFG::putByVal): * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jsc.cpp: (GlobalObject::finishCreation): * llint/LLIntSlowPaths.cpp: (JSC::LLInt::LLINT_SLOW_PATH_DECL): * runtime/Arguments.cpp: (JSC::Arguments::putByIndex): * runtime/Arguments.h: (Arguments): * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncPush): (JSC::arrayProtoFuncReverse): (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSort): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): * runtime/ClassInfo.h: (MethodTable): * runtime/JSArray.cpp: (JSC::SparseArrayValueMap::put): (JSC::JSArray::put): (JSC::JSArray::putByIndex): (JSC::JSArray::putByIndexBeyondVectorLength): (JSC::JSArray::push): (JSC::JSArray::shiftCount): (JSC::JSArray::unshiftCount): * runtime/JSArray.h: (SparseArrayValueMap): (JSArray): * runtime/JSByteArray.cpp: (JSC::JSByteArray::putByIndex): * runtime/JSByteArray.h: (JSByteArray): * runtime/JSCell.cpp: (JSC::JSCell::putByIndex): * runtime/JSCell.h: (JSCell): * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::putByIndex): * runtime/JSNotAnObject.h: (JSNotAnObject): * runtime/JSONObject.cpp: (JSC::Walker::walk): * runtime/JSObject.cpp: (JSC::JSObject::putByIndex): * runtime/JSObject.h: (JSC::JSValue::putByIndex): * runtime/RegExpConstructor.cpp: (JSC::RegExpMatchesArray::fillArrayInstance): * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::putByIndex): * runtime/StringPrototype.cpp: (JSC::stringProtoFuncSplit): 2012-03-05 Yuqiang Xian <yuqiang.xian@intel.com> PredictNone is incorrectly treated as isDoublePrediction https://bugs.webkit.org/show_bug.cgi?id=80365 Reviewed by Filip Pizlo. Also it is incorrectly treated as isFixedIndexedStorageObjectPrediction. * bytecode/PredictedType.h: (JSC::isFixedIndexedStorageObjectPrediction): (JSC::isDoublePrediction): 2012-03-05 Filip Pizlo <fpizlo@apple.com> The LLInt should work even when the JIT is disabled https://bugs.webkit.org/show_bug.cgi?id=80340 <rdar://problem/10922235> Reviewed by Gavin Barraclough. * assembler/MacroAssemblerCodeRef.h: (JSC::MacroAssemblerCodePtr::createLLIntCodePtr): (MacroAssemblerCodeRef): (JSC::MacroAssemblerCodeRef::createLLIntCodeRef): * interpreter/Interpreter.cpp: (JSC::Interpreter::initialize): (JSC::Interpreter::execute): (JSC::Interpreter::executeCall): (JSC::Interpreter::executeConstruct): * jit/JIT.h: (JSC::JIT::compileCTINativeCall): * jit/JITStubs.h: (JSC::JITThunks::ctiNativeCall): (JSC::JITThunks::ctiNativeConstruct): * llint/LLIntEntrypoints.cpp: (JSC::LLInt::getFunctionEntrypoint): (JSC::LLInt::getEvalEntrypoint): (JSC::LLInt::getProgramEntrypoint): * llint/LLIntSlowPaths.cpp: (JSC::LLInt::LLINT_SLOW_PATH_DECL): (LLInt): * llint/LLIntSlowPaths.h: (LLInt): * llint/LowLevelInterpreter.h: * llint/LowLevelInterpreter32_64.asm: * runtime/Executable.h: (NativeExecutable): (JSC::NativeExecutable::create): (JSC::NativeExecutable::finishCreation): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: (JSGlobalData): * runtime/Options.cpp: (Options): (JSC::Options::parse): (JSC::Options::initializeOptions): * runtime/Options.h: (Options): * wtf/Platform.h: 2012-03-05 Yuqiang Xian <yuqiang.xian@intel.com> Checks for dead variables are not sufficient when fixing the expected values in DFG OSR entry https://bugs.webkit.org/show_bug.cgi?id=80371 Reviewed by Filip Pizlo. A dead variable should be identified when there's no node referencing it. But we currently failed to catch the case where there are some nodes referencing a variable but those nodes are actually not referenced by others so will be ignored in code generation. In such case we should also consider that variable to be a dead variable in the block and fix the expected values. This is performance neutral on SunSpider, V8 and Kraken. * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::noticeOSREntry): 2012-03-05 Oliver Hunt <oliver@apple.com> Fix Qt build. * assembler/AbstractMacroAssembler.h: * assembler/MacroAssembler.h: (MacroAssembler): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileArithSub): * jit/JITArithmetic32_64.cpp: (JSC::JIT::emitSub32Constant): 2012-03-05 Eric Seidel <eric@webkit.org> Update JavaScriptCore files to use fully-qualified WTF include paths https://bugs.webkit.org/show_bug.cgi?id=79960 Reviewed by Adam Barth. This change does 5 small/related things: 1. Updates JavaScriptCore.xcodeproj to install WTF headers into $BUILD/usr/local/include (WebCore, WebKit were already setup to look there, but JavaScriptCore.xcodeproj was not installing headers there.) 2. Makes JavaScriptCore targets include $BUILD/usr/local/include in their header search path, as that's where the WTF headers will be installed. 3. Similarly updates JavaScriptCore.vcproj/copy-files.cmd to copy WTF headers to PrivateHeaders/wtf/* in addition to the current behavior of flattening all headers to PrivateHeaders/*.h. 4. Updates a bunch of JSC files to use #include <wtf/Foo.h> instead of #include "Foo.h" since soon the WTF headers will not be part of the JavaScriptCore Xcode project. 5. Makes build-webkit build the WTF XCode project by default. * API/tests/JSNode.c: * API/tests/JSNodeList.c: * Configurations/Base.xcconfig: * assembler/MacroAssemblerCodeRef.h: * bytecompiler/BytecodeGenerator.h: * dfg/DFGOperations.cpp: * heap/GCAssertions.h: * heap/HandleHeap.h: * heap/HandleStack.h: * heap/MarkedSpace.h: * heap/PassWeak.h: * heap/Strong.h: * heap/Weak.h: * jit/HostCallReturnValue.cpp: * jit/JIT.cpp: * jit/JITStubs.cpp: * jit/ThunkGenerators.cpp: * parser/Lexer.cpp: * runtime/Completion.cpp: * runtime/Executable.cpp: * runtime/Identifier.h: * runtime/InitializeThreading.cpp: * runtime/JSDateMath.cpp: * runtime/JSGlobalObjectFunctions.cpp: * runtime/JSStringBuilder.h: * runtime/JSVariableObject.h: * runtime/NumberPrototype.cpp: * runtime/WriteBarrier.h: * tools/CodeProfile.cpp: * tools/TieredMMapArray.h: * yarr/YarrJIT.cpp: 2012-03-05 Oliver Hunt <oliver@apple.com> Add basic support for constant blinding to the JIT https://bugs.webkit.org/show_bug.cgi?id=80354 Reviewed by Filip Pizlo. This patch adds basic constant blinding support to the JIT, at the MacroAssembler level. This means all JITs in JSC (Yarr, baseline, and DFG) get constant blinding. Woo! This patch only introduces blinding for Imm32, a later patch will do similar for ImmPtr. In order to make misuse of Imm32 as a trusted type essentially impossible, we make TrustedImm32 a private parent of Imm32 and add an explicit accessor that's needed to access the actual value. This also means you cannot accidentally pass an untrusted value to a function that does not perform blinding. To make everything work sensibly, this patch also corrects some code that was using Imm32 when TrustedImm32 could be used, and refactors a few callers that use untrusted immediates, so that they call slightly different varaints of the functions that they used previously. This is largely necessary to deal with x86-32 not having sufficient registers to handle the additional work required when we choose to blind a constant. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::Imm32::asTrustedImm32): (Imm32): (JSC::AbstractMacroAssembler::beginUninterruptedSequence): (JSC::AbstractMacroAssembler::endUninterruptedSequence): (JSC::AbstractMacroAssembler::AbstractMacroAssembler): (AbstractMacroAssembler): (JSC::AbstractMacroAssembler::inUninterruptedSequence): (JSC::AbstractMacroAssembler::random): (JSC::AbstractMacroAssembler::scratchRegisterForBlinding): (JSC::AbstractMacroAssembler::shouldBlindForSpecificArch): * assembler/MacroAssembler.h: (JSC::MacroAssembler::addressForPoke): (MacroAssembler): (JSC::MacroAssembler::poke): (JSC::MacroAssembler::branchPtr): (JSC::MacroAssembler::branch32): (JSC::MacroAssembler::convertInt32ToDouble): (JSC::MacroAssembler::shouldBlind): (JSC::MacroAssembler::BlindedImm32::BlindedImm32): (BlindedImm32): (JSC::MacroAssembler::keyForConstant): (JSC::MacroAssembler::xorBlindConstant): (JSC::MacroAssembler::additionBlindedConstant): (JSC::MacroAssembler::andBlindedConstant): (JSC::MacroAssembler::orBlindedConstant): (JSC::MacroAssembler::loadXorBlindedConstant): (JSC::MacroAssembler::add32): (JSC::MacroAssembler::addPtr): (JSC::MacroAssembler::and32): (JSC::MacroAssembler::andPtr): (JSC::MacroAssembler::move): (JSC::MacroAssembler::or32): (JSC::MacroAssembler::store32): (JSC::MacroAssembler::sub32): (JSC::MacroAssembler::subPtr): (JSC::MacroAssembler::xor32): (JSC::MacroAssembler::branchAdd32): (JSC::MacroAssembler::branchMul32): (JSC::MacroAssembler::branchSub32): (JSC::MacroAssembler::trustedImm32ForShift): (JSC::MacroAssembler::lshift32): (JSC::MacroAssembler::rshift32): (JSC::MacroAssembler::urshift32): * assembler/MacroAssemblerARMv7.h: (MacroAssemblerARMv7): (JSC::MacroAssemblerARMv7::scratchRegisterForBlinding): (JSC::MacroAssemblerARMv7::shouldBlindForSpecificArch): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::branchSubPtr): (MacroAssemblerX86_64): (JSC::MacroAssemblerX86_64::scratchRegisterForBlinding): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::linkOSRExits): (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::compileArithSub): (JSC::DFG::SpeculativeJIT::compileStrictEqForConstant): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::callOperation): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compileDoubleCompare): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): (JSC::DFG::SpeculativeJIT::compileDoubleCompare): (JSC::DFG::SpeculativeJIT::compile): * jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases): (JSC::JIT::privateCompile): * jit/JITArithmetic.cpp: (JSC::JIT::compileBinaryArithOp): (JSC::JIT::emit_op_add): (JSC::JIT::emit_op_mul): (JSC::JIT::emit_op_div): * jit/JITArithmetic32_64.cpp: (JSC::JIT::emitAdd32Constant): (JSC::JIT::emitSub32Constant): (JSC::JIT::emitBinaryDoubleOp): (JSC::JIT::emitSlow_op_mul): (JSC::JIT::emit_op_div): * jit/JITCall.cpp: (JSC::JIT::compileLoadVarargs): * jit/JITCall32_64.cpp: (JSC::JIT::compileLoadVarargs): * jit/JITInlineMethods.h: (JSC::JIT::updateTopCallFrame): (JSC::JIT::emitValueProfilingSite): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emitSlow_op_jfalse): (JSC::JIT::emitSlow_op_jtrue): * jit/JITStubCall.h: (JITStubCall): (JSC::JITStubCall::addArgument): * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::backtrack): 2012-03-05 Gavin Barraclough <barraclough@apple.com> putByIndex should throw in strict mode https://bugs.webkit.org/show_bug.cgi?id=80335 Reviewed by Filip Pizlo. We'll need to pass an additional parameter. Part 1 - rename JSValue::put() for integer indices to JSValue::putByIndex() to match the method in the MethodTable, make this take a parameter indicating whether the put should throw. This fixes the cases where the base of the put is a primitive. * dfg/DFGOperations.cpp: (DFG): (JSC::DFG::putByVal): (JSC::DFG::operationPutByValInternal): * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * llint/LLIntSlowPaths.cpp: (JSC::LLInt::LLINT_SLOW_PATH_DECL): * runtime/JSObject.h: (JSC::JSValue::putByIndex): * runtime/JSValue.cpp: (JSC): * runtime/JSValue.h: (JSValue): 2012-03-05 Sam Weinig <sam@webkit.org> Add support for hosting layers in the window server in WebKit2 <rdar://problem/10400246> https://bugs.webkit.org/show_bug.cgi?id=80310 Reviewed by Anders Carlsson. * wtf/Platform.h: Add HAVE_LAYER_HOSTING_IN_WINDOW_SERVER. 2012-03-05 Filip Pizlo <fpizlo@apple.com> Unreviewed, attempted build fix for !ENABLE(JIT) after r109705. * bytecode/ExecutionCounter.cpp: (JSC::ExecutionCounter::applyMemoryUsageHeuristics): * bytecode/ExecutionCounter.h: 2012-03-05 Patrick Gansterer <paroga@webkit.org> Unreviewed. Build fix for !ENABLE(JIT) after r109705. * bytecode/ExecutionCounter.cpp: * bytecode/ExecutionCounter.h: 2012-03-05 Andy Wingo <wingo@igalia.com> Lexer: Specialize character predicates for LChar, UChar https://bugs.webkit.org/show_bug.cgi?id=79677 Reviewed by Oliver Hunt. This patch specializes isIdentStart, isIdentPart, isWhiteSpace, and isLineTerminator to perform a more limited number of checks if the lexer is being instantiated to work on LChar sequences. This is about a 1.5% win on the --parse-only suite, here. * parser/Lexer.cpp: (JSC::isLatin1): New static helper, specialized for LChar and UChar. (JSC::typesOfLatin1Characters): Rename from typesOfASCIICharacters, and expand to the range of the LChar type. All uses of isASCII are changed to use isLatin1. Generated using libunistring. (JSC::isNonLatin1IdentStart): (JSC::isIdentStart): (JSC::isNonLatin1IdentPart): (JSC::isIdentPart): (JSC::Lexer::shiftLineTerminator): (JSC::Lexer::parseIdentifier): (JSC::Lexer::parseIdentifierSlowCase): (JSC::Lexer::parseStringSlowCase): (JSC::Lexer::parseMultilineComment): (JSC::Lexer::lex): (JSC::Lexer::scanRegExp): (JSC::Lexer::skipRegExp): Sprinkle static_cast<T>(_) around. * parser/Lexer.h: (JSC::Lexer::isWhiteSpace): (JSC::Lexer::isLineTerminator): * KeywordLookupGenerator.py: (Trie.printAsC): Declare specialized isIdentPart static functions. 2012-03-05 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck. * GNUmakefile.list.am: Add missing header file. 2012-03-05 Andy Wingo <wingo@igalia.com> WTF: Micro-optimize cleanup of empty vectors and hash tables https://bugs.webkit.org/show_bug.cgi?id=79903 Reviewed by Michael Saboff and Geoffrey Garen. This patch speeds up cleanup of vectors and hash tables whose backing store was never allocated. This is the case by default for most vectors / hash tables that never had any entries added. The result for me is that calling checkSyntax 1000 times on concat-jquery-mootools-prototype.js goes from 6.234s to 6.068s, a 2.4% speedup. * wtf/HashTable.h: (WTF::HashTable::~HashTable): (WTF::::clear): Don't deallocate the storage or frob member variables if there is no backing storage. * wtf/Vector.h: (WTF::VectorBufferBase::deallocateBuffer): Likewise. 2012-03-04 Filip Pizlo <fpizlo@apple.com> JIT heuristics should be hyperbolic https://bugs.webkit.org/show_bug.cgi?id=80055 <rdar://problem/10922260> Reviewed by Oliver Hunt. Added tracking of the amount of executable memory typically used for a bytecode instruction. Modified the execution counter scheme to use this, and the amount of free memory, to determine how long to wait before invoking the JIT. The result is that even if we bomb the VM with more code than can fit in our executable memory pool, we still keep running and almost never run out of executable memory - which ensures that if we have to JIT something critical, then we'll likely have enough memory to do so. This also does not regress performance on the three main benchmarks. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::predictedMachineCodeSize): (JSC): (JSC::CodeBlock::usesOpcode): * bytecode/CodeBlock.h: (CodeBlock): (JSC::CodeBlock::checkIfJITThresholdReached): (JSC::CodeBlock::dontJITAnytimeSoon): (JSC::CodeBlock::jitAfterWarmUp): (JSC::CodeBlock::jitSoon): (JSC::CodeBlock::llintExecuteCounter): (JSC::CodeBlock::counterValueForOptimizeAfterWarmUp): (JSC::CodeBlock::counterValueForOptimizeAfterLongWarmUp): (JSC::CodeBlock::addressOfJITExecuteCounter): (JSC::CodeBlock::offsetOfJITExecuteCounter): (JSC::CodeBlock::offsetOfJITExecutionActiveThreshold): (JSC::CodeBlock::offsetOfJITExecutionTotalCount): (JSC::CodeBlock::jitExecuteCounter): (JSC::CodeBlock::checkIfOptimizationThresholdReached): (JSC::CodeBlock::optimizeNextInvocation): (JSC::CodeBlock::dontOptimizeAnytimeSoon): (JSC::CodeBlock::optimizeAfterWarmUp): (JSC::CodeBlock::optimizeAfterLongWarmUp): (JSC::CodeBlock::optimizeSoon): * bytecode/ExecutionCounter.cpp: Added. (JSC): (JSC::ExecutionCounter::ExecutionCounter): (JSC::ExecutionCounter::checkIfThresholdCrossedAndSet): (JSC::ExecutionCounter::setNewThreshold): (JSC::ExecutionCounter::deferIndefinitely): (JSC::ExecutionCounter::applyMemoryUsageHeuristics): (JSC::ExecutionCounter::applyMemoryUsageHeuristicsAndConvertToInt): (JSC::ExecutionCounter::hasCrossedThreshold): (JSC::ExecutionCounter::setThreshold): (JSC::ExecutionCounter::reset): * bytecode/ExecutionCounter.h: Added. (JSC): (ExecutionCounter): (JSC::ExecutionCounter::formattedTotalCount): * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * jit/ExecutableAllocator.cpp: (JSC::DemandExecutableAllocator::allocateNewSpace): (JSC::ExecutableAllocator::underMemoryPressure): (JSC): (JSC::ExecutableAllocator::memoryPressureMultiplier): * jit/ExecutableAllocator.h: * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::ExecutableAllocator::memoryPressureMultiplier): (JSC): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * llint/LLIntSlowPaths.cpp: (JSC::LLInt::jitCompileAndSetHeuristics): * llint/LowLevelInterpreter32_64.asm: * runtime/JSGlobalData.h: (JSGlobalData): * runtime/Options.cpp: (Options): (JSC::Options::initializeOptions): * runtime/Options.h: (Options): * wtf/SimpleStats.h: Added. (WTF): (SimpleStats): (WTF::SimpleStats::SimpleStats): (WTF::SimpleStats::add): (WTF::SimpleStats::operator!): (WTF::SimpleStats::count): (WTF::SimpleStats::sum): (WTF::SimpleStats::sumOfSquares): (WTF::SimpleStats::mean): (WTF::SimpleStats::variance): (WTF::SimpleStats::standardDeviation): 2012-03-04 Raphael Kubo da Costa <kubo@profusion.mobi> [CMake] Libraries are installed to /usr/lib and not /usr/lib64 on x86_64 https://bugs.webkit.org/show_bug.cgi?id=71507 Reviewed by Antonio Gomes. * CMakeLists.txt: Use ${LIB_INSTALL_DIR} instead of hardcoding "lib". 2012-03-04 David Kilzer <ddkilzer@apple.com> Fix build when the classic interpreter is enabled Reviewed by Gavin Barraclough. Fixes the following build error when running the "Generate Derived Sources" build phase script: offlineasm: Parsing JavaScriptCore/llint/LowLevelInterpreter.asm and ../../JSCLLIntOffsetsExtractor and creating assembly file LLIntAssembly.h. ./JavaScriptCore/offlineasm/offsets.rb:145:in `offsetsAndConfigurationIndex': unhandled exception from JavaScriptCore/offlineasm/asm.rb:131 Command /bin/sh failed with exit code 1 Gavin's fix in r109674 avoided the #error statement in JITStubs.h when compiling LLIntOffsetsExtractor.cpp, but it caused the "Generate Derived Sources" build phase script to fail when JavaScriptCore/offlineasm/asm.rb was run. The solution is to detect when the classic interpreter is being built and simply exit early from asm.rb in that case. * llint/LLIntOffsetsExtractor.cpp: (JSC::LLIntOffsetsExtractor::dummy): Return NULL pointer if the JIT is disabled. Note that offsets.rb doesn't care about the return value here, but instead it cares about finding the magic values in the binary. The magic values are no longer present when the JIT is disabled. * offlineasm/asm.rb: Catch MissingMagicValuesException and exit early with a status message. * offlineasm/offsets.rb: (MissingMagicValuesException): Add new exception class. (offsetsAndConfigurationIndex): Throw MissingMagicValuesException when no magic values are found. 2012-03-04 Jurij Smakov <jurij@wooyd.org> SPARC also needs aligned accesses. Rubber-stamped by Gustavo Noronha Silva. * wtf/Platform.h: 2012-03-04 Gavin Barraclough <barraclough@apple.com> Unreviewed build fix. * jit/JITStubs.h: - Move ENABLE(JIT) to head of file. 2012-03-03 Gavin Barraclough <barraclough@apple.com> Split JSArray's [[Put]] & [[DefineOwnProperty]] traps. https://bugs.webkit.org/show_bug.cgi?id=80217 Reviewed by Filip Pizlo. putByIndex() provides similar behavior to put(), but for indexed property names. Many places in ArrayPrototype call putByIndex() where they really mean to call [[DefineOwnProperty]]. This is only okay due to a bug – putByIndex should be calling numeric accessors (& respecting numeric read only properties) on the prototype chain, but isn't. Add a new putDirectIndex (matching JSObject's putDirect* methods), to correctly provide a fast [[DefineOwnProperty]] interface. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncConcat): (JSC::arrayProtoFuncSlice): (JSC::arrayProtoFuncFilter): (JSC::arrayProtoFuncMap): * runtime/JSArray.cpp: (JSC): (JSC::reject): (JSC::SparseArrayValueMap::putDirect): (JSC::JSArray::defineOwnNumericProperty): (JSC::JSArray::putByIndexBeyondVectorLength): (JSC::JSArray::putDirectIndexBeyondVectorLength): * runtime/JSArray.h: (SparseArrayValueMap): (JSArray): (JSC::JSArray::putDirectIndex): 2012-03-03 Benjamin Poulain <benjamin@webkit.org> Implement the basis of KURLWTFURL https://bugs.webkit.org/show_bug.cgi?id=79600 Reviewed by Adam Barth. Add an API to know if a ParsedURL is valid. * wtf/url/api/ParsedURL.cpp: (WTF::ParsedURL::ParsedURL): (WTF): (WTF::ParsedURL::isolatedCopy): This is needed by APIs moving URL objects between thread and by KURL's detach() on write. (WTF::ParsedURL::baseAsString): (WTF::ParsedURL::segment): Add a stronger constraint on accessors: the client of this API should never ask for the segments on an invalid URL. * wtf/url/api/ParsedURL.h: (WTF): (WTF::ParsedURL::ParsedURL): (ParsedURL): (WTF::ParsedURL::isValid): 2012-03-03 Hans Wennborg <hans@chromium.org> Implement Speech JavaScript API https://bugs.webkit.org/show_bug.cgi?id=80019 Reviewed by Adam Barth. Add ENABLE_SCRIPTED_SPEECH. * Configurations/FeatureDefines.xcconfig: 2012-03-02 Filip Pizlo <fpizlo@apple.com> When getting the line number of a call into a call frame with no code block, it's incorrect to rely on the returnPC https://bugs.webkit.org/show_bug.cgi?id=80195 Reviewed by Oliver Hunt. * interpreter/Interpreter.cpp: (JSC::getCallerInfo): * jit/JITCall.cpp: (JSC::JIT::compileLoadVarargs): 2012-03-02 Han Hojong <hojong.han@samsung.com> Expected results updated for checking type conversion https://bugs.webkit.org/show_bug.cgi?id=80138 Reviewed by Gavin Barraclough. * tests/mozilla/ecma/TypeConversion/9.3.1-3.js: 2012-03-02 Kenichi Ishibashi <bashi@chromium.org> Adding WebSocket per-frame DEFLATE extension https://bugs.webkit.org/show_bug.cgi?id=77522 Added USE(ZLIB) flag. Reviewed by Kent Tamura. * wtf/Platform.h: 2012-03-02 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix for platforms that have DFG_JIT disabled but PARALLEL_GC enabled. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): 2012-03-01 Filip Pizlo <fpizlo@apple.com> DFGCodeBlocks should not trace CodeBlocks that are also going to be traced by virtue of being in the transitive closure https://bugs.webkit.org/show_bug.cgi?id=80098 Reviewed by Anders Carlsson. If DFGCodeBlocks traces a CodeBlock that might also be traced via its owner Executable, then you might have the visitAggregate() method called concurrently by multiple threads. This is benign on 64-bit -- visitAggregate() and everything it calls turns out to be racy and slightly imprecise but not unsound. But on 32-bit, visitAggregate() may crash due to word tearing in ValueProfile bucket updates inside of computeUpdatedPrediction(). It would seem that the fix is just to have DFGCodeBlocks not trace CodeBlocks that are not jettisoned. But CodeBlocks may be jettisoned later during the GC, so it must trace any CodeBlock that it knows to be live by virtue of it being reachable from the stack. Hence the real fix is to make sure that concurrent calls into CodeBlock::visitAggregate() don't lead to two threads racing over each other as they clobber state. This patch achieves this with a simple CAS loop: whichever thread wins the CAS race (which is trivially linearizable) will get to trace the CodeBlock; all other threads give up and go home. Unfortunately there will be no new tests. It's possible to reproduce this maybe 1/10 times by running V8-v6's raytrace repeatedly, using the V8 harness hacked to rerun it even when it's gotten sufficient counts. But that takes a while - sometimes up to a minute to get a crash. I have no other reliable repro case. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): * bytecode/CodeBlock.h: (DFGData): * heap/DFGCodeBlocks.cpp: (JSC::DFGCodeBlocks::clearMarks): 2012-03-01 Filip Pizlo <fpizlo@apple.com> The JIT should not crash the entire process just because there is not enough executable memory, if the LLInt is enabled https://bugs.webkit.org/show_bug.cgi?id=79962 Reviewed by Csaba Osztrogonác. Fix for ARM, SH4. * assembler/AssemblerBufferWithConstantPool.h: (JSC::AssemblerBufferWithConstantPool::executableCopy): 2012-03-01 Ryosuke Niwa <rniwa@webkit.org> Revert my change. Broke builds. Source/JavaScriptCore/wtf/Atomics.h:188: error: redefinition of 'bool WTF::weakCompareAndSwap(volatile uintptr_t*, uintptr_t, uintptr_t)' Source/JavaScriptCore/wtf/Atomics.h:122: error: 'bool WTF::weakCompareAndSwap(volatile unsigned int*, unsigned int, unsigned i * wtf/Atomics.h: (WTF): (WTF::weakCompareAndSwap): 2012-03-01 Ryosuke Niwa <rniwa@webkit.org> Gcc build fix. Rubber-stamped by Filip Pizlo. * wtf/Atomics.h: (WTF): (WTF::weakCompareAndSwap): 2012-03-01 Gavin Barraclough <barraclough@apple.com> ES5.1-15.3.5.4. prohibits Function.caller from [[Get]]ting a strict caller https://bugs.webkit.org/show_bug.cgi?id=80011 Reviewed by Oliver Hunt. Also, fix getting the caller from within a bound function, for within a getter, or setter (make our implementation match other browsers). * interpreter/Interpreter.cpp: (JSC::getCallerInfo): - Allow this to get the caller of host functions. (JSC::Interpreter::retrieveCallerFromVMCode): - This should use getCallerInfo, and should skip over function bindings. * runtime/JSFunction.cpp: (JSC::JSFunction::callerGetter): - This should never return a strict-mode function. 2012-03-01 Yuqiang Xian <yuqiang.xian@intel.com> DFG local CSE for a node can be terminated earlier https://bugs.webkit.org/show_bug.cgi?id=80014 Reviewed by Filip Pizlo. When one of the node's childredn is met in the process of back traversing the nodes, we don't need to traverse the remaining nodes. This is performance neutral on SunSpider, V8 and Kraken. * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::pureCSE): (JSC::DFG::CSEPhase::impureCSE): (JSC::DFG::CSEPhase::getByValLoadElimination): (JSC::DFG::CSEPhase::checkFunctionElimination): (JSC::DFG::CSEPhase::checkStructureLoadElimination): (JSC::DFG::CSEPhase::getByOffsetLoadElimination): (JSC::DFG::CSEPhase::getPropertyStorageLoadElimination): (JSC::DFG::CSEPhase::getIndexedPropertyStorageLoadElimination): 2012-02-29 Yuqiang Xian <yuqiang.xian@intel.com> DFG BasicBlocks should not require that their nodes have continuous indices in the graph https://bugs.webkit.org/show_bug.cgi?id=79899 Reviewed by Filip Pizlo. This will make it more convenient to insert nodes into the DFG. With this capability we now place the Phi nodes in the corresponding blocks. Local CSE is modified to not to rely on the assumption of continuous node indices in a block. This is performance neutral on SunSpider, V8 and Kraken. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::AbstractState): (JSC::DFG::AbstractState::beginBasicBlock): (JSC::DFG::AbstractState::execute): (JSC::DFG::AbstractState::clobberStructures): (JSC::DFG::AbstractState::mergeToSuccessors): (JSC::DFG::AbstractState::dump): * dfg/DFGAbstractState.h: (JSC::DFG::AbstractState::forNode): (AbstractState): * dfg/DFGArithNodeFlagsInferencePhase.cpp: (ArithNodeFlagsInferencePhase): * dfg/DFGBasicBlock.h: (JSC::DFG::BasicBlock::BasicBlock): (BasicBlock): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::addToGraph): (ByteCodeParser): (JSC::DFG::ByteCodeParser::insertPhiNode): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::processPhiStack): (JSC::DFG::ByteCodeParser::linkBlock): (JSC::DFG::ByteCodeParser::determineReachability): (JSC::DFG::ByteCodeParser::parseCodeBlock): * dfg/DFGCFAPhase.cpp: (JSC::DFG::CFAPhase::performBlockCFA): (CFAPhase): * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::CSEPhase): (JSC::DFG::CSEPhase::endIndexForPureCSE): (JSC::DFG::CSEPhase::pureCSE): (JSC::DFG::CSEPhase::impureCSE): (JSC::DFG::CSEPhase::globalVarLoadElimination): (JSC::DFG::CSEPhase::getByValLoadElimination): (JSC::DFG::CSEPhase::checkFunctionElimination): (JSC::DFG::CSEPhase::checkStructureLoadElimination): (JSC::DFG::CSEPhase::getByOffsetLoadElimination): (JSC::DFG::CSEPhase::getPropertyStorageLoadElimination): (JSC::DFG::CSEPhase::getIndexedPropertyStorageLoadElimination): (JSC::DFG::CSEPhase::getScopeChainLoadElimination): (JSC::DFG::CSEPhase::performNodeCSE): (JSC::DFG::CSEPhase::performBlockCSE): (CSEPhase): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGPhase.cpp: (JSC::DFG::Phase::beginPhase): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativeCompare): (JSC::DFG::SpeculativeJIT::nonSpeculativeStrictEq): (JSC::DFG::SpeculativeJIT::compilePeepHoleBranch): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::compileStrictEqForConstant): (JSC::DFG::SpeculativeJIT::compileStrictEq): * dfg/DFGSpeculativeJIT.h: (SpeculativeJIT): (JSC::DFG::SpeculativeJIT::detectPeepHoleBranch): (JSC::DFG::SpeculativeJIT::SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativeCompareNull): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativeCompareNull): * dfg/DFGVirtualRegisterAllocationPhase.cpp: (JSC::DFG::VirtualRegisterAllocationPhase::run): 2012-02-29 Filip Pizlo <fpizlo@apple.com> The JIT should not crash the entire process just because there is not enough executable memory, if the LLInt is enabled https://bugs.webkit.org/show_bug.cgi?id=79962 <rdar://problem/10922215> Unreviewed, adding forgotten file. * jit/JITCompilationEffort.h: Added. (JSC): 2012-02-29 Filip Pizlo <fpizlo@apple.com> The JIT should not crash the entire process just because there is not enough executable memory, if the LLInt is enabled https://bugs.webkit.org/show_bug.cgi?id=79962 <rdar://problem/10922215> Reviewed by Gavin Barraclough. Added the notion of JITCompilationEffort. If we're JIT'ing as a result of a tier-up, then we set it to JITCompilationCanFail. Otherwise it's JITCompilationMustSucceed. This preserves the old behavior of LLInt is disabled or if we're compiling something that can't be interpreted (like an OSR exit stub). * JavaScriptCore.xcodeproj/project.pbxproj: * assembler/ARMAssembler.cpp: (JSC::ARMAssembler::executableCopy): * assembler/ARMAssembler.h: (ARMAssembler): * assembler/AssemblerBuffer.h: (JSC::AssemblerBuffer::executableCopy): * assembler/LinkBuffer.h: (JSC::LinkBuffer::LinkBuffer): (JSC::LinkBuffer::~LinkBuffer): (LinkBuffer): (JSC::LinkBuffer::didFailToAllocate): (JSC::LinkBuffer::isValid): (JSC::LinkBuffer::linkCode): (JSC::LinkBuffer::performFinalization): * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::executableCopy): * assembler/SH4Assembler.h: (JSC::SH4Assembler::executableCopy): * assembler/X86Assembler.h: (JSC::X86Assembler::executableCopy): (JSC::X86Assembler::X86InstructionFormatter::executableCopy): * bytecode/CodeBlock.cpp: (JSC::ProgramCodeBlock::jitCompileImpl): (JSC::EvalCodeBlock::jitCompileImpl): (JSC::FunctionCodeBlock::jitCompileImpl): * bytecode/CodeBlock.h: (JSC::CodeBlock::jitCompile): (CodeBlock): (ProgramCodeBlock): (EvalCodeBlock): (FunctionCodeBlock): * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compile): (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGJITCompiler.h: (JITCompiler): * jit/ExecutableAllocator.cpp: (JSC::DemandExecutableAllocator::allocateNewSpace): (JSC::ExecutableAllocator::allocate): * jit/ExecutableAllocator.h: (ExecutableAllocator): * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::ExecutableAllocator::allocate): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JIT.h: (JSC::JIT::compile): (JIT): * jit/JITCompilationEffort.h: Added. (JSC): * jit/JITDriver.h: (JSC::jitCompileIfAppropriate): (JSC::jitCompileFunctionIfAppropriate): * llint/LLIntSlowPaths.cpp: (LLInt): (JSC::LLInt::jitCompileAndSetHeuristics): (JSC::LLInt::entryOSR): (JSC::LLInt::LLINT_SLOW_PATH_DECL): * runtime/Executable.cpp: (JSC::EvalExecutable::jitCompile): (JSC::ProgramExecutable::jitCompile): (JSC::FunctionExecutable::jitCompileForCall): (JSC::FunctionExecutable::jitCompileForConstruct): * runtime/Executable.h: (EvalExecutable): (ProgramExecutable): (FunctionExecutable): (JSC::FunctionExecutable::jitCompileFor): * runtime/ExecutionHarness.h: (JSC::prepareForExecution): (JSC::prepareFunctionForExecution): 2012-02-29 No'am Rosenthal <noam.rosenthal@nokia.com> [Qt][WK2] Get rid of the #ifdef mess in LayerTreeHost[Proxy] https://bugs.webkit.org/show_bug.cgi?id=79501 Enable WTF_USE_UI_SIDE_COMPOSITING for Qt. Reviewed by Kenneth Rohde Christiansen. * wtf/Platform.h: 2012-02-29 Gavin Barraclough <barraclough@apple.com> Rubber stamped by Oliver Hunt. * tests/mozilla/ecma_2/RegExp/constructor-001.js: * tests/mozilla/ecma_2/RegExp/function-001.js: * tests/mozilla/ecma_2/RegExp/properties-001.js: - Check in new test cases results. 2012-02-29 Mark Rowe <mrowe@apple.com> Stop installing JSCLLIntOffsetsExtractor. Replace the separate TestRegExp and TestAPI xcconfig files with a single ToolExecutable xcconfig file that derives the product name from the target name. We can then use that xcconfig file for JSCLLIntOffsetsExtractor. This has the results of setting SKIP_INSTALL = YES for JSCLLIntOffsetsExtractor. While I was doing this fiddling I noticed that the JSCLLIntOffsetsExtractor target had a custom value for USER_HEADER_SEARCH_PATHS to allow it to find LLIntDesiredOffsets.h. A better way of doing that is to add LLIntDesiredOffsets.h to the Xcode project so that it'll be included in the header map. That allows us to remove the override of USER_HEADER_SEARCH_PATHS entirely. So I did that too! Reviewed by Filip Pizlo. * Configurations/TestRegExp.xcconfig: Removed. * Configurations/ToolExecutable.xcconfig: Renamed from Source/JavaScriptCore/Configurations/TestAPI.xcconfig. * JavaScriptCore.xcodeproj/project.pbxproj: 2012-02-28 Filip Pizlo <fpizlo@apple.com> RefCounted::deprecatedTurnOffVerifier() should not be deprecated https://bugs.webkit.org/show_bug.cgi?id=79864 Reviewed by Oliver Hunt. Removed the word "deprecated" from the name of this method, since this method should not be deprecated. It works just fine as it is, and there is simply no alternative to calling this method for many interesting JSC classes. * parser/SourceProvider.h: (JSC::SourceProvider::SourceProvider): * runtime/SymbolTable.h: (JSC::SharedSymbolTable::SharedSymbolTable): * wtf/MetaAllocator.cpp: (WTF::MetaAllocatorHandle::MetaAllocatorHandle): (WTF::MetaAllocator::allocate): * wtf/RefCounted.h: (RefCountedBase): (WTF::RefCountedBase::turnOffVerifier): 2012-02-29 Gavin Barraclough <barraclough@apple.com> 'source' property of RegExp instance cannot be "" https://bugs.webkit.org/show_bug.cgi?id=79938 Reviewed by Oliver Hunt. 15.10.6.4 specifies that RegExp.prototype.toString must return '/' + source + '/', and also states that the result must be a valid RegularExpressionLiteral. '//' is not a valid RegularExpressionLiteral (since it is a single line comment), and hence source cannot ever validly be "". If the source is empty, return a different Pattern that would match the same thing. * runtime/RegExpObject.cpp: (JSC::regExpObjectSource): - Do not return "" if the source is empty, this would lead to invalid behaviour in toString. * runtime/RegExpPrototype.cpp: (JSC::regExpProtoFuncToString): - No need to special case the empty string - this should be being done by 'source'. 2012-02-29 Gavin Barraclough <barraclough@apple.com> Writable attribute not set correctly when redefining an accessor to a data descriptor https://bugs.webkit.org/show_bug.cgi?id=79931 Reviewed by Oliver Hunt. * runtime/JSObject.cpp: (JSC::JSObject::defineOwnProperty): - use attributesOverridingCurrent instead of attributesWithOverride. * runtime/PropertyDescriptor.cpp: * runtime/PropertyDescriptor.h: - remove attributesWithOverride - attributesOverridingCurrent does the same thing. 2012-02-29 Kevin Ollivier <kevino@theolliviers.com> Add JSCore symbol exports needed by wx port https://bugs.webkit.org/show_bug.cgi?id=77280 Reviewed by Hajime Morita. * wtf/ArrayBufferView.h: * wtf/ExportMacros.h: 2012-02-28 Raphael Kubo da Costa <kubo@profusion.mobi> [CMake] Always build wtf as a static library. https://bugs.webkit.org/show_bug.cgi?id=79857 Reviewed by Eric Seidel. To help the efforts in bug 75673 to move WTF out of JavaScriptCore, act more like the other ports and remove the possibility of building WTF as a shared library. It does not make much sense to, for example, ship WTF as a separate .so with webkit-efl packages, and it should be small enough not to cause problems during linking. * wtf/CMakeLists.txt: 2012-02-28 Dmitry Lomov <dslomov@google.com> [JSC] Implement ArrayBuffer transfer https://bugs.webkit.org/show_bug.cgi?id=73493. Implement ArrayBuffer transfer, per Khronos spec: http://www.khronos.org/registry/typedarray/specs/latest/#9. This brings parity with V8 implementation of transferable typed arrays. Reviewed by Oliver Hunt. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Extra export. * wtf/ArrayBuffer.h: (ArrayBuffer): Added extra export. 2012-02-28 Kevin Ollivier <kevino@theolliviers.com> [wx] Unreviewed. Build fix after recent LLInt additions. * wscript: 2012-02-28 Mark Hahnenberg <mhahnenberg@apple.com> Refactor SpeculativeJIT::emitAllocateJSFinalObject https://bugs.webkit.org/show_bug.cgi?id=79801 Reviewed by Filip Pizlo. * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::emitAllocateBasicJSObject): Split emitAllocateJSFinalObject out to form this function, which is more generic in that it can allocate a variety of classes. (SpeculativeJIT): (JSC::DFG::SpeculativeJIT::emitAllocateJSFinalObject): Changed to use the new helper function. 2012-02-28 Gavin Barraclough <barraclough@apple.com> [[Get]]/[[Put]] for primitives should not wrap on strict accessor call https://bugs.webkit.org/show_bug.cgi?id=79588 Reviewed by Oliver Hunt. In the case of [[Get]], this is a pretty trivial bug - just don't wrap primitives at the point you call a getter. For setters, this is a little more involved, since we have already wrapped the value up in a synthesized object. Stop doing so. There is also a further subtely, that in strict mode all attempts to create a new data property on the object should throw. * runtime/JSCell.cpp: (JSC::JSCell::put): - [[Put]] to a string primitive should use JSValue::putToPrimitive. * runtime/JSObject.cpp: (JSC::JSObject::put): - Remove static function called in one place. * runtime/JSObject.h: (JSC::JSValue::put): - [[Put]] to a non-cell JSValue should use JSValue::putToPrimitive. * runtime/JSValue.cpp: (JSC::JSValue::synthesizePrototype): - Add support for synthesizing the prototype of strings. (JSC::JSValue::putToPrimitive): - Added, implements [[Put]] for primitive bases, per 8.7.2. * runtime/JSValue.h: (JSValue): - Add declaration for JSValue::putToPrimitive. * runtime/PropertySlot.cpp: (JSC::PropertySlot::functionGetter): - Don't call ToObject on primitive this values. 2012-02-28 Mark Hahnenberg <mhahnenberg@apple.com> Re-enable parallel GC on Mac https://bugs.webkit.org/show_bug.cgi?id=79837 Rubber stamped by Filip Pizlo. * runtime/Options.cpp: (JSC::Options::initializeOptions): We accidentally disabled parallel GC with this line, so we removed it and things should go back to normal. 2012-02-28 Filip Pizlo <fpizlo@apple.com> Some run-javascriptcore-tests broken for 32-bit debug https://bugs.webkit.org/show_bug.cgi?id=79844 Rubber stamped by Oliver Hunt. These assertions are just plain wrong for 32-bit. We could either have a massive assertion that depends on value representation, that has to be changed every time we change the JITs, resulting in a bug tail of debug-mode crashes, or we could get rid of the assertions. I pick the latter. * dfg/DFGOperations.cpp: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): 2012-02-28 Mark Hahnenberg <mhahnenberg@apple.com> Get rid of padding cruft in CopiedBlock https://bugs.webkit.org/show_bug.cgi?id=79686 Reviewed by Filip Pizlo. * heap/CopiedBlock.h: (CopiedBlock): Removed the extra padding that was used for alignment purposes until the calculation of the payload offset into CopiedBlocks was redone recently. 2012-02-28 Anders Carlsson <andersca@apple.com> Fix build with newer versions of clang. Clang now warns since we're not passing a CFString literal to CFStringCreateWithFormatAndArguments, but it's OK to ignore this warning since clang is also checking that the caller (vprintf_stderr_common) takes a string literal. * wtf/Assertions.cpp: 2012-02-28 Mario Sanchez Prada <msanchez@igalia.com> [GTK] Add GMainLoop and GMainContext to be handled by GRefPtr https://bugs.webkit.org/show_bug.cgi?id=79496 Reviewed by Martin Robinson. Handle GMainLoop and GMainContext in GRefPtr, by calling g_main_loop_(un)ref and g_main_context_(un)ref in the implementation of the refGPtr and derefGPtr template functions. * wtf/gobject/GRefPtr.cpp: (WTF::refGPtr): (WTF): (WTF::derefGPtr): * wtf/gobject/GRefPtr.h: (WTF): * wtf/gobject/GTypedefs.h: 2012-02-28 Yong Li <yoli@rim.com> JSString::resolveRope() should report extra memory cost to the heap. https://bugs.webkit.org/show_bug.cgi?id=79555 Reviewed by Michael Saboff. At the time a JSString is constructed with fibers, it doesn't report extra memory cost, which is reasonable because it hasn't allocate new memory. However when the rope is resolved, it should report meory cost for the new buffer. * runtime/JSString.cpp: (JSC::JSString::resolveRope): 2012-02-27 Oliver Hunt <oliver@apple.com> sputnik/Unicode/Unicode_500/S7.2_A1.6_T1.html crashes in the interpreter https://bugs.webkit.org/show_bug.cgi?id=79728 Reviewed by Gavin Barraclough. When initialising a chained get instruction we may end up in a state where the instruction stream says we have a scopechain, but it has not yet been set (eg. if allocating the StructureChain itself is what leads to the GC). We could re-order the allocation, but it occurs in a couple of places, so it seems less fragile simply to null check the scopechain slot before we actually visit the slot. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitStructures): 2012-02-27 Filip Pizlo <fpizlo@apple.com> Old JIT's style of JSVALUE64 strict equality is subtly wrong https://bugs.webkit.org/show_bug.cgi?id=79700 Reviewed by Oliver Hunt. * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::comparePtr): (MacroAssemblerX86_64): * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativeStrictEq): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeStrictEq): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeStrictEq): * jit/JITOpcodes.cpp: (JSC::JIT::compileOpStrictEq): (JSC::JIT::emitSlow_op_stricteq): (JSC::JIT::emitSlow_op_nstricteq): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): 2012-02-27 Gavin Barraclough <barraclough@apple.com> Implement support for op_negate and op_bitnot in the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=79617 Reviewed by Filip Pizlo. Add an ArithNegate op to the DFG JIT, to implement op_negate. This patch also adds support for op_negate to the JSVALUE64 baseline JIT (JSVALUE32_64 already had this), so that we can profile the slowpath usage. This is a 2.5%-3% Sunspider progression and a 1% win on Kraken. * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::sub_S): - Added sub_S from immediate. (ARMv7Assembler): (JSC::ARMv7Assembler::vneg): - Added double negate. * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::negateDouble): - Added double negate. (MacroAssemblerARMv7): (JSC::MacroAssemblerARMv7::branchNeg32): - Added. * assembler/MacroAssemblerX86.h: (MacroAssemblerX86): - moved loadDouble, absDouble to common. * assembler/MacroAssemblerX86Common.h: (MacroAssemblerX86Common): (JSC::MacroAssemblerX86Common::absDouble): - implementation can be shared. (JSC::MacroAssemblerX86Common::negateDouble): - Added. (JSC::MacroAssemblerX86Common::loadDouble): - allow absDouble to have a common implementation. * assembler/MacroAssemblerX86_64.h: (MacroAssemblerX86_64): - moved loadDouble, absDouble to common. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): - support ArithNegate. * dfg/DFGArithNodeFlagsInferencePhase.cpp: (JSC::DFG::ArithNodeFlagsInferencePhase::propagate): - support ArithNegate. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::makeSafe): - support ArithNegate. (JSC::DFG::ByteCodeParser::parseBlock): - support op_negate. * dfg/DFGCSEPhase.cpp: (JSC::DFG::CSEPhase::performNodeCSE): - support ArithNegate. * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): - support op_negate. * dfg/DFGGraph.h: (JSC::DFG::Graph::negateShouldSpeculateInteger): - support ArithNegate. * dfg/DFGNode.h: (JSC::DFG::Node::hasArithNodeFlags): - support ArithNegate. * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): - support ArithNegate. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileArithNegate): - support ArithNegate. * dfg/DFGSpeculativeJIT.h: (SpeculativeJIT): - support ArithNegate. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): - support ArithNegate. * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): - support ArithNegate. * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompileSlowCases): - Add support for op_negate in JSVALUE64. * jit/JITArithmetic.cpp: (JSC::JIT::emit_op_negate): (JSC::JIT::emitSlow_op_negate): - Add support for op_negate in JSVALUE64. 2012-02-27 Mahesh Kulkarni <mahesh.kulkarni@nokia.com> Unreviewed. Build fix for linux-bot (qt) after r109021. * runtime/Error.cpp: 2012-02-27 Oliver Hunt <oliver@apple.com> REGRESSION (r108112): AWS Management Console at amazon.com fails to initialize https://bugs.webkit.org/show_bug.cgi?id=79693 Reviewed by Filip Pizlo. Alas we can't provide the stack trace as an array, as despite everyone wanting an array, everyone arbitrarily creates the array by calling split on the stack trace. To create the array we would have provided them in the first place. This changes the exception's stack property to a \n separated string. To get the old array just do <exception>.stack.split("\n"). * runtime/Error.cpp: (JSC::addErrorInfo): 2012-02-27 Gavin Barraclough <barraclough@apple.com> RegExp lastIndex should behave as a regular property https://bugs.webkit.org/show_bug.cgi?id=79446 Reviewed by Sam Weinig. lastIndex should be a regular data descriptor, with the attributes configurable:false, enumerable:false, writable:true. As such, it should be possible to reconfigure writable as false. If the lastIndex property is reconfigured to be read-only, we should respect this correctly. * runtime/CommonIdentifiers.h: - Removed some unused identifiers, added lastIndex. * runtime/RegExpObject.cpp: (JSC::RegExpObject::getOwnPropertySlot): - lastIndex is no longer a static value, provided specific handling. (JSC::RegExpObject::getOwnPropertyDescriptor): - lastIndex is no longer a static value, provided specific handling. (JSC::RegExpObject::deleteProperty): - lastIndex is no longer a static value, provided specific handling. (JSC::RegExpObject::getOwnPropertyNames): - lastIndex is no longer a static value, provided specific handling. (JSC::RegExpObject::getPropertyNames): - lastIndex is no longer a static value, provided specific handling. (JSC::reject): - helper function for defineOwnProperty. (JSC::RegExpObject::defineOwnProperty): - lastIndex is no longer a static value, provided specific handling. (JSC::RegExpObject::put): - lastIndex is no longer a static value, provided specific handling. (JSC::RegExpObject::match): - Pass setLastIndex an ExecState, so it can throw if read-only. * runtime/RegExpObject.h: (JSC::RegExpObject::setLastIndex): - Pass setLastIndex an ExecState, so it can throw if read-only. (RegExpObjectData): - Added lastIndexIsWritable. * runtime/RegExpPrototype.cpp: (JSC::regExpProtoFuncCompile): - Pass setLastIndex an ExecState, so it can throw if read-only. 2012-02-27 Gavin Barraclough <barraclough@apple.com> Implement support for op_negate and op_bitnot in the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=79617 Reviewed by Sam Weinig. Remove op_bitnop - this is redundant, ~x === x^-1. This is a fractional (<1%) progression. Remove not32(X) from the MacroAssemblers - make this an optimization to add32(-1, X). Remove CanReuse from the result type - this was unused. Remove op_bitnot. * assembler/MacroAssemblerARM.h: (MacroAssemblerARM): (JSC::MacroAssemblerARM::xor32): * assembler/MacroAssemblerARMv7.h: (MacroAssemblerARMv7): (JSC::MacroAssemblerARMv7::xor32): * assembler/MacroAssemblerMIPS.h: (MacroAssemblerMIPS): (JSC::MacroAssemblerMIPS::xor32): * assembler/MacroAssemblerSH4.h: (MacroAssemblerSH4): (JSC::MacroAssemblerSH4::xor32): * assembler/MacroAssemblerX86Common.h: (MacroAssemblerX86Common): (JSC::MacroAssemblerX86Common::xor32): * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): * bytecode/Opcode.h: (JSC): (JSC::padOpcodeName): * bytecompiler/NodesCodegen.cpp: (JSC): (JSC::BitwiseNotNode::emitBytecode): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompileSlowCases): * jit/JIT.h: (JIT): * jit/JITArithmetic32_64.cpp: (JSC): * jit/JITOpcodes.cpp: (JSC): * jit/JITStubs.cpp: (JSC): * jit/JITStubs.h: * llint/LLIntSlowPaths.cpp: (LLInt): * llint/LLIntSlowPaths.h: (LLInt): * llint/LowLevelInterpreter32_64.asm: * parser/NodeConstructors.h: (JSC::NegateNode::NegateNode): (JSC::BitwiseNotNode::BitwiseNotNode): (JSC::MultNode::MultNode): (JSC::DivNode::DivNode): (JSC::ModNode::ModNode): (JSC::SubNode::SubNode): (JSC::UnsignedRightShiftNode::UnsignedRightShiftNode): * parser/Nodes.h: (BitwiseNotNode): (JSC::BitwiseNotNode::expr): (JSC): * parser/ResultType.h: (ResultType): (JSC::ResultType::numberTypeIsInt32): (JSC::ResultType::stringOrNumberType): (JSC::ResultType::forAdd): (JSC::ResultType::forBitOp): 2012-02-27 Michael Saboff <msaboff@apple.com> Error check regexp min quantifier https://bugs.webkit.org/show_bug.cgi?id=70648 Reviewed by Gavin Barraclough. Added checking for min or only quantifier being UINT_MAX. When encountered this becomes a SyntaxError during parsing. * yarr/YarrParser.h: (JSC::Yarr::Parser::parseQuantifier): (JSC::Yarr::Parser::parse): (Parser): 2012-02-27 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck. * GNUmakefile.list.am: Add missing files. 2012-02-26 Hajime Morrita <morrita@chromium.org> Move ChromeClient::showContextMenu() to ContextMenuClient https://bugs.webkit.org/show_bug.cgi?id=79427 Reviewed by Adam Barth. Added ACCESSIBILITY_CONTEXT_MENUS. * wtf/Platform.h: 2012-02-26 Filip Pizlo <fpizlo@apple.com> LayoutTests/fast/xpath/xpath-functional-test.html is crashing in the DFG https://bugs.webkit.org/show_bug.cgi?id=79616 Reviewed by Oliver Hunt. Guard against the fact that in JSVALUE64, JSValue().isCell() == true. * dfg/DFGAbstractValue.h: (JSC::DFG::AbstractValue::validate): 2012-02-26 Filip Pizlo <fpizlo@apple.com> DFG should support activations and nested functions https://bugs.webkit.org/show_bug.cgi?id=79554 Reviewed by Sam Weinig. Fix 32-bit. The 32-bit function+activation code had some really weird register reuse bugs. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-02-26 Filip Pizlo <fpizlo@apple.com> Getting the instruction stream for a code block should not require two loads https://bugs.webkit.org/show_bug.cgi?id=79608 Reviewed by Sam Weinig. Introduced the RefCountedArray class, which contains a single inline pointer to a ref-counted non-resizeable vector backing store. This satisfies the requirements of CodeBlock, which desires the ability to share instruction streams with other CodeBlocks. It also reduces the number of loads required for getting the instruction stream by one. This patch also gets rid of the bytecode discarding logic, since we don't use it anymore and it's unlikely to ever work right with DFG or LLInt. And I didn't feel like porting dead code to use RefCountedArray. * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.cpp: (JSC::instructionOffsetForNth): (JSC::CodeBlock::dump): (JSC::CodeBlock::CodeBlock): (JSC::CodeBlock::finalizeUnconditionally): (JSC::CodeBlock::handlerForBytecodeOffset): (JSC::CodeBlock::lineNumberForBytecodeOffset): (JSC::CodeBlock::expressionRangeForBytecodeOffset): (JSC::CodeBlock::shrinkToFit): * bytecode/CodeBlock.h: (CodeBlock): (JSC::CodeBlock::numberOfInstructions): (JSC::CodeBlock::instructions): (JSC::CodeBlock::instructionCount): (JSC::CodeBlock::valueProfileForBytecodeOffset): (JSC): * bytecompiler/BytecodeGenerator.cpp: (JSC::Label::setLocation): (JSC): (JSC::BytecodeGenerator::generate): (JSC::BytecodeGenerator::newLabel): * bytecompiler/BytecodeGenerator.h: (JSC): (BytecodeGenerator): (JSC::BytecodeGenerator::instructions): * bytecompiler/Label.h: (JSC::Label::Label): (Label): * dfg/DFGByteCodeCache.h: (JSC::DFG::ByteCodeCache::~ByteCodeCache): (JSC::DFG::ByteCodeCache::get): * jit/JITExceptions.cpp: (JSC::genericThrow): * llint/LowLevelInterpreter32_64.asm: * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::codeBlockWithBytecodeFor): (JSC::FunctionExecutable::produceCodeBlockFor): * wtf/RefCountedArray.h: Added. (WTF): (RefCountedArray): (WTF::RefCountedArray::RefCountedArray): (WTF::RefCountedArray::operator=): (WTF::RefCountedArray::~RefCountedArray): (WTF::RefCountedArray::size): (WTF::RefCountedArray::data): (WTF::RefCountedArray::begin): (WTF::RefCountedArray::end): (WTF::RefCountedArray::at): (WTF::RefCountedArray::operator[]): (Header): (WTF::RefCountedArray::Header::size): (WTF::RefCountedArray::Header::payload): (WTF::RefCountedArray::Header::fromPayload): * wtf/Platform.h: 2012-02-26 Yusuke Suzuki <utatane.tea@gmail.com> StringLiteral and NumericLiteral are allowed as ObjectLiteral getter / setter name https://bugs.webkit.org/show_bug.cgi?id=79571 Reviewed by Gavin Barraclough. * parser/ASTBuilder.h: (JSC::ASTBuilder::createGetterOrSetterProperty): * parser/Parser.cpp: (JSC::::parseProperty): * parser/SyntaxChecker.h: (JSC::SyntaxChecker::createGetterOrSetterProperty): 2012-02-26 Mark Hahnenberg <mhahnenberg@apple.com> Implement fast path for op_new_array in the baseline JIT https://bugs.webkit.org/show_bug.cgi?id=78612 Reviewed by Filip Pizlo. heap/CopiedAllocator.h: (CopiedAllocator): Friended the JIT to allow access to m_currentOffset. * heap/CopiedSpace.h: (CopiedSpace): Friended the JIT to allow access to isOversize. (JSC::CopiedSpace::allocator): * heap/Heap.h: (JSC::Heap::storageAllocator): Added a getter for the CopiedAllocator class so the JIT can use it for simple allocation i.e. when we can just bump the offset without having to do anything else. * jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases): Added new slow case for op_new_array for when we have to bail out because the fast allocation path fails for whatever reason. * jit/JIT.h: (JIT): * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateBasicStorage): Added utility function that allows objects to allocate generic backing stores. This function is used by emitAllocateJSArray. (JSC): (JSC::JIT::emitAllocateJSArray): Added utility function that allows the client to more easily allocate JSArrays. This function is used by emit_op_new_array and I expect it will also be used for emit_op_new_array_buffer. * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_new_array): Changed to do inline allocation of JSArrays. Still does a stub call for oversize arrays. (JSC): (JSC::JIT::emitSlow_op_new_array): New slow path that just bails out to a stub call if we fail in any way on the fast path. * runtime/JSArray.cpp: (JSC): * runtime/JSArray.h: Added lots of offset functions for all the fields that we need to initialize in the JIT. (ArrayStorage): (JSC::ArrayStorage::lengthOffset): (JSC::ArrayStorage::numValuesInVectorOffset): (JSC::ArrayStorage::allocBaseOffset): (JSC::ArrayStorage::vectorOffset): (JSArray): (JSC::JSArray::sparseValueMapOffset): (JSC::JSArray::subclassDataOffset): (JSC::JSArray::indexBiasOffset): (JSC): (JSC::JSArray::storageSize): Moved this function from being a static function in the cpp file to being a static function in the JSArray class. This move allows the JIT to call it to see what size it should allocate. 2012-02-26 Patrick Gansterer <paroga@webkit.org> Unreviewed. Build fix for ENABLE(CLASSIC_INTERPRETER) after r108681. * interpreter/Interpreter.cpp: (JSC::getLineNumberForCallFrame): (JSC::Interpreter::getStackTrace): 2012-02-26 Patrick Gansterer <paroga@webkit.org> Unreviewed. Build fix for !ENABLE(JIT) after r108681. * interpreter/Interpreter.cpp: (JSC::getLineNumberForCallFrame): 2012-02-25 Filip Pizlo <fpizlo@apple.com> LLInt assembly file should be split into 32-bit and 64-bit parts https://bugs.webkit.org/show_bug.cgi?id=79584 Reviewed by Sam Weinig. Moved LowLevelInterpreter.asm to LowLevelInterpreter32_64.asm. Gave offlineasm the ability to include files, and correctly track dependencies: it restricts the include mechanism to using the same directory as the source file, and uses the SHA1 hash of all .asm files in that directory as an input hash. * llint/LLIntOfflineAsmConfig.h: * llint/LowLevelInterpreter.asm: * llint/LowLevelInterpreter32_64.asm: Added. - This is just the entire contents of what was previously LowLevelInterpreter.asm * llint/LowLevelInterpreter64.asm: Added. * offlineasm/asm.rb: * offlineasm/ast.rb: * offlineasm/generate_offset_extractor.rb: * offlineasm/parser.rb: * offlineasm/self_hash.rb: 2012-02-25 Filip Pizlo <fpizlo@apple.com> Offlineasm should support X86_64 https://bugs.webkit.org/show_bug.cgi?id=79581 Reviewed by Oliver Hunt. * llint/LLIntOfflineAsmConfig.h: * offlineasm/backends.rb: * offlineasm/instructions.rb: * offlineasm/settings.rb: * offlineasm/x86.rb: 2012-02-25 Filip Pizlo <fpizlo@apple.com> DFG should support activations and nested functions https://bugs.webkit.org/show_bug.cgi?id=79554 Reviewed by Oliver Hunt. Wrote the simplest possible implementation of activations. Big speed-up on code that uses activations, no speed-up on major benchmarks (SunSpider, V8, Kraken) because they do not appear to have sufficient coverage over code that uses activations. * bytecode/PredictedType.cpp: (JSC::predictionToString): (JSC::predictionFromValue): * bytecode/PredictedType.h: (JSC): (JSC::isEmptyPrediction): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (ByteCodeParser): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::buildOperandMapsIfNecessary): (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): (JSC::DFG::canInlineOpcode): * dfg/DFGGraph.h: (JSC::DFG::Graph::needsActivation): * dfg/DFGNode.h: (DFG): (JSC::DFG::Node::storageAccessDataIndex): (Node): (JSC::DFG::Node::hasFunctionDeclIndex): (JSC::DFG::Node::functionDeclIndex): (JSC::DFG::Node::hasFunctionExprIndex): (JSC::DFG::Node::functionExprIndex): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileNewFunctionNoCheck): (DFG): (JSC::DFG::SpeculativeJIT::compileNewFunctionExpression): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::callOperation): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-02-25 Benjamin Poulain <benjamin@webkit.org> Add an empty skeleton of KURL for WTFURL https://bugs.webkit.org/show_bug.cgi?id=78990 Reviewed by Adam Barth. * JavaScriptCore.xcodeproj/project.pbxproj: Export the relevant classes from WTFURL so that can use them in WebCore. 2012-02-25 Filip Pizlo <fpizlo@apple.com> Unreviewed, fix build for DFG disabled and LLInt enabled. * jit/JIT.cpp: (JSC::JIT::privateCompile): * llint/LLIntSlowPaths.cpp: (LLInt): (JSC::LLInt::LLINT_SLOW_PATH_DECL): 2012-02-25 Mark Hahnenberg <mhahnenberg@apple.com> Fix the CopiedBlock offset alignment in a cross platform fashion https://bugs.webkit.org/show_bug.cgi?id=79556 Reviewed by Filip Pizlo. Replaced m_payload with a payload() method that calculates the offset of the payload with the proper alignment. This change allows us to avoid alignment-related issues in a cross-platform manner. * heap/CopiedAllocator.h: (JSC::CopiedAllocator::currentUtilization): * heap/CopiedBlock.h: (JSC::CopiedBlock::CopiedBlock): (JSC::CopiedBlock::payload): (CopiedBlock): * heap/CopiedSpace.cpp: (JSC::CopiedSpace::doneFillingBlock): * heap/CopiedSpaceInlineMethods.h: (JSC::CopiedSpace::borrowBlock): (JSC::CopiedSpace::allocateFromBlock): 2012-02-24 Michael Saboff <msaboff@apple.com> Unreviewed, Windows build fix. Changed signature in export to match change made in r108858. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-02-24 Filip Pizlo <fpizlo@apple.com> DFG support for op_new_regexp should be enabled https://bugs.webkit.org/show_bug.cgi?id=79538 Reviewed by Oliver Hunt. No performance change. * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGCommon.h: 2012-02-24 Michael Saboff <msaboff@apple.com> ASSERT(position < 0) in JSC::Yarr::Interpreter::InputStream::readChecked https://bugs.webkit.org/show_bug.cgi?id=73728 Reviewed by Gavin Barraclough. Fixed the mixing of signed and unsigned character indeces in YARR interpreter. * runtime/RegExp.cpp: (JSC::RegExp::match): Added code to check for match longer than 2^31 and return no match after resetting the offsets. * yarr/YarrInterpreter.cpp: Changed to use unsigned for all character index handling except when matching back references. (JSC::Yarr::Interpreter::InputStream::readChecked): (JSC::Yarr::Interpreter::InputStream::checkInput): (JSC::Yarr::Interpreter::InputStream::uncheckInput): (JSC::Yarr::Interpreter::InputStream::atStart): (JSC::Yarr::Interpreter::InputStream::atEnd): (JSC::Yarr::Interpreter::InputStream::isAvailableInput): (JSC::Yarr::Interpreter::checkCharacter): (JSC::Yarr::Interpreter::checkCasedCharacter): (JSC::Yarr::Interpreter::checkCharacterClass): (JSC::Yarr::Interpreter::tryConsumeBackReference): (JSC::Yarr::Interpreter::matchAssertionBOL): (JSC::Yarr::Interpreter::matchAssertionWordBoundary): (JSC::Yarr::Interpreter::backtrackPatternCharacter): (JSC::Yarr::Interpreter::backtrackPatternCasedCharacter): (JSC::Yarr::Interpreter::matchCharacterClass): (JSC::Yarr::Interpreter::backtrackCharacterClass): (JSC::Yarr::Interpreter::matchParenthesesOnceBegin): (JSC::Yarr::Interpreter::matchDisjunction): (JSC::Yarr::Interpreter::interpret): (JSC::Yarr::ByteCompiler::assertionBOL): (JSC::Yarr::ByteCompiler::assertionEOL): (JSC::Yarr::ByteCompiler::assertionWordBoundary): (JSC::Yarr::ByteCompiler::atomPatternCharacter): (JSC::Yarr::ByteCompiler::atomCharacterClass): (JSC::Yarr::ByteCompiler::atomBackReference): (JSC::Yarr::ByteCompiler::atomParenthesesOnceBegin): (JSC::Yarr::ByteCompiler::atomParenthesesTerminalBegin): (JSC::Yarr::ByteCompiler::atomParenthesesSubpatternBegin): (JSC::Yarr::ByteCompiler::atomParentheticalAssertionEnd): (JSC::Yarr::ByteCompiler::emitDisjunction): * yarr/YarrInterpreter.h: 2012-02-24 Filip Pizlo <fpizlo@apple.com> Unreviewed, build fix for builds where the DFG is disabled but the LLInt is enabled. * llint/LLIntOfflineAsmConfig.h: * llint/LowLevelInterpreter.asm: 2012-02-24 Filip Pizlo <fpizlo@apple.com> DFG should be able to handle variables getting captured https://bugs.webkit.org/show_bug.cgi?id=79469 Reviewed by Oliver Hunt. Made captured variables work by placing a Flush on the SetLocal and forcing the emission of the GetLocal even if copy propagation tells us who has the value. Changed the CFA and various prediction codes to understand that we can't really prove anything about captured variables. Well, we could in the future by just looking at what side effects are happening, but in this first cut we just assume that we can't reason about captured variables. Also added a mode where the DFG pretends that all variables and arguments got captured. Used this mode to harden the code. This is performance neutral. Capturing all variables is a slow down, but not too big of one. This seems to predict that when we add activation support, the amount of speed benefit we'll get from increased coverage will far outweigh the pessimism that we'll have to endure for captured variables. * bytecode/CodeType.h: (JSC::codeTypeToString): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::initialize): (JSC::DFG::AbstractState::endBasicBlock): (JSC::DFG::AbstractState::execute): (JSC::DFG::AbstractState::merge): * dfg/DFGAbstractState.h: (AbstractState): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getLocal): (JSC::DFG::ByteCodeParser::setLocal): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::setArgument): (JSC::DFG::ByteCodeParser::flushArgument): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::processPhiStack): (JSC::DFG::ByteCodeParser::parseCodeBlock): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGCapabilities.h: (JSC::DFG::mightInlineFunctionForCall): (JSC::DFG::mightInlineFunctionForConstruct): * dfg/DFGCommon.h: * dfg/DFGGraph.h: (JSC::DFG::Graph::needsActivation): (Graph): (JSC::DFG::Graph::argumentIsCaptured): (JSC::DFG::Graph::localIsCaptured): (JSC::DFG::Graph::isCaptured): * dfg/DFGNode.h: (JSC::DFG::Node::shouldGenerate): * dfg/DFGPredictionPropagationPhase.cpp: (JSC::DFG::PredictionPropagationPhase::propagate): (JSC::DFG::PredictionPropagationPhase::doRoundOfDoubleVoting): * dfg/DFGSpeculativeJIT.cpp: (DFG): (JSC::DFG::ValueSource::dump): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (ValueSource): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGVirtualRegisterAllocationPhase.cpp: (JSC::DFG::VirtualRegisterAllocationPhase::run): 2012-02-24 Gavin Barraclough <barraclough@apple.com> Should not allow malformed \x escapes https://bugs.webkit.org/show_bug.cgi?id=79462 Reviewed by Oliver Hunt. * parser/Lexer.cpp: (JSC::::parseString): (JSC::::parseStringSlowCase): - Prohibit malformed '\x' escapes * tests/mozilla/ecma/Array/15.4.5.1-1.js: * tests/mozilla/ecma/LexicalConventions/7.7.4.js: * tests/mozilla/ecma_2/RegExp/hex-001.js: * tests/mozilla/js1_2/regexp/hexadecimal.js: - Remove erroneous test cases (correct behaviour is tested by LayoutTests/sputnik). 2012-02-24 Daniel Bates <dbates@webkit.org> Fix change log entry for changeset r108819; add bug URL https://bugs.webkit.org/show_bug.cgi?id=79504 Changeset r108819 is associated with bug #79504. * ChangeLog 2012-02-24 Daniel Bates <dbates@webkit.org> Substitute ENABLE(CLASSIC_INTERPRETER) for ENABLE(INTERPRETER) in Interpreter.cpp https://bugs.webkit.org/show_bug.cgi?id=79504 Reviewed by Oliver Hunt. There are a few places in Interpreter.cpp that need to be updated to use ENABLE(CLASSIC_INTERPRETER) following the renaming of ENABLE_INTERPRETER to ENABLE_CLASSIC_INTERPRETER in changeset <http://trac.webkit.org/changeset/108020> (https://bugs.webkit.org/show_bug.cgi?id=78791). * interpreter/Interpreter.cpp: (JSC::getLineNumberForCallFrame): (JSC::getCallerInfo): (JSC::getSourceURLFromCallFrame): 2012-02-24 Adam Roben <aroben@apple.com> Undo the BUILDING_WTF part of r108808 This broke the build, which is obviously worse than the linker warning it was trying to solve. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: 2012-02-24 Adam Roben <aroben@apple.com> Fix linker warnings on Windows * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Removed symbols that are already exported via JS_EXPORTDATA. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: Define BUILDING_WTF. We aren't actually building WTF, but we are statically linking it, so we need to define this symbol so that we export WTF's exports. 2012-02-24 Philippe Normand <pnormand@igalia.com> Fix GTK WebAudio build for WebKitGTK 1.7.90. Patch by Priit Laes <plaes@plaes.org> on 2012-02-24 Rubber-stamped by Philippe Normand. * GNUmakefile.list.am: Add Complex.h to the list of files so it gets disted in the tarballs. 2012-02-24 Zoltan Herczeg <zherczeg@webkit.org> [Qt] Buildfix for "Zero out CopiedBlocks on initialization". https://bugs.webkit.org/show_bug.cgi?id=79199 Ruber stamped by Csaba Osztrogonác. Temporary fix since the new member wastes a little space on 64 bit systems. Although it is harmless, it is only needed for 32 bit systems. * heap/CopiedBlock.h: (CopiedBlock): 2012-02-24 Han Hojong <hojong.han@samsung.com> Remove useless jump instructions for short circuit https://bugs.webkit.org/show_bug.cgi?id=75602 Reviewed by Michael Saboff. Jump instruction is inserted to make short circuit, however it does nothing but moving to the next instruction. Therefore useless jump instructions are removed, and jump list is moved into the case not for a short circuit, so that only necessary instructions are added to JIT code unless it has a 16 bit pattern character and an 8 bit string. * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::generatePatternCharacterGreedy): (JSC::Yarr::YarrGenerator::backtrackPatternCharacterNonGreedy): 2012-02-24 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r108731. http://trac.webkit.org/changeset/108731 https://bugs.webkit.org/show_bug.cgi?id=79464 Broke Chromium Win tests (Requested by bashi on #webkit). * wtf/Platform.h: 2012-02-24 Andrew Lo <anlo@rim.com> [BlackBerry] Enable requestAnimationFrame https://bugs.webkit.org/show_bug.cgi?id=79408 Use timer implementation of requestAnimationFrame on BlackBerry. Reviewed by Rob Buis. * wtf/Platform.h: 2012-02-24 Mathias Bynens <mathias@qiwi.be> `\u200c` and `\u200d` should be allowed in IdentifierPart, as per ES5 https://bugs.webkit.org/show_bug.cgi?id=78908 Add additional checks for zero-width non-joiner (0x200C) and zero-width joiner (0x200D) characters. Reviewed by Michael Saboff. * parser/Lexer.cpp: (JSC::isNonASCIIIdentPart) * runtime/LiteralParser.cpp: (JSC::::Lexer::lexIdentifier) 2012-02-23 Kenichi Ishibashi <bashi@chromium.org> Adding WebSocket per-frame DEFLATE extension https://bugs.webkit.org/show_bug.cgi?id=77522 Added USE(ZLIB) flag. Reviewed by Kent Tamura. * wtf/Platform.h: 2012-02-23 Mark Hahnenberg <mhahnenberg@apple.com> Zero out CopiedBlocks on initialization https://bugs.webkit.org/show_bug.cgi?id=79199 Reviewed by Filip Pizlo. Made CopyBlocks zero their payloads during construction. This allows JSArray to avoid having to manually clear its backing store upon allocation and also alleviates any future pain with regard to the garbage collector trying to mark what it thinks are values in what is actually uninitialized memory. * heap/CopiedBlock.h: (JSC::CopiedBlock::CopiedBlock): * runtime/JSArray.cpp: (JSC::JSArray::finishCreation): (JSC::JSArray::tryFinishCreationUninitialized): (JSC::JSArray::increaseVectorLength): (JSC::JSArray::unshiftCountSlowCase): 2012-02-23 Oliver Hunt <oliver@apple.com> Make Interpreter::getStackTrace be able to generate the line number for the top callframe if none is provided https://bugs.webkit.org/show_bug.cgi?id=79407 Reviewed by Gavin Barraclough. Outside of exception handling, we don't know what our source line number is. This change allows us to pass -1 is as the initial line number, and get the correct line number in the resultant stack trace. We can't completely elide the initial line number (yet) due to some idiosyncrasies of the exception handling machinery. * interpreter/Interpreter.cpp: (JSC::getLineNumberForCallFrame): (JSC): (JSC::Interpreter::getStackTrace): 2012-02-22 Filip Pizlo <fpizlo@apple.com> DFG OSR exit value profiling should have graceful handling of local variables and arguments https://bugs.webkit.org/show_bug.cgi?id=79310 Reviewed by Gavin Barraclough. Previously, if we OSR exited because a prediction in a local was wrong, we'd only realize what the true type of the local was if the regular value profiling kicked in and told us. Unless the local was block-locally copy propagated, in which case we'd know from an OSR exit profile. This patch adds OSR exit profiling to all locals and arguments. Now, if we OSR exit because of a mispredicted local or argument type, we'll know what the type of the local or argument should be immediately upon exiting. The way that local variable OSR exit profiling works is that we now have a lazily added set of OSR-exit-only value profiles for exit sites that are BadType and that cited a GetLocal as their value source. The value profiles are only added if the OSR exit is taken, and are keyed by CodeBlock, bytecode index of the GetLocal, and operand. The look-up is performed by querying the CompressedLazyOperandValueProfileHolder in the CodeBlock, using a key that contains the bytecode index and the operand. Because the value profiles are added at random times, they are not sorted; instead they are just stored in an arbitrarily-ordered SegmentedVector. Look-ups are made fast by "decompressing": the DFG::ByteCodeParser creates a LazyOperandValueProfileParser, which turns the CompressedLazyOperandValueProfileHolder's contents into a HashMap for the duration of DFG parsing. Previously, OSR exits had a pointer to the ValueProfile that had the specFailBucket into which values observed during OSR exit would be placed. Now it uses a lazy thunk for a ValueProfile. I call this the MethodOfGettingAValueProfile. It may either contain a ValueProfile inside it (which works for previous uses of OSR exit profiling) or it may just have knowledge of how to go about creating the LazyOperandValueProfile in the case that the OSR exit is actually taken. This ensures that we never have to create NumOperands*NumBytecodeIndices*NumCodeBlocks value profiling buckets unless we actually did OSR exit on every single operand, in every single instruction, in each code block (that's probably unlikely). This appears to be neutral on the major benchmarks, but is a double-digit speed-up on code deliberately written to have data flow that spans basic blocks and where the code exhibits post-optimization polymorphism in a local variable. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::stronglyVisitStrongReferences): * bytecode/CodeBlock.h: (CodeBlock): (JSC::CodeBlock::lazyOperandValueProfiles): * bytecode/LazyOperandValueProfile.cpp: Added. (JSC): (JSC::CompressedLazyOperandValueProfileHolder::CompressedLazyOperandValueProfileHolder): (JSC::CompressedLazyOperandValueProfileHolder::~CompressedLazyOperandValueProfileHolder): (JSC::CompressedLazyOperandValueProfileHolder::computeUpdatedPredictions): (JSC::CompressedLazyOperandValueProfileHolder::add): (JSC::LazyOperandValueProfileParser::LazyOperandValueProfileParser): (JSC::LazyOperandValueProfileParser::~LazyOperandValueProfileParser): (JSC::LazyOperandValueProfileParser::getIfPresent): (JSC::LazyOperandValueProfileParser::prediction): * bytecode/LazyOperandValueProfile.h: Added. (JSC): (LazyOperandValueProfileKey): (JSC::LazyOperandValueProfileKey::LazyOperandValueProfileKey): (JSC::LazyOperandValueProfileKey::operator!): (JSC::LazyOperandValueProfileKey::operator==): (JSC::LazyOperandValueProfileKey::hash): (JSC::LazyOperandValueProfileKey::bytecodeOffset): (JSC::LazyOperandValueProfileKey::operand): (JSC::LazyOperandValueProfileKey::isHashTableDeletedValue): (JSC::LazyOperandValueProfileKeyHash::hash): (JSC::LazyOperandValueProfileKeyHash::equal): (LazyOperandValueProfileKeyHash): (WTF): (JSC::LazyOperandValueProfile::LazyOperandValueProfile): (LazyOperandValueProfile): (JSC::LazyOperandValueProfile::key): (CompressedLazyOperandValueProfileHolder): (LazyOperandValueProfileParser): * bytecode/MethodOfGettingAValueProfile.cpp: Added. (JSC): (JSC::MethodOfGettingAValueProfile::fromLazyOperand): (JSC::MethodOfGettingAValueProfile::getSpecFailBucket): * bytecode/MethodOfGettingAValueProfile.h: Added. (JSC): (MethodOfGettingAValueProfile): (JSC::MethodOfGettingAValueProfile::MethodOfGettingAValueProfile): (JSC::MethodOfGettingAValueProfile::operator!): * bytecode/ValueProfile.cpp: Removed. * bytecode/ValueProfile.h: (JSC): (ValueProfileBase): (JSC::ValueProfileBase::ValueProfileBase): (JSC::ValueProfileBase::dump): (JSC::ValueProfileBase::computeUpdatedPrediction): (JSC::MinimalValueProfile::MinimalValueProfile): (ValueProfileWithLogNumberOfBuckets): (JSC::ValueProfileWithLogNumberOfBuckets::ValueProfileWithLogNumberOfBuckets): (JSC::ValueProfile::ValueProfile): (JSC::getValueProfileBytecodeOffset): (JSC::getRareCaseProfileBytecodeOffset): * dfg/DFGByteCodeParser.cpp: (ByteCodeParser): (JSC::DFG::ByteCodeParser::injectLazyOperandPrediction): (JSC::DFG::ByteCodeParser::getLocal): (JSC::DFG::ByteCodeParser::getArgument): (InlineStackEntry): (JSC::DFG::ByteCodeParser::fixVariableAccessPredictions): (DFG): (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGGraph.h: (JSC::DFG::Graph::valueProfileFor): (JSC::DFG::Graph::methodOfGettingAValueProfileFor): (Graph): * dfg/DFGNode.h: (Node): * dfg/DFGOSRExit.cpp: (JSC::DFG::OSRExit::OSRExit): * dfg/DFGOSRExit.h: (OSRExit): * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGPhase.cpp: (JSC::DFG::Phase::beginPhase): (JSC::DFG::Phase::endPhase): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::checkArgumentTypes): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::speculationCheck): * dfg/DFGVariableAccessData.h: (JSC::DFG::VariableAccessData::nonUnifiedPrediction): (VariableAccessData): 2012-02-23 Filip Pizlo <fpizlo@apple.com> Build fix. * llint/LLIntOffsetsExtractor.cpp: 2012-02-23 Kevin Ollivier <kevino@theolliviers.com> [wx] Build fix, disable LLINT for now and fix ENABLE defines for it. * llint/LLIntOffsetsExtractor.cpp: * wtf/Platform.h: 2012-02-23 Kevin Ollivier <kevino@theolliviers.com> [wx] Build fix for non-Mac wx builds. * runtime/DatePrototype.cpp: 2012-02-22 Filip Pizlo <fpizlo@apple.com> DFG's logic for emitting a Flush is too convoluted and contains an inaccurate comment https://bugs.webkit.org/show_bug.cgi?id=79334 Reviewed by Oliver Hunt. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getLocal): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::flush): 2012-02-23 Gavin Barraclough <barraclough@apple.com> Object.isSealed / Object.isFrozen don't work for native objects https://bugs.webkit.org/show_bug.cgi?id=79331 Reviewed by Sam Weinig. Need to inspect all properties, including static ones. This exposes a couple of bugs in Array & Arguments: - getOwnPropertyDescriptor doesn't correctly report the writable attribute of array length. - Arguments object's defineOwnProperty does not handle callee/caller/length correctly. * runtime/Arguments.cpp: (JSC::Arguments::defineOwnProperty): - Add handling for callee/caller/length. * runtime/JSArray.cpp: (JSC::JSArray::getOwnPropertyDescriptor): - report length's writability correctly. * runtime/ObjectConstructor.cpp: (JSC::objectConstructorSeal): (JSC::objectConstructorFreeze): (JSC::objectConstructorIsSealed): (JSC::objectConstructorIsFrozen): - Add spec-based implementation for non-final objects. 2012-02-23 Gavin Barraclough <barraclough@apple.com> pop of array hole should get from the prototype chain https://bugs.webkit.org/show_bug.cgi?id=79338 Reviewed by Sam Weinig. * runtime/JSArray.cpp: (JSC::JSArray::pop): - If the fast fast vector case fails, more closely follow the spec. 2012-02-23 Yong Li <yoli@rim.com> JSString::outOfMemory() should ASSERT(isRope()) rather than !isRope() https://bugs.webkit.org/show_bug.cgi?id=79268 Reviewed by Michael Saboff. resolveRope() is the only caller of outOfMemory(), and it calls outOfMemory() after it fails to allocate a buffer for m_value. So outOfMemory() should assert isRope() rather than !isRope(). * runtime/JSString.cpp: (JSC::JSString::outOfMemory): 2012-02-23 Patrick Gansterer <paroga@webkit.org> [CMake] Add WEBKIT_INCLUDE_CONFIG_FILES_IF_EXISTS macro https://bugs.webkit.org/show_bug.cgi?id=79371 Reviewed by Daniel Bates. * CMakeLists.txt: * shell/CMakeLists.txt: * wtf/CMakeLists.txt: 2012-02-23 Aron Rosenberg <arosenberg@logitech.com> Fix the PRI macros used in WTF::String formatters to be compatible with Qt and Visual Studio 2005 and newer. https://bugs.webkit.org/show_bug.cgi?id=76210 Add compile time check for Visual Studio 2005 or newer. Reviewed by Simon Hausmann. * os-win32/inttypes.h: 2012-02-22 Gavin Barraclough <barraclough@apple.com> Implement [[DefineOwnProperty]] for the arguments object https://bugs.webkit.org/show_bug.cgi?id=79309 Reviewed by Sam Weinig. * runtime/Arguments.cpp: (JSC::Arguments::deletePropertyByIndex): (JSC::Arguments::deleteProperty): - Deleting an argument should also delete the copy on the object, if any. (JSC::Arguments::defineOwnProperty): - Defining a property may override the live mapping. * runtime/Arguments.h: (Arguments): 2012-02-22 Gavin Barraclough <barraclough@apple.com> Fix Object.freeze for non-final objects. https://bugs.webkit.org/show_bug.cgi?id=79286 Reviewed by Oliver Hunt. For vanilla objects we implement this with a single transition, for objects with special properties we should just follow the spec defined algorithm. * runtime/JSArray.cpp: (JSC::SparseArrayValueMap::put): - this does need to handle inextensible objects. * runtime/ObjectConstructor.cpp: (JSC::objectConstructorSeal): (JSC::objectConstructorFreeze): - Implement spec defined algorithm for non-final objects. * runtime/Structure.cpp: (JSC::Structure::Structure): (JSC::Structure::freezeTransition): - freeze should set m_hasReadOnlyOrGetterSetterPropertiesExcludingProto. * runtime/Structure.h: (JSC::Structure::hasReadOnlyOrGetterSetterPropertiesExcludingProto): (JSC::Structure::setHasGetterSetterProperties): (JSC::Structure::setContainsReadOnlyProperties): (Structure): - renamed m_hasReadOnlyOrGetterSetterPropertiesExcludingProto. 2012-02-22 Mark Hahnenberg <mhahnenberg@apple.com> Allocations from CopiedBlocks should always be 8-byte aligned https://bugs.webkit.org/show_bug.cgi?id=79271 Reviewed by Geoffrey Garen. * heap/CopiedAllocator.h: (JSC::CopiedAllocator::allocate): * heap/CopiedBlock.h: Changed to add padding so that the start of the payload is always guaranteed to be 8 byte aligned on both 64- and 32-bit platforms. (CopiedBlock): * heap/CopiedSpace.cpp: Changed all assertions of isPointerAligned to is8ByteAligned. (JSC::CopiedSpace::tryAllocateOversize): (JSC::CopiedSpace::getFreshBlock): * heap/CopiedSpaceInlineMethods.h: (JSC::CopiedSpace::allocateFromBlock): * runtime/JSArray.h: (ArrayStorage): Added padding for ArrayStorage to make sure that it is always 8 byte aligned on both 64- and 32-bit platforms. * wtf/StdLibExtras.h: (WTF::is8ByteAligned): Added new utility function that functions similarly to the way isPointerAligned does, but it just always checks for 8 byte alignment. (WTF): 2012-02-22 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r108456. http://trac.webkit.org/changeset/108456 https://bugs.webkit.org/show_bug.cgi?id=79223 Broke fast/regex/pcre-test-4.html and cannot find anyone on IRC (Requested by zherczeg on #webkit). * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::backtrackPatternCharacterGreedy): 2012-02-22 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r108468. http://trac.webkit.org/changeset/108468 https://bugs.webkit.org/show_bug.cgi?id=79219 Broke Chromium Win release build (Requested by bashi on #webkit). * wtf/Platform.h: 2012-02-22 Kenichi Ishibashi <bashi@chromium.org> Adding WebSocket per-frame DEFLATE extension https://bugs.webkit.org/show_bug.cgi?id=77522 Added USE(ZLIB) flag. Reviewed by Kent Tamura. * wtf/Platform.h: 2012-02-22 Hojong Han <hojong.han@samsung.com> Short circuit fixed for a 16 bt pattern character and an 8 bit string. https://bugs.webkit.org/show_bug.cgi?id=75602 Reviewed by Gavin Barraclough. * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::backtrackPatternCharacterGreedy): 2012-02-21 Filip Pizlo <fpizlo@apple.com> Build fix for systems with case sensitive disks. * llint/LLIntOfflineAsmConfig.h: 2012-02-21 Filip Pizlo <fpizlo@apple.com> JSC should be a triple-tier VM https://bugs.webkit.org/show_bug.cgi?id=75812 <rdar://problem/10079694> Reviewed by Gavin Barraclough. Implemented an interpreter that uses the JIT's calling convention. This interpreter is called LLInt, or the Low Level Interpreter. JSC will now will start by executing code in LLInt and will only tier up to the old JIT after the code is proven hot. LLInt is written in a modified form of our macro assembly. This new macro assembly is compiled by an offline assembler (see offlineasm), which implements many modern conveniences such as a Turing-complete CPS-based macro language and direct access to relevant C++ type information (basically offsets of fields and sizes of structs/classes). Code executing in LLInt appears to the rest of the JSC world "as if" it were executing in the old JIT. Hence, things like exception handling and cross-execution-engine calls just work and require pretty much no additional overhead. This interpreter is 2-2.5x faster than our old interpreter on SunSpider, V8, and Kraken. With triple-tiering turned on, we're neutral on SunSpider, V8, and Kraken, but appear to get a double-digit improvement on real-world websites due to a huge reduction in the amount of JIT'ing. * CMakeLists.txt: * GNUmakefile.am: * GNUmakefile.list.am: * JavaScriptCore.pri: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * assembler/LinkBuffer.h: * assembler/MacroAssemblerCodeRef.h: (MacroAssemblerCodePtr): (JSC::MacroAssemblerCodePtr::createFromExecutableAddress): * bytecode/BytecodeConventions.h: Added. * bytecode/CallLinkStatus.cpp: (JSC::CallLinkStatus::computeFromLLInt): (JSC): (JSC::CallLinkStatus::computeFor): * bytecode/CallLinkStatus.h: (JSC::CallLinkStatus::isSet): (JSC::CallLinkStatus::operator!): (CallLinkStatus): * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): (JSC::CodeBlock::CodeBlock): (JSC::CodeBlock::~CodeBlock): (JSC::CodeBlock::finalizeUnconditionally): (JSC::CodeBlock::stronglyVisitStrongReferences): (JSC): (JSC::CodeBlock::unlinkCalls): (JSC::CodeBlock::unlinkIncomingCalls): (JSC::CodeBlock::bytecodeOffset): (JSC::ProgramCodeBlock::jettison): (JSC::EvalCodeBlock::jettison): (JSC::FunctionCodeBlock::jettison): (JSC::ProgramCodeBlock::jitCompileImpl): (JSC::EvalCodeBlock::jitCompileImpl): (JSC::FunctionCodeBlock::jitCompileImpl): * bytecode/CodeBlock.h: (JSC): (CodeBlock): (JSC::CodeBlock::baselineVersion): (JSC::CodeBlock::linkIncomingCall): (JSC::CodeBlock::bytecodeOffset): (JSC::CodeBlock::jitCompile): (JSC::CodeBlock::hasOptimizedReplacement): (JSC::CodeBlock::addPropertyAccessInstruction): (JSC::CodeBlock::addGlobalResolveInstruction): (JSC::CodeBlock::addLLIntCallLinkInfo): (JSC::CodeBlock::addGlobalResolveInfo): (JSC::CodeBlock::numberOfMethodCallLinkInfos): (JSC::CodeBlock::valueProfilePredictionForBytecodeOffset): (JSC::CodeBlock::likelyToTakeSlowCase): (JSC::CodeBlock::couldTakeSlowCase): (JSC::CodeBlock::likelyToTakeSpecialFastCase): (JSC::CodeBlock::likelyToTakeDeepestSlowCase): (JSC::CodeBlock::likelyToTakeAnySlowCase): (JSC::CodeBlock::addFrequentExitSite): (JSC::CodeBlock::dontJITAnytimeSoon): (JSC::CodeBlock::jitAfterWarmUp): (JSC::CodeBlock::jitSoon): (JSC::CodeBlock::llintExecuteCounter): (ProgramCodeBlock): (EvalCodeBlock): (FunctionCodeBlock): * bytecode/GetByIdStatus.cpp: (JSC::GetByIdStatus::computeFromLLInt): (JSC): (JSC::GetByIdStatus::computeFor): * bytecode/GetByIdStatus.h: (JSC::GetByIdStatus::GetByIdStatus): (JSC::GetByIdStatus::wasSeenInJIT): (GetByIdStatus): * bytecode/Instruction.h: (JSC): (JSC::Instruction::Instruction): (Instruction): * bytecode/LLIntCallLinkInfo.h: Added. (JSC): (JSC::LLIntCallLinkInfo::LLIntCallLinkInfo): (LLIntCallLinkInfo): (JSC::LLIntCallLinkInfo::~LLIntCallLinkInfo): (JSC::LLIntCallLinkInfo::isLinked): (JSC::LLIntCallLinkInfo::unlink): * bytecode/MethodCallLinkStatus.cpp: (JSC::MethodCallLinkStatus::computeFor): * bytecode/Opcode.cpp: (JSC): * bytecode/Opcode.h: (JSC): (JSC::padOpcodeName): * bytecode/PutByIdStatus.cpp: (JSC::PutByIdStatus::computeFromLLInt): (JSC): (JSC::PutByIdStatus::computeFor): * bytecode/PutByIdStatus.h: (PutByIdStatus): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitResolve): (JSC::BytecodeGenerator::emitResolveWithBase): (JSC::BytecodeGenerator::emitGetById): (JSC::BytecodeGenerator::emitPutById): (JSC::BytecodeGenerator::emitDirectPutById): (JSC::BytecodeGenerator::emitCall): (JSC::BytecodeGenerator::emitConstruct): (JSC::BytecodeGenerator::emitCatch): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getPredictionWithoutOSRExit): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGOSRExitCompiler.cpp: * dfg/DFGOperations.cpp: * heap/Heap.h: (JSC): (JSC::Heap::firstAllocatorWithoutDestructors): (Heap): * heap/MarkStack.cpp: (JSC::visitChildren): * heap/MarkedAllocator.h: (JSC): (MarkedAllocator): * heap/MarkedSpace.h: (JSC): (MarkedSpace): (JSC::MarkedSpace::firstAllocator): * interpreter/CallFrame.cpp: (JSC): (JSC::CallFrame::bytecodeOffsetForNonDFGCode): (JSC::CallFrame::setBytecodeOffsetForNonDFGCode): (JSC::CallFrame::currentVPC): (JSC::CallFrame::setCurrentVPC): (JSC::CallFrame::trueCallerFrame): * interpreter/CallFrame.h: (JSC::ExecState::hasReturnPC): (JSC::ExecState::clearReturnPC): (ExecState): (JSC::ExecState::bytecodeOffsetForNonDFGCode): (JSC::ExecState::currentVPC): (JSC::ExecState::setCurrentVPC): * interpreter/Interpreter.cpp: (JSC::Interpreter::Interpreter): (JSC::Interpreter::~Interpreter): (JSC): (JSC::Interpreter::initialize): (JSC::Interpreter::isOpcode): (JSC::Interpreter::unwindCallFrame): (JSC::getCallerInfo): (JSC::Interpreter::privateExecute): (JSC::Interpreter::retrieveLastCaller): * interpreter/Interpreter.h: (JSC): (Interpreter): (JSC::Interpreter::getOpcode): (JSC::Interpreter::getOpcodeID): (JSC::Interpreter::classicEnabled): * interpreter/RegisterFile.h: (JSC): (RegisterFile): * jit/ExecutableAllocator.h: (JSC): * jit/HostCallReturnValue.cpp: Added. (JSC): (JSC::getHostCallReturnValueWithExecState): * jit/HostCallReturnValue.h: Added. (JSC): (JSC::initializeHostCallReturnValue): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompileSlowCases): (JSC::JIT::privateCompile): * jit/JITCode.h: (JSC::JITCode::isOptimizingJIT): (JITCode): (JSC::JITCode::isBaselineCode): (JSC::JITCode::JITCode): * jit/JITDriver.h: (JSC::jitCompileIfAppropriate): (JSC::jitCompileFunctionIfAppropriate): * jit/JITExceptions.cpp: (JSC::jitThrow): * jit/JITInlineMethods.h: (JSC::JIT::updateTopCallFrame): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): (JSC): * jit/JITStubs.h: (JSC): * jit/JSInterfaceJIT.h: * llint: Added. * llint/LLIntCommon.h: Added. * llint/LLIntData.cpp: Added. (LLInt): (JSC::LLInt::Data::Data): (JSC::LLInt::Data::performAssertions): (JSC::LLInt::Data::~Data): * llint/LLIntData.h: Added. (JSC): (LLInt): (Data): (JSC::LLInt::Data::exceptionInstructions): (JSC::LLInt::Data::opcodeMap): (JSC::LLInt::Data::performAssertions): * llint/LLIntEntrypoints.cpp: Added. (LLInt): (JSC::LLInt::getFunctionEntrypoint): (JSC::LLInt::getEvalEntrypoint): (JSC::LLInt::getProgramEntrypoint): * llint/LLIntEntrypoints.h: Added. (JSC): (LLInt): (JSC::LLInt::getEntrypoint): * llint/LLIntExceptions.cpp: Added. (LLInt): (JSC::LLInt::interpreterThrowInCaller): (JSC::LLInt::returnToThrowForThrownException): (JSC::LLInt::returnToThrow): (JSC::LLInt::callToThrow): * llint/LLIntExceptions.h: Added. (JSC): (LLInt): * llint/LLIntOfflineAsmConfig.h: Added. * llint/LLIntOffsetsExtractor.cpp: Added. (JSC): (LLIntOffsetsExtractor): (JSC::LLIntOffsetsExtractor::dummy): (main): * llint/LLIntSlowPaths.cpp: Added. (LLInt): (JSC::LLInt::llint_trace_operand): (JSC::LLInt::llint_trace_value): (JSC::LLInt::LLINT_SLOW_PATH_DECL): (JSC::LLInt::traceFunctionPrologue): (JSC::LLInt::shouldJIT): (JSC::LLInt::entryOSR): (JSC::LLInt::resolveGlobal): (JSC::LLInt::getByVal): (JSC::LLInt::handleHostCall): (JSC::LLInt::setUpCall): (JSC::LLInt::genericCall): * llint/LLIntSlowPaths.h: Added. (JSC): (LLInt): * llint/LLIntThunks.cpp: Added. (LLInt): (JSC::LLInt::generateThunkWithJumpTo): (JSC::LLInt::functionForCallEntryThunkGenerator): (JSC::LLInt::functionForConstructEntryThunkGenerator): (JSC::LLInt::functionForCallArityCheckThunkGenerator): (JSC::LLInt::functionForConstructArityCheckThunkGenerator): (JSC::LLInt::evalEntryThunkGenerator): (JSC::LLInt::programEntryThunkGenerator): * llint/LLIntThunks.h: Added. (JSC): (LLInt): * llint/LowLevelInterpreter.asm: Added. * llint/LowLevelInterpreter.cpp: Added. * llint/LowLevelInterpreter.h: Added. * offlineasm: Added. * offlineasm/armv7.rb: Added. * offlineasm/asm.rb: Added. * offlineasm/ast.rb: Added. * offlineasm/backends.rb: Added. * offlineasm/generate_offset_extractor.rb: Added. * offlineasm/instructions.rb: Added. * offlineasm/offset_extractor_constants.rb: Added. * offlineasm/offsets.rb: Added. * offlineasm/opt.rb: Added. * offlineasm/parser.rb: Added. * offlineasm/registers.rb: Added. * offlineasm/self_hash.rb: Added. * offlineasm/settings.rb: Added. * offlineasm/transform.rb: Added. * offlineasm/x86.rb: Added. * runtime/CodeSpecializationKind.h: Added. (JSC): * runtime/CommonSlowPaths.h: (JSC::CommonSlowPaths::arityCheckFor): (CommonSlowPaths): * runtime/Executable.cpp: (JSC::jettisonCodeBlock): (JSC): (JSC::EvalExecutable::jitCompile): (JSC::samplingDescription): (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::jitCompile): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::baselineCodeBlockFor): (JSC::FunctionExecutable::jitCompileForCall): (JSC::FunctionExecutable::jitCompileForConstruct): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): * runtime/Executable.h: (JSC): (EvalExecutable): (ProgramExecutable): (FunctionExecutable): (JSC::FunctionExecutable::jitCompileFor): * runtime/ExecutionHarness.h: Added. (JSC): (JSC::prepareForExecution): (JSC::prepareFunctionForExecution): * runtime/JSArray.h: (JSC): (JSArray): * runtime/JSCell.h: (JSC): (JSCell): * runtime/JSFunction.h: (JSC): (JSFunction): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: (JSC): (JSGlobalData): * runtime/JSGlobalObject.h: (JSC): (JSGlobalObject): * runtime/JSObject.h: (JSC): (JSObject): (JSFinalObject): * runtime/JSPropertyNameIterator.h: (JSC): (JSPropertyNameIterator): * runtime/JSString.h: (JSC): (JSString): * runtime/JSTypeInfo.h: (JSC): (TypeInfo): * runtime/JSValue.cpp: (JSC::JSValue::description): * runtime/JSValue.h: (LLInt): (JSValue): * runtime/JSVariableObject.h: (JSC): (JSVariableObject): * runtime/Options.cpp: (Options): (JSC::Options::initializeOptions): * runtime/Options.h: (Options): * runtime/ScopeChain.h: (JSC): (ScopeChainNode): * runtime/Structure.cpp: (JSC::Structure::addPropertyTransition): * runtime/Structure.h: (JSC): (Structure): * runtime/StructureChain.h: (JSC): (StructureChain): * wtf/InlineASM.h: * wtf/Platform.h: * wtf/SentinelLinkedList.h: (SentinelLinkedList): (WTF::SentinelLinkedList::isEmpty): * wtf/text/StringImpl.h: (JSC): (StringImpl): 2012-02-21 Oliver Hunt <oliver@apple.com> Unbreak double-typed arrays on ARMv7 https://bugs.webkit.org/show_bug.cgi?id=79177 Reviewed by Gavin Barraclough. The existing code had completely broken address arithmetic. * JSCTypedArrayStubs.h: (JSC): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::storeDouble): (JSC::MacroAssemblerARMv7::storeFloat): 2012-02-21 Gavin Barraclough <barraclough@apple.com> Should be able to reconfigure a non-configurable property as read-only https://bugs.webkit.org/show_bug.cgi?id=79170 Reviewed by Sam Weinig. See ES5.1 8.12.9 10.a.i - the spec prohibits making a read-only property writable, but does not inhibit making a writable property read-only. * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: (JSC::JSGlobalData::setInDefineOwnProperty): (JSGlobalData): (JSC::JSGlobalData::isInDefineOwnProperty): - Added flag, tracking whether we are in JSObject::defineOwnProperty. * runtime/JSObject.cpp: (JSC::JSObject::deleteProperty): (DefineOwnPropertyScope): - Always allow properties to be deleted by DefineOwnProperty - assume it knows what it is doing! (JSC::DefineOwnPropertyScope::DefineOwnPropertyScope): (JSC::DefineOwnPropertyScope::~DefineOwnPropertyScope): - Added RAII helper. (JSC::JSObject::defineOwnProperty): - Track on the globalData when we are in this method. 2012-02-21 Oliver Hunt <oliver@apple.com> Make TypedArrays be available in commandline jsc https://bugs.webkit.org/show_bug.cgi?id=79163 Reviewed by Gavin Barraclough. Adds a compile time option to have jsc support a basic implementation of the TypedArrays available in WebCore. This lets us test the typed array logic in the JIT witout having to build webcore. * JSCTypedArrayStubs.h: Added. (JSC): * JavaScriptCore.xcodeproj/project.pbxproj: * jsc.cpp: (GlobalObject::finishCreation): (GlobalObject): (GlobalObject::addConstructableFunction): * runtime/JSGlobalData.h: (JSGlobalData): 2012-02-21 Tom Sepez <tsepez@chromium.org> equalIgnoringNullity() only comparing half the bytes for equality https://bugs.webkit.org/show_bug.cgi?id=79135 Reviewed by Adam Barth. * wtf/text/StringImpl.h: (WTF::equalIgnoringNullity): 2012-02-21 Roland Takacs <takacs.roland@stud.u-szeged.hu> Unnecessary preprocessor macros in MainThread.h/cpp https://bugs.webkit.org/show_bug.cgi?id=79083 Removed invalid/wrong PLATFORM(WINDOWS) preprocessor macro. * wtf/MainThread.cpp: (WTF): * wtf/MainThread.h: (WTF): 2012-02-21 Sam Weinig <sam@webkit.org> Attempt to fix the Snow Leopard build. * Configurations/Base.xcconfig: 2012-02-21 Sam Weinig <sam@webkit.org> Use libc++ when building with Clang on Mac https://bugs.webkit.org/show_bug.cgi?id=78981 Reviewed by Dan Bernstein. * Configurations/Base.xcconfig: 2012-02-21 Adam Roben <aroben@apple.com> Roll out r108309, r108323, and r108326 They broke the 32-bit Lion build. Original bugs is <http://webkit.org/b/75812> <rdar://problem/10079694>. * CMakeLists.txt: * GNUmakefile.am: * GNUmakefile.list.am: * JavaScriptCore.pri: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * assembler/LinkBuffer.h: * assembler/MacroAssemblerCodeRef.h: * bytecode/BytecodeConventions.h: Removed. * bytecode/CallLinkStatus.cpp: * bytecode/CallLinkStatus.h: * bytecode/CodeBlock.cpp: * bytecode/CodeBlock.h: * bytecode/GetByIdStatus.cpp: * bytecode/GetByIdStatus.h: * bytecode/Instruction.h: * bytecode/LLIntCallLinkInfo.h: Removed. * bytecode/MethodCallLinkStatus.cpp: * bytecode/Opcode.cpp: * bytecode/Opcode.h: * bytecode/PutByIdStatus.cpp: * bytecode/PutByIdStatus.h: * bytecompiler/BytecodeGenerator.cpp: * dfg/DFGByteCodeParser.cpp: * dfg/DFGCapabilities.h: * dfg/DFGOSRExitCompiler.cpp: * dfg/DFGOperations.cpp: * heap/Heap.h: * heap/MarkStack.cpp: * heap/MarkedAllocator.h: * heap/MarkedSpace.h: * interpreter/CallFrame.cpp: * interpreter/CallFrame.h: * interpreter/Interpreter.cpp: * interpreter/Interpreter.h: * interpreter/RegisterFile.h: * jit/ExecutableAllocator.h: * jit/HostCallReturnValue.cpp: Removed. * jit/HostCallReturnValue.h: Removed. * jit/JIT.cpp: * jit/JITCode.h: * jit/JITDriver.h: * jit/JITExceptions.cpp: * jit/JITInlineMethods.h: * jit/JITStubs.cpp: * jit/JITStubs.h: * jit/JSInterfaceJIT.h: * llint/LLIntCommon.h: Removed. * llint/LLIntData.cpp: Removed. * llint/LLIntData.h: Removed. * llint/LLIntEntrypoints.cpp: Removed. * llint/LLIntEntrypoints.h: Removed. * llint/LLIntExceptions.cpp: Removed. * llint/LLIntExceptions.h: Removed. * llint/LLIntOfflineAsmConfig.h: Removed. * llint/LLIntOffsetsExtractor.cpp: Removed. * llint/LLIntSlowPaths.cpp: Removed. * llint/LLIntSlowPaths.h: Removed. * llint/LLIntThunks.cpp: Removed. * llint/LLIntThunks.h: Removed. * llint/LowLevelInterpreter.asm: Removed. * llint/LowLevelInterpreter.cpp: Removed. * llint/LowLevelInterpreter.h: Removed. * offlineasm/armv7.rb: Removed. * offlineasm/asm.rb: Removed. * offlineasm/ast.rb: Removed. * offlineasm/backends.rb: Removed. * offlineasm/generate_offset_extractor.rb: Removed. * offlineasm/instructions.rb: Removed. * offlineasm/offset_extractor_constants.rb: Removed. * offlineasm/offsets.rb: Removed. * offlineasm/opt.rb: Removed. * offlineasm/parser.rb: Removed. * offlineasm/registers.rb: Removed. * offlineasm/self_hash.rb: Removed. * offlineasm/settings.rb: Removed. * offlineasm/transform.rb: Removed. * offlineasm/x86.rb: Removed. * runtime/CodeSpecializationKind.h: Removed. * runtime/CommonSlowPaths.h: * runtime/Executable.cpp: * runtime/Executable.h: * runtime/ExecutionHarness.h: Removed. * runtime/JSArray.h: * runtime/JSCell.h: * runtime/JSFunction.h: * runtime/JSGlobalData.cpp: * runtime/JSGlobalData.h: * runtime/JSGlobalObject.h: * runtime/JSObject.h: * runtime/JSPropertyNameIterator.h: * runtime/JSString.h: * runtime/JSTypeInfo.h: * runtime/JSValue.cpp: * runtime/JSValue.h: * runtime/JSVariableObject.h: * runtime/Options.cpp: * runtime/Options.h: * runtime/ScopeChain.h: * runtime/Structure.cpp: * runtime/Structure.h: * runtime/StructureChain.h: * wtf/InlineASM.h: * wtf/Platform.h: * wtf/SentinelLinkedList.h: * wtf/text/StringImpl.h: 2012-02-21 Gustavo Noronha Silva <kov@debian.org> and Bob Tracy <rct@frus.com> Does not build on IA64, SPARC and Alpha https://bugs.webkit.org/show_bug.cgi?id=79047 Rubber-stamped by Kent Tamura. * wtf/dtoa/utils.h: these architectures also have correct double operations, so add them to the appropriate side of the check. 2012-02-21 Filip Pizlo <fpizlo@apple.com> Fix massive crashes in all tests introduced by previous build fix, and fix non-DFG build. https://bugs.webkit.org/show_bug.cgi?id=75812 Reviewed by Csaba Osztrogonác. * dfg/DFGOperations.cpp: (JSC): * jit/HostCallReturnValue.h: (JSC::initializeHostCallReturnValue): 2012-02-21 Filip Pizlo <fpizlo@apple.com> Attempted build fix for ELF platforms. * dfg/DFGOperations.cpp: (JSC): (JSC::getHostCallReturnValueWithExecState): * jit/HostCallReturnValue.cpp: (JSC): * jit/HostCallReturnValue.h: (JSC::initializeHostCallReturnValue): 2012-02-20 Filip Pizlo <fpizlo@apple.com> JSC should be a triple-tier VM https://bugs.webkit.org/show_bug.cgi?id=75812 <rdar://problem/10079694> Reviewed by Gavin Barraclough. Implemented an interpreter that uses the JIT's calling convention. This interpreter is called LLInt, or the Low Level Interpreter. JSC will now will start by executing code in LLInt and will only tier up to the old JIT after the code is proven hot. LLInt is written in a modified form of our macro assembly. This new macro assembly is compiled by an offline assembler (see offlineasm), which implements many modern conveniences such as a Turing-complete CPS-based macro language and direct access to relevant C++ type information (basically offsets of fields and sizes of structs/classes). Code executing in LLInt appears to the rest of the JSC world "as if" it were executing in the old JIT. Hence, things like exception handling and cross-execution-engine calls just work and require pretty much no additional overhead. This interpreter is 2-2.5x faster than our old interpreter on SunSpider, V8, and Kraken. With triple-tiering turned on, we're neutral on SunSpider, V8, and Kraken, but appear to get a double-digit improvement on real-world websites due to a huge reduction in the amount of JIT'ing. * CMakeLists.txt: * GNUmakefile.am: * GNUmakefile.list.am: * JavaScriptCore.pri: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * assembler/LinkBuffer.h: * assembler/MacroAssemblerCodeRef.h: (MacroAssemblerCodePtr): (JSC::MacroAssemblerCodePtr::createFromExecutableAddress): * bytecode/BytecodeConventions.h: Added. * bytecode/CallLinkStatus.cpp: (JSC::CallLinkStatus::computeFromLLInt): (JSC): (JSC::CallLinkStatus::computeFor): * bytecode/CallLinkStatus.h: (JSC::CallLinkStatus::isSet): (JSC::CallLinkStatus::operator!): (CallLinkStatus): * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): (JSC::CodeBlock::CodeBlock): (JSC::CodeBlock::~CodeBlock): (JSC::CodeBlock::finalizeUnconditionally): (JSC::CodeBlock::stronglyVisitStrongReferences): (JSC): (JSC::CodeBlock::unlinkCalls): (JSC::CodeBlock::unlinkIncomingCalls): (JSC::CodeBlock::bytecodeOffset): (JSC::ProgramCodeBlock::jettison): (JSC::EvalCodeBlock::jettison): (JSC::FunctionCodeBlock::jettison): (JSC::ProgramCodeBlock::jitCompileImpl): (JSC::EvalCodeBlock::jitCompileImpl): (JSC::FunctionCodeBlock::jitCompileImpl): * bytecode/CodeBlock.h: (JSC): (CodeBlock): (JSC::CodeBlock::baselineVersion): (JSC::CodeBlock::linkIncomingCall): (JSC::CodeBlock::bytecodeOffset): (JSC::CodeBlock::jitCompile): (JSC::CodeBlock::hasOptimizedReplacement): (JSC::CodeBlock::addPropertyAccessInstruction): (JSC::CodeBlock::addGlobalResolveInstruction): (JSC::CodeBlock::addLLIntCallLinkInfo): (JSC::CodeBlock::addGlobalResolveInfo): (JSC::CodeBlock::numberOfMethodCallLinkInfos): (JSC::CodeBlock::valueProfilePredictionForBytecodeOffset): (JSC::CodeBlock::likelyToTakeSlowCase): (JSC::CodeBlock::couldTakeSlowCase): (JSC::CodeBlock::likelyToTakeSpecialFastCase): (JSC::CodeBlock::likelyToTakeDeepestSlowCase): (JSC::CodeBlock::likelyToTakeAnySlowCase): (JSC::CodeBlock::addFrequentExitSite): (JSC::CodeBlock::dontJITAnytimeSoon): (JSC::CodeBlock::jitAfterWarmUp): (JSC::CodeBlock::jitSoon): (JSC::CodeBlock::llintExecuteCounter): (ProgramCodeBlock): (EvalCodeBlock): (FunctionCodeBlock): * bytecode/GetByIdStatus.cpp: (JSC::GetByIdStatus::computeFromLLInt): (JSC): (JSC::GetByIdStatus::computeFor): * bytecode/GetByIdStatus.h: (JSC::GetByIdStatus::GetByIdStatus): (JSC::GetByIdStatus::wasSeenInJIT): (GetByIdStatus): * bytecode/Instruction.h: (JSC): (JSC::Instruction::Instruction): (Instruction): * bytecode/LLIntCallLinkInfo.h: Added. (JSC): (JSC::LLIntCallLinkInfo::LLIntCallLinkInfo): (LLIntCallLinkInfo): (JSC::LLIntCallLinkInfo::~LLIntCallLinkInfo): (JSC::LLIntCallLinkInfo::isLinked): (JSC::LLIntCallLinkInfo::unlink): * bytecode/MethodCallLinkStatus.cpp: (JSC::MethodCallLinkStatus::computeFor): * bytecode/Opcode.cpp: (JSC): * bytecode/Opcode.h: (JSC): (JSC::padOpcodeName): * bytecode/PutByIdStatus.cpp: (JSC::PutByIdStatus::computeFromLLInt): (JSC): (JSC::PutByIdStatus::computeFor): * bytecode/PutByIdStatus.h: (PutByIdStatus): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitResolve): (JSC::BytecodeGenerator::emitResolveWithBase): (JSC::BytecodeGenerator::emitGetById): (JSC::BytecodeGenerator::emitPutById): (JSC::BytecodeGenerator::emitDirectPutById): (JSC::BytecodeGenerator::emitCall): (JSC::BytecodeGenerator::emitConstruct): (JSC::BytecodeGenerator::emitCatch): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getPredictionWithoutOSRExit): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGOSRExitCompiler.cpp: * dfg/DFGOperations.cpp: * heap/Heap.h: (JSC): (JSC::Heap::firstAllocatorWithoutDestructors): (Heap): * heap/MarkStack.cpp: (JSC::visitChildren): * heap/MarkedAllocator.h: (JSC): (MarkedAllocator): * heap/MarkedSpace.h: (JSC): (MarkedSpace): (JSC::MarkedSpace::firstAllocator): * interpreter/CallFrame.cpp: (JSC): (JSC::CallFrame::bytecodeOffsetForNonDFGCode): (JSC::CallFrame::setBytecodeOffsetForNonDFGCode): (JSC::CallFrame::currentVPC): (JSC::CallFrame::setCurrentVPC): (JSC::CallFrame::trueCallerFrame): * interpreter/CallFrame.h: (JSC::ExecState::hasReturnPC): (JSC::ExecState::clearReturnPC): (ExecState): (JSC::ExecState::bytecodeOffsetForNonDFGCode): (JSC::ExecState::currentVPC): (JSC::ExecState::setCurrentVPC): * interpreter/Interpreter.cpp: (JSC::Interpreter::Interpreter): (JSC::Interpreter::~Interpreter): (JSC): (JSC::Interpreter::initialize): (JSC::Interpreter::isOpcode): (JSC::Interpreter::unwindCallFrame): (JSC::getCallerInfo): (JSC::Interpreter::privateExecute): (JSC::Interpreter::retrieveLastCaller): * interpreter/Interpreter.h: (JSC): (Interpreter): (JSC::Interpreter::getOpcode): (JSC::Interpreter::getOpcodeID): (JSC::Interpreter::classicEnabled): * interpreter/RegisterFile.h: (JSC): (RegisterFile): * jit/ExecutableAllocator.h: (JSC): * jit/HostCallReturnValue.cpp: Added. (JSC): (JSC::getHostCallReturnValueWithExecState): * jit/HostCallReturnValue.h: Added. (JSC): (JSC::initializeHostCallReturnValue): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompileSlowCases): (JSC::JIT::privateCompile): * jit/JITCode.h: (JSC::JITCode::isOptimizingJIT): (JITCode): (JSC::JITCode::isBaselineCode): (JSC::JITCode::JITCode): * jit/JITDriver.h: (JSC::jitCompileIfAppropriate): (JSC::jitCompileFunctionIfAppropriate): * jit/JITExceptions.cpp: (JSC::jitThrow): * jit/JITInlineMethods.h: (JSC::JIT::updateTopCallFrame): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): (JSC): * jit/JITStubs.h: (JSC): * jit/JSInterfaceJIT.h: * llint: Added. * llint/LLIntCommon.h: Added. * llint/LLIntData.cpp: Added. (LLInt): (JSC::LLInt::Data::Data): (JSC::LLInt::Data::performAssertions): (JSC::LLInt::Data::~Data): * llint/LLIntData.h: Added. (JSC): (LLInt): (Data): (JSC::LLInt::Data::exceptionInstructions): (JSC::LLInt::Data::opcodeMap): (JSC::LLInt::Data::performAssertions): * llint/LLIntEntrypoints.cpp: Added. (LLInt): (JSC::LLInt::getFunctionEntrypoint): (JSC::LLInt::getEvalEntrypoint): (JSC::LLInt::getProgramEntrypoint): * llint/LLIntEntrypoints.h: Added. (JSC): (LLInt): (JSC::LLInt::getEntrypoint): * llint/LLIntExceptions.cpp: Added. (LLInt): (JSC::LLInt::interpreterThrowInCaller): (JSC::LLInt::returnToThrowForThrownException): (JSC::LLInt::returnToThrow): (JSC::LLInt::callToThrow): * llint/LLIntExceptions.h: Added. (JSC): (LLInt): * llint/LLIntOfflineAsmConfig.h: Added. * llint/LLIntOffsetsExtractor.cpp: Added. (JSC): (LLIntOffsetsExtractor): (JSC::LLIntOffsetsExtractor::dummy): (main): * llint/LLIntSlowPaths.cpp: Added. (LLInt): (JSC::LLInt::llint_trace_operand): (JSC::LLInt::llint_trace_value): (JSC::LLInt::LLINT_SLOW_PATH_DECL): (JSC::LLInt::traceFunctionPrologue): (JSC::LLInt::shouldJIT): (JSC::LLInt::entryOSR): (JSC::LLInt::resolveGlobal): (JSC::LLInt::getByVal): (JSC::LLInt::handleHostCall): (JSC::LLInt::setUpCall): (JSC::LLInt::genericCall): * llint/LLIntSlowPaths.h: Added. (JSC): (LLInt): * llint/LLIntThunks.cpp: Added. (LLInt): (JSC::LLInt::generateThunkWithJumpTo): (JSC::LLInt::functionForCallEntryThunkGenerator): (JSC::LLInt::functionForConstructEntryThunkGenerator): (JSC::LLInt::functionForCallArityCheckThunkGenerator): (JSC::LLInt::functionForConstructArityCheckThunkGenerator): (JSC::LLInt::evalEntryThunkGenerator): (JSC::LLInt::programEntryThunkGenerator): * llint/LLIntThunks.h: Added. (JSC): (LLInt): * llint/LowLevelInterpreter.asm: Added. * llint/LowLevelInterpreter.cpp: Added. * llint/LowLevelInterpreter.h: Added. * offlineasm: Added. * offlineasm/armv7.rb: Added. * offlineasm/asm.rb: Added. * offlineasm/ast.rb: Added. * offlineasm/backends.rb: Added. * offlineasm/generate_offset_extractor.rb: Added. * offlineasm/instructions.rb: Added. * offlineasm/offset_extractor_constants.rb: Added. * offlineasm/offsets.rb: Added. * offlineasm/opt.rb: Added. * offlineasm/parser.rb: Added. * offlineasm/registers.rb: Added. * offlineasm/self_hash.rb: Added. * offlineasm/settings.rb: Added. * offlineasm/transform.rb: Added. * offlineasm/x86.rb: Added. * runtime/CodeSpecializationKind.h: Added. (JSC): * runtime/CommonSlowPaths.h: (JSC::CommonSlowPaths::arityCheckFor): (CommonSlowPaths): * runtime/Executable.cpp: (JSC::jettisonCodeBlock): (JSC): (JSC::EvalExecutable::jitCompile): (JSC::samplingDescription): (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::jitCompile): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::baselineCodeBlockFor): (JSC::FunctionExecutable::jitCompileForCall): (JSC::FunctionExecutable::jitCompileForConstruct): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): * runtime/Executable.h: (JSC): (EvalExecutable): (ProgramExecutable): (FunctionExecutable): (JSC::FunctionExecutable::jitCompileFor): * runtime/ExecutionHarness.h: Added. (JSC): (JSC::prepareForExecution): (JSC::prepareFunctionForExecution): * runtime/JSArray.h: (JSC): (JSArray): * runtime/JSCell.h: (JSC): (JSCell): * runtime/JSFunction.h: (JSC): (JSFunction): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: (JSC): (JSGlobalData): * runtime/JSGlobalObject.h: (JSC): (JSGlobalObject): * runtime/JSObject.h: (JSC): (JSObject): (JSFinalObject): * runtime/JSPropertyNameIterator.h: (JSC): (JSPropertyNameIterator): * runtime/JSString.h: (JSC): (JSString): * runtime/JSTypeInfo.h: (JSC): (TypeInfo): * runtime/JSValue.cpp: (JSC::JSValue::description): * runtime/JSValue.h: (LLInt): (JSValue): * runtime/JSVariableObject.h: (JSC): (JSVariableObject): * runtime/Options.cpp: (Options): (JSC::Options::initializeOptions): * runtime/Options.h: (Options): * runtime/ScopeChain.h: (JSC): (ScopeChainNode): * runtime/Structure.cpp: (JSC::Structure::addPropertyTransition): * runtime/Structure.h: (JSC): (Structure): * runtime/StructureChain.h: (JSC): (StructureChain): * wtf/InlineASM.h: * wtf/Platform.h: * wtf/SentinelLinkedList.h: (SentinelLinkedList): (WTF::SentinelLinkedList::isEmpty): * wtf/text/StringImpl.h: (JSC): (StringImpl): 2012-02-20 Filip Pizlo <fpizlo@apple.com> Unreviewed, rolling out http://trac.webkit.org/changeset/108291 It completely broke the 32-bit JIT. * heap/CopiedAllocator.h: * heap/CopiedSpace.h: (CopiedSpace): * heap/Heap.h: (JSC::Heap::allocatorForObjectWithDestructor): * jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases): * jit/JIT.h: (JIT): * jit/JITInlineMethods.h: (JSC): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_new_array): * runtime/JSArray.cpp: (JSC::storageSize): (JSC): * runtime/JSArray.h: (ArrayStorage): (JSArray): 2012-02-20 Gavin Barraclough <barraclough@apple.com> [[Put]] should throw if prototype chain contains a readonly property. https://bugs.webkit.org/show_bug.cgi?id=79069 Reviewed by Oliver Hunt. Currently we only check the base of the put, not the prototype chain. Fold this check in with the test for accessors. * runtime/JSObject.cpp: (JSC::JSObject::put): - Updated to test all objects in the propotype chain for readonly properties. (JSC::JSObject::putDirectAccessor): (JSC::putDescriptor): - Record the presence of readonly properties on the structure. * runtime/Structure.cpp: (JSC::Structure::Structure): - hasGetterSetterPropertiesExcludingProto expanded to hasReadOnlyOrGetterSetterPropertiesExcludingProto. * runtime/Structure.h: (JSC::Structure::hasReadOnlyOrGetterSetterPropertiesExcludingProto): (JSC::Structure::setHasGetterSetterProperties): - hasGetterSetterPropertiesExcludingProto expanded to hasReadOnlyOrGetterSetterPropertiesExcludingProto. (JSC::Structure::setContainsReadOnlyProperties): - Added. 2012-02-20 Mark Hahnenberg <mhahnenberg@apple.com> Implement fast path for op_new_array in the baseline JIT https://bugs.webkit.org/show_bug.cgi?id=78612 Reviewed by Filip Pizlo. * heap/CopiedAllocator.h: (CopiedAllocator): Friended the JIT to allow access to m_currentOffset. * heap/CopiedSpace.h: (CopiedSpace): Friended the JIT to allow access to (JSC::CopiedSpace::allocator): * heap/Heap.h: (JSC::Heap::storageAllocator): Added a getter for the CopiedAllocator class so the JIT can use it for simple allocation i.e. when we can just bump the offset without having to do anything else. * jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases): Added new slow case for op_new_array for when we have to bail out because the fast allocation path fails for whatever reason. * jit/JIT.h: (JIT): * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateBasicStorage): Added utility function that allows objects to allocate generic backing stores. This function is used by emitAllocateJSArray. (JSC): (JSC::JIT::emitAllocateJSArray): Added utility function that allows the client to more easily allocate JSArrays. This function is used by emit_op_new_array and I expect it will also be used for emit_op_new_array_buffer. * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_new_array): Changed to do inline allocation of JSArrays. Still does a stub call for oversize arrays. (JSC): (JSC::JIT::emitSlow_op_new_array): Just bails out to a stub call if we fail in any way on the fast path. * runtime/JSArray.cpp: (JSC): * runtime/JSArray.h: Added lots of offset functions for all the fields that we need to initialize in the JIT. (ArrayStorage): (JSC::ArrayStorage::lengthOffset): (JSC::ArrayStorage::numValuesInVectorOffset): (JSC::ArrayStorage::allocBaseOffset): (JSC::ArrayStorage::vectorOffset): (JSArray): (JSC::JSArray::sparseValueMapOffset): (JSC::JSArray::subclassDataOffset): (JSC::JSArray::indexBiasOffset): (JSC): (JSC::JSArray::storageSize): Moved this function from being a static function in the cpp file to being a static function in the JSArray class. This move allows the JIT to call it to see what size it should allocate. 2012-02-20 Gavin Barraclough <barraclough@apple.com> DefineOwnProperty fails with numeric properties & Object.prototype https://bugs.webkit.org/show_bug.cgi?id=79059 Reviewed by Oliver Hunt. ObjectPrototype caches whether it contains any numeric properties (m_hasNoPropertiesWithUInt32Names), calls to defineOwnProperty need to update this cache. * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::put): (JSC::ObjectPrototype::defineOwnProperty): (JSC): (JSC::ObjectPrototype::getOwnPropertySlotByIndex): * runtime/ObjectPrototype.h: (ObjectPrototype): 2012-02-20 Pino Toscano <pino@debian.org> Does not build on GNU Hurd https://bugs.webkit.org/show_bug.cgi?id=79045 Reviewed by Gustavo Noronha Silva. * wtf/Platform.h: define WTF_OS_HURD. * wtf/ThreadIdentifierDataPthreads.cpp: adds a band-aid fix for the lack of PTHREAD_KEYS_MAX definition, with a value which should not cause issues. 2012-02-20 Gavin Barraclough <barraclough@apple.com> Unreviewed windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-02-20 Mark Hahnenberg <mhahnenberg@apple.com> Undoing accidental changes * heap/Heap.cpp: (JSC::Heap::collectAllGarbage): 2012-02-20 Mark Hahnenberg <mhahnenberg@apple.com> Factor out allocation in CopySpace into a separate CopyAllocator https://bugs.webkit.org/show_bug.cgi?id=78610 Reviewed by Oliver Hunt. Added a new CopyAllocator class, which allows us to do allocations without having to load the current offset and store the current offset in the current block. This change will allow us to easily do inline assembly in the JIT for array allocations. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * heap/CopiedAllocator.h: Added. (JSC): (CopiedAllocator): (JSC::CopiedAllocator::currentBlock): (JSC::CopiedAllocator::CopiedAllocator): (JSC::CopiedAllocator::allocate): (JSC::CopiedAllocator::fitsInCurrentBlock): (JSC::CopiedAllocator::wasLastAllocation): (JSC::CopiedAllocator::startedCopying): (JSC::CopiedAllocator::resetCurrentBlock): (JSC::CopiedAllocator::currentUtilization): (JSC::CopiedAllocator::resetLastAllocation): * heap/CopiedBlock.h: (CopiedBlock): * heap/CopiedSpace.cpp: Moved some stuff from CopiedSpaceInlineMethods to here because we weren't really getting any benefits from having such big functions in a header file. (JSC::CopiedSpace::CopiedSpace): (JSC): (JSC::CopiedSpace::init): (JSC::CopiedSpace::tryAllocateSlowCase): (JSC::CopiedSpace::tryAllocateOversize): (JSC::CopiedSpace::tryReallocate): (JSC::CopiedSpace::tryReallocateOversize): (JSC::CopiedSpace::doneFillingBlock): (JSC::CopiedSpace::doneCopying): (JSC::CopiedSpace::getFreshBlock): * heap/CopiedSpace.h: (CopiedSpace): * heap/CopiedSpaceInlineMethods.h: (JSC): (JSC::CopiedSpace::startedCopying): (JSC::CopiedSpace::addNewBlock): (JSC::CopiedSpace::allocateNewBlock): (JSC::CopiedSpace::fitsInBlock): (JSC::CopiedSpace::tryAllocate): (JSC::CopiedSpace::allocateFromBlock): * heap/Heap.cpp: (JSC::Heap::collectAllGarbage): * heap/HeapBlock.h: (HeapBlock): 2012-02-20 Patrick Gansterer <paroga@webkit.org> Fix Visual Studio 2010 build. * bytecompiler/NodesCodegen.cpp: (JSC::PropertyListNode::emitBytecode): 2012-02-16 Gavin Barraclough <barraclough@apple.com> Move special __proto__ property to Object.prototype https://bugs.webkit.org/show_bug.cgi?id=78409 Reviewed by Oliver Hunt. Re-implement this as a regular accessor property. This has three key benefits: 1) It makes it possible for objects to be given properties named __proto__. 2) Object.prototype.__proto__ can be deleted, preventing object prototypes from being changed. 3) This largely removes the magic used the implement __proto__, it can just be made a regular accessor property. * parser/Parser.cpp: (JSC::::parseFunctionInfo): - No need to prohibit functions named __proto__. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): - Add __proto__ accessor to Object.prototype. * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncProtoGetter): (JSC::globalFuncProtoSetter): - Definition of the __proto__ accessor functions. * runtime/JSGlobalObjectFunctions.h: - Declaration of the __proto__ accessor functions. * runtime/JSObject.cpp: (JSC::JSObject::put): - Remove the special handling for __proto__, there is still a check to allow for a fast guard for accessors excluding __proto__. (JSC::JSObject::putDirectAccessor): - Track on the structure whether an object contains accessors other than one for __proto__. (JSC::JSObject::defineOwnProperty): - No need to prohibit definition of own properties named __proto__. * runtime/JSObject.h: (JSC::JSObject::inlineGetOwnPropertySlot): - Remove the special handling for __proto__. (JSC::JSValue::get): - Remove the special handling for __proto__. * runtime/JSString.cpp: (JSC::JSString::getOwnPropertySlot): - Remove the special handling for __proto__. * runtime/JSValue.h: (JSValue): - Made synthesizePrototype public (this may be needed by the __proto__ getter). * runtime/ObjectConstructor.cpp: (JSC::objectConstructorGetPrototypeOf): - Perform the security check & call prototype() directly. * runtime/Structure.cpp: (JSC::Structure::Structure): - Added 'ExcludingProto' variant of the 'hasGetterSetterProperties' state. * runtime/Structure.h: (JSC::Structure::hasGetterSetterPropertiesExcludingProto): (JSC::Structure::setHasGetterSetterProperties): (Structure): - Added 'ExcludingProto' variant of the 'hasGetterSetterProperties' state. 2012-02-20 Michael Saboff <msaboff@apple.com> Update toLower and toUpper tests for Unicode 6.1 changes https://bugs.webkit.org/show_bug.cgi?id=78923 Reviewed by Oliver Hunt. * tests/mozilla/ecma/String/15.5.4.11-2.js: Updated the test to handle a third set of results for updated Unicode 6.1 changes. (getTestCases): (TestCaseMultiExpected): (writeTestCaseResultMultiExpected): (getTestCaseResultMultiExpected): (test): (GetUnicodeValues): (DecimalToHexString): 2012-02-20 Andy Wingo <wingo@igalia.com> Remove unused features from CodeFeatures https://bugs.webkit.org/show_bug.cgi?id=78804 Reviewed by Gavin Barraclough. * parser/Nodes.h: * parser/ASTBuilder.h: (JSC::ClosureFeature): (JSC::ASTBuilder::createFunctionBody): (JSC::ASTBuilder::usesClosures): Remove "ClosureFeature". Since we track captured variables more precisely, this bit doesn't do us any good. (JSC::AssignFeature): (JSC::ASTBuilder::makeAssignNode): (JSC::ASTBuilder::makePrefixNode): (JSC::ASTBuilder::makePostfixNode): (JSC::ASTBuilder::usesAssignment): Similarly, remove AssignFeature. It is unused. 2012-02-19 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck issues. * GNUmakefile.list.am: Add missing files. 2012-02-18 Sam Weinig <sam@webkit.org> Fix style issues in DFG Phase classes https://bugs.webkit.org/show_bug.cgi?id=78983 Reviewed by Ryosuke Niwa. * dfg/DFGArithNodeFlagsInferencePhase.cpp: * dfg/DFGCFAPhase.cpp: * dfg/DFGCSEPhase.cpp: * dfg/DFGPredictionPropagationPhase.cpp: * dfg/DFGVirtualRegisterAllocationPhase.cpp: Add a space before the colon in class declarations. 2012-02-18 Filip Pizlo <fpizlo@apple.com> Attempt to fix Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-02-18 Sam Weinig <sam@webkit.org> Fix the libc++ build. Reviewed by Anders Carlsson. * heap/Weak.h: Libc++'s nullptr emulation does not allow default construction of the nullptr_t type. Work around this with the arguably clearer just returning nullptr. 2012-02-18 Filip Pizlo <fpizlo@apple.com> DFGPropagator.cpp has too many things https://bugs.webkit.org/show_bug.cgi?id=78956 Reviewed by Oliver Hunt. Added the notion of a DFG::Phase. Removed DFG::Propagator, and took its various things and put them into separate files. These new phases follow the naming convention "DFG<name>Phase" where <name> is a noun. They are called via functions of the form "perform<name>". * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * dfg/DFGArithNodeFlagsInferencePhase.cpp: Added. (DFG): (JSC::DFG::performArithNodeFlagsInference): * dfg/DFGArithNodeFlagsInferencePhase.h: Added. (DFG): * dfg/DFGCFAPhase.cpp: Added. (DFG): (JSC::DFG::performCFA): * dfg/DFGCFAPhase.h: Added. (DFG): * dfg/DFGCSEPhase.cpp: Added. (DFG): (JSC::DFG::performCSE): * dfg/DFGCSEPhase.h: Added. (DFG): * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGPhase.cpp: Added. (DFG): (JSC::DFG::Phase::beginPhase): (JSC::DFG::Phase::endPhase): * dfg/DFGPhase.h: Added. (DFG): (Phase): (JSC::DFG::Phase::Phase): (JSC::DFG::Phase::~Phase): (JSC::DFG::Phase::globalData): (JSC::DFG::Phase::codeBlock): (JSC::DFG::Phase::profiledBlock): (JSC::DFG::Phase::beginPhase): (JSC::DFG::Phase::endPhase): (JSC::DFG::runPhase): * dfg/DFGPredictionPropagationPhase.cpp: Added. (DFG): (JSC::DFG::performPredictionPropagation): * dfg/DFGPredictionPropagationPhase.h: Added. (DFG): * dfg/DFGPropagator.cpp: Removed. * dfg/DFGPropagator.h: Removed. * dfg/DFGVirtualRegisterAllocationPhase.cpp: Added. (DFG): (JSC::DFG::performVirtualRegisterAllocation): * dfg/DFGVirtualRegisterAllocationPhase.h: Added. (DFG): 2012-02-17 Filip Pizlo <fpizlo@apple.com> DFG::Graph should have references to JSGlobalData, the CodeBlock being compiled, and the CodeBlock that was used for profiling https://bugs.webkit.org/show_bug.cgi?id=78954 Reviewed by Gavin Barraclough. * bytecode/CodeBlock.h: (JSC::baselineCodeBlockForOriginAndBaselineCodeBlock): (JSC): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::AbstractState): (JSC::DFG::AbstractState::execute): * dfg/DFGAbstractState.h: * dfg/DFGAssemblyHelpers.h: (AssemblyHelpers): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::parse): * dfg/DFGByteCodeParser.h: (DFG): * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGGraph.h: (JSC::DFG::Graph::Graph): (Graph): (JSC::DFG::Graph::getJSConstantPrediction): (JSC::DFG::Graph::addShouldSpeculateInteger): (JSC::DFG::Graph::isInt32Constant): (JSC::DFG::Graph::isDoubleConstant): (JSC::DFG::Graph::isNumberConstant): (JSC::DFG::Graph::isBooleanConstant): (JSC::DFG::Graph::isFunctionConstant): (JSC::DFG::Graph::valueOfJSConstant): (JSC::DFG::Graph::valueOfInt32Constant): (JSC::DFG::Graph::valueOfNumberConstant): (JSC::DFG::Graph::valueOfBooleanConstant): (JSC::DFG::Graph::valueOfFunctionConstant): (JSC::DFG::Graph::baselineCodeBlockFor): (JSC::DFG::Graph::valueProfileFor): (JSC::DFG::Graph::addImmediateShouldSpeculateInteger): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::JITCompiler): (JITCompiler): * dfg/DFGOSRExit.cpp: (JSC::DFG::OSRExit::considerAddingAsFrequentExitSiteSlow): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::Propagator): (JSC::DFG::Propagator::isNotNegZero): (JSC::DFG::Propagator::isNotZero): (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::doRoundOfDoubleVoting): (JSC::DFG::Propagator::globalCFA): (JSC::DFG::propagate): * dfg/DFGPropagator.h: (DFG): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): (JSC::DFG::SpeculativeJIT::compileAdd): (JSC::DFG::SpeculativeJIT::compileArithSub): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isConstant): (JSC::DFG::SpeculativeJIT::isJSConstant): (JSC::DFG::SpeculativeJIT::isInt32Constant): (JSC::DFG::SpeculativeJIT::isDoubleConstant): (JSC::DFG::SpeculativeJIT::isNumberConstant): (JSC::DFG::SpeculativeJIT::isBooleanConstant): (JSC::DFG::SpeculativeJIT::isFunctionConstant): (JSC::DFG::SpeculativeJIT::valueOfInt32Constant): (JSC::DFG::SpeculativeJIT::valueOfNumberConstant): (JSC::DFG::SpeculativeJIT::valueOfJSConstant): (JSC::DFG::SpeculativeJIT::valueOfBooleanConstant): (JSC::DFG::SpeculativeJIT::valueOfFunctionConstant): (JSC::DFG::SpeculativeJIT::speculationCheck): (JSC::DFG::SpeculativeJIT::SpeculativeJIT): 2012-02-17 Ahmad Sharif <asharif.tools@gmail.com> There is a warning in memset in glibc that gets triggered through a warndecl when the fill-value of memset is a non-zero constant and the size is zero. This warning is enabled when building with -D_FORTIFY_SOURCE=2. This patch fixes the warning. https://bugs.webkit.org/show_bug.cgi?id=78513 Reviewed by Alexey Proskuryakov * wtf/Vector.h: 2012-02-17 Kalev Lember <kalevlember@gmail.com> Remove unused parameters from WTF threading API https://bugs.webkit.org/show_bug.cgi?id=78389 Reviewed by Adam Roben. waitForThreadCompletion() had an out param 'void **result' to get the 'void *' returned by ThreadFunction. However, the implementation in ThreadingWin.cpp ignored the out param, not filling it in. This had led to a situation where none of the client code made use of the param and just ignored it. To clean this up, the patch changes the signature of ThreadFunction to return void instead of void* and drops the the unused 'void **result' parameter from waitForThreadCompletion. Also, all client code is updated for the API change. As mentioned in https://bugs.webkit.org/show_bug.cgi?id=78389 , even though the change only affects internal API, Safari is using it directly and we'll need to keep the old versions around for ABI compatibility. For this, the patch adds compatibility wrappers with the old ABI. * JavaScriptCore.order: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * bytecode/SamplingTool.cpp: (JSC::SamplingThread::threadStartFunc): (JSC::SamplingThread::stop): * bytecode/SamplingTool.h: (SamplingThread): * heap/Heap.cpp: (JSC::Heap::~Heap): (JSC::Heap::blockFreeingThreadStartFunc): * heap/Heap.h: * heap/MarkStack.cpp: (JSC::MarkStackThreadSharedData::markingThreadStartFunc): (JSC::MarkStackThreadSharedData::~MarkStackThreadSharedData): * heap/MarkStack.h: (MarkStackThreadSharedData): * wtf/ParallelJobsGeneric.cpp: (WTF::ParallelEnvironment::ThreadPrivate::workerThread): * wtf/ParallelJobsGeneric.h: (ThreadPrivate): * wtf/ThreadFunctionInvocation.h: Update the signature of ThreadFunction. (WTF): * wtf/Threading.cpp: (WTF::threadEntryPoint): Update for ThreadFunction signature change. (WTF): (WTF::ThreadFunctionWithReturnValueInvocation::ThreadFunctionWithReturnValueInvocation): ABI compatibility function for Safari. (ThreadFunctionWithReturnValueInvocation): Ditto. (WTF::compatEntryPoint): Ditto. (WTF::createThread): Ditto. (WTF::waitForThreadCompletion): Ditto. * wtf/Threading.h: Update the signature of ThreadFunction and waitForThreadCompletion. (WTF): * wtf/ThreadingPthreads.cpp: Implement the new API. (WTF::wtfThreadEntryPoint): (WTF): (WTF::createThreadInternal): (WTF::waitForThreadCompletion): * wtf/ThreadingWin.cpp: Implement the new API. (WTF::wtfThreadEntryPoint): (WTF::waitForThreadCompletion): 2012-02-16 Oliver Hunt <oliver@apple.com> Implement Error.stack https://bugs.webkit.org/show_bug.cgi?id=66994 Reviewed by Gavin Barraclough. Implement support for stack traces on exception objects. This is a rewrite of the core portion of the last stack walking logic, but the mechanical work of adding the information to an exception comes from the original work by Juan Carlos Montemayor Elosua. * interpreter/Interpreter.cpp: (JSC::getCallerInfo): (JSC): (JSC::getSourceURLFromCallFrame): (JSC::getStackFrameCodeType): (JSC::Interpreter::getStackTrace): (JSC::Interpreter::throwException): (JSC::Interpreter::privateExecute): * interpreter/Interpreter.h: (JSC): (StackFrame): (JSC::StackFrame::toString): (Interpreter): * jsc.cpp: (GlobalObject::finishCreation): (functionJSCStack): * parser/Nodes.h: (JSC::FunctionBodyNode::setInferredName): * parser/Parser.h: (JSC::::parse): * runtime/CommonIdentifiers.h: * runtime/Error.cpp: (JSC::addErrorInfo): * runtime/Error.h: (JSC): 2012-02-17 Mark Hahnenberg <mhahnenberg@apple.com> Rename Bump* to Copy* https://bugs.webkit.org/show_bug.cgi?id=78573 Reviewed by Geoffrey Garen. Renamed anything with "Bump" in the name to have "Copied" instead. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * heap/BumpBlock.h: Removed. * heap/BumpSpace.cpp: Removed. * heap/BumpSpace.h: Removed. * heap/BumpSpaceInlineMethods.h: Removed. * heap/ConservativeRoots.cpp: (JSC::ConservativeRoots::ConservativeRoots): (JSC::ConservativeRoots::genericAddPointer): * heap/ConservativeRoots.h: (ConservativeRoots): * heap/CopiedBlock.h: Added. (JSC): (CopiedBlock): (JSC::CopiedBlock::CopiedBlock): * heap/CopiedSpace.cpp: Added. (JSC): (JSC::CopiedSpace::tryAllocateSlowCase): * heap/CopiedSpace.h: Added. (JSC): (CopiedSpace): (JSC::CopiedSpace::isInCopyPhase): (JSC::CopiedSpace::totalMemoryAllocated): (JSC::CopiedSpace::totalMemoryUtilized): * heap/CopiedSpaceInlineMethods.h: Added. (JSC): (JSC::CopiedSpace::CopiedSpace): (JSC::CopiedSpace::init): (JSC::CopiedSpace::contains): (JSC::CopiedSpace::pin): (JSC::CopiedSpace::startedCopying): (JSC::CopiedSpace::doneCopying): (JSC::CopiedSpace::doneFillingBlock): (JSC::CopiedSpace::recycleBlock): (JSC::CopiedSpace::getFreshBlock): (JSC::CopiedSpace::borrowBlock): (JSC::CopiedSpace::addNewBlock): (JSC::CopiedSpace::allocateNewBlock): (JSC::CopiedSpace::fitsInBlock): (JSC::CopiedSpace::fitsInCurrentBlock): (JSC::CopiedSpace::tryAllocate): (JSC::CopiedSpace::tryAllocateOversize): (JSC::CopiedSpace::allocateFromBlock): (JSC::CopiedSpace::tryReallocate): (JSC::CopiedSpace::tryReallocateOversize): (JSC::CopiedSpace::isOversize): (JSC::CopiedSpace::isPinned): (JSC::CopiedSpace::oversizeBlockFor): (JSC::CopiedSpace::blockFor): * heap/Heap.cpp: * heap/Heap.h: (JSC): (Heap): * heap/MarkStack.cpp: (JSC::MarkStackThreadSharedData::MarkStackThreadSharedData): (JSC::SlotVisitor::drainFromShared): (JSC::SlotVisitor::startCopying): (JSC::SlotVisitor::allocateNewSpace): (JSC::SlotVisitor::doneCopying): * heap/MarkStack.h: (MarkStackThreadSharedData): * heap/SlotVisitor.h: (SlotVisitor): * runtime/JSArray.cpp: * runtime/JSObject.cpp: 2012-02-16 Yuqiang Xian <yuqiang.xian@intel.com> Add JSC code profiling support on Linux x86 https://bugs.webkit.org/show_bug.cgi?id=78871 Reviewed by Gavin Barraclough. We don't unwind the stack for now as we cannot guarantee all the libraries are compiled without -fomit-frame-pointer. * tools/CodeProfile.cpp: (JSC::CodeProfile::sample): * tools/CodeProfiling.cpp: (JSC): (JSC::profilingTimer): (JSC::CodeProfiling::begin): (JSC::CodeProfiling::end): 2012-02-16 Csaba Osztrogonác <ossy@webkit.org> Unreviewed. Rolling out r107980, because it broke 32 bit platforms. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::Interpreter::throwException): (JSC::Interpreter::privateExecute): * interpreter/Interpreter.h: (JSC): (Interpreter): * jsc.cpp: (GlobalObject::finishCreation): * parser/Nodes.h: (JSC::FunctionBodyNode::setInferredName): * parser/Parser.h: (JSC::::parse): * runtime/CommonIdentifiers.h: * runtime/Error.cpp: (JSC::addErrorInfo): * runtime/Error.h: (JSC): 2012-02-16 Filip Pizlo <fpizlo@apple.com> ENABLE_INTERPRETER should be ENABLE_CLASSIC_INTERPRETER https://bugs.webkit.org/show_bug.cgi?id=78791 Rubber stamped by Oliver Hunt. Just a renaming, nothing more. Also renamed COMPUTED_GOTO_INTERPRETER to COMPUTED_GOTO_CLASSIC_INTERPRETER. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): (JSC::CodeBlock::stronglyVisitStrongReferences): (JSC): (JSC::CodeBlock::shrinkToFit): * bytecode/CodeBlock.h: (CodeBlock): * bytecode/Instruction.h: (JSC::Instruction::Instruction): * bytecode/Opcode.h: (JSC::padOpcodeName): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitResolve): (JSC::BytecodeGenerator::emitResolveWithBase): (JSC::BytecodeGenerator::emitGetById): (JSC::BytecodeGenerator::emitPutById): (JSC::BytecodeGenerator::emitDirectPutById): * interpreter/AbstractPC.cpp: (JSC::AbstractPC::AbstractPC): * interpreter/AbstractPC.h: (AbstractPC): * interpreter/CallFrame.h: (ExecState): * interpreter/Interpreter.cpp: (JSC): (JSC::Interpreter::initialize): (JSC::Interpreter::isOpcode): (JSC::Interpreter::unwindCallFrame): (JSC::Interpreter::execute): (JSC::Interpreter::privateExecute): (JSC::Interpreter::retrieveLastCaller): * interpreter/Interpreter.h: (JSC::Interpreter::getOpcode): (JSC::Interpreter::getOpcodeID): (Interpreter): * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::FixedVMPoolExecutableAllocator::FixedVMPoolExecutableAllocator): * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): * runtime/Executable.h: (NativeExecutable): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::getHostFunction): * runtime/JSGlobalData.h: (JSGlobalData): * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveAndCommit): * wtf/Platform.h: 2012-02-15 Geoffrey Garen <ggaren@apple.com> Made Weak<T> single-owner, adding PassWeak<T> https://bugs.webkit.org/show_bug.cgi?id=78740 Reviewed by Sam Weinig. This works basically the same way as OwnPtr<T> and PassOwnPtr<T>. This clarifies the semantics of finalizers: It's ambiguous and probably a bug to copy a finalizer (i.e., it's a bug to run a C++ destructor twice), so I've made Weak<T> non-copyable. Anywhere we used to copy a Weak<T>, we now use PassWeak<T>. This also makes Weak<T> HashMaps more efficient. * API/JSClassRef.cpp: (OpaqueJSClass::prototype): Use PassWeak<T> instead of set(), since set() is gone now. * JavaScriptCore.xcodeproj/project.pbxproj: Export! * heap/PassWeak.h: Added. (JSC): (PassWeak): (JSC::PassWeak::PassWeak): (JSC::PassWeak::~PassWeak): (JSC::PassWeak::get): (JSC::::leakHandle): (JSC::adoptWeak): (JSC::operator==): (JSC::operator!=): This is the Weak<T> version of PassOwnPtr<T>. * heap/Weak.h: (Weak): (JSC::Weak::Weak): (JSC::Weak::release): (JSC::Weak::hashTableDeletedValue): (JSC::=): (JSC): Changed to be non-copyable, removing a lot of copying-related APIs. Added hash traits so hash maps still work. * jit/JITStubs.cpp: (JSC::JITThunks::hostFunctionStub): * runtime/RegExpCache.cpp: (JSC::RegExpCache::lookupOrCreate): Use PassWeak<T>, as required by our new hash map API. 2012-02-16 Mark Hahnenberg <mhahnenberg@apple.com> Fix the broken viewport tests https://bugs.webkit.org/show_bug.cgi?id=78774 Reviewed by Kenneth Rohde Christiansen. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/text/WTFString.cpp: (WTF): (WTF::toDoubleType): Template-ized to allow other functions to specify whether they want to allow trailing junk or not when calling strtod. (WTF::charactersToDouble): (WTF::charactersToFloat): (WTF::charactersToFloatIgnoringJunk): Created new version of charactersToFloat that allows trailing junk. * wtf/text/WTFString.h: (WTF): 2012-02-16 Oliver Hunt <oliver@apple.com> Implement Error.stack https://bugs.webkit.org/show_bug.cgi?id=66994 Reviewed by Gavin Barraclough. Implement support for stack traces on exception objects. This is a rewrite of the core portion of the last stack walking logic, but the mechanical work of adding the information to an exception comes from the original work by Juan Carlos Montemayor Elosua. * interpreter/Interpreter.cpp: (JSC::getCallerInfo): (JSC): (JSC::getSourceURLFromCallFrame): (JSC::getStackFrameCodeType): (JSC::Interpreter::getStackTrace): (JSC::Interpreter::throwException): (JSC::Interpreter::privateExecute): * interpreter/Interpreter.h: (JSC): (StackFrame): (JSC::StackFrame::toString): (Interpreter): * jsc.cpp: (GlobalObject::finishCreation): (functionJSCStack): * parser/Nodes.h: (JSC::FunctionBodyNode::setInferredName): * parser/Parser.h: (JSC::::parse): * runtime/CommonIdentifiers.h: * runtime/Error.cpp: (JSC::addErrorInfo): * runtime/Error.h: (JSC): 2012-02-15 Gavin Barraclough <barraclough@apple.com> Numerous trivial bugs in Object.defineProperty https://bugs.webkit.org/show_bug.cgi?id=78777 Reviewed by Sam Weinig. There are a handful of really trivial bugs, related to Object.defineProperty: * Redefining an accessor with different attributes changes the attributes, but not the get/set functions! * Calling an undefined setter should only throw in strict mode. * When redefining an accessor to a data decriptor, if writable is not specified we should default to false. * Any attempt to redefine a non-configurable property of an array as configurable should be rejected. * Object.defineProperties should call toObject on 'Properties' argument, rather than throwing if it is not an object. * If preventExtensions has been called on an array, subsequent assignment beyond array bounds should fail. * 'isFrozen' shouldn't be checking the ReadOnly bit for accessor descriptors (we presently always keep this bit as 'false'). * Should be able to redefine an non-writable, non-configurable property, with the same value and attributes. * Should be able to define an non-configurable accessor. These are mostly all one-line changes, e.g. inverted boolean checks, masking against wrong attribute. * runtime/JSArray.cpp: (JSC::SparseArrayValueMap::put): - Added ASSERT. - Calling an undefined setter should only throw in strict mode. (JSC::JSArray::putDescriptor): - Should be able to define an non-configurable accessor. (JSC::JSArray::defineOwnNumericProperty): - Any attempt to redefine a non-configurable property of an array as configurable should be rejected. (JSC::JSArray::putByIndexBeyondVectorLength): - If preventExtensions has been called on an array, subsequent assignment beyond array bounds should fail. * runtime/JSArray.h: (JSArray): - made enterDictionaryMode public, called from JSObject. * runtime/JSObject.cpp: (JSC::JSObject::put): - Calling an undefined setter should only throw in strict mode. (JSC::JSObject::preventExtensions): - Put array objects into dictionary mode to handle this! (JSC::JSObject::defineOwnProperty): - Should be able to redefine an non-writable, non-configurable property, with the same value and attributes. - Redefining an accessor with different attributes changes the attributes, but not the get/set functions! * runtime/ObjectConstructor.cpp: (JSC::objectConstructorDefineProperties): - Object.defineProperties should call toObject on 'Properties' argument, rather than throwing if it is not an object. * runtime/PropertyDescriptor.cpp: (JSC::PropertyDescriptor::attributesWithOverride): - When redefining an accessor to a data decriptor, if writable is not specified we should default to false. (JSC::PropertyDescriptor::attributesOverridingCurrent): - When redefining an accessor to a data decriptor, if writable is not specified we should default to false. * runtime/Structure.cpp: (JSC::Structure::freezeTransition): - 'freezeTransition' shouldn't be setting the ReadOnly bit for accessor descriptors (we presently always keep this bit as 'false'). (JSC::Structure::isFrozen): - 'isFrozen' shouldn't be checking the ReadOnly bit for accessor descriptors (we presently always keep this bit as 'false'). 2012-02-13 Filip Pizlo <fpizlo@apple.com> DFG should not check the types of arguments that are dead https://bugs.webkit.org/show_bug.cgi?id=78518 Reviewed by Geoff Garen. The argument checks are now elided if the corresponding SetArgument is dead, and the abstract value of the argument is set to bottom (None, []). This is performance neutral on the benchmarks we currently track. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::initialize): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::checkArgumentTypes): 2012-02-15 Oliver Hunt <oliver@apple.com> Ensure that the DFG JIT always plants a CodeOrigin when making calls https://bugs.webkit.org/show_bug.cgi?id=78763 Reviewed by Gavin Barraclough. Make all calls plant a CodeOrigin prior to the actual call. Also clobbers the Interpreter with logic to ensure that the interpreter always plants a bytecode offset. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGJITCompiler.h: (CallBeginToken): (JSC::DFG::JITCompiler::beginJSCall): (JSC::DFG::JITCompiler::beginCall): * dfg/DFGRepatch.cpp: (JSC::DFG::tryBuildGetByIDList): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::appendCallWithExceptionCheck): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): * interpreter/AbstractPC.cpp: (JSC::AbstractPC::AbstractPC): * interpreter/CallFrame.cpp: (JSC::CallFrame::trueCallFrame): * interpreter/CallFrame.h: (JSC::ExecState::bytecodeOffsetForNonDFGCode): (ExecState): (JSC::ExecState::setBytecodeOffsetForNonDFGCode): (JSC::ExecState::codeOriginIndexForDFG): 2012-02-14 Oliver Hunt <oliver@apple.com> Fix Interpreter. * runtime/Executable.cpp: (JSC): * runtime/Executable.h: (ExecutableBase): 2012-02-14 Matt Lilek <mrl@apple.com> Don't ENABLE_DASHBOARD_SUPPORT unconditionally on all Mac platforms https://bugs.webkit.org/show_bug.cgi?id=78629 Reviewed by David Kilzer. * Configurations/FeatureDefines.xcconfig: 2012-02-14 Filip Pizlo <fpizlo@apple.com> Unreviewed, build fix for non-DFG platforms. * assembler/MacroAssembler.h: (MacroAssembler): 2012-02-14 Filip Pizlo <fpizlo@apple.com> Unreviewed, fix build and configuration goof. * assembler/MacroAssembler.h: (JSC::MacroAssembler::invert): * dfg/DFGCommon.h: 2012-02-13 Filip Pizlo <fpizlo@apple.com> DFG should be able to emit code on control flow edges https://bugs.webkit.org/show_bug.cgi?id=78515 Reviewed by Gavin Barraclough. This gets us a few steps closer to being able to perform global register allocation, by allowing us to have landing pads on control flow edges. This will let us reshuffle registers if it happens to be necessary due to different reg alloc decisions in differen blocks. This also introduces the notion of a landing pad for OSR entry, which will allow us to emit code that places data into registers when we're entering into the DFG from the old JIT. Finally, this patch introduces a verification mode that checks that the landing pads are actually emitted and do actually work as advertised. When verification is disabled, this has no effect on behavior. * assembler/MacroAssembler.h: (MacroAssembler): (JSC::MacroAssembler::invert): (JSC::MacroAssembler::isInvertible): * dfg/DFGCommon.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compile): (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleDoubleBranch): (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality): (JSC::DFG::SpeculativeJIT::compilePeepHoleIntegerBranch): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::createOSREntries): (DFG): (JSC::DFG::SpeculativeJIT::linkOSREntries): (JSC::DFG::SpeculativeJIT::compileStrictEqForConstant): * dfg/DFGSpeculativeJIT.h: (SpeculativeJIT): (JSC::DFG::SpeculativeJIT::branchDouble): (JSC::DFG::SpeculativeJIT::branchDoubleNonZero): (JSC::DFG::SpeculativeJIT::branch32): (JSC::DFG::SpeculativeJIT::branchTest32): (JSC::DFG::SpeculativeJIT::branchPtr): (JSC::DFG::SpeculativeJIT::branchTestPtr): (JSC::DFG::SpeculativeJIT::branchTest8): (JSC::DFG::SpeculativeJIT::jump): (JSC::DFG::SpeculativeJIT::haveEdgeCodeToEmit): (JSC::DFG::SpeculativeJIT::emitEdgeCode): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranch): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeStrictEq): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranch): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeStrictEq): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): 2012-02-14 Filip Pizlo <fpizlo@apple.com> Assertion failure under JSC::DFG::AbstractState::execute loading economist.com https://bugs.webkit.org/show_bug.cgi?id=78153 <rdar://problem/10861712> <rdar://problem/10861947> Reviewed by Oliver Hunt. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileAdd): 2012-02-14 Eric Seidel <eric@webkit.org> Upstream Android's additions to Platform.h https://bugs.webkit.org/show_bug.cgi?id=78536 Reviewed by Adam Barth. * wtf/Platform.h: 2012-02-12 Mark Hahnenberg <mhahnenberg@apple.com> Replace old strtod with new strtod https://bugs.webkit.org/show_bug.cgi?id=68044 Reviewed by Geoffrey Garen. * parser/Lexer.cpp: Added template argument. This version allows junk after numbers. (JSC::::lex): * runtime/JSGlobalObjectFunctions.cpp: Ditto. (JSC::parseInt): (JSC::jsStrDecimalLiteral): * runtime/LiteralParser.cpp: Ditto. (JSC::::Lexer::lexNumber): * wtf/dtoa.cpp: Replaced old strtod with a new version that uses the new StringToDoubleConverter. It takes a template argument to allow clients to determine statically whether it should allow junk after the numbers or not. (WTF): (WTF::strtod): * wtf/dtoa.h: (WTF): * wtf/text/WTFString.cpp: Added template argument. This version does not allow junk after numbers. (WTF::toDoubleType): 2012-02-13 Mark Hahnenberg <mhahnenberg@apple.com> More windows build fixing * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-02-13 Oliver Hunt <oliver@apple.com> Executing out of bounds in JSC::Yarr::YarrCodeBlock::execute / JSC::RegExp::match https://bugs.webkit.org/show_bug.cgi?id=76315 Reviewed by Gavin Barraclough. Perform a 3 byte compare using two comparisons, rather than trying to perform the operation with a four byte load. * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::generatePatternCharacterOnce): 2012-02-13 Mark Hahnenberg <mhahnenberg@apple.com> Windows build fix * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-02-12 Mark Hahnenberg <mhahnenberg@apple.com> Replace old strtod with new strtod https://bugs.webkit.org/show_bug.cgi?id=68044 Reviewed by Geoffrey Garen. * parser/Lexer.cpp: Added template argument. This version allows junk after numbers. (JSC::::lex): * runtime/JSGlobalObjectFunctions.cpp: Ditto. (JSC::parseInt): (JSC::jsStrDecimalLiteral): * runtime/LiteralParser.cpp: Ditto. (JSC::::Lexer::lexNumber): * wtf/dtoa.cpp: Replaced old strtod with a new version that uses the new StringToDoubleConverter. It takes a template argument to allow clients to determine statically whether it should allow junk after the numbers or not. (WTF): (WTF::strtod): * wtf/dtoa.h: (WTF): * wtf/text/WTFString.cpp: Added template argument. This version does not allow junk after numbers. (WTF::toDoubleType): 2012-02-13 Sam Weinig <sam@webkit.org> Move JSC related assertions out of Assertions.h and into their own header https://bugs.webkit.org/show_bug.cgi?id=78508 Reviewed by Gavin Barraclough. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: Add GCAssertions.h * heap/GCAssertions.h: Added. Move assertions here. * runtime/WriteBarrier.h: Add #include of GCAssertions.h * wtf/Assertions.h: Remove JSC related assertions. * wtf/Compiler.h: Add compiler check for __has_trivial_destructor. 2012-02-13 Chao-ying Fu <fu@mips.com> Update MIPS patchOffsetGetByIdSlowCaseCall https://bugs.webkit.org/show_bug.cgi?id=78392 Reviewed by Gavin Barraclough. * jit/JIT.h: (JIT): 2012-02-13 Patrick Gansterer <paroga@webkit.org> Remove obsolete #if from ThreadSpecific.h https://bugs.webkit.org/show_bug.cgi?id=78485 Reviewed by Adam Roben. Since alle platform use either pthread or Win32 for threading, we can remove all PLATFORM() preprocessor statements. * wtf/ThreadSpecific.h: (ThreadSpecific): 2012-02-13 Jessie Berlin <jberlin@apple.com> Fix the Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-02-13 Sam Weinig <sam@webkit.org> Use C11's _Static_assert for COMPILE_ASSERT if it is available https://bugs.webkit.org/show_bug.cgi?id=78506 Rubber-stamped by Antti Koivisto. Use C11's _Static_assert for COMPILE_ASSERT if it is available to give slightly better error messages. * wtf/Assertions.h: Use _Static_assert if it is available. * wtf/Compiler.h: Add COMPILER_SUPPORTS support for _Static_assert when using the LLVM Compiler. 2012-02-13 Mario Sanchez Prada <msanchez@igalia.com> [GTK] Add GSList to the list of GObject types in GOwnPtr https://bugs.webkit.org/show_bug.cgi?id=78487 Reviewed by Philippe Normand. Handle the GSList type in GOwnPtr, by calling g_slist_free in the implementation of the freeOwnedGPtr template function. * wtf/gobject/GOwnPtr.cpp: (WTF::GSList): (WTF): * wtf/gobject/GOwnPtr.h: (WTF): * wtf/gobject/GTypedefs.h: 2012-02-06 Raphael Kubo da Costa <kubo@profusion.mobi> [EFL] Drop support for the Curl network backend. https://bugs.webkit.org/show_bug.cgi?id=77874 Reviewed by Eric Seidel. Nobody seems to be maintaining the Curl backend in WebCore, the EFL port developers all seem to be using the Soup backend and the port itself has many features which are only implemented for the latter. * wtf/PlatformEfl.cmake: Always build the gobject-dependent source files. 2012-02-13 Patrick Gansterer <paroga@webkit.org> Unreviewed. Build fix for !ENABLE(JIT) after r107485. * bytecode/PolymorphicPutByIdList.cpp: 2012-02-13 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=78434 Unreviewed - temporarily reverting r107498 will I fix a couple of testcases. * parser/Parser.cpp: (JSC::::parseFunctionInfo): * runtime/ClassInfo.h: (MethodTable): (JSC): * runtime/JSCell.cpp: (JSC): * runtime/JSCell.h: (JSCell): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): * runtime/JSGlobalObjectFunctions.cpp: (JSC): * runtime/JSGlobalObjectFunctions.h: (JSC): * runtime/JSObject.cpp: (JSC::JSObject::put): (JSC): (JSC::JSObject::putDirectAccessor): (JSC::JSObject::defineOwnProperty): * runtime/JSObject.h: (JSC::JSObject::inlineGetOwnPropertySlot): (JSC::JSValue::get): * runtime/JSString.cpp: (JSC::JSString::getOwnPropertySlot): * runtime/JSValue.h: (JSValue): * runtime/ObjectConstructor.cpp: (JSC::objectConstructorGetPrototypeOf): * runtime/Structure.cpp: (JSC::Structure::Structure): * runtime/Structure.h: (JSC::Structure::setHasGetterSetterProperties): (Structure): 2012-02-12 Ashod Nakashian <ashodnakashian@yahoo.com> KeywordLookupGenerator.py script fails in some cases https://bugs.webkit.org/show_bug.cgi?id=77886 Reviewed by Benjamin Poulain. * parser/Keywords.table: Converted to LF-only. 2012-02-12 Shinya Kawanaka <shinyak@google.com> Introduce ShadowRootList. https://bugs.webkit.org/show_bug.cgi?id=78069 Reviewed by Hajime Morita. DoublyLinkedList should have tail() method to take the last element. * wtf/DoublyLinkedList.h: (DoublyLinkedList): (WTF::::tail): (WTF): 2012-02-12 Raphael Kubo da Costa <kubo@profusion.mobi> [CMake] Move source files in WTF_HEADERS to WTF_SOURCES. https://bugs.webkit.org/show_bug.cgi?id=78436 Reviewed by Daniel Bates. * wtf/CMakeLists.txt: Move .cpp files from WTF_HEADERS to WTF_SOURCES, and correctly sort the files which start with 'M'. 2012-02-12 Sam Weinig <sam@webkit.org> Move the NumberOfCores.h/cpp files into the WTF group of JavaScriptCore.xcodeproj. Rubber-stamped by Anders Carlsson. * JavaScriptCore.xcodeproj/project.pbxproj: 2012-02-12 Raphael Kubo da Costa <kubo@profusion.mobi> [CMake] Remove unused or empty variable definitions. https://bugs.webkit.org/show_bug.cgi?id=78437 Reviewed by Daniel Bates. * CMakeLists.txt: Remove unused JavaScriptCore_HEADERS definition. * shell/CMakeLists.txt: Remove unused JSC_HEADERS definition. * wtf/CMakeLists.txt: Remove empty WTF_LIBRARIES definition, it will be defined later by Platform*.cmake via LIST(APPEND WTF_LIBRARIES). 2012-02-12 Filip Pizlo <fpizlo@apple.com> DFG::SpeculativeJIT calls fprintf() instead of dataLog in terminateSpeculativeExecution() https://bugs.webkit.org/show_bug.cgi?id=78431 Reviewed by Gavin Barraclough. * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution): 2012-02-11 Benjamin Poulain <benjamin@webkit.org> Add back WTFURL to WebKit https://bugs.webkit.org/show_bug.cgi?id=77291 Reviewed by Adam Barth. WTFURL was removed from WebKit in r86787. This patch adds the code back to WTF with the following changes: -Guard the feature with USE(WTFURL). -Change the typename CHAR to CharacterType to follow recent WebKit conventions. -Fix some coding style to make check-webkit-style happy. * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/Platform.h: * wtf/url/api/ParsedURL.cpp: Added. (WTF): (WTF::ParsedURL::ParsedURL): (WTF::ParsedURL::scheme): (WTF::ParsedURL::username): (WTF::ParsedURL::password): (WTF::ParsedURL::host): (WTF::ParsedURL::port): (WTF::ParsedURL::path): (WTF::ParsedURL::query): (WTF::ParsedURL::fragment): (WTF::ParsedURL::segment): * wtf/url/api/ParsedURL.h: Added. (WTF): (ParsedURL): (WTF::ParsedURL::spec): * wtf/url/api/URLString.h: Added. (WTF): (URLString): (WTF::URLString::URLString): (WTF::URLString::string): * wtf/url/src/RawURLBuffer.h: Added. (WTF): (RawURLBuffer): (WTF::RawURLBuffer::RawURLBuffer): (WTF::RawURLBuffer::~RawURLBuffer): (WTF::RawURLBuffer::resize): * wtf/url/src/URLBuffer.h: Added. (WTF): (URLBuffer): (WTF::URLBuffer::URLBuffer): (WTF::URLBuffer::~URLBuffer): (WTF::URLBuffer::at): (WTF::URLBuffer::set): (WTF::URLBuffer::capacity): (WTF::URLBuffer::length): (WTF::URLBuffer::data): (WTF::URLBuffer::setLength): (WTF::URLBuffer::append): (WTF::URLBuffer::grow): * wtf/url/src/URLCharacterTypes.cpp: Added. (WTF): (): * wtf/url/src/URLCharacterTypes.h: Added. (WTF): (URLCharacterTypes): (WTF::URLCharacterTypes::isQueryChar): (WTF::URLCharacterTypes::isIPv4Char): (WTF::URLCharacterTypes::isHexChar): (): (WTF::URLCharacterTypes::isCharOfType): * wtf/url/src/URLComponent.h: Added. (WTF): (URLComponent): (WTF::URLComponent::URLComponent): (WTF::URLComponent::fromRange): (WTF::URLComponent::isValid): (WTF::URLComponent::isNonEmpty): (WTF::URLComponent::isEmptyOrInvalid): (WTF::URLComponent::reset): (WTF::URLComponent::operator==): (WTF::URLComponent::begin): (WTF::URLComponent::setBegin): (WTF::URLComponent::length): (WTF::URLComponent::setLength): (WTF::URLComponent::end): * wtf/url/src/URLEscape.cpp: Added. (WTF): (): * wtf/url/src/URLEscape.h: Added. (WTF): (WTF::appendURLEscapedCharacter): * wtf/url/src/URLParser.h: Added. (WTF): (URLParser): (): (WTF::URLParser::isPossibleAuthorityTerminator): (WTF::URLParser::parseAuthority): (WTF::URLParser::extractScheme): (WTF::URLParser::parseAfterScheme): (WTF::URLParser::parseStandardURL): (WTF::URLParser::parsePath): (WTF::URLParser::parsePathURL): (WTF::URLParser::parseMailtoURL): (WTF::URLParser::parsePort): (WTF::URLParser::extractFileName): (WTF::URLParser::extractQueryKeyValue): (WTF::URLParser::isURLSlash): (WTF::URLParser::shouldTrimFromURL): (WTF::URLParser::trimURL): (WTF::URLParser::consecutiveSlashes): (WTF::URLParser::isPortDigit): (WTF::URLParser::nextAuthorityTerminator): (WTF::URLParser::parseUserInfo): (WTF::URLParser::parseServerInfo): * wtf/url/src/URLQueryCanonicalizer.h: Added. (WTF): (URLQueryCanonicalizer): (WTF::URLQueryCanonicalizer::canonicalize): (WTF::URLQueryCanonicalizer::isAllASCII): (WTF::URLQueryCanonicalizer::isRaw8Bit): (WTF::URLQueryCanonicalizer::appendRaw8BitQueryString): (WTF::URLQueryCanonicalizer::convertToQueryEncoding): * wtf/url/src/URLSegments.cpp: Added. (WTF): (WTF::URLSegments::length): (WTF::URLSegments::charactersBefore): * wtf/url/src/URLSegments.h: Added. (WTF): (URLSegments): (): (WTF::URLSegments::URLSegments): 2012-02-11 Filip Pizlo <fpizlo@apple.com> Old JIT put_by_id profiling counts every put_by_id_transition as taking slow path https://bugs.webkit.org/show_bug.cgi?id=78430 <rdar://problem/10849469> <rdar://problem/10849684> Reviewed by Gavin Barraclough. The old JIT's put_by_id transition caching involves repatching the slow call to a generated stub. That means that the call is counted as "slow case". So, this patch inserts code to decrement the slow case count if the stub succeeds. Looks like a ~1% speed-up on V8. * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompilePutByIdTransition): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::privateCompilePutByIdTransition): 2012-02-11 Filip Pizlo <fpizlo@apple.com> Build fix for Qt. * wtf/DataLog.h: 2012-02-11 Filip Pizlo <fpizlo@apple.com> It should be possible to send all JSC debug logging to a file https://bugs.webkit.org/show_bug.cgi?id=78418 Reviewed by Sam Weinig. Introduced wtf/DataLog, which defines WTF::dataFile, WTF::dataLog, and WTF::dataLogV. Changed all debugging- and profiling-related printfs to use WTF::dataLog() or one of its friends. By default, debug logging goes to stderr, unless you change the setting in wtf/DataLog.cpp. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * assembler/LinkBuffer.h: (JSC::LinkBuffer::dumpLinkStatistics): (JSC::LinkBuffer::dumpCode): * assembler/SH4Assembler.h: (JSC::SH4Assembler::vprintfStdoutInstr): * bytecode/CodeBlock.cpp: (JSC::CodeBlock::printUnaryOp): (JSC::CodeBlock::printBinaryOp): (JSC::CodeBlock::printConditionalJump): (JSC::CodeBlock::printGetByIdOp): (JSC::CodeBlock::printCallOp): (JSC::CodeBlock::printPutByIdOp): (JSC::printGlobalResolveInfo): (JSC::printStructureStubInfo): (JSC::CodeBlock::printStructure): (JSC::CodeBlock::printStructures): (JSC::CodeBlock::dump): (JSC::CodeBlock::dumpStatistics): (JSC::CodeBlock::finalizeUnconditionally): (JSC::CodeBlock::shouldOptimizeNow): (JSC::CodeBlock::tallyFrequentExitSites): (JSC::CodeBlock::dumpValueProfiles): * bytecode/Opcode.cpp: (JSC::OpcodeStats::~OpcodeStats): * bytecode/SamplingTool.cpp: (JSC::SamplingFlags::stop): (JSC::SamplingRegion::dumpInternal): (JSC::SamplingTool::dump): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::endBasicBlock): (JSC::DFG::AbstractState::mergeStateAtTail): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getPredictionWithoutOSRExit): (JSC::DFG::ByteCodeParser::makeSafe): (JSC::DFG::ByteCodeParser::makeDivSafe): (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::processPhiStack): (JSC::DFG::ByteCodeParser::linkBlock): (JSC::DFG::ByteCodeParser::parseCodeBlock): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGCommon.h: * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGGraph.cpp: (JSC::DFG::printWhiteSpace): (JSC::DFG::Graph::dumpCodeOrigin): (JSC::DFG::Graph::dump): (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGOSREntry.cpp: (JSC::DFG::prepareOSREntry): * dfg/DFGOSRExitCompiler.cpp: * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOperations.cpp: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::fixpoint): (JSC::DFG::Propagator::propagateArithNodeFlags): (JSC::DFG::Propagator::propagateArithNodeFlagsForward): (JSC::DFG::Propagator::propagateArithNodeFlagsBackward): (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::propagatePredictionsForward): (JSC::DFG::Propagator::propagatePredictionsBackward): (JSC::DFG::Propagator::doRoundOfDoubleVoting): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::fixup): (JSC::DFG::Propagator::startIndexForChildren): (JSC::DFG::Propagator::endIndexForPureCSE): (JSC::DFG::Propagator::setReplacement): (JSC::DFG::Propagator::eliminate): (JSC::DFG::Propagator::performNodeCSE): (JSC::DFG::Propagator::localCSE): (JSC::DFG::Propagator::allocateVirtualRegisters): (JSC::DFG::Propagator::performBlockCFA): (JSC::DFG::Propagator::performForwardCFA): * dfg/DFGRegisterBank.h: (JSC::DFG::RegisterBank::dump): * dfg/DFGScoreBoard.h: (JSC::DFG::ScoreBoard::dump): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::dump): (JSC::DFG::SpeculativeJIT::checkConsistency): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): * heap/Heap.cpp: (JSC::Heap::destroy): * heap/MarkedBlock.h: * interpreter/CallFrame.cpp: (JSC::CallFrame::dumpCaller): * interpreter/Interpreter.cpp: (JSC::Interpreter::dumpRegisters): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompileSlowCases): (JSC::JIT::privateCompile): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * profiler/Profile.cpp: (JSC::Profile::debugPrintData): (JSC::Profile::debugPrintDataSampleStyle): * profiler/ProfileNode.cpp: (JSC::ProfileNode::debugPrintData): (JSC::ProfileNode::debugPrintDataSampleStyle): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::dumpRegExpTrace): * runtime/RegExp.cpp: (JSC::RegExp::matchCompareWithInterpreter): * runtime/SamplingCounter.cpp: (JSC::AbstractSamplingCounter::dump): * runtime/SamplingCounter.h: (JSC::DeletableSamplingCounter::~DeletableSamplingCounter): * runtime/ScopeChain.cpp: (JSC::ScopeChainNode::print): * runtime/Structure.cpp: (JSC::Structure::dumpStatistics): (JSC::PropertyMapStatisticsExitLogger::~PropertyMapStatisticsExitLogger): * tools/CodeProfile.cpp: (JSC::CodeProfile::report): * tools/ProfileTreeNode.h: (JSC::ProfileTreeNode::dumpInternal): * wtf/CMakeLists.txt: * wtf/DataLog.cpp: Added. (WTF): (WTF::initializeLogFileOnce): (WTF::initializeLogFile): (WTF::dataFile): (WTF::dataLogV): (WTF::dataLog): * wtf/DataLog.h: Added. (WTF): * wtf/HashTable.cpp: (WTF::HashTableStats::~HashTableStats): * wtf/MetaAllocator.cpp: (WTF::MetaAllocator::dumpProfile): * wtf/text/WTFString.cpp: (String::show): * yarr/YarrInterpreter.cpp: (JSC::Yarr::ByteCompiler::dumpDisjunction): 2012-02-11 Gavin Barraclough <barraclough@apple.com> Move special __proto__ property to Object.prototype https://bugs.webkit.org/show_bug.cgi?id=78409 Reviewed by Oliver Hunt. Re-implement this as a regular accessor property. This has three key benefits: 1) It makes it possible for objects to be given properties named __proto__. 2) Object.prototype.__proto__ can be deleted, preventing object prototypes from being changed. 3) This largely removes the magic used the implement __proto__, it can just be made a regular accessor property. * parser/Parser.cpp: (JSC::::parseFunctionInfo): - No need to prohibit functions named __proto__. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): - Add __proto__ accessor to Object.prototype. * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncProtoGetter): (JSC::globalFuncProtoSetter): - Definition of the __proto__ accessor functions. * runtime/JSGlobalObjectFunctions.h: - Declaration of the __proto__ accessor functions. * runtime/JSObject.cpp: (JSC::JSObject::put): - Remove the special handling for __proto__, there is still a check to allow for a fast guard for accessors excluding __proto__. (JSC::JSObject::putDirectAccessor): - Track on the structure whether an object contains accessors other than one for __proto__. (JSC::JSObject::defineOwnProperty): - No need to prohibit definition of own properties named __proto__. * runtime/JSObject.h: (JSC::JSObject::inlineGetOwnPropertySlot): - Remove the special handling for __proto__. (JSC::JSValue::get): - Remove the special handling for __proto__. * runtime/JSString.cpp: (JSC::JSString::getOwnPropertySlot): - Remove the special handling for __proto__. * runtime/JSValue.h: (JSValue): - Made synthesizePrototype public (this may be needed by the __proto__ getter). * runtime/ObjectConstructor.cpp: (JSC::objectConstructorGetPrototypeOf): - Perform the security check & call prototype() directly. * runtime/Structure.cpp: (JSC::Structure::Structure): - Added 'ExcludingProto' variant of the 'hasGetterSetterProperties' state. * runtime/Structure.h: (JSC::Structure::hasGetterSetterPropertiesExcludingProto): (JSC::Structure::setHasGetterSetterProperties): (Structure): - Added 'ExcludingProto' variant of the 'hasGetterSetterProperties' state. 2012-02-11 Filip Pizlo <fpizlo@apple.com> DFG CFA assumes that a WeakJSConstant's structure is known https://bugs.webkit.org/show_bug.cgi?id=78428 <rdar://problem/10849492> <rdar://problem/10849621> Reviewed by Gavin Barraclough. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): 2012-02-11 Mark Hahnenberg <mhahnenberg@apple.com> Qt debug build fix * heap/MarkedBlock.cpp: (JSC::MarkedBlock::callDestructor): Platforms that don't use clang will allocate JSFinalObjects in the destuctor subspace, so we should remove this assert so it doesn't cause crashes. 2012-02-11 Filip Pizlo <fpizlo@apple.com> Old 32_64 JIT should assert that its use of map() is consistent with the DFG OSR exit's expectations https://bugs.webkit.org/show_bug.cgi?id=78419 <rdar://problem/10817121> Reviewed by Oliver Hunt. * jit/JITInlineMethods.h: (JSC::JIT::map): 2012-02-11 Mark Hahnenberg <mhahnenberg@apple.com> Reduce the reentrancy limit of the interpreter for the iOS simulator https://bugs.webkit.org/show_bug.cgi?id=78400 Reviewed by Gavin Barraclough. * interpreter/Interpreter.h: Lowered the maximum reentrancy limit for large thread stacks. (JSC): 2012-02-11 Filip Pizlo <fpizlo@apple.com> [DFG] Misuse of WeakJSConstants in silentFillGPR code. https://bugs.webkit.org/show_bug.cgi?id=78423 <rdar://problem/10849353> <rdar://problem/10804043> Reviewed by Sam Weinig. The code was using Node::isConstant(), when it was supposed to use Node::hasConstant(). This patch is a surgical fix; the bigger problem is: why do we have isConstant() and hasConstant() when hasConstant() is correct and isConstant() is almost always wrong? * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::silentFillGPR): 2012-02-11 Sam Weinig <sam@webkit.org> Prepare JavaScriptCore to build with libc++ <rdar://problem/10426673> https://bugs.webkit.org/show_bug.cgi?id=78424 Reviewed by Anders Carlsson. * wtf/NullPtr.cpp: * wtf/NullPtr.h: libc++ provides std::nullptr emulation, so we don't have to. 2012-02-07 Filip Pizlo <fpizlo@apple.com> DFG should have polymorphic put_by_id caching https://bugs.webkit.org/show_bug.cgi?id=78062 <rdar://problem/10326439> <rdar://problem/10824839> Reviewed by Oliver Hunt. Implemented polymorphic put_by_id caching in the DFG, and added much of the machinery that would be needed to implement it in the old JIT as well. I decided against using the old PolymorphicAccessStructureList mechanism as this didn't quite fit with put_by_id. In particular, I wanted the ability to have one list that captured all relevant cases (including proto put_by_id if we ever decided to do it). And I wanted the code to have better encapsulation. And I didn't want to get confused by the fact that the original (non-list) put_by_id cache may itself consist of a stub routine. This code is still sub-optimal (for example adding a replace to a list whose previous elements are all transitions should just repatch the original code, but here it will generate a stub) but it already generates a >20% speed-up on V8-splay, leading to a 2% win overall in splay. Neutral elsewhere. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * bytecode/PolymorphicPutByIdList.cpp: Added. (JSC): (JSC::PutByIdAccess::fromStructureStubInfo): (JSC::PutByIdAccess::visitWeak): (JSC::PolymorphicPutByIdList::PolymorphicPutByIdList): (JSC::PolymorphicPutByIdList::from): (JSC::PolymorphicPutByIdList::~PolymorphicPutByIdList): (JSC::PolymorphicPutByIdList::isFull): (JSC::PolymorphicPutByIdList::isAlmostFull): (JSC::PolymorphicPutByIdList::addAccess): (JSC::PolymorphicPutByIdList::visitWeak): * bytecode/PolymorphicPutByIdList.h: Added. (JSC): (PutByIdAccess): (JSC::PutByIdAccess::PutByIdAccess): (JSC::PutByIdAccess::transition): (JSC::PutByIdAccess::replace): (JSC::PutByIdAccess::isSet): (JSC::PutByIdAccess::operator!): (JSC::PutByIdAccess::type): (JSC::PutByIdAccess::isTransition): (JSC::PutByIdAccess::isReplace): (JSC::PutByIdAccess::oldStructure): (JSC::PutByIdAccess::structure): (JSC::PutByIdAccess::newStructure): (JSC::PutByIdAccess::chain): (JSC::PutByIdAccess::stubRoutine): (PolymorphicPutByIdList): (JSC::PolymorphicPutByIdList::currentSlowPathTarget): (JSC::PolymorphicPutByIdList::isEmpty): (JSC::PolymorphicPutByIdList::size): (JSC::PolymorphicPutByIdList::at): (JSC::PolymorphicPutByIdList::operator[]): (JSC::PolymorphicPutByIdList::kind): * bytecode/PutKind.h: Added. (JSC): * bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::deref): (JSC::StructureStubInfo::visitWeakReferences): * bytecode/StructureStubInfo.h: (JSC): (JSC::isPutByIdAccess): (JSC::StructureStubInfo::initPutByIdList): (StructureStubInfo): (JSC::StructureStubInfo::reset): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: (DFG): * dfg/DFGRepatch.cpp: (JSC::DFG::appropriateGenericPutByIdFunction): (JSC::DFG::appropriateListBuildingPutByIdFunction): (DFG): (JSC::DFG::emitPutReplaceStub): (JSC::DFG::emitPutTransitionStub): (JSC::DFG::tryCachePutByID): (JSC::DFG::dfgRepatchPutByID): (JSC::DFG::tryBuildPutByIdList): (JSC::DFG::dfgBuildPutByIdList): (JSC::DFG::dfgResetPutByID): * dfg/DFGRepatch.h: (DFG): * runtime/WriteBarrier.h: (WriteBarrierBase): (JSC::WriteBarrierBase::copyFrom): 2012-02-10 Vineet Chaudhary <rgf748@motorola.com> https://bugs.webkit.org/show_bug.cgi?id=72756 DOMHTMLElement’s accessKey property is declared as available in WebKit version that didn’t have it Reviewed by Timothy Hatcher. * API/WebKitAvailability.h: Added AVAILABLE_AFTER_WEBKIT_VERSION_5_1 and AVAILABLE_WEBKIT_VERSION_1_3_AND_LATER_BUT_DEPRECATED_AFTER_WEBKIT_VERSION_5_1 for the new versions. 2012-02-10 Mark Hahnenberg <mhahnenberg@apple.com> Fixing windows build Unreviewed build fix * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-02-10 Adam Klein <adamk@chromium.org> Enable MUTATION_OBSERVERS by default on all platforms https://bugs.webkit.org/show_bug.cgi?id=78196 Reviewed by Ojan Vafai. * Configurations/FeatureDefines.xcconfig: 2012-02-10 Yong Li <yoli@rim.com> ENABLE(ASSEMBLER_WX_EXCLUSIVE): LinkBuffer can leave pages not marked as executable. https://bugs.webkit.org/show_bug.cgi?id=76724 Reviewed by Rob Buis. This issue only exists when both ENABLE(ASSEMBLER_WX_EXCLUSIVE) and ENABLE(BRANCH_COMPACTION) are on. The size used to call makeExecutable can be smaller than the one that was used for makeWritable. So it can leave pages behind that are not set back to default flags. When an assembly on one of those pages is executed or JIT returns to those pages in the case it was already executing from there, the software will crash. * assembler/LinkBuffer.h: Add m_initialSize and use it in performFinalization(). (JSC::LinkBuffer::LinkBuffer): (JSC::LinkBuffer::linkCode): (JSC::LinkBuffer::performFinalization): (LinkBuffer): 2012-02-10 Mark Hahnenberg <mhahnenberg@apple.com> Split MarkedSpace into destructor and destructor-free subspaces https://bugs.webkit.org/show_bug.cgi?id=77761 Reviewed by Geoffrey Garen. * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::emitAllocateJSFinalObject): Switched over to use destructor-free space. * heap/Heap.h: (JSC::Heap::allocatorForObjectWithoutDestructor): Added to give clients (e.g. the JIT) the ability to pick which subspace they want to allocate out of. (JSC::Heap::allocatorForObjectWithDestructor): Ditto. (Heap): (JSC::Heap::allocateWithDestructor): Added private function for CellAllocator to use. (JSC): (JSC::Heap::allocateWithoutDestructor): Ditto. * heap/MarkedAllocator.cpp: Added the cellsNeedDestruction flag to allocators so that they can allocate their MarkedBlocks correctly. (JSC::MarkedAllocator::allocateBlock): * heap/MarkedAllocator.h: (JSC::MarkedAllocator::cellsNeedDestruction): (MarkedAllocator): (JSC::MarkedAllocator::MarkedAllocator): (JSC): (JSC::MarkedAllocator::init): Replaced custom set functions, which were only used upon initialization, with an init function that does all of that stuff in fewer lines. * heap/MarkedBlock.cpp: (JSC::MarkedBlock::create): (JSC::MarkedBlock::recycle): (JSC::MarkedBlock::MarkedBlock): (JSC::MarkedBlock::callDestructor): Templatized, along with specializedSweep and sweepHelper, to make checking the m_cellsNeedDestructor flag faster and cleaner looking. (JSC): (JSC::MarkedBlock::specializedSweep): (JSC::MarkedBlock::sweep): (JSC::MarkedBlock::sweepHelper): * heap/MarkedBlock.h: (MarkedBlock): (JSC::MarkedBlock::cellsNeedDestruction): (JSC): * heap/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::resetAllocators): (JSC::MarkedSpace::canonicalizeCellLivenessData): (JSC::TakeIfUnmarked::operator()): * heap/MarkedSpace.h: (MarkedSpace): (Subspace): (JSC::MarkedSpace::allocatorFor): Needed function to differentiate between the two broad subspaces of allocators. (JSC): (JSC::MarkedSpace::destructorAllocatorFor): Ditto. (JSC::MarkedSpace::allocateWithoutDestructor): Ditto. (JSC::MarkedSpace::allocateWithDestructor): Ditto. (JSC::MarkedSpace::forEachBlock): * jit/JIT.h: * jit/JITInlineMethods.h: Modified to use the proper allocator for JSFinalObjects and others. (JSC::JIT::emitAllocateBasicJSObject): (JSC::JIT::emitAllocateJSFinalObject): (JSC::JIT::emitAllocateJSFunction): * runtime/JSArray.cpp: (JSC): * runtime/JSArray.h: (JSArray): (JSC::JSArray::create): (JSC): (JSC::JSArray::tryCreateUninitialized): * runtime/JSCell.h: (JSCell): (JSC): (NeedsDestructor): Template struct that calculates at compile time whether the class in question requires destruction or not using the compiler type trait __has_trivial_destructor. allocateCell then checks this constant to decide whether to allocate in the destructor or destructor-free parts of the heap. (JSC::allocateCell): * runtime/JSFunction.cpp: (JSC): * runtime/JSFunction.h: (JSFunction): * runtime/JSObject.cpp: (JSC): * runtime/JSObject.h: (JSNonFinalObject): (JSC): (JSFinalObject): (JSC::JSFinalObject::create): 2012-02-10 Adrienne Walker <enne@google.com> Remove implicit copy constructor usage in HashMaps with OwnPtr https://bugs.webkit.org/show_bug.cgi?id=78071 Reviewed by Darin Adler. Change the return type of emptyValue() in PairHashTraits to be the actual type returned rather than the trait type to avoid an implicit generation of the OwnPtr copy constructor. This happens for hash traits involving OwnPtr where the empty value is not zero and each hash bucket needs to be initialized with emptyValue(). Also, update StructureTransitionTable to use default hash traits rather than rolling its own, in order to update it to handle EmptyValueType. Test: patch from bug 74154 compiles on Clang with this patch * runtime/StructureTransitionTable.h: (StructureTransitionTable): * wtf/HashTraits.h: (GenericHashTraits): (PairHashTraits): (WTF::PairHashTraits::emptyValue): 2012-02-10 Aron Rosenberg <arosenberg@logitech.com> [Qt] Fix compiler warning in Visual Studio 2010 about TR1 https://bugs.webkit.org/show_bug.cgi?id=63642 Reviewed by Simon Hausmann. * JavaScriptCore.pri: 2012-02-10 Michael Saboff <msaboff@apple.com> Yarr assert with regexp where alternative in *-quantified group matches empty https://bugs.webkit.org/show_bug.cgi?id=67752 Reviewed by Gavin Barraclough. Added backtracking for the prior alternative if it matched but didn't consume any input characters. * yarr/YarrJIT.cpp: (YarrOp): New jump. (JSC::Yarr::YarrGenerator::generate): Emit conditional jump when an alternative matches and no input was consumed. Moved the zero length match check for a set of alternatives to the alternative code from the parentheses cases to the alternative end cases. Converted the existing zero length checks in the parentheses cases to runtime assertion checks. (JSC::Yarr::YarrGenerator::backtrack): Link new jump to backtrack to prior term. 2012-02-10 Roland Takacs <takacs.roland@stud.u-szeged.hu> [Qt] GC should be parallel on Qt platform https://bugs.webkit.org/show_bug.cgi?id=73309 Reviewed by Zoltan Herczeg. These changes made the parallel gc feature available for Qt port. The implementation of "registerGCThread" and "isMainThreadOrGCThread", and a local static function [initializeGCThreads] is moved from MainThreadMac.mm to the common MainThread.cpp to make them available for other platforms. Measurement results: V8 speed-up: 1.025x as fast [From: 663.4ms To: 647.0ms ] V8 Splay speed-up: 1.185x as fast [From: 138.4ms To: 116.8ms ] Tested on Intel(R) Core(TM) i5-2320 CPU @ 3.00GHz with 4-core. * JavaScriptCore.order: * wtf/MainThread.cpp: (WTF::initializeMainThread): (WTF): (WTF::initializeGCThreads): (WTF::registerGCThread): (WTF::isMainThreadOrGCThread): * wtf/MainThread.h: (WTF): * wtf/Platform.h: * wtf/mac/MainThreadMac.mm: (WTF): 2012-02-09 Andy Wingo <wingo@igalia.com> Eliminate dead code in BytecodeGenerator::resolve() https://bugs.webkit.org/show_bug.cgi?id=78242 Reviewed by Gavin Barraclough. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::resolve): BytecodeGenerator::shouldOptimizeLocals() is only true for FunctionCode, and thus cannot be true for GlobalCode. 2012-02-09 Andy Wingo <wingo@igalia.com> Remove BytecodeGenerator::isLocal https://bugs.webkit.org/show_bug.cgi?id=78241 Minor refactor to BytecodeGenerator. Reviewed by Gavin Barraclough. * bytecompiler/BytecodeGenerator.h: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::isLocal): (JSC::BytecodeGenerator::isLocalConstant): Remove now-unused methods. * bytecompiler/NodesCodegen.cpp: (JSC::ResolveNode::isPure): Use the ResolveResult mechanism instead of isLocal. This will recognize more resolve nodes as being pure. (JSC::PrefixResolveNode::emitBytecode): Use isReadOnly on the location instead of isLocalConstant. 2012-02-09 Oliver Hunt <oliver@apple.com> The JS Parser scope object needs a VectorTrait specialization https://bugs.webkit.org/show_bug.cgi?id=78308 Reviewed by Gavin Barraclough. This showed up as a periodic crash in various bits of generated code originally, but I've added an assertion in the bytecode generator that makes the effected code much more crash-happy should it go wrong again. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): (JSC::BytecodeGenerator::resolve): * parser/Parser.cpp: * parser/Parser.h: (JSC): * runtime/JSActivation.h: (JSC::JSActivation::isValidScopedLookup): (JSActivation): 2012-02-08 Oliver Hunt <oliver@apple.com> Whoops, fix the build. * runtime/Executable.cpp: (JSC::FunctionExecutable::FunctionExecutable): 2012-02-08 Oliver Hunt <oliver@apple.com> Fix issue encountered while debugging stacktraces https://bugs.webkit.org/show_bug.cgi?id=78147 Reviewed by Gavin Barraclough. Debugging is easier if we always ensure that we have a non-null inferred name. * runtime/Executable.cpp: (JSC::FunctionExecutable::FunctionExecutable): 2012-02-08 Oliver Hunt <oliver@apple.com> updateTopCallframe in the baseline JIT doesn't provide enough information to the stubs https://bugs.webkit.org/show_bug.cgi?id=78145 Reviewed by Gavin Barraclough. Fix the updateTopCallFrame helper to store additional information that becomes necessary when we are trying to provide more stack frame information. * interpreter/CallFrame.h: (JSC::ExecState::bytecodeOffsetForBaselineJIT): (ExecState): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JIT.h: (JSC::JIT::compileGetByIdProto): (JSC::JIT::compileGetByIdSelfList): (JSC::JIT::compileGetByIdProtoList): (JSC::JIT::compileGetByIdChainList): (JSC::JIT::compileGetByIdChain): (JSC::JIT::compilePutByIdTransition): (JIT): * jit/JITInlineMethods.h: (JSC::JIT::updateTopCallFrame): 2012-02-07 Robert Kroeger <rjkroege@chromium.org> [chromium] Remove the enable marcro for the no longer necessary Chromium gesture recognizer. https://bugs.webkit.org/show_bug.cgi?id=77492 Reviewed by Adam Barth. * wtf/Platform.h: 2012-02-07 Tony Chang <tony@chromium.org> merge DashboardSupportCSSPropertyNames.in into CSSPropertyNames.in https://bugs.webkit.org/show_bug.cgi?id=78036 Reviewed by Darin Adler. * Configurations/FeatureDefines.xcconfig: Add ENABLE_DASHBOARD_SUPPORT to FEATURE_DEFINES. 2012-02-07 Gyuyoung Kim <gyuyoung.kim@samsung.com> [CMAKE] Use *bin* and *lib* directories for executable and libraries. https://bugs.webkit.org/show_bug.cgi?id=77928 Reviewed by Daniel Bates. CMake has used *Programs* directory for executable. In addition, shared libraries are being built in source directory. It is better to set common places in order to maintain executable and libraries. *bin* is for executable and *lib* is for library. * shell/CMakeLists.txt: Change *Programs* with *bin*. 2012-02-07 Gavin Barraclough <barraclough@apple.com> Crash on http://www.rickshawbags.com/ https://bugs.webkit.org/show_bug.cgi?id=78045 Reviewed by Darin Adler. Problem URL is: http://www.rickshawbags.com/customize/custom-bag#!thl=rickshaw/bag() This is a bug introduced by https://bugs.webkit.org/show_bug.cgi?id=71933, isVariableObject() checks were excluding StaticScopeObjects, this patch inadvertently changed them to be included. * runtime/JSType.h: - sort JSType enum such that StaticScopeObjectType comes before VariableObjectType, and thus is excluded from isVariableObject() checks. 2012-02-06 Jer Noble <jer.noble@apple.com> Use CMClock as a timing source for PlatformClock where available. https://bugs.webkit.org/show_bug.cgi?id=77885 Reviewed by Eric Carlson. * wtf/Platform.h: Added WTF_USE_COREMEDIA. 2012-02-06 Filip Pizlo <fpizlo@apple.com> ValueToNumber and ValueToDouble nodes don't do anything and should be removed https://bugs.webkit.org/show_bug.cgi?id=77855 <rdar://problem/10811325> Reviewed by Gavin Barraclough. Removed ValueToNumber and ValueToDouble, because the only thing they were doing was wasting registers. This looks like a 1% win on V8 (with a 5% win on crypto) and a 2-3% win on Kraken, mostly due to a >10% win on gaussian-blur. No win anywhere else. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getToInt32): (ByteCodeParser): (JSC::DFG::ByteCodeParser::handleMinMax): (JSC::DFG::ByteCodeParser::handleIntrinsic): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGNode.h: (DFG): (JSC::DFG::Node::hasArithNodeFlags): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateArithNodeFlags): (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::vote): (JSC::DFG::Propagator::doRoundOfDoubleVoting): (Propagator): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::canonicalize): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-02-06 Patrick Gansterer <paroga@webkit.org> Unreviewed WinCE build fix after r106197. * tools/CodeProfiling.cpp: (JSC::CodeProfiling::notifyAllocator): getenv() isn't supported by WinCE. Don't call it. 2012-02-05 Gavin Barraclough <barraclough@apple.com> Remove JSObject defineGetter/defineSetter lookupGetter/lookupSetter https://bugs.webkit.org/show_bug.cgi?id=77451 Reviewed by Sam Weinig. These can now all be implemented in terms of defineOwnProperty & getPropertyDescriptor. Also remove initializeGetterSetterProperty, since this is equivalent to putDirectAccessor. * JavaScriptCore.exp: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::defineOwnProperty): * debugger/DebuggerActivation.h: (DebuggerActivation): * runtime/ClassInfo.h: (MethodTable): (JSC): * runtime/JSBoundFunction.cpp: (JSC::JSBoundFunction::finishCreation): * runtime/JSCell.cpp: (JSC): * runtime/JSCell.h: (JSCell): * runtime/JSFunction.cpp: (JSC::JSFunction::getOwnPropertySlot): (JSC::JSFunction::getOwnPropertyDescriptor): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::defineOwnProperty): (JSC): * runtime/JSGlobalObject.h: (JSGlobalObject): * runtime/JSObject.cpp: (JSC): * runtime/JSObject.h: (JSObject): * runtime/ObjectPrototype.cpp: (JSC::objectProtoFuncDefineGetter): (JSC::objectProtoFuncDefineSetter): (JSC::objectProtoFuncLookupGetter): (JSC::objectProtoFuncLookupSetter): 2012-02-06 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck. * GNUmakefile.list.am: Add missing files. 2012-02-05 Filip Pizlo <fpizlo@apple.com> DFG's child references from one node to another should have room for type information https://bugs.webkit.org/show_bug.cgi?id=77797 Reviewed by Oliver Hunt. The DFG::Node::child fields now contain both a DFG::NodeIndex (which is just an unsigned) and a DFG::UseKind (which is currently an effectively empty enum). They are encapsulated together as a DFG::NodeUse, which can in most cases still be used as an index (for example DFG::Graph, AbstractState, and SpeculativeJIT all accept NodeUse in most places where they really want a NodeIndex). The NodeUse stores both the index and the UseKind without bloating the memory usage of DFG::Node, since we really don't need full 32 bits for the NodeIndex (a DFG::Node is roughly 11 words, so if we assume that we never want to use more than 1GB to DFG compile something - likely a sensible assumption! - then we will only be able to have room for about 24 million nodes, which means we only need about 24.5 bits for the node index). Currently the DFG::NodeUse allocates 4 bits for the UseKind and 28 bits for the index, but stores the index as a signed number to make NoNode work naturally. Hence we really just have 27 bits for the index. This is performance-neutral on all benchmarks we track. * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGAbstractState.h: (JSC::DFG::AbstractState::forNode): (AbstractState): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getLocal): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::toInt32): (JSC::DFG::ByteCodeParser::addVarArgChild): (JSC::DFG::ByteCodeParser::processPhiStack): * dfg/DFGCommon.h: * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): (DFG): * dfg/DFGGraph.h: (Graph): (JSC::DFG::Graph::operator[]): (JSC::DFG::Graph::at): (JSC::DFG::Graph::ref): (JSC::DFG::Graph::deref): (JSC::DFG::Graph::clearAndDerefChild1): (JSC::DFG::Graph::clearAndDerefChild2): (JSC::DFG::Graph::clearAndDerefChild3): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::getPrediction): * dfg/DFGNode.h: (JSC::DFG::Node::Node): (JSC::DFG::Node::child1): (JSC::DFG::Node::child1Unchecked): (JSC::DFG::Node::child2): (JSC::DFG::Node::child3): (JSC::DFG::Node::firstChild): (JSC::DFG::Node::numChildren): (JSC::DFG::Node::dumpChildren): (Node): * dfg/DFGNodeReferenceBlob.h: Added. (DFG): (NodeReferenceBlob): (JSC::DFG::NodeReferenceBlob::NodeReferenceBlob): (JSC::DFG::NodeReferenceBlob::child): (JSC::DFG::NodeReferenceBlob::child1): (JSC::DFG::NodeReferenceBlob::child2): (JSC::DFG::NodeReferenceBlob::child3): (JSC::DFG::NodeReferenceBlob::child1Unchecked): (JSC::DFG::NodeReferenceBlob::initialize): (JSC::DFG::NodeReferenceBlob::firstChild): (JSC::DFG::NodeReferenceBlob::setFirstChild): (JSC::DFG::NodeReferenceBlob::numChildren): (JSC::DFG::NodeReferenceBlob::setNumChildren): * dfg/DFGNodeUse.h: Added. (DFG): (NodeUse): (JSC::DFG::NodeUse::NodeUse): (JSC::DFG::NodeUse::indexUnchecked): (JSC::DFG::NodeUse::index): (JSC::DFG::NodeUse::setIndex): (JSC::DFG::NodeUse::useKind): (JSC::DFG::NodeUse::setUseKind): (JSC::DFG::NodeUse::isSet): (JSC::DFG::NodeUse::operator!): (JSC::DFG::NodeUse::operator==): (JSC::DFG::NodeUse::operator!=): (JSC::DFG::NodeUse::shift): (JSC::DFG::NodeUse::makeWord): (JSC::DFG::operator==): (JSC::DFG::operator!=): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateArithNodeFlags): (JSC::DFG::Propagator::vote): (JSC::DFG::Propagator::toDouble): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::canonicalize): (JSC::DFG::Propagator::startIndex): (JSC::DFG::Propagator::globalVarLoadElimination): (JSC::DFG::Propagator::getByValLoadElimination): (JSC::DFG::Propagator::getByOffsetLoadElimination): (JSC::DFG::Propagator::performSubstitution): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGScoreBoard.h: (JSC::DFG::ScoreBoard::use): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::useChildren): (JSC::DFG::SpeculativeJIT::writeBarrier): (JSC::DFG::SpeculativeJIT::nonSpeculativeStrictEq): (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality): (JSC::DFG::SpeculativeJIT::compilePeepHoleIntegerBranch): (JSC::DFG::SpeculativeJIT::compileMovHint): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): (JSC::DFG::SpeculativeJIT::compilePutByValForFloatTypedArray): (JSC::DFG::SpeculativeJIT::compileSoftModulo): (JSC::DFG::SpeculativeJIT::compileAdd): (JSC::DFG::SpeculativeJIT::compileArithSub): (JSC::DFG::SpeculativeJIT::compileStrictEqForConstant): (JSC::DFG::SpeculativeJIT::compileStrictEq): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::at): (JSC::DFG::SpeculativeJIT::canReuse): (JSC::DFG::SpeculativeJIT::use): (SpeculativeJIT): (JSC::DFG::SpeculativeJIT::detectPeepHoleBranch): (JSC::DFG::SpeculativeJIT::speculationCheck): (JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution): (JSC::DFG::IntegerOperand::IntegerOperand): (JSC::DFG::DoubleOperand::DoubleOperand): (JSC::DFG::JSValueOperand::JSValueOperand): (JSC::DFG::StorageOperand::StorageOperand): (JSC::DFG::SpeculateIntegerOperand::SpeculateIntegerOperand): (JSC::DFG::SpeculateStrictInt32Operand::SpeculateStrictInt32Operand): (JSC::DFG::SpeculateDoubleOperand::SpeculateDoubleOperand): (JSC::DFG::SpeculateCellOperand::SpeculateCellOperand): (JSC::DFG::SpeculateBooleanOperand::SpeculateBooleanOperand): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativeValueToNumber): (JSC::DFG::SpeculativeJIT::nonSpeculativeValueToInt32): (JSC::DFG::SpeculativeJIT::cachedPutById): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompareNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull): (JSC::DFG::SpeculativeJIT::nonSpeculativeCompareNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranch): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompare): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeStrictEq): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeStrictEq): (JSC::DFG::SpeculativeJIT::emitCall): (JSC::DFG::SpeculativeJIT::compileValueAdd): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativeValueToNumber): (JSC::DFG::SpeculativeJIT::nonSpeculativeValueToInt32): (JSC::DFG::SpeculativeJIT::cachedPutById): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompareNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull): (JSC::DFG::SpeculativeJIT::nonSpeculativeCompareNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranch): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompare): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeStrictEq): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeStrictEq): (JSC::DFG::SpeculativeJIT::emitCall): (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compileValueAdd): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): 2012-02-05 Gyuyoung Kim <gyuyoung.kim@samsung.com> [CMAKE] Support javascriptcore test for EFL port. https://bugs.webkit.org/show_bug.cgi?id=77425 Reviewed by Daniel Bates. Efl and WinCE as well as Blackberry port are now using Cmake as its build system and they are share the make file to create jsc excutable. In order to run "run-javascriptcore-tests", EFL port needs to change jsc installation configuration with executable output directory(e.g. Programs). So, this patch change jsc installation configuration only for EFL port. * shell/CMakeLists.txt: 2012-02-04 Gavin Barraclough <barraclough@apple.com> Rubber stamped by Sam Weinig. * yarr/YarrPattern.cpp: (JSC::Yarr::YarrPatternConstructor::quantifyAtom): - Fix comment. 2012-02-04 Kalev Lember <kalevlember@gmail.com> [GTK] CurrentTime: Reorder headers for win32 https://bugs.webkit.org/show_bug.cgi?id=77808 Reviewed by Martin Robinson. In GTK+ win32 port, monotonicallyIncreasingTime() implementation is based on g_get_monotonic_time(). Reorder headers to make sure glib.h gets included even when the platform is win32. CurrentTime.cpp: In function 'double WTF::monotonicallyIncreasingTime()': CurrentTime.cpp:321:53: error: 'g_get_monotonic_time' was not declared in this scope CurrentTime.cpp:322:1: warning: control reaches end of non-void function [-Wreturn-type] * wtf/CurrentTime.cpp: 2012-02-03 Anders Carlsson <andersca@apple.com> Prefix the typedef in WTF_MAKE_FAST_ALLOCATED with underscores https://bugs.webkit.org/show_bug.cgi?id=77788 Reviewed by Andreas Kling. The current typedef name, 'ThisIsHereToForceASemicolonAfterThisMacro', shows up when trying to code-complete 'this' in Xcode. Prefix the typedef with two underscores to stop this from happening. * wtf/FastAllocBase.h: 2012-02-03 Rob Buis <rbuis@rim.com> Fix alignment warnings in ARMv7 https://bugs.webkit.org/show_bug.cgi?id=55368 Reviewed by Filip Pizlo. Use reinterpret_cast_ptr and static_cast to get rid of alignment issues in ARMv7 code. * heap/HandleTypes.h: (JSC::HandleTypes::getFromSlot): * heap/MarkedBlock.cpp: (JSC::MarkedBlock::specializedSweep): * heap/MarkedBlock.h: (JSC::MarkedBlock::forEachCell): * runtime/WriteBarrier.h: (JSC::WriteBarrierBase::get): (JSC::WriteBarrierBase::unvalidatedGet): 2012-02-03 Mark Hahnenberg <mhahnenberg@apple.com> Build fix Unreviewed build fix Forgot to add a couple files. * heap/MarkedAllocator.cpp: Added. (JSC): (JSC::MarkedAllocator::tryAllocateHelper): (JSC::MarkedAllocator::tryAllocate): (JSC::MarkedAllocator::allocateSlowCase): (JSC::MarkedAllocator::allocateBlock): (JSC::MarkedAllocator::addBlock): (JSC::MarkedAllocator::removeBlock): * heap/MarkedAllocator.h: Added. (JSC): (DFG): (MarkedAllocator): (JSC::MarkedAllocator::cellSize): (JSC::MarkedAllocator::heap): (JSC::MarkedAllocator::setHeap): (JSC::MarkedAllocator::setCellSize): (JSC::MarkedAllocator::setMarkedSpace): (JSC::MarkedAllocator::MarkedAllocator): (JSC::MarkedAllocator::allocate): (JSC::MarkedAllocator::reset): (JSC::MarkedAllocator::zapFreeList): (JSC::MarkedAllocator::forEachBlock): 2012-02-03 Mark Hahnenberg <mhahnenberg@apple.com> Refactor MarkedBlock::SizeClass into a separate class https://bugs.webkit.org/show_bug.cgi?id=77600 Reviewed by Geoffrey Garen. We pulled SizeClass out into its own class, named MarkedAllocator, and gave it the responsibility of allocating objects from the collection of MarkedBlocks that it manages. Also limited the amount of coupling to internal data fields from other places, although it's mostly unavoidable in the JIT code. Eventually MarkedAllocator will implement various policies to do with object management, e.g. whether or not to run destructors on objects that it manages. MarkedSpace will manage a collection of MarkedAllocators with varying policies, as it does now but to a larger extent. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::emitAllocateJSFinalObject): * heap/Heap.cpp: (JSC::Heap::collect): (JSC::Heap::resetAllocators): * heap/Heap.h: (JSC::Heap::allocatorForObject): (Heap): * heap/MarkedAllocator.cpp: Added. (JSC): (JSC::MarkedAllocator::tryAllocateHelper): (JSC::MarkedAllocator::tryAllocate): (JSC::MarkedAllocator::allocateSlowCase): (JSC::MarkedAllocator::allocateBlock): (JSC::MarkedAllocator::addBlock): (JSC::MarkedAllocator::removeBlock): * heap/MarkedAllocator.h: Added. (JSC): (DFG): (MarkedAllocator): (JSC::MarkedAllocator::cellSize): (JSC::MarkedAllocator::heap): (JSC::MarkedAllocator::setHeap): (JSC::MarkedAllocator::setCellSize): (JSC::MarkedAllocator::setMarkedSpace): (JSC::MarkedAllocator::MarkedAllocator): (JSC::MarkedAllocator::allocate): (JSC::MarkedAllocator::reset): (JSC::MarkedAllocator::zapFreeList): (JSC::MarkedAllocator::forEachBlock): * heap/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::resetAllocators): (JSC::MarkedSpace::canonicalizeCellLivenessData): (JSC::TakeIfUnmarked::operator()): * heap/MarkedSpace.h: (MarkedSpace): (JSC::MarkedSpace::allocatorFor): (JSC::MarkedSpace::allocate): (JSC::MarkedSpace::forEachBlock): (JSC::MarkedSpace::didAddBlock): (JSC::MarkedSpace::didConsumeFreeList): * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateBasicJSObject): 2012-02-03 Simon Hausmann <simon.hausmann@nokia.com> [Qt] Replace GNU linker script for exports with export macros in WTF/JSC https://bugs.webkit.org/show_bug.cgi?id=77723 Reviewed by Tor Arne Vestbø. * wtf/Platform.h: Enable use of export macros. 2012-02-02 Hajime Morrita <morrita@chromium.org> Unreviewed, removing an unnecessarily JS_PRIVATE_EXPORT annotation. * interpreter/Interpreter.h: (Interpreter): 2012-01-31 Hajime Morrita <morrita@chromium.org> [Mac] eliminate JavaScriptCore.exp https://bugs.webkit.org/show_bug.cgi?id=72854 Reviewed by Darin Adler. - Removed exp files and corresponding makefile entries. - Changed the build configuration no to use exp file. * Configurations/JavaScriptCore.xcconfig: * DerivedSources.make: * JavaScriptCore.JSVALUE32_64only.exp: Removed. * JavaScriptCore.JSVALUE64only.exp: Removed. * JavaScriptCore.exp: Removed. * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/Platform.h: 2012-02-02 Benjamin Poulain <bpoulain@apple.com> Running a Web Worker on about:blank crashes the interpreter https://bugs.webkit.org/show_bug.cgi?id=77593 Reviewed by Michael Saboff. The method Interpreter::execute() was crashing on empty programs because the assumption is made the source is not null. This patch shortcut the execution when the String is null to avoid invalid memory access. * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): 2012-02-02 Kalev Lember <kalevlember@gmail.com> [GTK] Use win32 native threading https://bugs.webkit.org/show_bug.cgi?id=77676 Reviewed by Martin Robinson. r97269 switched from glib threading to pthreads, breaking win32 GTK+. This is a follow up, removing some leftovers in ThreadSpecific.h and switching win32 to use the native threading in ThreadingWin.cpp. * GNUmakefile.list.am: Compile in win32 native threading support * wtf/ThreadSpecific.h: Remove GTK+-specific definitions (ThreadSpecific): (WTF::::destroy): 2012-02-02 Filip Pizlo <fpizlo@apple.com> retrieveCallerFromVMCode should call trueCallerFrame https://bugs.webkit.org/show_bug.cgi?id=77684 Reviewed by Oliver Hunt. * interpreter/Interpreter.cpp: (JSC::Interpreter::retrieveCallerFromVMCode): 2012-02-02 Kalev Lember <kalevlember@gmail.com> [GTK] Implement current executable path finding for win32 https://bugs.webkit.org/show_bug.cgi?id=77677 Reviewed by Martin Robinson. The WTF helper for getting the binary path that was added in r101710 left out the win32 implementation. Fix this. * wtf/gobject/GlibUtilities.cpp: (getCurrentExecutablePath): 2012-02-02 Filip Pizlo <fpizlo@apple.com> Throwing away bytecode and then reparsing during DFG optimization is just plain wrong and makes things crash https://bugs.webkit.org/show_bug.cgi?id=77680 <rdar://problem/10798490> Reviewed by Oliver Hunt. This is the minimal surgical fix: it removes the code that triggered bytecode throw-away. Once we're confident that this is a good idea, we can kill all of the code that implements the feature. * bytecode/CodeBlock.h: (JSC::CodeBlock::discardBytecodeLater): (JSC::CodeBlock::addValueProfile): * jit/JITDriver.h: (JSC::jitCompileIfAppropriate): (JSC::jitCompileFunctionIfAppropriate): 2012-02-02 Filip Pizlo <fpizlo@apple.com> Release build debugging should be easier https://bugs.webkit.org/show_bug.cgi?id=77669 Reviewed by Gavin Barraclough. * assembler/ARMAssembler.h: (ARMAssembler): (JSC::ARMAssembler::debugOffset): * assembler/ARMv7Assembler.h: (ARMv7Assembler): (JSC::ARMv7Assembler::debugOffset): (ARMInstructionFormatter): (JSC::ARMv7Assembler::ARMInstructionFormatter::debugOffset): * assembler/AbstractMacroAssembler.h: (AbstractMacroAssembler): (JSC::AbstractMacroAssembler::debugOffset): * assembler/AssemblerBuffer.h: (AssemblerBuffer): (JSC::AssemblerBuffer::debugOffset): * assembler/LinkBuffer.h: (LinkBuffer): (JSC::LinkBuffer::debugSize): * assembler/MIPSAssembler.h: (MIPSAssembler): (JSC::MIPSAssembler::debugOffset): * assembler/X86Assembler.h: (X86Assembler): (JSC::X86Assembler::debugOffset): (X86InstructionFormatter): (JSC::X86Assembler::X86InstructionFormatter::debugOffset): * bytecode/CodeBlock.cpp: (JSC): * bytecode/CodeBlock.h: (CodeBlock): * bytecode/CodeOrigin.h: (CodeOrigin): (JSC): (JSC::CodeOrigin::inlineStack): * bytecode/DFGExitProfile.h: (JSC::DFG::exitKindToString): * bytecode/DataFormat.h: (JSC::dataFormatToString): * bytecode/PredictedType.cpp: (JSC): (JSC::predictionToString): * bytecode/PredictedType.h: (JSC): * bytecode/ValueRecovery.h: (ValueRecovery): (JSC::ValueRecovery::dump): * bytecompiler/BytecodeGenerator.cpp: (JSC): (JSC::BytecodeGenerator::setDumpsGeneratedCode): (JSC::BytecodeGenerator::dumpsGeneratedCode): (JSC::BytecodeGenerator::generate): * dfg/DFGAbstractValue.h: (StructureAbstractValue): (JSC::DFG::StructureAbstractValue::dump): (AbstractValue): (JSC::DFG::AbstractValue::dump): * dfg/DFGAssemblyHelpers.h: (DFG): (AssemblyHelpers): (JSC::DFG::AssemblyHelpers::debugCall): * dfg/DFGFPRInfo.h: (FPRInfo): (JSC::DFG::FPRInfo::debugName): * dfg/DFGGPRInfo.h: (GPRInfo): (JSC::DFG::GPRInfo::debugName): * dfg/DFGGraph.cpp: (DFG): * dfg/DFGGraph.h: (Graph): * dfg/DFGNode.h: (DFG): (JSC::DFG::arithNodeFlagsAsString): (Node): (JSC::DFG::Node::hasIdentifier): (JSC::DFG::Node::dumpChildren): * dfg/DFGOSRExit.cpp: (DFG): (JSC::DFG::OSRExit::dump): * dfg/DFGOSRExit.h: (OSRExit): * runtime/JSValue.cpp: (JSC): (JSC::JSValue::description): * runtime/JSValue.h: (JSValue): * wtf/BitVector.cpp: (WTF): (WTF::BitVector::dump): * wtf/BitVector.h: (BitVector): 2012-02-02 Oliver Hunt <oliver@apple.com> Getters and setters cause line numbers in errors/console.log to be offset for the whole file https://bugs.webkit.org/show_bug.cgi?id=77675 Reviewed by Timothy Hatcher. Our default literal parsing logic doesn't handle the extra work required for getters and setters. When it encounters one, it rolls back the lexer and then switches to a more complete parsing function. Unfortunately it was only winding back the character position, and was ignoring the line number and other lexer data. This led to every getter and setter causing the line number to be incorrectly incremented leading to increasingly incorrect numbers for the rest of the file. * parser/Parser.cpp: (JSC::::parseObjectLiteral): 2012-02-02 Andy Wingo <wingo@igalia.com> Fix type punning warning in HashTable.h debug builds https://bugs.webkit.org/show_bug.cgi?id=77422 Reviewed by Gavin Barraclough. * wtf/HashTable.h (WTF::HashTable::checkKey): Fix type punning warning appearing in debug builds with gcc-4.6.2 on GNU/Linux. 2012-02-01 Michael Saboff <msaboff@apple.com> Yarr crash with regexp replace https://bugs.webkit.org/show_bug.cgi?id=67454 Reviewed by Gavin Barraclough. Properly handle the case of a back reference to an unmatched subpattern by always matching without consuming any characters. * yarr/YarrInterpreter.cpp: (JSC::Yarr::Interpreter::matchBackReference): (JSC::Yarr::Interpreter::backtrackBackReference): 2012-02-01 Gavin Barraclough <barraclough@apple.com> calling function on catch block scope containing an eval result in wrong this value being passed https://bugs.webkit.org/show_bug.cgi?id=77581 Reviewed by Oliver Hunt. javascript:function F(){ return 'F' in this; }; try { throw F; } catch (e) { eval(""); alert(e()); } * bytecompiler/NodesCodegen.cpp: (JSC::TryNode::emitBytecode): * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): * parser/ASTBuilder.h: (JSC::ASTBuilder::createTryStatement): * parser/NodeConstructors.h: (JSC::TryNode::TryNode): * parser/Nodes.h: (TryNode): * parser/Parser.cpp: (JSC::::parseTryStatement): * parser/SyntaxChecker.h: (JSC::SyntaxChecker::createTryStatement): * runtime/JSObject.h: (JSObject): (JSC::JSObject::isStaticScopeObject): (JSC): 2012-02-01 Oliver Hunt <oliver@apple.com> Add support for inferred function names https://bugs.webkit.org/show_bug.cgi?id=77579 Reviewed by Gavin Barraclough. Add new "inferred" names to function expressions, getters, and setters. This property is not exposed to JS, so is only visible in the debugger and profiler. * JavaScriptCore.exp: * bytecompiler/BytecodeGenerator.h: (JSC::BytecodeGenerator::makeFunction): * debugger/DebuggerCallFrame.cpp: (JSC::DebuggerCallFrame::calculatedFunctionName): * parser/ASTBuilder.h: (JSC::ASTBuilder::createAssignResolve): (JSC::ASTBuilder::createGetterOrSetterProperty): (JSC::ASTBuilder::createProperty): (JSC::ASTBuilder::makeAssignNode): * parser/Nodes.h: (JSC::FunctionBodyNode::setInferredName): (JSC::FunctionBodyNode::inferredName): (FunctionBodyNode): * profiler/Profiler.cpp: (JSC): (JSC::Profiler::createCallIdentifier): (JSC::createCallIdentifierFromFunctionImp): * runtime/Executable.cpp: (JSC::FunctionExecutable::FunctionExecutable): (JSC::FunctionExecutable::fromGlobalCode): * runtime/Executable.h: (JSC::FunctionExecutable::create): (JSC::FunctionExecutable::inferredName): (FunctionExecutable): * runtime/JSFunction.cpp: (JSC::JSFunction::calculatedDisplayName): (JSC): (JSC::getCalculatedDisplayName): * runtime/JSFunction.h: (JSC): 2012-02-01 Filip Pizlo <fpizlo@apple.com> DFG should fold double-to-int conversions https://bugs.webkit.org/show_bug.cgi?id=77532 Reviewed by Oliver Hunt. Performance neutral on major benchmarks. But it makes calling V8's Math.random() 4x faster. * bytecode/CodeBlock.cpp: (JSC): (JSC::CodeBlock::addOrFindConstant): * bytecode/CodeBlock.h: (JSC::CodeBlock::addConstant): (CodeBlock): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::toInt32): (ByteCodeParser): (JSC::DFG::ByteCodeParser::getJSConstantForValue): (JSC::DFG::ByteCodeParser::isInt32Constant): * dfg/DFGGraph.h: (JSC::DFG::Graph::addShouldSpeculateInteger): (Graph): (JSC::DFG::Graph::addImmediateShouldSpeculateInteger): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::doRoundOfDoubleVoting): (JSC::DFG::Propagator::fixupNode): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileAdd): (DFG): (JSC::DFG::SpeculativeJIT::compileArithSub): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::valueOfNumberConstantAsInt32): (SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/JSValueInlineMethods.h: (JSC::JSValue::asDouble): 2012-02-01 Filip Pizlo <fpizlo@apple.com> DFG graph dump for GetScopedVar should show the correct prediction https://bugs.webkit.org/show_bug.cgi?id=77530 Reviewed by Geoff Garen. GetScopedVar has a heap prediction, not a variable prediction. But it does have a variable. Hence we need to check for heap predictions before checking for variable predictions. * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): 2012-02-01 Mark Hahnenberg <mhahnenberg@apple.com> Replace JSArray destructor with finalizer https://bugs.webkit.org/show_bug.cgi?id=77488 Reviewed by Geoffrey Garen. * JavaScriptCore.exp: * runtime/JSArray.cpp: (JSC::JSArray::finalize): Added finalizer. (JSC::JSArray::allocateSparseMap): Factored out code for allocating new sparse maps. (JSC): (JSC::JSArray::deallocateSparseMap): Factored out code for deallocating sparse maps. (JSC::JSArray::enterDictionaryMode): Renamed enterSparseMode to enterDictionaryMode because the old name was confusing because we could have a sparse array that never called enterSparseMode. (JSC::JSArray::defineOwnNumericProperty): (JSC::JSArray::setLengthWritable): (JSC::JSArray::putByIndexBeyondVectorLength): (JSC::JSArray::setLength): (JSC::JSArray::pop): (JSC::JSArray::sort): (JSC::JSArray::compactForSorting): * runtime/JSArray.h: (JSArray): 2012-02-01 Andy Wingo <wingo@igalia.com> Refactor identifier resolution in BytecodeGenerator https://bugs.webkit.org/show_bug.cgi?id=76285 Reviewed by Geoffrey Garen. * bytecompiler/BytecodeGenerator.h: (JSC::ResolveResult): New class, to describe the storage location corresponding to an identifier in a program. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::resolve): New function, replacing findScopedProperty. (JSC::BytecodeGenerator::resolveConstDecl): New function, encapsulating what ConstDeclNode::emitBytecode used to do. (JSC::BytecodeGenerator::emitGetStaticVar): (JSC::BytecodeGenerator::emitPutStaticVar): New functions, corresponding to the old emitGetScopedVar and emitPutScopedVar. (JSC::BytecodeGenerator::registerFor): Remove version that took an Identifier&; replaced by ResolveResult::local(). (JSC::BytecodeGenerator::emitResolve): (JSC::BytecodeGenerator::emitResolveBase): (JSC::BytecodeGenerator::emitResolveBaseForPut): (JSC::BytecodeGenerator::emitResolveWithBase): (JSC::BytecodeGenerator::emitResolveWithThis): Change to accept a "resolveResult" argument. This is more clear, and reduces the amount of double analysis happening at compile-time. * bytecompiler/NodesCodegen.cpp: (JSC::ResolveNode::emitBytecode): (JSC::EvalFunctionCallNode::emitBytecode): (JSC::FunctionCallResolveNode::emitBytecode): (JSC::PostfixResolveNode::emitBytecode): (JSC::DeleteResolveNode::emitBytecode): (JSC::TypeOfResolveNode::emitBytecode): (JSC::PrefixResolveNode::emitBytecode): (JSC::ReadModifyResolveNode::emitBytecode): (JSC::AssignResolveNode::emitBytecode): (JSC::ConstDeclNode::emitCodeSingle): (JSC::ForInNode::emitBytecode): Refactor to use the new ResolveResult structure. 2012-02-01 Csaba Osztrogonác <ossy@webkit.org> Implement Error.stack https://bugs.webkit.org/show_bug.cgi?id=66994 Unreviewed, rolling out r106407. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/AbstractPC.cpp: (JSC::AbstractPC::AbstractPC): * interpreter/Interpreter.cpp: (JSC::Interpreter::throwException): * interpreter/Interpreter.h: (JSC): (Interpreter): * jsc.cpp: (GlobalObject::finishCreation): * parser/Parser.h: (JSC::::parse): * runtime/CommonIdentifiers.h: * runtime/Error.cpp: (JSC::addErrorInfo): * runtime/Error.h: (JSC): 2012-01-31 Hajime Morrita <morrita@chromium.org> Add missing JS_PRIVATE_EXPORTs https://bugs.webkit.org/show_bug.cgi?id=77507 Reviewed by Kevin Ollivier. * heap/MarkedSpace.h: (MarkedSpace): * interpreter/Interpreter.h: (Interpreter): * runtime/JSValue.h: (JSValue): * wtf/text/AtomicString.h: (WTF::AtomicString::add): * wtf/text/WTFString.h: (WTF): 2012-01-31 Geoffrey Garen <ggaren@apple.com> Stop using -fomit-frame-pointer https://bugs.webkit.org/show_bug.cgi?id=77403 Reviewed by Filip Pizlo. JavaScriptCore is too fast. I'm just the man to fix it. * Configurations/JavaScriptCore.xcconfig: 2012-01-31 Michael Saboff <msaboff@apple.com> StringProtoFuncToUpperCase should call StringImpl::upper similar to StringProtoToLowerCase https://bugs.webkit.org/show_bug.cgi?id=76647 Reviewed by Darin Adler. Changed stringProtoFuncToUpperCase to call StringImpl::upper() in a manor similar to stringProtoFuncToLowerCase(). Fixed StringImpl::upper() to handle to special cases. One case is s-sharp (0xdf) which converts to "SS". The other case is for characters which become 16 bit values when converted to upper case. For those, we up convert the the source string and use the 16 bit path. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncToUpperCase): * wtf/text/StringImpl.cpp: (WTF::StringImpl::upper): * wtf/unicode/CharacterNames.h: (smallLetterSharpS): New constant 2012-01-31 Oliver Hunt <oliver@apple.com> Remove unneeded sourceId property https://bugs.webkit.org/show_bug.cgi?id=77495 Reviewed by Filip Pizlo. sourceId isn't used anymore, so we'll just remove it. * runtime/Error.cpp: (JSC): (JSC::addErrorInfo): (JSC::hasErrorInfo): 2012-01-31 Oliver Hunt <oliver@apple.com> Implement Error.stack https://bugs.webkit.org/show_bug.cgi?id=66994 Reviewed by Gavin Barraclough. Original patch by Juan Carlos Montemayor Elosua: This patch utilizes topCallFrame to create a stack trace when an error is thrown. Users will also be able to use the stack() command in jsc to get arrays with stack trace information. Modified to be correct on ToT, with a variety of correctness, performance, and security improvements. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::getCallerLine): (JSC::getSourceURLFromCallFrame): (JSC::getStackFrameCodeType): (JSC::Interpreter::getStackTrace): (JSC::Interpreter::throwException): * interpreter/Interpreter.h: (JSC::StackFrame::toString): * jsc.cpp: (GlobalObject::finishCreation): (functionJSCStack): * parser/Parser.h: (JSC::Parser::parse): * runtime/CommonIdentifiers.h: * runtime/Error.cpp: (JSC::addErrorInfo): * runtime/Error.h: 2012-01-31 Scott Graham <scottmg@chromium.org> [Chromium] Remove references to gyp cygwin build target https://bugs.webkit.org/show_bug.cgi?id=77253 Reviewed by Julien Chaffraix. Target dependency is no longer required, it's done earlier in the build process. * JavaScriptCore.gyp/JavaScriptCore.gyp: 2012-01-31 Michael Saboff <msaboff@apple.com> ASSERT(m_jumpsToLink.isEmpty()) failing in ARMv7Assembler dtor https://bugs.webkit.org/show_bug.cgi?id=77443 Reviewed by Gavin Barraclough. Removed failing ASSERT() and thus destructor. The ASSERT isn't needed. We are hitting it in the YARR JIT case where we bail out and go to the interpreter with a partially JIT'ed function. Since we haven't linked the JIT'ed code, there is likely to be some unresolved jumps in the vector when the ARMv7Assembler destructor is called. For the case where we complete the JIT process, we clear the vector at the end of LinkBuffer::linkCode (LinkBuffer.h:292). * assembler/ARMv7Assembler.h: (ARMv7Assembler): 2012-01-31 Anders Carlsson <andersca@apple.com> Vector<T>::operator== shouldn't require T to have operator!= https://bugs.webkit.org/show_bug.cgi?id=77448 Reviewed by Andreas Kling. Change VectorComparer::compare to use !(a == b) instead of a != b since it makes more sense for Vector::operator== to use the element's operator==. * wtf/Vector.h: 2012-01-30 Oliver Hunt <oliver@apple.com> get_by_val_arguments is broken in the interpreter https://bugs.webkit.org/show_bug.cgi?id=77389 Reviewed by Gavin Barraclough. When get_by_val had wad a value profile added, the same slot was not added to get_by_val_arguments. This broke the interpreter as the interpreter falls back on its regular get_by_val implementation. No tests are added as the interpreter is fairly broken in its current state (multiple tests fail due to this bug). * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): * bytecode/Opcode.h: (JSC): (): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitGetArgumentByVal): 2012-01-30 Oliver Hunt <oliver@apple.com> Unexpected syntax error https://bugs.webkit.org/show_bug.cgi?id=77340 Reviewed by Gavin Barraclough. Function calls and new expressions have the same semantics for assignment, so should simply share their lhs handling. * parser/Parser.cpp: (JSC::::parseMemberExpression): 2012-01-30 Gavin Barraclough <barraclough@apple.com> Unreviewed ARMv7 build fix. * tools/CodeProfiling.cpp: (JSC): (JSC::setProfileTimer): (JSC::CodeProfiling::begin): (JSC::CodeProfiling::end): 2012-01-30 David Levin <levin@chromium.org> Using OS(WIN) or OS(MAC) should cause a build error. https://bugs.webkit.org/show_bug.cgi?id=77162 Reviewed by Darin Adler. * wtf/Platform.h: Expand them into something that will cause a compile error. 2012-01-30 Yong Li <yoli@rim.com> [BlackBerry] OS(QNX) also has TM_GMTOFF, TM_ZONE, and TIMEGM https://bugs.webkit.org/show_bug.cgi?id=77360 Reviewed by Rob Buis. Turn on HAVE(TM_GMTOFF), HAVE(TM_ZONE), and HAVE(TIMEGM) for OS(QNX). * wtf/Platform.h: 2012-01-30 Gavin Barraclough <barraclough@apple.com> Speculative Windows build fix. * assembler/MacroAssemblerCodeRef.h: (FunctionPtr): 2012-01-30 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=77163 MacroAssemblerCodeRef.h uses OS(WIN) instead of OS(WINDOWS) Rubber stamped by Geoff Garen * assembler/MacroAssemblerCodeRef.h: 2012-01-30 Gavin Barraclough <barraclough@apple.com> Unreviewed build fix for interpreter builds. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::CodeBlock): * bytecode/CodeBlock.h: (CodeBlock): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * tools/CodeProfile.cpp: (JSC::CodeProfile::sample): 2012-01-30 Gavin Barraclough <barraclough@apple.com> Unreviewed build fix following bug#76855 * JavaScriptCore.exp: 2012-01-30 Michael Saboff <msaboff@apple.com> CaseFoldingHash::hash() doesn't handle 8 bit strings directly https://bugs.webkit.org/show_bug.cgi?id=76652 Reviewed by Andreas Kling. * wtf/text/StringHash.h: (WTF::CaseFoldingHash::hash): Added 8 bit string code path. 2012-01-30 Michael Saboff <msaboff@apple.com> stringProtoFuncReplace converts 8 bit strings to 16 bit during replacement https://bugs.webkit.org/show_bug.cgi?id=76651 Reviewed by Geoffrey Garen. Made local function substituteBackreferencesSlow a template function based on character width. Cleaned up getCharacters() in both UString and StringImpl. Changed getCharacters<UChar> to up convert an 8 bit string to 16 bits if necessary. * runtime/StringPrototype.cpp: (JSC::substituteBackreferencesSlow): (JSC::substituteBackreferences): * runtime/UString.h: (JSC::LChar): (JSC::UChar): * wtf/text/StringImpl.h: (WTF::UChar): 2012-01-30 Gavin Barraclough <barraclough@apple.com> Clean up putDirect https://bugs.webkit.org/show_bug.cgi?id=76232 Reviewed by Sam Weinig. Part 3 - merge op_put_getter & op_put_setter. Putting these separately is inefficient (and makes future optimiation, e.g. making GetterSetter immutable) harder. Change to emit a single op_put_getter_setter bytecode op. Ultimately we should probably be able to merge this with put direct, to create a common op to initialize object literal properties. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): * bytecode/Opcode.h: (JSC): (): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitPutGetterSetter): * bytecompiler/BytecodeGenerator.h: (BytecodeGenerator): * bytecompiler/NodesCodegen.cpp: (JSC::PropertyListNode::emitBytecode): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): * jit/JIT.h: (JIT): * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_put_getter_setter): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_put_getter_setter): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jit/JITStubs.h: (): * runtime/JSObject.cpp: (JSC::JSObject::putDirectVirtual): (JSC::JSObject::putDirectAccessor): (JSC): (JSC::putDescriptor): (JSC::JSObject::defineOwnProperty): * runtime/JSObject.h: (): (JSC::JSObject::putDirectInternal): (JSC::JSObject::putDirect): (JSC::JSObject::putDirectWithoutTransition): 2012-01-30 Michael Saboff <msaboff@apple.com> Dromaeo tests call parseSimpleLengthValue() on 8 bit strings https://bugs.webkit.org/show_bug.cgi?id=76649 Reviewed by Geoffrey Garen. * JavaScriptCore.exp: Added export for charactersToDouble. 2012-01-30 Michael Saboff <msaboff@apple.com> WebCore decodeEscapeSequences unnecessarily converts 8 bit strings to 16 bit when decoding. https://bugs.webkit.org/show_bug.cgi?id=76648 Reviewed by Geoffrey Garen. Added a new overloaded append member that takes a String& argument, an offest and a length to do direct sub string appending to a StringBuilder. * wtf/text/StringBuilder.h: (WTF::StringBuilder::append): 2012-01-29 Zoltan Herczeg <zherczeg@webkit.org> Custom written CSS lexer https://bugs.webkit.org/show_bug.cgi?id=70107 Reviewed by Antti Koivisto and Oliver Hunt. Add new helper functions for the custom written CSS lexer. * wtf/ASCIICType.h: (WTF::toASCIILowerUnchecked): (WTF): (WTF::isASCIIAlphaCaselessEqual): 2012-01-29 Filip Pizlo <fpizlo@apple.com> REGRESSION (r105576-r105582): Web Inspector Crash in JSC::JSValue::toString(JSC::ExecState*) const https://bugs.webkit.org/show_bug.cgi?id=77146 <rdar://problem/10770586> Reviewed by Oliver Hunt. The old JIT expects that the result of the last operation is in the lastResultRegister. The DFG JIT is designed to correctly track the lastResultRegister by looking at SetLocal nodes. However, when the DFG JIT inlines a code block, it forgets that the inlined code block's result would have been placed in the lastResultRegister. Hence if we OSR exit on the first node following the end of an inlined code block that had a return value, and that first node uses the return value, the old JIT will get massively confused. This patch takes a surgical approach: instead of making the DFG smarter, it makes the old JIT slightly dumber. * jit/JITCall.cpp: (JSC::JIT::emit_op_call_put_result): 2012-01-29 Filip Pizlo <fpizlo@apple.com> Build fix for Mac non-x64 platforms. * tools/CodeProfiling.cpp: (JSC): 2012-01-28 Gavin Barraclough <barraclough@apple.com> Reserve 'let' https://bugs.webkit.org/show_bug.cgi?id=77293 Rubber stamped by Oliver Hunt. 'let' may become a keyword in ES6. We're going to try experimentally reserving it, to see if this breaks the web. * parser/Keywords.table: 2012-01-27 Gavin Barraclough <barraclough@apple.com> Implement a JIT-code aware sampling profiler for JSC https://bugs.webkit.org/show_bug.cgi?id=76855 Reviewed by Oliver Hunt. To enable the profiler, set the JSC_CODE_PROFILING environment variable to 1 (no tracing the C stack), 2 (trace one level of C code) or 3 (recursively trace all samples). The profiler requires -fomit-frame-pointer to be removed from the build flags. * JavaScriptCore.exp: - Removed an export. * JavaScriptCore.xcodeproj/project.pbxproj: - Added new files * bytecode/CodeBlock.cpp: - For baseline codeblocks, cache the result of canCompileWithDFG. * bytecode/CodeBlock.h: - For baseline codeblocks, cache the result of canCompileWithDFG. * jit/ExecutableAllocator.cpp: (JSC::ExecutableAllocator::initializeAllocator): - Notify the profiler when the allocator is created. (JSC::ExecutableAllocator::allocate): - Inform the allocated of the ownerUID. * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::ExecutableAllocator::initializeAllocator): - Notify the profiler when the allocator is created. (JSC::ExecutableAllocator::allocate): - Inform the allocated of the ownerUID. * jit/JITStubs.cpp: - If profiling, don't mask the return address in JIT code. (We do so to provide nicer backtraces in debug builds). * runtime/Completion.cpp: (JSC::evaluate): - Notify the profiler of script evaluations. * tools: Added. * tools/CodeProfile.cpp: Added. (JSC::symbolName): - Helper function to get the name of a symbol in the framework. (JSC::truncateTrace): - Helper to truncate traces into methods know to have uninformatively deep stacks. (JSC::CodeProfile::sample): - Record a stack trace classifying samples. (JSC::CodeProfile::report): - {Print profiler output. * tools/CodeProfile.h: Added. - new class, captures a set of samples associated with an evaluated script, and nested to record samples from subscripts. * tools/CodeProfiling.cpp: Added. (JSC::CodeProfiling::profilingTimer): - callback fired then a timer event occurs. (JSC::CodeProfiling::notifyAllocator): - called when the executable allocator is constructed. (JSC::CodeProfiling::getOwnerUIDForPC): - helper to lookup the codeblock from an address in JIT code (JSC::CodeProfiling::begin): - enter a profiling scope. (JSC::CodeProfiling::end): - exit a profiling scope. * tools/CodeProfiling.h: Added. - new class, instantialed from Completion to define a profiling scope. * tools/ProfileTreeNode.h: Added. - new class, used to construct a tree of samples. * tools/TieredMMapArray.h: Added. - new class, a malloc-free vector (can be used while the main thread is suspended, possibly holding the malloc heap lock). * wtf/MetaAllocator.cpp: (WTF::MetaAllocatorHandle::MetaAllocatorHandle): (WTF::MetaAllocator::allocate): - Allow allocation handles to track information about their owner. * wtf/MetaAllocator.h: (MetaAllocator): - Allow allocation handles to track information about their owner. * wtf/MetaAllocatorHandle.h: (MetaAllocatorHandle): (WTF::MetaAllocatorHandle::ownerUID): - Allow allocation handles to track information about their owner. * wtf/OSAllocator.h: (WTF::OSAllocator::reallocateCommitted): - reallocate an existing, committed memory allocation. 2012-01-28 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r106187. http://trac.webkit.org/changeset/106187 https://bugs.webkit.org/show_bug.cgi?id=77276 The last rollout was a false charge. (Requested by morrita on #webkit). * runtime/ExceptionHelpers.h: (InterruptedExecutionError): * runtime/JSBoundFunction.h: (JSBoundFunction): * runtime/RegExp.h: (RegExp): * runtime/RegExpMatchesArray.h: (RegExpMatchesArray): 2012-01-28 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r106151. http://trac.webkit.org/changeset/106151 https://bugs.webkit.org/show_bug.cgi?id=77275 may break windows build (Requested by morrita on #webkit). * runtime/ExceptionHelpers.h: (InterruptedExecutionError): * runtime/JSBoundFunction.h: (JSBoundFunction): * runtime/RegExp.h: (RegExp): * runtime/RegExpMatchesArray.h: (RegExpMatchesArray): 2012-01-28 Filip Pizlo <fpizlo@apple.com> GC invoked while doing an old JIT property storage reallocation may lead to an object that refers to a dead structure https://bugs.webkit.org/show_bug.cgi?id=77273 <rdar://problem/10770565> Reviewed by Gavin Barraclough. The put_by_id transition was already saving the old structure by virtue of having the object on the stack, so that wasn't going to get deleted. But the new structure was unprotected in the transition. I've now changed the transition code to save the new structure, ensuring that the GC will know it to be marked if invoked from within put_by_id_transition_realloc. * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompilePutByIdTransition): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::privateCompilePutByIdTransition): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jit/JITStubs.h: (JSC): (): 2012-01-27 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r106167. http://trac.webkit.org/changeset/106167 https://bugs.webkit.org/show_bug.cgi?id=77264 broke LayoutTests/fast/js/string-capitalization.html (Requested by msaboff on #webkit). * runtime/StringPrototype.cpp: (JSC::stringProtoFuncToLowerCase): (JSC::stringProtoFuncToUpperCase): * wtf/text/StringImpl.cpp: (WTF::StringImpl::upper): 2012-01-27 Filip Pizlo <fpizlo@apple.com> Build fix for interpreter platforms. * interpreter/AbstractPC.cpp: (JSC::AbstractPC::AbstractPC): 2012-01-27 Michael Saboff <msaboff@apple.com> StringProtoFuncToUpperCase should call StringImpl::upper similar to StringProtoToLowerCase https://bugs.webkit.org/show_bug.cgi?id=76647 Reviewed by Geoffrey Garen. Changed stringProtoFuncToUpperCase to call StringImpl::upper() is a manor similar to stringProtoFuncToLowerCase(). Fixed StringImpl::upper() to handle the two 8 bit characters that when converted to upper case become 16 bit characters. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncToLowerCase): Removed extra trailing whitespace. (JSC::stringProtoFuncToUpperCase): * wtf/text/StringImpl.cpp: (WTF::StringImpl::upper): 2012-01-27 Hajime Morita <morrita@google.com> [JSC] ThunkGenerators.cpp should hide its asm-defined symbols https://bugs.webkit.org/show_bug.cgi?id=77244 Reviewed by Filip Pizlo. * jit/ThunkGenerators.cpp: Added HIDE_SYMBOLS() * wtf/InlineASM.h: Moved some duplicated macros from ThunkGenerators.cpp 2012-01-27 Simon Hausmann <simon.hausmann@nokia.com> [JSC] Asm-originated symbols should be marked as hidden https://bugs.webkit.org/show_bug.cgi?id=77150 Reviewed by Filip Pizlo. * dfg/DFGOperations.cpp: The HIDE_SYMBOLS macros were present in the CPU(ARM) preprocessor branches, but they were missing in the CPU(X86) and the CPU(X86_64) cases. 2012-01-27 MORITA Hajime <morrita@google.com> [JSC] Some JS_EXPORTDATA may not be necessary. https://bugs.webkit.org/show_bug.cgi?id=77145 Reviewed by Darin Adler. Removed JS_EXPORTDATA attributes whose attributing symbols are not exported on Mac port. * runtime/ExceptionHelpers.h: (InterruptedExecutionError): * runtime/JSBoundFunction.h: (JSBoundFunction): * runtime/RegExp.h: (RegExp): * runtime/RegExpMatchesArray.h: (RegExpMatchesArray): 2012-01-27 MORITA Hajime <morrita@google.com> [WTF] WTFString.h has some extra JS_EXPORT_PRIVATEs https://bugs.webkit.org/show_bug.cgi?id=77113 Reviewed by Darin Adler. * wtf/text/WTFString.h: Removed some WTF_EXPORT_PRIVATE attributes which we don't need to export. 2012-01-27 Zeno Albisser <zeno@webkit.org> [Qt][Mac] Build fails after adding ICU support (r105997). https://bugs.webkit.org/show_bug.cgi?id=77118 Use Apple code path for unicode date formats on mac. Reviewed by Tor Arne Vestbø. * runtime/DatePrototype.cpp: (): 2012-01-27 Carlos Garcia Campos <cgarcia@igalia.com> [GTK] Add a GKeyFile especialization to GOwnPtr https://bugs.webkit.org/show_bug.cgi?id=77191 Reviewed by Martin Robinson. * wtf/gobject/GOwnPtr.cpp: (WTF::GKeyFile): Implement freeOwnedGPtr for GKeyFile. * wtf/gobject/GOwnPtr.h: Add GKeyFile template. * wtf/gobject/GTypedefs.h: Add forward declaration for GKeyFile. 2012-01-25 Yury Semikhatsky <yurys@chromium.org> Web Inspector: should be possible to open function declaration from script popover https://bugs.webkit.org/show_bug.cgi?id=76913 Added display function name and source location to the popover in scripts panel. Now when a function is hovered user can navigate to its definition. Reviewed by Pavel Feldman. * JavaScriptCore/JavaScriptCore.exp * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSFunction.h: (JSFunction): 2012-01-26 Kevin Ollivier <kevino@theolliviers.com> [wx] Unreviewed. Build fix, wx uses the Mac ICU headers so we must match Mac behavior. * runtime/DatePrototype.cpp: (): 2012-01-26 Mark Hahnenberg <mhahnenberg@apple.com> Merge AllocationSpace into MarkedSpace https://bugs.webkit.org/show_bug.cgi?id=77116 Reviewed by Geoffrey Garen. Merging AllocationSpace and MarkedSpace in preparation for future refactoring/enhancement to MarkedSpace allocation. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * heap/AllocationSpace.cpp: Removed. * heap/AllocationSpace.h: Removed. * heap/BumpSpace.h: (BumpSpace): * heap/Heap.h: (JSC::Heap::objectSpace): (Heap): (): * heap/HeapBlock.h: (): * heap/MarkedSpace.cpp: (JSC::MarkedSpace::tryAllocateHelper): (JSC): (JSC::MarkedSpace::tryAllocate): (JSC::MarkedSpace::allocateSlowCase): (JSC::MarkedSpace::allocateBlock): (JSC::MarkedSpace::freeBlocks): (TakeIfUnmarked): (JSC::TakeIfUnmarked::TakeIfUnmarked): (JSC::TakeIfUnmarked::operator()): (JSC::TakeIfUnmarked::returnValue): (JSC::MarkedSpace::shrink): (GatherDirtyCells): (JSC::GatherDirtyCells::returnValue): (JSC::GatherDirtyCells::GatherDirtyCells): (JSC::GatherDirtyCells::operator()): (JSC::MarkedSpace::gatherDirtyCells): * heap/MarkedSpace.h: (MarkedSpace): (JSC::MarkedSpace::blocks): (JSC::MarkedSpace::forEachCell): (JSC): (JSC::MarkedSpace::allocate): 2012-01-26 Oliver Hunt <oliver@apple.com> MSVC bug fix. <rdar://problem/10703671> MSVC generates bad code for enum compare. RS=Geoff Make bitfield large enough to work around MSVC's desire to make enums signed types. * bytecode/CallLinkInfo.h: (CallLinkInfo): 2012-01-26 Filip Pizlo <fpizlo@apple.com> All DFG helpers that may call out to arbitrary JS code must know where they were called from due to inlining and call stack walking https://bugs.webkit.org/show_bug.cgi?id=77070 <rdar://problem/10750834> Reviewed by Geoff Garen. Changed the DFG to always record a code origin index in the tag of the argument count (which we previously left blank for the benefit of LLInt, but is still otherwise unused by the DFG), so that if we ever need to walk the stack accurately we know where to start. In particular, if the current ExecState* points several semantic call frames away from the true semantic call frame because we had performed inlining, having the code origin index recorded means that we can reify those call frames as necessary to give runtime/library code an accurate view of the current JS state. This required several large but mechanical changes: - Calling a function from the DFG now plants a store32 instruction to store the code origin index. But the indices of code origins were previously picked by the DFG::JITCompiler after code generation completed. I changed this somewhat; even though the code origins are put into the CodeBlock after code gen, the code gen now knows a priori what their indices will be. Extensive assertions are in place to ensure that the two don't get out of sync, in the form of the DFG::CallBeginToken. Note that this mechanism has almost no effect on JS calls; those don't need the code origin index set in the call frame because we can get it by doing a binary search on the return PC. - Stack walking now always calls trueCallFrame() first before beginning the walk, since even the top call frame may be wrong. It still calls trueCallerFrame() as before to get to the next frame, though trueCallerFrame() is now mostly a wrapper around callerFrame()->trueCallFrame(). - Because the mechanism for getting the code origin of a call frame is bimodal (either the call frame knows its code origin because the code origin index was set, or it's necessary to use the callee frame's return PC), I put in extra mechanisms to determine whether your caller, or your callee, corresponds to a call out of C++ code. Previously we just had the host call flag, but this is insufficient as it does not cover the case of someone calling JSC::call(). But luckily we can determine this just by looking at the return PC: if the return PC is in range of the ctiTrampiline, then two things are true: this call frame's PC will tell you nothing about where you came from in your caller, and the caller already knows where it's at because it must have set the code origin index (unless it's not DFG code, in which case we don't care because there is no inlining to worry about). - During testing this revealed a simple off-by-one goof in DFG::ByteCodeParser's inlining code, so I fixed it. - Finally because I was tired of doing random #if's for checking if I should be passing around an Instruction* or a ReturnAddressPtr, I created a class called AbstractPC that holds whatever notion of a PC is appropriate for the current execution environment. It's designed to work gracefully even if both the interpreter and the JIT are compiled in, and should integrate nicely with the LLInt. This is neutral on all benchmarks and fixes some nasty corner-case regressions of evil code that uses combinations of getters/setters and function.arguments. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * bytecode/CodeBlock.h: (JSC::CodeBlock::codeOrigin): (CodeBlock): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleInlining): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: (CallBeginToken): (JSC::DFG::CallBeginToken::CallBeginToken): (JSC::DFG::CallBeginToken::assertCodeOriginIndex): (JSC::DFG::CallBeginToken::assertNoCodeOriginIndex): (DFG): (JSC::DFG::CallExceptionRecord::CallExceptionRecord): (CallExceptionRecord): (JSC::DFG::JITCompiler::JITCompiler): (JITCompiler): (JSC::DFG::JITCompiler::nextCallBeginToken): (JSC::DFG::JITCompiler::beginCall): (JSC::DFG::JITCompiler::notifyCall): (JSC::DFG::JITCompiler::addExceptionCheck): (JSC::DFG::JITCompiler::addFastExceptionCheck): * dfg/DFGOperations.cpp: (): * dfg/DFGRepatch.cpp: (JSC::DFG::tryBuildGetByIDList): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::appendCallWithExceptionCheck): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): * interpreter/AbstractPC.cpp: Added. (JSC): (JSC::AbstractPC::AbstractPC): * interpreter/AbstractPC.h: Added. (JSC): (AbstractPC): (JSC::AbstractPC::AbstractPC): (JSC::AbstractPC::hasJITReturnAddress): (JSC::AbstractPC::jitReturnAddress): (JSC::AbstractPC::hasInterpreterReturnAddress): (JSC::AbstractPC::interpreterReturnAddress): (JSC::AbstractPC::isSet): (JSC::AbstractPC::operator!): (): * interpreter/CallFrame.cpp: (JSC): (JSC::CallFrame::trueCallFrame): (JSC::CallFrame::trueCallerFrame): * interpreter/CallFrame.h: (JSC::ExecState::abstractReturnPC): (JSC::ExecState::codeOriginIndexForDFGWithInlining): (ExecState): (JSC::ExecState::trueCallFrame): (JSC::ExecState::trueCallFrameFromVMCode): * interpreter/Interpreter.cpp: (JSC::Interpreter::retrieveArgumentsFromVMCode): (JSC::Interpreter::retrieveCallerFromVMCode): (JSC::Interpreter::findFunctionCallFrameFromVMCode): * interpreter/Interpreter.h: (Interpreter): (): * jit/JITStubs.cpp: (JSC): (): * jit/JITStubs.h: (JSC): (JSC::returnAddressIsInCtiTrampoline): * runtime/JSFunction.cpp: (JSC::JSFunction::argumentsGetter): (JSC::JSFunction::callerGetter): (JSC::JSFunction::getOwnPropertyDescriptor): 2012-01-26 Peter Varga <pvarga@webkit.org> Fix build when VERBOSE_SPECULATION_FAILURE is enabled in DFG https://bugs.webkit.org/show_bug.cgi?id=77104 Reviewed by Filip Pizlo. * dfg/DFGOperations.cpp: (): 2012-01-26 Michael Saboff <msaboff@apple.com> String::latin1() should take advantage of 8 bit strings https://bugs.webkit.org/show_bug.cgi?id=76646 Reviewed by Geoffrey Garen. * wtf/text/WTFString.cpp: (WTF::String::latin1): For 8 bit strings, use existing buffer without conversion. 2012-01-26 Michael Saboff <msaboff@apple.com> Dromaeo tests usage of StringImpl find routines cause 8->16 bit conversions https://bugs.webkit.org/show_bug.cgi?id=76645 Reviewed by Geoffrey Garen. * wtf/text/StringImpl.cpp: (WTF::equalIgnoringCase): New LChar version. (WTF::findInner): New helper function. (WTF::StringImpl::find): Added 8 bit path. (WTF::reverseFindInner): New helper funciton. (WTF::StringImpl::reverseFind): Added 8 bit path. (WTF::StringImpl::reverseFindIgnoringCase): Added 8 bit path. * wtf/text/StringImpl.h: (WTF): 2012-01-26 Csaba Osztrogonác <ossy@webkit.org> [Qt][Win] One more speculative buildfix after r105970. * JavaScriptCore.pri: 2012-01-26 Csaba Osztrogonác <ossy@webkit.org> [Qt][Win] Speculative buildfix after r105970. * JavaScriptCore.pri: Link lgdi for DeleteObject() and DeleteDC(). 2012-01-26 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r105982. http://trac.webkit.org/changeset/105982 https://bugs.webkit.org/show_bug.cgi?id=77090 breaks the world (Requested by WildFox on #webkit). * wtf/MainThread.cpp: (WTF): * wtf/Platform.h: * wtf/mac/MainThreadMac.mm: (WTF): (WTF::registerGCThread): (WTF::isMainThreadOrGCThread): 2012-01-26 Roland Takacs <takacs.roland@stud.u-szeged.hu> [Qt] GC should be parallel on Qt platform https://bugs.webkit.org/show_bug.cgi?id=73309 Reviewed by Zoltan Herczeg. These changes made the parallel gc feature available for Qt port. The implementation of "registerGCThread" and "isMainThreadOrGCThread" is moved from MainThreadMac.mm to the common MainThread.cpp to make them available for other platforms. Measurement results: V8 speed-up: 1.071x as fast [From: 746.1ms To: 696.4ms ] WindScorpion speed-up: 1.082x as fast [From: 3490.4ms To: 3226.7ms] V8 Splay speed-up: 1.158x as fast [From: 145.8ms To: 125.9ms ] Tested on Intel(R) Core(TM) i5-2320 CPU @ 3.00GHz with 4-core. * wtf/MainThread.cpp: (WTF): (WTF::registerGCThread): (WTF::isMainThreadOrGCThread): * wtf/Platform.h: * wtf/mac/MainThreadMac.mm: 2012-01-26 Andy Estes <aestes@apple.com> REGRESSION (r105555): Incorrect use of OS() macro breaks OwnPtr when used with Win32 data types https://bugs.webkit.org/show_bug.cgi?id=77073 Reviewed by Ryosuke Niwa. r105555 changed PLATFORM(WIN) to OS(WIN), but WTF_OS_WIN isn't defined. This should have been changed to OS(WINDOWS). This causes the preprocessor to strip out Win32 data type overrides for deleteOwnedPtr, causing allocations made by Win32 to be deleted by fastmalloc. * wtf/OwnPtrCommon.h: (WTF): Use OS(WINDOWS) instead of OS(WIN). 2012-01-25 Mark Rowe <mrowe@apple.com> Attempted Mac build fix after r105939. * runtime/DatePrototype.cpp: Don't #include unicode/udat.h on Mac or iOS. It isn't used on these platforms and isn't available in the ICU headers for Mac. 2012-01-25 Mark Rowe <mrowe@apple.com> Build in to an alternate location when USE_STAGING_INSTALL_PATH is set. <rdar://problem/10609417> Adopt USE_STAGING_INSTALL_PATH Reviewed by David Kilzer. * Configurations/Base.xcconfig: Define NORMAL_JAVASCRIPTCORE_FRAMEWORKS_DIR, which contains the path where JavaScriptCore is normally installed. Update JAVASCRIPTCORE_FRAMEWORKS_DIR to point to the staged frameworks directory when USE_STAGING_INSTALL_PATH is set. * Configurations/JavaScriptCore.xcconfig: Always set the framework's install name based on the normal framework location. This prevents an incorrect install name from being used when installing in to the staged frameworks directory. 2012-01-25 Eli Fidler <efidler@rim.com> Implement Date.toLocaleString() using ICU https://bugs.webkit.org/show_bug.cgi?id=76714 Reviewed by Darin Adler. * runtime/DatePrototype.cpp: (JSC::formatLocaleDate): 2012-01-25 Hajime Morita <morrita@google.com> ENABLE_SHADOW_DOM should be available via build-webkit --shadow-dom https://bugs.webkit.org/show_bug.cgi?id=76863 Reviewed by Dimitri Glazkov. Added a feature flag. * Configurations/FeatureDefines.xcconfig: 2012-01-25 Yong Li <yoli@rim.com> [BlackBerry] Implement OSAllocator::commit/decommit. BlackBerry port should support virtual memory decommiting. https://bugs.webkit.org/show_bug.cgi?id=77013 Reviewed by Rob Buis. * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveUncommitted): (WTF::OSAllocator::commit): (WTF::OSAllocator::decommit): * wtf/Platform.h: 2012-01-24 Oliver Hunt <oliver@apple.com> Make DFG update topCallFrame https://bugs.webkit.org/show_bug.cgi?id=76969 Reviewed by Filip Pizlo. Add NativeCallFrameTracer to manage topCallFrame assignment in the DFG operations, and make use of it. * dfg/DFGOperations.cpp: (JSC::DFG::operationPutByValInternal): (): * interpreter/Interpreter.h: (JSC): (NativeCallFrameTracer): (JSC::NativeCallFrameTracer::NativeCallFrameTracer): 2012-01-24 Filip Pizlo <fpizlo@apple.com> Inlining breaks call frame walking when the walking is done from outside the inlinee, but inside a code block that had inlining https://bugs.webkit.org/show_bug.cgi?id=76978 <rdar://problem/10720904> Reviewed by Oliver Hunt. * bytecode/CodeBlock.h: (JSC::CodeBlock::codeOriginForReturn): * interpreter/CallFrame.cpp: (JSC::CallFrame::trueCallerFrame): 2012-01-24 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=76855 Implement a JIT-code aware sampling profiler for JSC Reviewed by Oliver Hunt. Add support to MetaAllocator.cpp to track all live handles in a map, allowing lookup based on any address within the allocation. * wtf/MetaAllocator.cpp: (WTF::MetaAllocatorTracker::notify): (WTF::MetaAllocatorTracker::release): - Track live handle objects in a map. (WTF::MetaAllocator::release): - Removed support for handles with null m_allocator (no longer used). - Notify the tracker of handles being released. (WTF::MetaAllocatorHandle::~MetaAllocatorHandle): - Moved functionality out into MetaAllocator::release. (WTF::MetaAllocatorHandle::shrink): - Removed support for handles with null m_allocator (no longer used). (WTF::MetaAllocator::MetaAllocator): - Initialize m_tracker. (WTF::MetaAllocator::allocate): - Notify the tracker of new allocations. * wtf/MetaAllocator.h: (WTF::MetaAllocatorTracker::find): - Lookup a MetaAllocatorHandle based on an address inside the allocation. (WTF::MetaAllocator::trackAllocations): - Register a callback object to track allocation state. * wtf/MetaAllocatorHandle.h: - Remove unused createSelfManagedHandle/constructor. (WTF::MetaAllocatorHandle::key): - Added, for use in RedBlackTree. 2012-01-24 Mark Hahnenberg <mhahnenberg@apple.com> Use copying collector for out-of-line JSObject property storage https://bugs.webkit.org/show_bug.cgi?id=76665 Reviewed by Geoffrey Garen. * runtime/JSObject.cpp: (JSC::JSObject::visitChildren): Changed to use copyAndAppend whenever the property storage is out-of-line. Also added a temporary variable to avoid warnings from GCC. (JSC::JSObject::allocatePropertyStorage): Changed to use tryAllocateStorage/tryReallocateStorage as opposed to operator new. Also added a temporary variable to avoid warnings from GCC. * runtime/JSObject.h: 2012-01-24 Geoffrey Garen <ggaren@apple.com> JSValue::toString() should return a JSString* instead of a UString https://bugs.webkit.org/show_bug.cgi?id=76861 Fixed two failing layout tests after my last patch. Reviewed by Gavin Barraclough. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncSort): Call value() after calling toString(), as in all other cases. I missed this case because the JSString* type has a valid operator<, so the compiler didn't complain. 2012-01-24 Kenichi Ishibashi <bashi@chromium.org> [V8] Add Uint8ClampedArray support https://bugs.webkit.org/show_bug.cgi?id=76803 Reviewed by Kenneth Russell. * wtf/ArrayBufferView.h: (WTF::ArrayBufferView::isUnsignedByteClampedArray): Added. * wtf/Uint8ClampedArray.h: (WTF::Uint8ClampedArray::isUnsignedByteClampedArray): Overridden to return true. 2012-01-23 Carlos Garcia Campos <cgarcia@igalia.com> [GTK] Add WebKitDownload to WebKit2 GTK+ API https://bugs.webkit.org/show_bug.cgi?id=72949 Reviewed by Martin Robinson. * wtf/gobject/GOwnPtr.cpp: (WTF::GTimer): Use g_timer_destroy() to free a GTimer. * wtf/gobject/GOwnPtr.h: Add GTimer template. * wtf/gobject/GTypedefs.h: Add GTimer forward declaration. 2012-01-24 Ilya Tikhonovsky <loislo@chromium.org> Unreviewed build fix for Qt LinuxSH4 build after r105698. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): 2012-01-23 Geoffrey Garen <ggaren@apple.com> JSValue::toString() should return a JSString* instead of a UString https://bugs.webkit.org/show_bug.cgi?id=76861 Reviewed by Gavin Barraclough. This makes the common case -- toString() on a string -- faster and inline-able. (Not a measureable speedup, but we can now remove a bunch of duplicate hand-rolled code for this optimization.) This also clarifies the boundary between "C++ strings" and "JS strings". In all cases other than true, false, null, undefined, and multi-digit numbers, the JS runtime was just retrieving a UString from a JSString, so returning a JSString* is strictly better. In the other cases, we can optimize to avoid creating a new JSString if we care to, but it doesn't seem to be a big deal. * JavaScriptCore.exp: Export! * jsc.cpp: (functionPrint): (functionDebug): (functionRun): (functionLoad): (functionCheckSyntax): (runWithScripts): (runInteractive): * API/JSValueRef.cpp: (JSValueToStringCopy): * bytecode/CodeBlock.cpp: (JSC::valueToSourceString): Call value() after calling toString(), to convert from "JS string" (JSString*) to "C++ string" (UString), since toString() no longer returns a "C++ string". * dfg/DFGOperations.cpp: (JSC::DFG::operationValueAddNotNumber): * jit/JITStubs.cpp: (op_add): Updated for removal of toPrimitiveString(): all '+' operands can use toString(), except for object operands, which need to take a slow path to call toPrimitive(). * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncToString): (JSC::arrayProtoFuncToLocaleString): (JSC::arrayProtoFuncJoin): (JSC::arrayProtoFuncPush): * runtime/CommonSlowPaths.h: (JSC::CommonSlowPaths::opIn): * runtime/DateConstructor.cpp: (JSC::dateParse): * runtime/DatePrototype.cpp: (JSC::formatLocaleDate): Call value() after calling toString(), as above. * runtime/ErrorInstance.h: (JSC::ErrorInstance::create): Simplified down to one canonical create() function, to make string handling easier. * runtime/ErrorPrototype.cpp: (JSC::errorProtoFuncToString): * runtime/ExceptionHelpers.cpp: (JSC::createInvalidParamError): (JSC::createNotAConstructorError): (JSC::createNotAFunctionError): (JSC::createNotAnObjectError): * runtime/FunctionConstructor.cpp: (JSC::constructFunctionSkippingEvalEnabledCheck): * runtime/FunctionPrototype.cpp: (JSC::functionProtoFuncBind): * runtime/JSArray.cpp: (JSC::JSArray::sort): Call value() after calling toString(), as above. * runtime/JSCell.cpp: * runtime/JSCell.h: Removed JSCell::toString() because JSValue does this job now. Doing it in JSCell is slower (requires extra type checking), and creates the misimpression that language-defined toString() behavior is an implementation detail of JSCell. * runtime/JSGlobalObjectFunctions.cpp: (JSC::encode): (JSC::decode): (JSC::globalFuncEval): (JSC::globalFuncParseInt): (JSC::globalFuncParseFloat): (JSC::globalFuncEscape): (JSC::globalFuncUnescape): Call value() after calling toString(), as above. * runtime/JSONObject.cpp: (JSC::unwrapBoxedPrimitive): (JSC::Stringifier::Stringifier): (JSC::JSONProtoFuncParse): Removed some manual optimization that toString() takes care of. * runtime/JSObject.cpp: (JSC::JSObject::toString): * runtime/JSObject.h: Updated to return JSString*. * runtime/JSString.cpp: * runtime/JSString.h: (JSC::JSValue::toString): Removed, since I removed JSCell::toString(). * runtime/JSValue.cpp: (JSC::JSValue::toStringSlowCase): Removed toPrimitiveString(), and re- spawned toStringSlowCase() from its zombie corpse, since toPrimitiveString() basically did what we want all the time. (Note that the toPrimitive() preference changes from NoPreference to PreferString, because that's how ToString is defined in the language. op_add does not want this behavior.) * runtime/NumberPrototype.cpp: (JSC::numberProtoFuncToString): (JSC::numberProtoFuncToLocaleString): A little simpler, now that toString() returns a JSString*. * runtime/ObjectConstructor.cpp: (JSC::objectConstructorGetOwnPropertyDescriptor): (JSC::objectConstructorDefineProperty): * runtime/ObjectPrototype.cpp: (JSC::objectProtoFuncHasOwnProperty): (JSC::objectProtoFuncDefineGetter): (JSC::objectProtoFuncDefineSetter): (JSC::objectProtoFuncLookupGetter): (JSC::objectProtoFuncLookupSetter): (JSC::objectProtoFuncPropertyIsEnumerable): More calls to value(), as above. * runtime/Operations.cpp: (JSC::jsAddSlowCase): Need to check for object before taking the toString() fast path becuase adding an object to a string requires calling toPrimitive() on the object, not toString(). (They differ in their preferred conversion type.) * runtime/Operations.h: (JSC::jsString): (JSC::jsStringFromArguments): This code gets simpler, now that toString() does the right thing. (JSC::jsAdd): Now checks for object, just like jsAddSlowCase(). * runtime/RegExpConstructor.cpp: (JSC::setRegExpConstructorInput): (JSC::constructRegExp): * runtime/RegExpObject.cpp: (JSC::RegExpObject::match): * runtime/RegExpPrototype.cpp: (JSC::regExpProtoFuncCompile): (JSC::regExpProtoFuncToString): More calls to value(), as above. * runtime/StringConstructor.cpp: (JSC::constructWithStringConstructor): (JSC::callStringConstructor): This code gets simpler, now that toString() does the right thing. * runtime/StringPrototype.cpp: (JSC::replaceUsingRegExpSearch): (JSC::replaceUsingStringSearch): (JSC::stringProtoFuncReplace): (JSC::stringProtoFuncCharAt): (JSC::stringProtoFuncCharCodeAt): (JSC::stringProtoFuncConcat): (JSC::stringProtoFuncIndexOf): (JSC::stringProtoFuncLastIndexOf): (JSC::stringProtoFuncMatch): (JSC::stringProtoFuncSearch): (JSC::stringProtoFuncSlice): (JSC::stringProtoFuncSplit): (JSC::stringProtoFuncSubstr): (JSC::stringProtoFuncSubstring): (JSC::stringProtoFuncToLowerCase): (JSC::stringProtoFuncToUpperCase): (JSC::stringProtoFuncLocaleCompare): (JSC::stringProtoFuncBig): (JSC::stringProtoFuncSmall): (JSC::stringProtoFuncBlink): (JSC::stringProtoFuncBold): (JSC::stringProtoFuncFixed): (JSC::stringProtoFuncItalics): (JSC::stringProtoFuncStrike): (JSC::stringProtoFuncSub): (JSC::stringProtoFuncSup): (JSC::stringProtoFuncFontcolor): (JSC::stringProtoFuncFontsize): (JSC::stringProtoFuncAnchor): (JSC::stringProtoFuncLink): (JSC::trimString): Some of this code gets simpler, now that toString() does the right thing. More calls to value(), as above. 2012-01-23 Luke Macpherson <macpherson@chromium.org> Unreviewed, rolling out r105676. http://trac.webkit.org/changeset/105676 https://bugs.webkit.org/show_bug.cgi?id=76665 Breaks build on max due to compile warnings. * runtime/JSObject.cpp: (JSC::JSObject::finalize): (JSC::JSObject::visitChildren): (JSC::JSObject::allocatePropertyStorage): * runtime/JSObject.h: 2012-01-23 Mark Hahnenberg <mhahnenberg@apple.com> Use copying collector for out-of-line JSObject property storage https://bugs.webkit.org/show_bug.cgi?id=76665 Reviewed by Geoffrey Garen. * runtime/JSObject.cpp: (JSC::JSObject::visitChildren): Changed to use copyAndAppend whenever the property storage is out-of-line. (JSC::JSObject::allocatePropertyStorage): Changed to use tryAllocateStorage/tryReallocateStorage as opposed to operator new. * runtime/JSObject.h: 2012-01-23 Brian Weinstein <bweinstein@apple.com> More build fixing after r105646. * JavaScriptCore.exp: 2012-01-23 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=76855 Implement a JIT-code aware sampling profiler for JSC Reviewed by Geoff Garen. Step 2: generalize RedBlackTree. The profiler is going to want tio use a RedBlackTree, allow this class to work with subclasses of RedBlackTree::Node, Node should not need to know the names of the m_key and m_value fields (the subclass can provide a key() accessor), and RedBlackTree does not need to know anything about ValueType. * JavaScriptCore.exp: * wtf/MetaAllocator.cpp: (WTF::MetaAllocator::findAndRemoveFreeSpace): (WTF::MetaAllocator::debugFreeSpaceSize): (WTF::MetaAllocator::addFreeSpace): * wtf/MetaAllocator.h: (WTF::MetaAllocator::FreeSpaceNode::FreeSpaceNode): (WTF::MetaAllocator::FreeSpaceNode::key): * wtf/MetaAllocatorHandle.h: (WTF::MetaAllocatorHandle::key): * wtf/RedBlackTree.h: (WTF::RedBlackTree::Node::successor): (WTF::RedBlackTree::Node::predecessor): (WTF::RedBlackTree::Node::parent): (WTF::RedBlackTree::Node::setParent): (WTF::RedBlackTree::Node::left): (WTF::RedBlackTree::Node::setLeft): (WTF::RedBlackTree::Node::right): (WTF::RedBlackTree::Node::setRight): (WTF::RedBlackTree::insert): (WTF::RedBlackTree::remove): (WTF::RedBlackTree::findExact): (WTF::RedBlackTree::findLeastGreaterThanOrEqual): (WTF::RedBlackTree::findGreatestLessThanOrEqual): (WTF::RedBlackTree::first): (WTF::RedBlackTree::last): (WTF::RedBlackTree::size): (WTF::RedBlackTree::treeMinimum): (WTF::RedBlackTree::treeMaximum): (WTF::RedBlackTree::treeInsert): (WTF::RedBlackTree::leftRotate): (WTF::RedBlackTree::rightRotate): (WTF::RedBlackTree::removeFixup): 2012-01-23 Andy Estes <aestes@apple.com> Fix the build after r105635. * JavaScriptCore.exp: 2012-01-23 Mark Hahnenberg <mhahnenberg@apple.com> Remove StackBounds from JSGlobalData https://bugs.webkit.org/show_bug.cgi?id=76310 Reviewed by Sam Weinig. Removed StackBounds and the stack() function from JSGlobalData since it no longer accessed any members of JSGlobalData. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): * heap/MachineStackMarker.cpp: (JSC::MachineThreads::addCurrentThread): (JSC::MachineThreads::gatherFromCurrentThread): * parser/Parser.cpp: (JSC::::Parser): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: 2012-01-23 Gavin Barraclough <barraclough@apple.com> Implement a JIT-code aware sampling profiler for JSC https://bugs.webkit.org/show_bug.cgi?id=76855 Rubber stanmped by Geoff Garen. Mechanical change - pass CodeBlock through to the executable allocator, such that we will be able to map ranges of JIT code back to their owner. * assembler/ARMAssembler.cpp: (JSC::ARMAssembler::executableCopy): * assembler/ARMAssembler.h: * assembler/AssemblerBuffer.h: (JSC::AssemblerBuffer::executableCopy): * assembler/AssemblerBufferWithConstantPool.h: (JSC::AssemblerBufferWithConstantPool::executableCopy): * assembler/LinkBuffer.h: (JSC::LinkBuffer::LinkBuffer): (JSC::LinkBuffer::linkCode): * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::executableCopy): * assembler/SH4Assembler.h: (JSC::SH4Assembler::executableCopy): * assembler/X86Assembler.h: (JSC::X86Assembler::executableCopy): (JSC::X86Assembler::X86InstructionFormatter::executableCopy): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compile): (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGOSRExitCompiler.cpp: * dfg/DFGRepatch.cpp: (JSC::DFG::generateProtoChainAccessStub): (JSC::DFG::tryCacheGetByID): (JSC::DFG::tryBuildGetByIDList): (JSC::DFG::tryCachePutByID): * dfg/DFGThunks.cpp: (JSC::DFG::osrExitGenerationThunkGenerator): * jit/ExecutableAllocator.cpp: (JSC::ExecutableAllocator::allocate): * jit/ExecutableAllocator.h: * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::ExecutableAllocator::allocate): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::privateCompileCTINativeCall): * jit/JITPropertyAccess.cpp: (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): * jit/JITStubs.cpp: * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::finalize): * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::compile): 2012-01-23 Xianzhu Wang <wangxianzhu@chromium.org> Basic enhancements to StringBuilder https://bugs.webkit.org/show_bug.cgi?id=67081 This change contains the following enhancements to StringBuilder, for convenience, performance, testability, etc.: - Change toStringPreserveCapacity() to const - new public methods: capacity(), swap(), toAtomicString(), canShrink() and append(const StringBuilder&) - == and != opearators to compare StringBuilders and a StringBuilder/String Unit tests: Tools/TestWebKitAPI/Tests/WTF/StringBuilder.cpp Reviewed by Darin Adler. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/text/AtomicString.cpp: (WTF::SubstringTranslator::hash): (WTF::SubstringTranslator::equal): (WTF::SubstringTranslator::translate): (WTF::AtomicString::add): (WTF::AtomicString::addSlowCase): * wtf/text/AtomicString.h: (WTF::AtomicString::AtomicString): (WTF::AtomicString::add): * wtf/text/StringBuilder.cpp: (WTF::StringBuilder::reifyString): (WTF::StringBuilder::resize): (WTF::StringBuilder::canShrink): (WTF::StringBuilder::shrinkToFit): * wtf/text/StringBuilder.h: (WTF::StringBuilder::append): (WTF::StringBuilder::toString): (WTF::StringBuilder::toStringPreserveCapacity): (WTF::StringBuilder::toAtomicString): (WTF::StringBuilder::isEmpty): (WTF::StringBuilder::capacity): (WTF::StringBuilder::is8Bit): (WTF::StringBuilder::swap): (WTF::equal): (WTF::operator==): (WTF::operator!=): * wtf/text/StringImpl.h: 2012-01-23 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck. * GNUmakefile.list.am: Add missing files, remove deleted files and fix indentation. 2012-01-22 Filip Pizlo <fpizlo@apple.com> Build fix for non-DFG platforms that error out on warn-unused-parameter. * bytecode/CallLinkStatus.cpp: (JSC::CallLinkStatus::computeFor): * bytecode/GetByIdStatus.cpp: (JSC::GetByIdStatus::computeFor): * bytecode/MethodCallLinkStatus.cpp: (JSC::MethodCallLinkStatus::computeFor): * bytecode/PutByIdStatus.cpp: (JSC::PutByIdStatus::computeFor): 2012-01-22 Filip Pizlo <fpizlo@apple.com> Build fix for non-DFG platforms. * bytecode/CallLinkStatus.cpp: (JSC::CallLinkStatus::computeFor): * bytecode/GetByIdStatus.cpp: (JSC::GetByIdStatus::computeFor): * bytecode/MethodCallLinkStatus.cpp: (JSC::MethodCallLinkStatus::computeFor): * bytecode/PutByIdStatus.cpp: (JSC::PutByIdStatus::computeFor): 2012-01-20 Filip Pizlo <fpizlo@apple.com> DFG should not have code that directly decodes the states of old JIT inline cache data structures https://bugs.webkit.org/show_bug.cgi?id=76768 Reviewed by Sam Weinig. Introduced new classes (like GetByIdStatus) that encapsulate the set of things that the DFG would like to know about property accesses and calls. Whereas it previously got this information by directly decoding the data structures used by the old JIT for inline caching, it now uses these classes, which do the work for it. This should make it somewhat more straight forward to introduce new ways of profiling the same information. Also hoisted StructureSet into bytecode/ from dfg/, because it's now used by code in bytecode/. Making this work right involved carefully ensuring that the heuristics for choosing how to handle property accesses was at least as good as what we had before, since I completely restructured that code. Currently the performance looks neutral. Since I rewrote the code I did change some things that I never liked before, like previously if a put_bu_id had executed exactly once then we'd compile it as if it had taken slow-path. Executing once is special because then the inline cache is not baked in, so there is no information about how the DFG should optimize the code. Now this is rationalized: if the put_by_id does not offer enough information to be optimized (i.e. had executed 0 or 1 times) then we turn it into a forced OSR exit (i.e. a patch point). However, get_by_id still has the old behavior; I left it that way because I didn't want to make too many changes at once. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * bytecode/CallLinkStatus.cpp: Added. (JSC::CallLinkStatus::computeFor): * bytecode/CallLinkStatus.h: Added. (JSC::CallLinkStatus::CallLinkStatus): (JSC::CallLinkStatus::isSet): (JSC::CallLinkStatus::operator!): (JSC::CallLinkStatus::couldTakeSlowPath): (JSC::CallLinkStatus::callTarget): * bytecode/GetByIdStatus.cpp: Added. (JSC::GetByIdStatus::computeFor): * bytecode/GetByIdStatus.h: Added. (JSC::GetByIdStatus::GetByIdStatus): (JSC::GetByIdStatus::state): (JSC::GetByIdStatus::isSet): (JSC::GetByIdStatus::operator!): (JSC::GetByIdStatus::isSimpleDirect): (JSC::GetByIdStatus::takesSlowPath): (JSC::GetByIdStatus::makesCalls): (JSC::GetByIdStatus::structureSet): (JSC::GetByIdStatus::offset): * bytecode/MethodCallLinkStatus.cpp: Added. (JSC::MethodCallLinkStatus::computeFor): * bytecode/MethodCallLinkStatus.h: Added. (JSC::MethodCallLinkStatus::MethodCallLinkStatus): (JSC::MethodCallLinkStatus::isSet): (JSC::MethodCallLinkStatus::operator!): (JSC::MethodCallLinkStatus::needsPrototypeCheck): (JSC::MethodCallLinkStatus::structure): (JSC::MethodCallLinkStatus::prototypeStructure): (JSC::MethodCallLinkStatus::function): (JSC::MethodCallLinkStatus::prototype): * bytecode/PutByIdStatus.cpp: Added. (JSC::PutByIdStatus::computeFor): * bytecode/PutByIdStatus.h: Added. (JSC::PutByIdStatus::PutByIdStatus): (JSC::PutByIdStatus::state): (JSC::PutByIdStatus::isSet): (JSC::PutByIdStatus::operator!): (JSC::PutByIdStatus::isSimpleReplace): (JSC::PutByIdStatus::isSimpleTransition): (JSC::PutByIdStatus::takesSlowPath): (JSC::PutByIdStatus::oldStructure): (JSC::PutByIdStatus::newStructure): (JSC::PutByIdStatus::structureChain): (JSC::PutByIdStatus::offset): * bytecode/StructureSet.h: Added. (JSC::StructureSet::StructureSet): (JSC::StructureSet::clear): (JSC::StructureSet::add): (JSC::StructureSet::addAll): (JSC::StructureSet::remove): (JSC::StructureSet::contains): (JSC::StructureSet::isSubsetOf): (JSC::StructureSet::isSupersetOf): (JSC::StructureSet::size): (JSC::StructureSet::at): (JSC::StructureSet::operator[]): (JSC::StructureSet::last): (JSC::StructureSet::predictionFromStructures): (JSC::StructureSet::operator==): (JSC::StructureSet::dump): * dfg/DFGAbstractValue.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGStructureSet.h: Removed. 2012-01-20 Filip Pizlo <fpizlo@apple.com> JIT compilation should not require ExecState https://bugs.webkit.org/show_bug.cgi?id=76729 <rdar://problem/10731545> Reviewed by Gavin Barraclough. Changed the relevant JIT driver functions to take JSGlobalData& instead of ExecState*, since really they just needed the global data. * dfg/DFGDriver.cpp: (JSC::DFG::compile): (JSC::DFG::tryCompile): (JSC::DFG::tryCompileFunction): * dfg/DFGDriver.h: (JSC::DFG::tryCompile): (JSC::DFG::tryCompileFunction): * jit/JITDriver.h: (JSC::jitCompileIfAppropriate): (JSC::jitCompileFunctionIfAppropriate): * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): 2012-01-20 David Levin <levin@chromium.org> Make OwnPtr<HDC> work for the Chromium Windows port. https://bugs.webkit.org/show_bug.cgi?id=76738 Reviewed by Jian Li. * JavaScriptCore.gyp/JavaScriptCore.gyp: Added OwnPtrWin.cpp to the Chromium Windows build. * wtf/OwnPtrCommon.h: Changed from platform WIN to OS WIN for OwnPtr<HDC> and similar constructs. 2012-01-19 Geoffrey Garen <ggaren@apple.com> Removed some regexp entry boilerplate code https://bugs.webkit.org/show_bug.cgi?id=76687 Reviewed by Darin Adler. 1% - 2% speedup on regexp tests, no change overall. * runtime/RegExp.cpp: (JSC::RegExp::match): - ASSERT that our startIndex is non-negative, because anything less would be uncivilized. - ASSERT that our input is not the null string for the same reason. - No need to test for startOffset being past the end of the string, since the regular expression engine will do this test for us. - No need to initialize the output vector, since the regular expression engine will fill it in for us. * yarr/YarrInterpreter.cpp: (JSC::Yarr::Interpreter::interpret): * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::compile): RegExp used to do these jobs for us, but now we do them for ourselves because it's a better separation of concerns, and the JIT can do them more efficiently than C++ code: - Test for "past the end" before doing any matching -- otherwise a* will match with zero length past the end of the string, which is wrong. - Initialize the output vector before doing any matching. 2012-01-20 Filip Pizlo <fpizlo@apple.com> Build fix for no-DFG configuration. Needed for <rdar://problem/10727689>. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitProfiledOpcode): * jit/JIT.h: (JSC::JIT::emitValueProfilingSite): 2012-01-19 Filip Pizlo <fpizlo@apple.com> Bytecode instructions that may have value profiling should have a direct inline link to the ValueProfile instance https://bugs.webkit.org/show_bug.cgi?id=76682 <rdar://problem/10727689> Reviewed by Sam Weinig. Each opcode that gets value profiled now has a link to its ValueProfile. This required rationalizing the emission of value profiles for opcode combos, like op_method_check/op_get_by_id and op_call/op_call_put_result. It only makes sense for one of them to have a value profile link, and it makes most sense for it to be the one that actually sets the result. The previous behavior was to have op_method_check profile for op_get_by_id when they were used together, but otherwise for op_get_by_id to have its own profiles. op_call already did the right thing; all profiling was done by op_call_put_result. But rationalizing this code required breaking some of the natural boundaries that the code had; for instance the code in DFG that emits a GetById in place of both op_method_check and op_get_by_id must now know that it's the latter of those that has the value profile, while the first of those constitutes the OSR target. Hence each CodeOrigin must now have two bytecode indices - one for OSR exit and one for profiling. Finally this change required some refiddling of our optimization heuristics, because now all code blocks have "more instructions" due to the value profile slots. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::printGetByIdOp): (JSC::CodeBlock::dump): * bytecode/CodeBlock.h: (JSC::CodeBlock::valueProfileForBytecodeOffset): * bytecode/CodeOrigin.h: (JSC::CodeOrigin::CodeOrigin): (JSC::CodeOrigin::bytecodeIndexForValueProfile): * bytecode/Instruction.h: (JSC::Instruction::Instruction): * bytecode/Opcode.h: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitProfiledOpcode): (JSC::BytecodeGenerator::emitResolve): (JSC::BytecodeGenerator::emitGetScopedVar): (JSC::BytecodeGenerator::emitResolveBase): (JSC::BytecodeGenerator::emitResolveBaseForPut): (JSC::BytecodeGenerator::emitResolveWithBase): (JSC::BytecodeGenerator::emitResolveWithThis): (JSC::BytecodeGenerator::emitGetById): (JSC::BytecodeGenerator::emitGetByVal): (JSC::BytecodeGenerator::emitCall): (JSC::BytecodeGenerator::emitCallVarargs): (JSC::BytecodeGenerator::emitConstruct): * bytecompiler/BytecodeGenerator.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::currentCodeOrigin): (JSC::DFG::ByteCodeParser::addCall): (JSC::DFG::ByteCodeParser::getPredictionWithoutOSRExit): (JSC::DFG::ByteCodeParser::getPrediction): (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGraph.h: (JSC::DFG::Graph::valueProfileFor): * jit/JIT.h: (JSC::JIT::emitValueProfilingSite): * jit/JITCall.cpp: (JSC::JIT::emit_op_call_put_result): * jit/JITCall32_64.cpp: (JSC::JIT::emit_op_call_put_result): * jit/JITInlineMethods.h: (JSC::JIT::emitValueProfilingSite): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_resolve): (JSC::JIT::emit_op_resolve_base): (JSC::JIT::emit_op_resolve_skip): (JSC::JIT::emit_op_resolve_global): (JSC::JIT::emitSlow_op_resolve_global): (JSC::JIT::emit_op_resolve_with_base): (JSC::JIT::emit_op_resolve_with_this): (JSC::JIT::emitSlow_op_resolve_global_dynamic): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_resolve): (JSC::JIT::emit_op_resolve_base): (JSC::JIT::emit_op_resolve_skip): (JSC::JIT::emit_op_resolve_global): (JSC::JIT::emitSlow_op_resolve_global): (JSC::JIT::emit_op_resolve_with_base): (JSC::JIT::emit_op_resolve_with_this): * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_get_by_val): (JSC::JIT::emitSlow_op_get_by_val): (JSC::JIT::emit_op_method_check): (JSC::JIT::emitSlow_op_method_check): (JSC::JIT::emit_op_get_by_id): (JSC::JIT::emitSlow_op_get_by_id): (JSC::JIT::emit_op_get_scoped_var): (JSC::JIT::emit_op_get_global_var): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_method_check): (JSC::JIT::emitSlow_op_method_check): (JSC::JIT::emit_op_get_by_val): (JSC::JIT::emitSlow_op_get_by_val): (JSC::JIT::emit_op_get_by_id): (JSC::JIT::emitSlow_op_get_by_id): (JSC::JIT::emit_op_get_scoped_var): (JSC::JIT::emit_op_get_global_var): * jit/JITStubCall.h: (JSC::JITStubCall::callWithValueProfiling): * runtime/Options.cpp: (JSC::Options::initializeOptions): 2012-01-20 ChangSeok Oh <shivamidow@gmail.com> undefined reference to symbol eina_module_free https://bugs.webkit.org/show_bug.cgi?id=76681 Reviewed by Martin Robinson. eina_module_free has been used without including eina libraries after r104936. * wtf/PlatformEfl.cmake: Add EINA_LIBRARIES. 2012-01-19 Tony Chang <tony@chromium.org> [chromium] Remove an obsolete comment about features.gypi https://bugs.webkit.org/show_bug.cgi?id=76643 There can be only one features.gypi. Reviewed by James Robinson. * JavaScriptCore.gyp/JavaScriptCore.gyp: 2012-01-19 Geoffrey Garen <ggaren@apple.com> Implicit creation of a regular expression should eagerly check for syntax errors https://bugs.webkit.org/show_bug.cgi?id=76642 Reviewed by Oliver Hunt. This is a correctness fix and a slight optimization. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncMatch): (JSC::stringProtoFuncSearch): Check for syntax errors because that's the correct behavior. * runtime/RegExp.cpp: (JSC::RegExp::match): ASSERT that we aren't a syntax error. (One line of code change, many lines of indentation change.) Since we have no clients that try to match a RegExp that is a syntax error, let's optimize out the check. 2012-01-19 Mark Hahnenberg <mhahnenberg@apple.com> Implement a new allocator for backing stores https://bugs.webkit.org/show_bug.cgi?id=75181 Reviewed by Filip Pizlo. We want to move away from using fastMalloc for the backing stores for some of our objects (e.g. JSArray, JSObject, JSString, etc). These backing stores have a nice property in that they only have a single owner (i.e. a single pointer to them at any one time). One way that we can take advantage of this property is to implement a simple bump allocator/copying collector, which will run alongside our normal mark/sweep collector, that only needs to update the single owner pointer rather than having to redirect an arbitrary number of pointers in from-space to to-space. This plan can give us a number of benefits. We can beat fastMalloc in terms of both performance and memory usage, we can track how much memory we're using far more accurately than our rough estimation now through the use of reportExtraMemoryCost, and we can allocate arbitrary size objects (as opposed to being limited to size classes like we have been historically). This is also another step toward moving away from lazy destruction, which will improve our memory footprint. We start by creating said allocator and moving the ArrayStorage for JSArray to use it rather than fastMalloc. The design of the collector is as follows: Allocation: -The collector allocates 64KB chunks from the OS to use for object allocation. -Each chunk contains an offset, a flag indicating if the block has been pinned, and a payload, along with next and prev pointers so that they can be put in DoublyLinkedLists. -Any allocation greater than 64KB gets its own separate oversize block, which is managed separately from the rest. -If the allocator receives a request for more than the remaining amount in the current block, it grabs a fresh block. -Grabbing a fresh block means grabbing one off of the global free list (which is now shared between the mark/sweep allocator and the bump allocator) if there is one. If there isn't a new one we do one of two things: allocate a new block from the OS if we're not ready for a GC yet, or run a GC and then try again. If we still don't have enough space after the GC, we allocate a new block from the OS. Garbage collection: -At the start of garbage collection during conservative stack scanning, if we encounter what appears to be a pointer to a bump-allocated block of memory, we pin that block so that it will not be copied for this round of collection. -We also pin any oversize blocks that we encounter, which effectively doubles as a "mark bit" for that block. Any oversize blocks that aren't pinned at the end of copying are given back to the OS. -Marking threads are now also responsible for copying bump-allocated objects to newSpace -Each marking thread has a private 64KB block into which it copies bump-allocated objects that it encounters. -When that block fills up, the marking thread gives it back to the allocator and requests a new one. -When all marking has concluded, each thread gives back its copy block, even if it isn't full. -At the conclusion of copying (which is done by the end of the marking phase), we un-pin any pinned blocks and give any blocks left in from-space to the global free list. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * heap/AllocationSpace.cpp: (JSC::AllocationSpace::allocateSlowCase): (JSC::AllocationSpace::allocateBlock): (JSC::AllocationSpace::freeBlocks): * heap/AllocationSpace.h: (JSC::AllocationSpace::waterMark): * heap/BumpBlock.h: Added. (JSC::BumpBlock::BumpBlock): * heap/BumpSpace.cpp: Added. (JSC::BumpSpace::tryAllocateSlowCase): * heap/BumpSpace.h: Added. (JSC::BumpSpace::isInCopyPhase): (JSC::BumpSpace::totalMemoryAllocated): (JSC::BumpSpace::totalMemoryUtilized): * heap/BumpSpaceInlineMethods.h: Added. (JSC::BumpSpace::BumpSpace): (JSC::BumpSpace::init): (JSC::BumpSpace::contains): (JSC::BumpSpace::pin): (JSC::BumpSpace::startedCopying): (JSC::BumpSpace::doneCopying): (JSC::BumpSpace::doneFillingBlock): (JSC::BumpSpace::recycleBlock): (JSC::BumpSpace::getFreshBlock): (JSC::BumpSpace::borrowBlock): (JSC::BumpSpace::addNewBlock): (JSC::BumpSpace::allocateNewBlock): (JSC::BumpSpace::fitsInBlock): (JSC::BumpSpace::fitsInCurrentBlock): (JSC::BumpSpace::tryAllocate): (JSC::BumpSpace::tryAllocateOversize): (JSC::BumpSpace::allocateFromBlock): (JSC::BumpSpace::tryReallocate): (JSC::BumpSpace::tryReallocateOversize): (JSC::BumpSpace::isOversize): (JSC::BumpSpace::isPinned): (JSC::BumpSpace::oversizeBlockFor): (JSC::BumpSpace::blockFor): * heap/ConservativeRoots.cpp: (JSC::ConservativeRoots::ConservativeRoots): (JSC::ConservativeRoots::genericAddPointer): (JSC::ConservativeRoots::add): * heap/ConservativeRoots.h: * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::blockFreeingThreadMain): (JSC::Heap::reportExtraMemoryCostSlowCase): (JSC::Heap::getConservativeRegisterRoots): (JSC::Heap::markRoots): (JSC::Heap::collect): (JSC::Heap::releaseFreeBlocks): * heap/Heap.h: (JSC::Heap::waterMark): (JSC::Heap::highWaterMark): (JSC::Heap::setHighWaterMark): (JSC::Heap::tryAllocateStorage): (JSC::Heap::tryReallocateStorage): * heap/HeapBlock.h: Added. (JSC::HeapBlock::HeapBlock): * heap/MarkStack.cpp: (JSC::MarkStackThreadSharedData::MarkStackThreadSharedData): (JSC::SlotVisitor::drain): (JSC::SlotVisitor::drainFromShared): (JSC::SlotVisitor::startCopying): (JSC::SlotVisitor::allocateNewSpace): (JSC::SlotVisitor::copy): (JSC::SlotVisitor::copyAndAppend): (JSC::SlotVisitor::doneCopying): * heap/MarkStack.h: * heap/MarkedBlock.cpp: (JSC::MarkedBlock::recycle): (JSC::MarkedBlock::MarkedBlock): * heap/MarkedBlock.h: * heap/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): * heap/MarkedSpace.h: (JSC::MarkedSpace::allocate): (JSC::MarkedSpace::forEachBlock): (JSC::MarkedSpace::SizeClass::resetAllocator): * heap/SlotVisitor.h: (JSC::SlotVisitor::SlotVisitor): * heap/TinyBloomFilter.h: (JSC::TinyBloomFilter::reset): * runtime/JSArray.cpp: (JSC::JSArray::JSArray): (JSC::JSArray::finishCreation): (JSC::JSArray::tryFinishCreationUninitialized): (JSC::JSArray::~JSArray): (JSC::JSArray::enterSparseMode): (JSC::JSArray::defineOwnNumericProperty): (JSC::JSArray::setLengthWritable): (JSC::JSArray::getOwnPropertySlotByIndex): (JSC::JSArray::getOwnPropertyDescriptor): (JSC::JSArray::putByIndexBeyondVectorLength): (JSC::JSArray::deletePropertyByIndex): (JSC::JSArray::getOwnPropertyNames): (JSC::JSArray::increaseVectorLength): (JSC::JSArray::unshiftCountSlowCase): (JSC::JSArray::setLength): (JSC::JSArray::pop): (JSC::JSArray::unshiftCount): (JSC::JSArray::visitChildren): (JSC::JSArray::sortNumeric): (JSC::JSArray::sort): (JSC::JSArray::compactForSorting): (JSC::JSArray::subclassData): (JSC::JSArray::setSubclassData): (JSC::JSArray::checkConsistency): * runtime/JSArray.h: (JSC::JSArray::inSparseMode): (JSC::JSArray::isLengthWritable): * wtf/CheckedBoolean.h: Added. (CheckedBoolean::CheckedBoolean): (CheckedBoolean::~CheckedBoolean): (CheckedBoolean::operator bool): * wtf/DoublyLinkedList.h: (WTF::::push): * wtf/StdLibExtras.h: (WTF::isPointerAligned): 2012-01-19 Joi Sigurdsson <joi@chromium.org> Enable use of precompiled headers in Chromium port on Windows. Bug 76381 - Use precompiled headers in Chromium port on Windows https://bugs.webkit.org/show_bug.cgi?id=76381 Reviewed by Tony Chang. * JavaScriptCore.gyp/JavaScriptCore.gyp: Include WinPrecompile.gypi. 2012-01-18 Roland Takacs <takacs.roland@stud.u-szeged.hu> Cross-platform processor core counter fix https://bugs.webkit.org/show_bug.cgi?id=76540 Reviewed by Zoltan Herczeg. I attached "OS(FREEBSD)" to "#if OS(DARWIN) || OS(OPENBSD) || OS(NETBSD)" and I removed the OS checking macros from ParallelJobsGeneric.cpp because the NumberOfCores.cpp contains them for counting CPU cores. The processor core counter patch located at https://bugs.webkit.org/show_bug.cgi?id=76530 * wtf/NumberOfCores.cpp: * wtf/ParallelJobsGeneric.cpp: 2012-01-18 Csaba Osztrogonác <ossy@webkit.org> Cross-platform processor core counter https://bugs.webkit.org/show_bug.cgi?id=76530 Unreviewed cross-MinGW buildfix after r105270. * wtf/NumberOfCores.cpp: Use windows.h instead of Windows.h. 2012-01-18 Roland Takacs <takacs.roland@stud.u-szeged.hu> Cross-platform processor core counter https://bugs.webkit.org/show_bug.cgi?id=76530 Reviewed by Zoltan Herczeg. Two files have been created that include the processor core counter function. It used to be in ParallelJobsGeneric.h/cpp before. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * runtime/Options.cpp: (JSC::Options::initializeOptions): * wtf/CMakeLists.txt: * wtf/NumberOfCores.cpp: Added. (WTF::numberOfProcessorCores): * wtf/NumberOfCores.h: Added. * wtf/ParallelJobsGeneric.cpp: (WTF::ParallelEnvironment::ParallelEnvironment): * wtf/ParallelJobsGeneric.h: 2012-01-18 Balazs Kelemen <kbalazs@webkit.org> [Qt] Consolidate layout test crash logging https://bugs.webkit.org/show_bug.cgi?id=75088 Reviewed by Simon Hausmann. Move backtrace generating logic into WTFReportBacktrace and add a way to deinstall signal handlers if we know that we have already printed the backtrace. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/Assertions.cpp: (WTFLogLocker::WTFReportBacktrace): (WTFLogLocker::WTFSetCrashHook): (WTFLogLocker::WTFInvokeCrashHook): * wtf/Assertions.h: 2012-01-17 Geoffrey Garen <ggaren@apple.com> Factored out some code into a helper function. I think this might help getting rid of omit-frame-pointer. Reviewed by Sam Weinig. No benchmark change. * runtime/StringPrototype.cpp: (JSC::removeUsingRegExpSearch): Moved to here... (JSC::replaceUsingRegExpSearch): ...from here. 2012-01-17 Caio Marcelo de Oliveira Filho <caio.oliveira@openbossa.org> Uint8ClampedArray support https://bugs.webkit.org/show_bug.cgi?id=74455 Reviewed by Filip Pizlo. * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/PredictedType.cpp: (JSC::predictionToString): (JSC::predictionFromClassInfo): * bytecode/PredictedType.h: (JSC::isUint8ClampedArrayPrediction): (JSC::isActionableMutableArrayPrediction): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::initialize): (JSC::DFG::AbstractState::execute): * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateUint8ClampedArray): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::clampDoubleToByte): (JSC::DFG::compileClampIntegerToByte): (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): (JSC::DFG::SpeculativeJIT::compileGetIndexedPropertyStorage): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/JSCell.h: * runtime/JSGlobalData.h: * wtf/Forward.h: * wtf/Uint8Array.h: * wtf/Uint8ClampedArray.h: Added. (WTF::Uint8ClampedArray::set): (WTF::Uint8ClampedArray::create): (WTF::Uint8ClampedArray::Uint8ClampedArray): (WTF::Uint8ClampedArray::subarray): 2012-01-17 Sam Weinig <sam@webkit.org> Add helper macro for forward declaring objective-c classes https://bugs.webkit.org/show_bug.cgi?id=76485 Reviewed by Anders Carlsson. * wtf/Compiler.h: Add OBJC_CLASS macro which helps reduce code when forward declaring an objective-c class in a header which can be included from both Objective-C and non-Objective-C files. 2012-01-17 Filip Pizlo <fpizlo@apple.com> DFG should be able to do JS and custom getter caching https://bugs.webkit.org/show_bug.cgi?id=76361 Reviewed by Csaba Osztrogonác. Fix for 32-bit. * dfg/DFGRepatch.cpp: (JSC::DFG::tryBuildGetByIDList): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-01-15 Filip Pizlo <fpizlo@apple.com> DFG should be able to do JS and custom getter caching https://bugs.webkit.org/show_bug.cgi?id=76361 <rdar://problem/10698060> Reviewed by Geoff Garen. Added the ability to cache JS getter calls and custom getter calls in the DFG. Most of this is pretty mundane, since the old JIT supported this functionality as well. But a couple interesting things had to happen: - There are now two variants of GetById: GetById, which works as before, and GetByIdFlush, which flushes registers prior to doing the GetById. Only GetByIdFlush can be used for caching getters. We detect which GetById style to use by looking at the inline caches of the old JIT. - Exception handling for getter calls planted in stubs uses a separate lookup handler routine, which uses the CodeOrigin stored in the StructureStubInfo. This is a 40% speed-up in the Dromaeo DOM Traversal average. It removes all of the DFG regressions we saw in Dromaeo. This is neutral on SunSpider, V8, and Kraken. * bytecode/StructureStubInfo.h: * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::emitExceptionCheck): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::willNeedFlush): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCCallHelpers.h: (JSC::DFG::CCallHelpers::setupResults): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: (JSC::DFG::PropertyAccessRecord::PropertyAccessRecord): (JSC::DFG::JITCompiler::addExceptionCheck): * dfg/DFGNode.h: (JSC::DFG::Node::hasIdentifier): (JSC::DFG::Node::hasHeapPrediction): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGRepatch.cpp: (JSC::DFG::tryCacheGetByID): (JSC::DFG::tryBuildGetByIDList): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::appendCallWithExceptionCheckSetResult): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::cachedGetById): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::cachedGetById): (JSC::DFG::SpeculativeJIT::compile): 2012-01-16 Jon Lee <jonlee@apple.com> Build fix for r105086. * Configurations/FeatureDefines.xcconfig: * wtf/Platform.h: 2012-01-16 Jon Lee <jonlee@apple.com> Remove HTML notifications support on Mac https://bugs.webkit.org/show_bug.cgi?id=76401 <rdar://problem/10589881> Reviewed by Sam Weinig. * wtf/Platform.h: Define ENABLE_HTML_NOTIFICATIONS macro. 2012-01-16 Zeno Albisser <zeno@webkit.org> [Qt] Fix QT_VERSION related warnings when building on Mac OS X https://bugs.webkit.org/show_bug.cgi?id=76340 This bug was caused by r104826. As already mentioned for https://bugs.webkit.org/show_bug.cgi?id=57239 we should not use "using namespace WebCore" in header files, because it might cause ambiguous references. This patch reverts the changes from r104826 and r104981 and removes the "using namespace WebCore" statement from two header files. Reviewed by Tor Arne Vestbø. * wtf/Platform.h: 2012-01-16 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck. * GNUmakefile.list.am: Fix typo. 2012-01-16 Pavel Heimlich <tropikhajma@gmail.com> Solaris Studio supports alignment macros too https://bugs.webkit.org/show_bug.cgi?id=75453 Reviewed by Hajime Morita. * wtf/Alignment.h: 2012-01-16 Yuqiang Xian <yuqiang.xian@intel.com> Build fix on 32bit if verbose debug is enabled in DFG https://bugs.webkit.org/show_bug.cgi?id=76351 Reviewed by Hajime Morita. Mostly change "%lu" to "%zu" to print a "size_t" variable. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::endBasicBlock): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::parseCodeBlock): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGOSREntry.cpp: (JSC::DFG::prepareOSREntry): 2012-01-15 Filip Pizlo <fpizlo@apple.com> The C calling convention logic in DFG::SpeculativeJIT should be available even when not generating code for the DFG speculative path https://bugs.webkit.org/show_bug.cgi?id=76355 Reviewed by Dan Bernstein. Moved all of the logic for placing C call arguments into the right place (stack or registers) into a new class, DFG::CCallHelpers. This class inherits from AssemblyHelpers, another DFG grab-bag of helper functions. I could have moved this code into AssemblyHelpers, but decided against it, because I wanted to limit the number of methods each class in the JIT has. Hence now we have a slightly odd organization of JIT classes in DFG: MacroAssembler (basic instruction emission) <= AssemblyHelpers (some additional JS smarts) <= CCallHelpers (understands calls to C functions) <= JITCompiler (can compile a graph to machine code). Each of these except for JITCompiler can be reused for stub compilation. * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGCCallHelpers.h: Added. (JSC::DFG::CCallHelpers::CCallHelpers): (JSC::DFG::CCallHelpers::resetCallArguments): (JSC::DFG::CCallHelpers::addCallArgument): (JSC::DFG::CCallHelpers::setupArguments): (JSC::DFG::CCallHelpers::setupArgumentsExecState): (JSC::DFG::CCallHelpers::setupArgumentsWithExecState): (JSC::DFG::CCallHelpers::setupTwoStubArgs): (JSC::DFG::CCallHelpers::setupStubArguments): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::JITCompiler): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::callOperation): 2012-01-15 Pablo Flouret <pablof@motorola.com> Fix compilation errors on build-webkit --debug --no-video on mac. https://bugs.webkit.org/show_bug.cgi?id=75867 Reviewed by Philippe Normand. Make ENABLE_VIDEO_TRACK conditional on ENABLE_VIDEO, video track feature doesn't build without video. * wtf/Platform.h: 2012-01-14 David Levin <levin@chromium.org> HWndDC should be in platform/win instead of wtf. https://bugs.webkit.org/show_bug.cgi?id=76314 Reviewed by Sam Weinig. * JavaScriptCore.gyp/JavaScriptCore.gyp: * JavaScriptCore.gypi: 2012-01-13 David Levin <levin@chromium.org> check-webkit-style: should encourage the use of Own* classes for Windows DC. https://bugs.webkit.org/show_bug.cgi?id=76227 Reviewed by Dirk Pranke. * wtf/win/HWndDCWin.h: (WTF::HwndDC::HwndDC): Add a way to do GetDCEx. There are no users, but I want to catch this in check-webkit-style and tell any users to use HwndDC to avoid leaks. 2012-01-13 David Levin <levin@chromium.org> Header file is missing header guard. Reviewed by Dirk Pranke. * wtf/win/HWndDCWin.h: Added the guards. 2012-01-13 Andy Wingo <wingo@igalia.com> Eval in strict mode does not need dynamic checks https://bugs.webkit.org/show_bug.cgi?id=76286 Reviewed by Oliver Hunt. * runtime/JSActivation.cpp (JSC::JSActivation::JSActivation): Eval in strict mode cannot introduce variables, so it not impose the need for dynamic checks. 2012-01-13 David Levin <levin@chromium.org> HWndDC is a better name than HwndDC. https://bugs.webkit.org/show_bug.cgi?id=76281 Reviewed by Darin Adler. * JavaScriptCore.gyp/JavaScriptCore.gyp: * JavaScriptCore.gypi: * wtf/win/HWndDCWin.h: Renamed from Source/JavaScriptCore/wtf/win/HwndDCWin.h. (WTF::HWndDC::HWndDC): (WTF::HWndDC::~HWndDC): (WTF::HWndDC::operator HDC): 2012-01-13 YoungTaeck Song <youngtaeck.song@samsung.com> [EFL] Add OwnPtr specialization for Eina_Module. https://bugs.webkit.org/show_bug.cgi?id=76255 Reviewed by Andreas Kling. Add an overload for deleteOwnedPtr(Eina_Module*) on EFL port. * wtf/OwnPtrCommon.h: * wtf/efl/OwnPtrEfl.cpp: (WTF::deleteOwnedPtr): 2012-01-13 Yuqiang Xian <yuqiang.xian@intel.com> Unreviewed build fix after r104787 if JIT_VERBOSE_OSR is defined * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): 2012-01-12 Hajime Morrita <morrita@chromium.org> JavaScriptCore: Mark all exported symbols in the header file automatically. https://bugs.webkit.org/show_bug.cgi?id=72855 Reviewed by Darin Adler. Added WTF_EXPORT_PRIVATE and JS_EXPORT_PRIVATE based on JavaScriptCore.exp files. The change is generated by a tool calledListExportables (https://github.com/omo/ListExportables) * API/OpaqueJSString.h: * bytecode/CodeBlock.h: * bytecode/SamplingTool.h: * debugger/Debugger.h: * debugger/DebuggerActivation.h: * debugger/DebuggerCallFrame.h: * heap/AllocationSpace.h: * heap/HandleHeap.h: * heap/Heap.h: * heap/MachineStackMarker.h: * heap/MarkStack.h: * heap/VTableSpectrum.h: * heap/WriteBarrierSupport.h: * parser/Nodes.h: * parser/ParserArena.h: * profiler/Profile.h: * runtime/ArgList.h: * runtime/CallData.h: * runtime/Completion.h: * runtime/ConstructData.h: * runtime/DateInstance.h: * runtime/Error.h: * runtime/ExceptionHelpers.h: * runtime/FunctionConstructor.h: * runtime/Identifier.h: * runtime/InitializeThreading.h: * runtime/InternalFunction.h: * runtime/JSArray.h: * runtime/JSByteArray.h: * runtime/JSCell.h: * runtime/JSFunction.h: * runtime/JSGlobalData.cpp: * runtime/JSGlobalData.h: * runtime/JSGlobalObject.h: * runtime/JSGlobalThis.h: * runtime/JSLock.h: * runtime/JSObject.h: * runtime/JSString.h: * runtime/JSValue.h: * runtime/JSVariableObject.h: * runtime/Lookup.h: * runtime/MemoryStatistics.h: * runtime/ObjectPrototype.h: * runtime/Options.h: * runtime/PropertyDescriptor.h: * runtime/PropertyNameArray.h: * runtime/PropertySlot.h: * runtime/RegExp.h: * runtime/RegExpObject.h: * runtime/SamplingCounter.h: * runtime/SmallStrings.h: * runtime/StringObject.h: * runtime/Structure.h: * runtime/TimeoutChecker.h: * runtime/UString.h: * runtime/WriteBarrier.h: * wtf/ArrayBufferView.h: * wtf/ByteArray.h: * wtf/CryptographicallyRandomNumber.h: * wtf/CurrentTime.h: * wtf/DateMath.h: * wtf/DecimalNumber.h: * wtf/FastMalloc.cpp: * wtf/FastMalloc.h: * wtf/MD5.h: * wtf/MainThread.h: * wtf/MetaAllocator.h: * wtf/MetaAllocatorHandle.h: * wtf/OSAllocator.h: * wtf/PageBlock.h: * wtf/RandomNumber.h: * wtf/RefCountedLeakCounter.h: * wtf/SHA1.h: * wtf/Threading.cpp: * wtf/Threading.h: * wtf/ThreadingPrimitives.h: * wtf/WTFThreadData.h: * wtf/dtoa.h: * wtf/text/AtomicString.h: * wtf/text/CString.h: * wtf/text/StringBuilder.h: * wtf/text/StringImpl.h: * wtf/text/WTFString.h: * wtf/unicode/Collator.h: * wtf/unicode/UTF8.h: * yarr/Yarr.h: * yarr/YarrPattern.h: 2012-01-12 MORITA Hajime <morrita@google.com> [Chromium] JSExportMacros.h should be visible. https://bugs.webkit.org/show_bug.cgi?id=76147 Reviewed by Tony Chang. * config.h: 2012-01-12 David Levin <levin@chromium.org> HwndDC is a better name than OwnGetDC. https://bugs.webkit.org/show_bug.cgi?id=76235 Reviewed by Dmitry Titov. This is a better name for two reasons: 1. "Own" implies "delete". In this case, the final call is a release (ReleaseDC). 2. "Ref" would be a better name due to the release but the RefPtr (and OwnPtr) classes always take something to hold on to. In this case, the object (the DC) is created by the class once it is given a Window to ensure that the HDC was actually created using GetDC. * JavaScriptCore.gyp/JavaScriptCore.gyp: * JavaScriptCore.gypi: * wtf/win/HwndDCWin.h: Renamed from Source/JavaScriptCore/wtf/win/OwnGetDCWin.h. (WTF::HwndDC::HwndDC): (WTF::HwndDC::~HwndDC): (WTF::HwndDC::operator HDC): 2012-01-12 Gavin Barraclough <barraclough@apple.com> Clean up putDirect (part 2) https://bugs.webkit.org/show_bug.cgi?id=76232 Reviewed by Sam Weinig. Rename putWithAttributes to putDirectVirtual, to identify that this has the same unchecked-DefineOwnProperty behaviour, change putDirectInternal to be templated on an enum indicating which behaviour it is supposed to be implementing, and change clients that are defining properties to call putDirectInternal correctly. * API/JSObjectRef.cpp: (JSObjectSetProperty): * JavaScriptCore.exp: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::putDirectVirtual): * debugger/DebuggerActivation.h: * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): * runtime/ClassInfo.h: * runtime/Error.cpp: (JSC::addErrorInfo): * runtime/JSActivation.cpp: (JSC::JSActivation::putDirectVirtual): * runtime/JSActivation.h: * runtime/JSCell.cpp: (JSC::JSCell::putDirectVirtual): * runtime/JSCell.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::putDirectVirtual): * runtime/JSGlobalObject.h: * runtime/JSObject.cpp: (JSC::JSObject::put): (JSC::JSObject::putDirectVirtual): (JSC::JSObject::defineGetter): (JSC::JSObject::initializeGetterSetterProperty): (JSC::JSObject::defineSetter): (JSC::putDescriptor): * runtime/JSObject.h: (JSC::JSObject::putDirectInternal): (JSC::JSObject::putOwnDataProperty): (JSC::JSObject::putDirect): * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::putDirectVirtual): * runtime/JSStaticScopeObject.h: * runtime/JSVariableObject.cpp: (JSC::JSVariableObject::putDirectVirtual): * runtime/JSVariableObject.h: 2012-01-12 Gavin Barraclough <barraclough@apple.com> Clean up putDirect (part 1) https://bugs.webkit.org/show_bug.cgi?id=76232 Reviewed by Sam Weinig. putDirect has ambiguous semantics, clean these up a bit. putDirect generally behaves a bit like a fast defineOwnProperty, but one that always creates the property, with no checking to validate the put it permitted. It also encompasses two slightly different behaviors. (1) a fast form of put for JSActivation, which doesn't have to handle searching the prototype chain, getter/setter properties, or the magic __proto__ value. Break this out as a new method, 'putOwnDataProperty'. (2) the version of putDirect on JSValue will also check for overwriting ReadOnly values, in strict mode. This is, however, not so smart on a few level, since it is only called from op_put_by_id with direct set, which is only used with an object as the base, and is only used to put new properties onto objects. * dfg/DFGOperations.cpp: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/JSActivation.cpp: (JSC::JSActivation::put): * runtime/JSFunction.cpp: (JSC::JSFunction::getOwnPropertySlot): * runtime/JSObject.h: (JSC::JSObject::putOwnDataProperty): * runtime/JSValue.h: 2012-01-12 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=76141 defineSetter/defineGetter may fail to update Accessor attribute Reviewed by Oliver Hunt. * runtime/JSObject.cpp: (JSC::JSObject::defineGetter): (JSC::JSObject::initializeGetterSetterProperty): (JSC::JSObject::defineSetter): * runtime/Structure.cpp: (JSC::Structure::attributeChangeTransition): * runtime/Structure.h: 2012-01-12 David Levin <levin@chromium.org> [chromium] Fix DC leak in WebScreenInfoFactory. https://bugs.webkit.org/show_bug.cgi?id=76203 Reviewed by Dmitry Titov. * JavaScriptCore.gyp/JavaScriptCore.gyp: Added OwnGetDCWin.h * JavaScriptCore.gypi: Added OwnGetDCWin.h * JavaScriptCore/wtf/win/OwnGetDCWin.h: Made an owner class for GetDC which needs ReleaseDC as opposed to DeleteDC. 2012-01-11 Gavin Barraclough <barraclough@apple.com> Allow accessor get/set property to be set to undefined https://bugs.webkit.org/show_bug.cgi?id=76148 Reviewed by Oliver Hunt. AccessorDescriptor properties may have their get & set properties defined to reference a function (Callable object) or be set to undefined. Valid PropertyDescriptors created by toPropertyDescriptor (defined from JS code via Object.defineProperty, etc) have get and set properties that are in one of three states (1) nonexistent, (2) set to undefined, or (3) a function (any Callable object). On the PropertyDescriptor object these three states are represneted by JSValue(), jsUndefined(), and any JSObject* (with a constraint that this must be callable). Logically the get/set property of an accessor descriptor on an object might be in any of the three states above, but in practice there is no way to distinguish between the first two states. As such we stor the get/set values in property storage in a JSObject* field, with 0 indicating absent or undefined. When unboxing to a PropertyDescriptor, map this back to a JS undefined value. * runtime/GetterSetter.h: (JSC::GetterSetter::setGetter): (JSC::GetterSetter::setSetter): - Allow the getter/setter to be cleared. * runtime/JSArray.cpp: (JSC::JSArray::putDescriptor): - Changed to call getterObject/setterObject. (JSC::JSArray::defineOwnNumericProperty): - Added ASSERT. * runtime/JSObject.cpp: (JSC::putDescriptor): (JSC::JSObject::defineOwnProperty): - Changed to call getterObject/setterObject. * runtime/ObjectConstructor.cpp: (JSC::objectConstructorGetOwnPropertyDescriptor): - getter/setter values read from properties on object are never missing, they will now be set as undefined by 'setDescriptor'. (JSC::toPropertyDescriptor): - Do not translate undefined->empty, this loses an important distinction between a get/set property being absent, or being explicitly set to undefined. * runtime/PropertyDescriptor.cpp: (JSC::PropertyDescriptor::getterObject): (JSC::PropertyDescriptor::setterObject): - Accessors to convert the get/set property to an object pointer, converting undefined to 0. (JSC::PropertyDescriptor::setDescriptor): (JSC::PropertyDescriptor::setAccessorDescriptor): - Translate a getter/setter internally represented at 0 to undefined, indicating that it is present. * runtime/PropertyDescriptor.h: - Declare getterObject/setterObject. 2012-01-12 Zeno Albisser <zeno@webkit.org> [Qt][WK2][Mac] Conflict of MacTypes.h defining a Fixed type after r104560. https://bugs.webkit.org/show_bug.cgi?id=76175 Defining ENABLE_CSS_FILTERS leads to ambiguous references due to MacTypes.h being included. Defining CF_OPEN_SOURCE works around this problem. Reviewed by Simon Hausmann. * wtf/Platform.h: 2012-01-12 Simon Hausmann <simon.hausmann@nokia.com> Make the new WTF module build on Qt https://bugs.webkit.org/show_bug.cgi?id=76163 Reviewed by Tor Arne Vestbø. * JavaScriptCore.pro: Removed wtf from the subdirs to build. 2012-01-11 Filip Pizlo <fpizlo@apple.com> CodeBlock::m_executeCounter should be renamed to CodeBlock::m_jitExecuteCounter https://bugs.webkit.org/show_bug.cgi?id=76144 <rdar://problem/10681711> Rubber stamped by Gavin Barraclough. * bytecode/CodeBlock.h: (JSC::CodeBlock::addressOfJITExecuteCounter): (JSC::CodeBlock::offsetOfJITExecuteCounter): (JSC::CodeBlock::jitExecuteCounter): (JSC::CodeBlock::optimizeNextInvocation): (JSC::CodeBlock::dontOptimizeAnytimeSoon): (JSC::CodeBlock::optimizeAfterWarmUp): (JSC::CodeBlock::optimizeAfterLongWarmUp): (JSC::CodeBlock::optimizeSoon): * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * jit/JIT.cpp: (JSC::JIT::emitOptimizationCheck): 2012-01-11 Gavin Barraclough <barraclough@apple.com> Merge 'Getter'/'Setter' attributes into 'Accessor' https://bugs.webkit.org/show_bug.cgi?id=76141 Reviewed by Filip Pizlo. These are currently ambiguous (and used inconsistently). It would logically appear that either being bit set implies that the corresponding type of accessor is present but (a) we don't correctly enforce this, and (b) this means the attributes would not be able to distinguish between a data descriptor and an accessor descriptor with neither a getter nor setter defined (which is a descriptor permissible under the spec). This ambiguity would lead to unsafe property caching behavior (though this does not represent an actual current bug, since we are currently unable to create descriptors that have neither a getter nor setter, it just prevents us from doing so). * runtime/Arguments.cpp: (JSC::Arguments::createStrictModeCallerIfNecessary): (JSC::Arguments::createStrictModeCalleeIfNecessary): * runtime/JSArray.cpp: (JSC::SparseArrayValueMap::put): (JSC::JSArray::putDescriptor): * runtime/JSBoundFunction.cpp: (JSC::JSBoundFunction::finishCreation): * runtime/JSFunction.cpp: (JSC::JSFunction::getOwnPropertySlot): (JSC::JSFunction::getOwnPropertyDescriptor): * runtime/JSObject.cpp: (JSC::JSObject::defineGetter): (JSC::JSObject::initializeGetterSetterProperty): (JSC::JSObject::defineSetter): (JSC::putDescriptor): (JSC::JSObject::defineOwnProperty): * runtime/JSObject.h: * runtime/ObjectConstructor.cpp: (JSC::objectConstructorDefineProperty): * runtime/PropertyDescriptor.cpp: (JSC::PropertyDescriptor::setDescriptor): (JSC::PropertyDescriptor::setAccessorDescriptor): (JSC::PropertyDescriptor::setSetter): (JSC::PropertyDescriptor::setGetter): (JSC::PropertyDescriptor::attributesOverridingCurrent): 2012-01-11 Gavin Barraclough <barraclough@apple.com> Object.defineProperty([], 'length', {}) should not make length read-only https://bugs.webkit.org/show_bug.cgi?id=76097 Reviewed by Oliver Hunt. * runtime/JSArray.cpp: (JSC::JSArray::defineOwnProperty): - We should be checking writablePresent(). 2012-01-11 Filip Pizlo <fpizlo@apple.com> Code duplication for invoking the JIT and DFG should be reduced https://bugs.webkit.org/show_bug.cgi?id=76117 <rdar://problem/10680189> Rubber stamped by Geoff Garen. * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * jit/JITDriver.h: Added. (JSC::jitCompileIfAppropriate): (JSC::jitCompileFunctionIfAppropriate): * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): 2012-01-11 Geoffrey Garen <ggaren@apple.com> Bytecode dumping is broken for call opcodes (due to two new operands) https://bugs.webkit.org/show_bug.cgi?id=75886 Reviewed by Oliver Hunt. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::printCallOp): Made a helper function, so I wouldn't have to fix this more than once. The helper function skips the extra two operands at the end of the opcode, used for optimization. (JSC::CodeBlock::dump): Used the helper function. * bytecode/CodeBlock.h: Declared the helper function. 2012-01-09 Geoffrey Garen <ggaren@apple.com> REGRESSION: d3 Bullet Charts demo doesn't work (call with argument assignment is broken) https://bugs.webkit.org/show_bug.cgi?id=75911 * bytecompiler/BytecodeGenerator.h: (JSC::BytecodeGenerator::emitNodeForLeftHandSide): Cleanup: No need to explicitly cast to our return type in C++. * bytecompiler/NodesCodegen.cpp: (JSC::FunctionCallResolveNode::emitBytecode): (JSC::ApplyFunctionCallDotNode::emitBytecode): Make sure to copy our function into a temporary register before evaluating our arguments, since argument evaluation might include function calls or assignments that overwrite our callee by name. 2012-01-11 Michael Saboff <msaboff@apple.com> v8-regexp spends 35% of its time allocating and copying internal regexp results data https://bugs.webkit.org/show_bug.cgi?id=76079 Reviewed by Geoffrey Garen. Added a new RegExpResults struct that has the input string, the number of subexpressions and the output vector. Changed RegExpConstructor to include a RegExpConstructorPrivate instead of having a reference to one. Changed RegExpMatchesArray to include a RegExpResults instead of a reference to a RegExpConstructorPrivate. Created an overloaded assignment operator to assign a RegExpConstructorPrivate to a RegExpResults. Collectively this change is worth 24% performance improvement to v8-regexp. * runtime/RegExpConstructor.cpp: (JSC::RegExpResult::operator=): (JSC::RegExpConstructor::RegExpConstructor): (JSC::RegExpMatchesArray::RegExpMatchesArray): (JSC::RegExpMatchesArray::finishCreation): (JSC::RegExpMatchesArray::~RegExpMatchesArray): (JSC::RegExpMatchesArray::fillArrayInstance): (JSC::RegExpConstructor::arrayOfMatches): (JSC::RegExpConstructor::getBackref): (JSC::RegExpConstructor::getLastParen): (JSC::RegExpConstructor::getLeftContext): (JSC::RegExpConstructor::getRightContext): (JSC::RegExpConstructor::setInput): (JSC::RegExpConstructor::input): (JSC::RegExpConstructor::setMultiline): (JSC::RegExpConstructor::multiline): * runtime/RegExpConstructor.h: (JSC::RegExpResult::RegExpResult): (JSC::RegExpConstructor::performMatch): * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::create): (JSC::RegExpMatchesArray::getOwnPropertySlot): (JSC::RegExpMatchesArray::getOwnPropertySlotByIndex): (JSC::RegExpMatchesArray::getOwnPropertyDescriptor): (JSC::RegExpMatchesArray::put): (JSC::RegExpMatchesArray::putByIndex): (JSC::RegExpMatchesArray::deleteProperty): (JSC::RegExpMatchesArray::deletePropertyByIndex): (JSC::RegExpMatchesArray::getOwnPropertyNames): 2012-01-11 Eugene Girard <girard@google.com> Typo in error message: Unexpected token 'defualt' https://bugs.webkit.org/show_bug.cgi?id=75105 Reviewed by Simon Fraser. * parser/Parser.h: (JSC::Parser::getTokenName): 2012-01-11 Anders Carlsson <andersca@apple.com> Assertion failure in JSC::allocateCell trying to allocate a JSString https://bugs.webkit.org/show_bug.cgi?id=76101 Reviewed by Adam Roben. Remove the ExecutableBase::s_info and JSString::s_info static member variables from the .def file and export them explicitly using the JS_EXPORTDATA macro. member variables explicitly using * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/Executable.h: * runtime/JSString.h: 2012-01-10 Mark Rowe <mrowe@apple.com> <rdar://problem/10673792> jsc should install directly in to versioned Resources subfolder This ensures that jsc ends up in a consistent location whether built in to the same DSTROOT as JavaScriptCore.framework or in to a different one. Rubber-stamped by Dan Bernstein. * Configurations/JSC.xcconfig: Update INSTALL_PATH. 2012-01-10 Filip Pizlo <fpizlo@apple.com> DFG inlining block linking compares BlockIndex against bytecode index https://bugs.webkit.org/show_bug.cgi?id=76018 <rdar://problem/10671979> Reviewed by Gavin Barraclough. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseCodeBlock): 2012-01-10 Filip Pizlo <fpizlo@apple.com> CodeBlock.h declares too many things https://bugs.webkit.org/show_bug.cgi?id=76001 Rubber stamped by Gavin Barraclough. Removed all non-CodeBlock type declarations from CodeBlock.h, and put them into separate header files. Also removed all non-CodeBlock method implementations from CodeBlock.cpp and put them into corresponding cpp files. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * assembler/RepatchBuffer.h: * bytecode/CallLinkInfo.cpp: Added. (JSC::CallLinkInfo::unlink): * bytecode/CallLinkInfo.h: Added. (JSC::CallLinkInfo::callTypeFor): (JSC::CallLinkInfo::CallLinkInfo): (JSC::CallLinkInfo::~CallLinkInfo): (JSC::CallLinkInfo::isLinked): (JSC::CallLinkInfo::seenOnce): (JSC::CallLinkInfo::setSeen): (JSC::getCallLinkInfoReturnLocation): (JSC::getCallLinkInfoBytecodeIndex): * bytecode/CallReturnOffsetToBytecodeOffset.h: Added. (JSC::CallReturnOffsetToBytecodeOffset::CallReturnOffsetToBytecodeOffset): (JSC::getCallReturnOffset): * bytecode/CodeBlock.cpp: * bytecode/CodeBlock.h: * bytecode/CodeType.h: Added. * bytecode/ExpressionRangeInfo.h: Added. * bytecode/GlobalResolveInfo.h: Added. (JSC::GlobalResolveInfo::GlobalResolveInfo): * bytecode/HandlerInfo.h: Added. * bytecode/LineInfo.h: Added. * bytecode/MethodCallLinkInfo.cpp: Added. (JSC::MethodCallLinkInfo::reset): * bytecode/MethodCallLinkInfo.h: Added. (JSC::MethodCallLinkInfo::MethodCallLinkInfo): (JSC::MethodCallLinkInfo::seenOnce): (JSC::MethodCallLinkInfo::setSeen): (JSC::getMethodCallLinkInfoReturnLocation): (JSC::getMethodCallLinkInfoBytecodeIndex): * bytecode/StructureStubInfo.h: (JSC::getStructureStubInfoReturnLocation): (JSC::getStructureStubInfoBytecodeIndex): 2012-01-10 Anders Carlsson <andersca@apple.com> Hang opening movie that requires authentication https://bugs.webkit.org/show_bug.cgi?id=75989 <rdar://problem/9601915> Reviewed by Sam Weinig. * wtf/Functional.h: Add function wrapper for a function that takes three parameters. 2012-01-10 Filip Pizlo <fpizlo@apple.com> CodeBlock::m_numParameters should be encapsulated https://bugs.webkit.org/show_bug.cgi?id=75985 <rdar://problem/10671020> Reviewed by Oliver Hunt. Encapsulated CodeBlock::m_numParameters and hooked argument profile creation into it. This appears to be performance neutral. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::CodeBlock): (JSC::CodeBlock::setNumParameters): (JSC::CodeBlock::addParameter): * bytecode/CodeBlock.h: (JSC::CodeBlock::numParameters): (JSC::CodeBlock::addressOfNumParameters): (JSC::CodeBlock::offsetOfNumParameters): (JSC::CodeBlock::numberOfArgumentValueProfiles): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): (JSC::BytecodeGenerator::addParameter): (JSC::BytecodeGenerator::emitReturn): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::AbstractState): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::checkArgumentTypes): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::SpeculativeJIT): * interpreter/Interpreter.cpp: (JSC::Interpreter::slideRegisterWindowForCall): (JSC::Interpreter::dumpRegisters): (JSC::Interpreter::execute): (JSC::Interpreter::prepareForRepeatCall): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JITStubs.cpp: (JSC::arityCheckFor): (JSC::lazyLinkFor): * runtime/Executable.cpp: (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): 2012-01-10 Gavin Barraclough <barraclough@apple.com> Build fix following https://bugs.webkit.org/show_bug.cgi?id=75935 Fix 32-bit builds. * runtime/JSArray.cpp: (JSC::JSArray::getOwnPropertyNames): (JSC::JSArray::setLength): 2012-01-10 Gavin Barraclough <barraclough@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-01-10 Gavin Barraclough <barraclough@apple.com> Do not allow Array length to be set if it is non-configurable https://bugs.webkit.org/show_bug.cgi?id=75935 Reviewed by Sam Weinig. Do not allow Array length to be set if it is non-configurable, and if the new length is less than the old length then intervening properties should removed in reverse order. Removal of properties should cease if an intervening indexed property being removed is non-configurable. * JavaScriptCore.exp: - Removed export for setLength. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncConcat): - JSArray::setLength now takes an ExecState* (JSC::arrayProtoFuncSlice): - JSArray::setLength now takes an ExecState* * runtime/JSArray.cpp: (JSC::JSArray::defineOwnProperty): - JSArray::setLength now takes an ExecState* (JSC::JSArray::put): - JSArray::setLength now takes an ExecState* (JSC::compareKeysForQSort): - Keys extracted from the map can be stored as unsigneds. (JSC::JSArray::getOwnPropertyNames): - Keys extracted from the map can be stored as unsigneds. (JSC::JSArray::setLength): - Check lengthIsReadOnly(), rather than copying the entire map to iterate over to determine which keys to remove, instead just copy the keys from the map to a Vector. When inSparseMode sort the keys in the Vector so that we can remove properties in reverse order. * runtime/JSArray.h: - JSArray::setLength now takes an ExecState* 2012-01-10 Gavin Barraclough <barraclough@apple.com> Use SameValue to compare property descriptor values https://bugs.webkit.org/show_bug.cgi?id=75975 Reviewed by Sam Weinig. Rather than strictEqual. * runtime/JSArray.cpp: (JSC::JSArray::defineOwnNumericProperty): - Missing configurablePresent() check. * runtime/JSObject.cpp: (JSC::JSObject::defineOwnProperty): - call sameValue. * runtime/PropertyDescriptor.cpp: (JSC::sameValue): - Moved from JSArray.cpp, fix NaN comparison. (JSC::PropertyDescriptor::equalTo): - call sameValue. * runtime/PropertyDescriptor.h: - Added declaration for sameValue. 2012-01-09 Gavin Barraclough <barraclough@apple.com> Error handling : in ISO8601 timezone https://bugs.webkit.org/show_bug.cgi?id=75919 Reviewed by Sam Weinig. * wtf/DateMath.cpp: (WTF::parseDateFromNullTerminatedCharacters): - need to increment the string position. 2012-01-09 Mark Rowe <mrowe@apple.com> JavaScriptCore executable targets shouldn't explicitly depend on the JavaScriptCore framework target <http://webkit.org/b/75907> / <rdar://problem/10659862> We'd like for it to be possible to build jsc without building JavaScriptCore.framework and the explicit dependencies prevent this. Reviewed by Dan Bernstein. * JavaScriptCore.xcodeproj/project.pbxproj: 2012-01-09 Adam Treat <atreat@rim.com> Log is a little to verbose for blackberry port https://bugs.webkit.org/show_bug.cgi?id=75728 The BlackBerry::Platform::Log* functions take care of the call to vfprintf which is resulting in unintentional noise in our logs. Add a conditional directive to fix. Change to using BlackBerry::Platform::logStreamV which does not insert threading info and newlines unlike BlackBerry::Platform::log. Finally, add log locking and unlocking which the BlackBerry platform uses to ensure that N threads do not trample on each other's logs. Reviewed by Rob Buis. * wtf/Assertions.cpp: (WTFLogLocker::WTFReportAssertionFailure): (WTFLogLocker::WTFReportAssertionFailureWithMessage): (WTFLogLocker::WTFReportArgumentAssertionFailure): (WTFLogLocker::WTFReportFatalError): (WTFLogLocker::WTFReportError): (WTFLogLocker::WTFLog): (WTFLogLocker::WTFLogVerbose): 2012-01-09 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=75789 defineOwnProperty not implemented for Array objects Reviewed by Sam Weinig. Implements support for getter/setter & non-default attribute properties on arrays, by forcing them into a dictionary-like 'SparseMode'. This fixes ~300 test-262 test failures. * JavaScriptCore.exp: - Updated exports. * dfg/DFGOperations.cpp: - JSArray::pop now requires an exec state. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncPop): - JSArray::pop now requires an exec state. * runtime/JSArray.cpp: (JSC::SparseArrayValueMap::add): - Add a potentially empty entry into the map. (JSC::SparseArrayValueMap::put): - Changed to call setter. (JSC::SparseArrayEntry::get): - calls getters. (JSC::SparseArrayEntry::getNonSparseMode): - does not call getters. (JSC::JSArray::enterSparseMode): - Convert into 'SparseMode' - removes the vectors, don't allow it to be recreated. (JSC::JSArray::putDescriptor): - Create a numeric property based on a descriptor. (JSC::sameValue): - See ES5.1 9.12. (JSC::reject): - Helper for the [[DefineOwnProperty]] algorithm. (JSC::JSArray::defineOwnNumericProperty): - Define an indexed property on an array object. (JSC::JSArray::setLengthWritable): - Marks the length read-only, enters SparseMode as necessary. (JSC::JSArray::defineOwnProperty): - Defines either an indexed property or 'length' on an array object. (JSC::JSArray::getOwnPropertySlotByIndex): - Updated to correctly handle accessor descriptors & attributes. (JSC::JSArray::getOwnPropertyDescriptor): - Updated to correctly handle accessor descriptors & attributes. (JSC::JSArray::put): - Pass strict mode flag to setLength. (JSC::JSArray::putByIndex): - putByIndexBeyondVectorLength requires an ExecState* rather than a JSGloablData&. (JSC::JSArray::putByIndexBeyondVectorLength): - Pass exec to SparseArrayValueMap::put. (JSC::JSArray::deletePropertyByIndex): - Do not allow deletion of non-configurable properties. (JSC::compareKeysForQSort): - used in implementation of getOwnPropertyNames. (JSC::JSArray::getOwnPropertyNames): - Properties in the sparse map should be iterated in order. (JSC::JSArray::setLength): - Updated to take a 'shouldThrow' flag, return a result indicating error. (JSC::JSArray::pop): - pop should throw an error if length is not writable, even if the array is empty. (JSC::JSArray::push): - putByIndexBeyondVectorLength requires an ExecState* rather than a JSGloablData&. (JSC::JSArray::sort): - Changed 'get' to 'getNonSparseMode' (can't be getters to call). (JSC::JSArray::compactForSorting): - Changed 'get' to 'getNonSparseMode' (can't be getters to call). * runtime/JSArray.h: (JSC::SparseArrayValueMap::lengthIsReadOnly): - Check if the length is read only. (JSC::SparseArrayValueMap::setLengthIsReadOnly): - Mark the length as read only. (JSC::SparseArrayValueMap::find): - Moved into header. (JSC::JSArray::isLengthWritable): - Wraps SparseArrayValueMap::lengthIsReadOnly. * runtime/JSObject.cpp: (JSC::JSObject::defineOwnProperty): - Should be returning the result of putDescriptor. * runtime/PropertyDescriptor.cpp: (JSC::PropertyDescriptor::attributesOverridingCurrent): - Added attributesOverridingCurrent - this should probably be merged with attributesWithOverride. * runtime/PropertyDescriptor.h: - Added attributesOverridingCurrent. 2012-01-09 Pavel Heimlich <tropikhajma@gmail.com> There is no support for fastcall in Solaris Studio. Fixes build on Solaris. https://bugs.webkit.org/show_bug.cgi?id=75736 Reviewed by Gavin Barraclough. * jit/JITStubs.h: 2012-01-09 Pavel Heimlich <tropikhajma@gmail.com> Fix build failure on Solaris https://bugs.webkit.org/show_bug.cgi?id=75733 Reviewed by Gavin Barraclough. * wtf/ByteArray.h: 2012-01-01 Raphael Kubo da Costa <kubo@profusion.mobi> [CMake] Clean up some cruft from WTF's CMakeLists.txt https://bugs.webkit.org/show_bug.cgi?id=75420 Reviewed by Daniel Bates. * wtf/CMakeLists.txt: Remove the unused WTF_PORT_FLAGS variable; add all needed paths to WTF_INCLUDE_DIRECTORIES in a single place. 2012-01-08 Xianzhu Wang <wangxianzhu@chromium.org> Fix compilation error about ListHashSetReverseIterator https://bugs.webkit.org/show_bug.cgi?id=75372 Reviewed by Darin Adler. There is a typo in class ListHashSetReverseIterator: typedef ListHashSetConstIterator<ValueArg, inlineCapacity, HashArg> const_reverse_iterator; Should be typedef ListHashSetConstReverseIterator<ValueArg, inlineCapacity, HashArg> const_reverse_iterator; * wtf/ListHashSet.h: 2012-01-08 Ryosuke Niwa <rniwa@webkit.org> WinCE build fix after r104415. * jit/JITExceptions.cpp: * jit/JITExceptions.h: 2012-01-08 Filip Pizlo <fpizlo@apple.com> The JIT's protocol for exception handling should be available to other parts of the system https://bugs.webkit.org/show_bug.cgi?id=75808 <rdar://problem/10661025> Reviewed by Oliver Hunt. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * jit/JITExceptions.cpp: Added. (JSC::genericThrow): (JSC::jitThrow): * jit/JITExceptions.h: Added. * jit/JITStubs.cpp: * runtime/JSGlobalData.h: 2012-01-06 Hajime Morrita <morrita@chromium.org> https://bugs.webkit.org/show_bug.cgi?id=75296 JSString should not have JS_EXPORTCLASS annotation Reviewed by Kevin Ollivier. * runtime/JSString.h: Removed JS_EXPORTCLASS annotation. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Added missing symbols which were hidden by JS_EXPORTCLASS. 2012-01-06 Michael Saboff <msaboff@apple.com> JSArray::pop() should compare SparseArrayValueMap::find() to SparseArrayValueMap::notFound() https://bugs.webkit.org/show_bug.cgi?id=75757 Reviewed by Gavin Barraclough. * runtime/JSArray.cpp: (JSC::JSArray::pop): Changed map->end() to map->notFound(). 2012-01-06 Filip Pizlo <fpizlo@apple.com> JIT stub slow paths that would be identical to that of an interpreter should be factored out https://bugs.webkit.org/show_bug.cgi?id=75743 <rdar://problem/10657024> Reviewed by Geoff Garen. * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/CommonSlowPaths.h: Added. (JSC::CommonSlowPaths::opInstanceOfSlow): (JSC::CommonSlowPaths::opIn): (JSC::CommonSlowPaths::opResolve): (JSC::CommonSlowPaths::opResolveSkip): (JSC::CommonSlowPaths::opResolveWithBase): (JSC::CommonSlowPaths::opResolveWithThis): 2012-01-06 Sam Weinig <sam@webkit.org> Fix windows build. * wtf/TypeTraits.cpp: 2012-01-05 Michael Saboff <msaboff@apple.com> Default HashTraits for Opcode don't work for Opcode = 0 https://bugs.webkit.org/show_bug.cgi?id=75595 Reviewed by Oliver Hunt. Removed the populating of the m_opcodeIDTable table in the case where the OpcodeID and Opcode are the same (m_enabled is false). Instead we just cast the one type to the other. * interpreter/Interpreter.cpp: (JSC::Interpreter::initialize): (JSC::Interpreter::isOpcode): * interpreter/Interpreter.h: (JSC::Interpreter::getOpcodeID): 2012-01-06 Sam Weinig <sam@webkit.org> Add a DecayArray type trait as a first step towards merging OwnPtr and OwnArrayPtr https://bugs.webkit.org/show_bug.cgi?id=75737 Reviewed by Anders Carlsson. * wtf/TypeTraits.cpp: * wtf/TypeTraits.h: Added a DecayArray trait, that can convert T[] and T[3] -> T*. DecayArray is composed of some helpers which are also exposed, Conditional<>, which can provide one type or another based on a boolean predicate, IsArray<> which can deduce array types, and RemoveExtent<>, which removes the extent from an array type. 2012-01-06 Oliver Hunt <oliver@apple.com> GetByteArrayLength is incorrect https://bugs.webkit.org/show_bug.cgi?id=75735 Reviewed by Filip Pizlo. Load the byte array length from the correct location. This stops an existing test from hanging. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2012-01-06 Filip Pizlo <fpizlo@apple.com> Fix build. * JavaScriptCore.xcodeproj/project.pbxproj: 2012-01-06 Oliver Hunt <oliver@apple.com> DFG no longer optimises CanvasPixelArray https://bugs.webkit.org/show_bug.cgi?id=75729 Reviewed by Gavin Barraclough. Rename ByteArray (in its ClassInfo) to Uint8ClampedArray to match the future name when we switch over to the new typed-array based ImageData specification. * runtime/JSByteArray.cpp: 2012-01-06 Caio Marcelo de Oliveira Filho <caio.oliveira@openbossa.org> Use HashMap<OwnPtr> for SourceProviderCache items https://bugs.webkit.org/show_bug.cgi?id=75346 Reviewed by Daniel Bates. * parser/Parser.cpp: * parser/SourceProviderCache.cpp: (JSC::SourceProviderCache::clear): (JSC::SourceProviderCache::add): * parser/SourceProviderCache.h: 2012-01-06 Sam Weinig <sam@webkit.org> Remove unused OwnFastMallocPtr class. https://bugs.webkit.org/show_bug.cgi?id=75722 Reviewed by Geoffrey Garen. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/CMakeLists.txt: * wtf/OwnFastMallocPtr.h: Removed. * wtf/text/StringImpl.h: * wtf/wtf.pro: 2012-01-06 Benjamin Poulain <bpoulain@webkit.org> [Mac] Sort the resources of JavaScriptCore.xcodeproj and remove duplicates https://bugs.webkit.org/show_bug.cgi?id=75631 Reviewed by Andreas Kling. * JavaScriptCore.xcodeproj/project.pbxproj: 2012-01-06 Eric Seidel <eric@webkit.org> and Gustavo Noronha Silva <gustavo.noronha@collabora.com> Make the new WTF module build on Gtk https://bugs.webkit.org/show_bug.cgi?id=75669 * GNUmakefile.am: 2012-01-06 Tor Arne Vestbø <tor.arne.vestbo@nokia.com> [Qt] Remove un-needed VPATHs from project includes Reviewed by Simon Hausmann. * JavaScriptCore.pri: * wtf/wtf.pri: 2012-01-06 Tor Arne Vestbø <tor.arne.vestbo@nokia.com> [Qt] Move listing of include paths and libs to pri files in sources Includepaths are sometimes modified by non-Qt contributors so keeping them in files inside Sources makes it more likely that they are updated along with project files for the other ports. Using pri files instead of prf files for this also has the benefit that the include() from the main target file can be parsed and followed by Qt Creator -- something that does not work with load(). Dependency from a target to a library through the WEBKIT variable are handled through forwarding-files in Tools/qmake/mkspecs/modules, which set the source root of the module and include the right pri file. Ideally we'd use the variant of include() that takes an optional namespace to read the variables into, or the fromfile() function, but both of these add an overhead of about 40% on the total qmake runtime, due to making a deep copy of all the variables in the project or re-reading all the prf files from scratch. Reviewed by Simon Hausmann. Reviewed by Ossy. * JavaScriptCore.pri: Renamed from Tools/qmake/mkspecs/features/javascriptcore.prf. * Target.pri: * wtf/wtf.pri: Renamed from Tools/qmake/mkspecs/features/wtf.prf. * wtf/wtf.pro: 2012-01-06 Hajime Morrita <morrita@chromium.org> WTF::String: Inline method shouldn't have WTF_EXPORT_PRIVATE https://bugs.webkit.org/show_bug.cgi?id=75612 Reviewed by Kevin Ollivier. * wtf/text/WTFString.h: (WTF::String::findIgnoringCase): (WTF::String::append): (WTF::String::fromUTF8): (WTF::String::fromUTF8WithLatin1Fallback): (WTF::String::isHashTableDeletedValue): 2012-01-05 Dan Bernstein <mitz@apple.com> <rdar://problem/10633760> Update copyright strings Reviewed by Mark Rowe. * Info.plist: 2012-01-05 Gavin Barraclough <barraclough@apple.com> Date constructor handles infinite values incorrectly. https://bugs.webkit.org/show_bug.cgi?id=70998 Reviewed by Filip Pizlo. * runtime/DateConstructor.cpp: (JSC::constructDate): - should be checking !finite rather then isnan. 2012-01-05 Gavin Barraclough <barraclough@apple.com> date.toISOString produces incorrect results for dates with ms prior to 1970 https://bugs.webkit.org/show_bug.cgi?id=75684 Reviewed by Sam Weinig. * runtime/DatePrototype.cpp: (JSC::dateProtoFuncToISOString): 2012-01-05 Gavin Barraclough <barraclough@apple.com> Array.prototype.lastIndexOf ignores undefined fromIndex. https://bugs.webkit.org/show_bug.cgi?id=75678 Reviewed by Sam Weinig. array.lastIndexOf(x, undefined) is equivalent to array.lastIndexOf(x, 0), not array.lastIndexOf(x) * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncLastIndexOf): - should check argumnet count, rather than checking agument value for undefined. 2012-01-05 Gavin Barraclough <barraclough@apple.com> Date parsing is too restrictive. https://bugs.webkit.org/show_bug.cgi?id=75671 Reviewed by Oliver Hunt. ES5 date parsing currently requires all fields to be present, which does not match the spec (ES5.1 15.9.1.15). The spec allow a date to be date only, or date + time. The date portion on the should match: (pseudocode!:) [(+|-)YY]YYYY[-MM[-DD]] though we are slightly more liberal (permitted by the spec), allowing: [+|-]Y+[-MM[-DD]] The time portion should match: THH:mm[:ss[.sss]][Z|(+|-)HH:mm] again we're slightly more liberal, allowing: THH:mm[:ss[.s+]][Z|(+|-)HH:mm] * wtf/DateMath.cpp: (WTF::parseES5DatePortion): - Month/day fields are optional, default to 01. (WTF::parseES5TimePortion): - Hours/Minutes are requires, seconds/timezone are optional. (WTF::parseES5DateFromNullTerminatedCharacters): - Dates may be date only, or date + time. 2012-01-05 Bruno Dilly <bdilly@profusion.mobi> [EFL] Undefined references to ICU_I18N symbols on WTF https://bugs.webkit.org/show_bug.cgi?id=75642 Unreviewed build fix. Add ${ICU_I18N_LIBRARIES} to WTF_LIBRARIES on wtf efl platform cmake. Some undefined references were ucol_setAttribute_44, ucol_close_44, ucol_getAttribute_44... * wtf/PlatformEfl.cmake: 2012-01-05 Geoffrey Garen <ggaren@apple.com> Refined the fast path for StringImpl::hash() https://bugs.webkit.org/show_bug.cgi?id=75178 Reviewed by Darin Adler. Moved the hash calculation code into an out-of-line function to clean up the hot path. No measurable benchmark change, but this knocks some samples off in Instruments, and I think this is a step toward removing -fomit-frame-pointer. * wtf/text/StringImpl.cpp: (WTF::StringImpl::hashSlowCase): * wtf/text/StringImpl.h: (WTF::StringImpl::hash): The patch. * wtf/text/StringStatics.cpp: (WTF::StringImpl::hashSlowCase): Abide by the cockamamie Windows build scheme, which requires all out-of-line StringImpl functions used by WebCore be defined in this file instead of StringImpl.cpp. (See http://trac.webkit.org/changeset/59187.) 2012-01-05 Gavin Barraclough <barraclough@apple.com> Literal tab in JSONString fails https://bugs.webkit.org/show_bug.cgi?id=71772 Reviewed by Oliver Hunt. rfc4627 does not allow literal tab characters in JSON source. * runtime/LiteralParser.cpp: (JSC::isSafeStringCharacter): - do not allow literal tab in StrictJSON mode. 2012-01-05 Gavin Barraclough <barraclough@apple.com> push/shift fifo may consume excessive memory https://bugs.webkit.org/show_bug.cgi?id=75610 Reviewed by Sam Weinig. Array object commonly store data in a vector, consisting of a portion that is in use, a pre-capacity (m_indexBias) and a post-capacity (the delta between m_length and m_vectorLength). Calls to shift with grow the pre-capacity, and the current algorithm for increaseVectorLength (used by push, or [[Put]]) will never shrink the pre-capacity, so a push/shift fifo may consume an inordinate amount of memory, whilst having a relatively small active length. * runtime/JSArray.cpp: (JSC::JSArray::increaseVectorLength): - If m_indexBias is non-zero, decay it over time. 2012-01-05 Csaba Osztrogonác <ossy@webkit.org> unshift/pop fifo may consume excessive memory https://bugs.webkit.org/show_bug.cgi?id=75588 Reviewed by Zoltan Herczeg. Buildfix after r104120. * runtime/JSArray.cpp: Remove useless asserts, baecause unsigned expression >= 0 is always true (JSC::JSArray::unshiftCount): 2012-01-05 Zoltan Herczeg <zherczeg@webkit.org> Unreviewed gardening after r104134. * wtf/Assertions.cpp: 2012-01-05 Zoltan Herczeg <zherczeg@webkit.org> Unreviewed gardening after r75605. Rubber stamped by NOBODY Csaba Osztrogonác. * wtf/Assertions.cpp: 2012-01-05 Benjamin Poulain <benjamin@webkit.org> Improve charactersAreAllASCII() to compare multiple characters at a time https://bugs.webkit.org/show_bug.cgi?id=74063 Reviewed by Darin Adler. A new header ASCIIFastPath.h contains the functions related to the detection of ASCII by using machine words. Part of it comes from WebCore's TextCodecASCIIFastPath.h. The function charactersAreAllASCII() is moved to TextCodecASCIIFastPath.h and is implemented with computer word comparison. The gain over the previous implementation of charactersAreAllASCII() is of the order of how many comparison are avoided (4x, 8x, 16x depending on the format and the CPU type). * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/text/ASCIIFastPath.h: Added. (WTF::isAlignedToMachineWord): (WTF::alignToMachineWord): (WTF::isAllASCII): (WTF::charactersAreAllASCII): * wtf/text/WTFString.h: * wtf/wtf.pro: 2012-01-05 Mark Rowe <mrowe@apple.com> <http://webkit.org/b/75606> [Mac] WTF logging functions should output to both stderr and ASL We should always log to both ASL and stderr on platforms where this won't result in launchd duplicating the messages. Reviewed by Dan Bernstein. * wtf/Assertions.cpp: (vprintf_stderr_common): 2012-01-05 Mark Rowe <mrowe@apple.com> <http://webkit.org/b/75605> WTF logging functions should call vprintf_stderr_common only once per line Several of the WTF logging functions make multiple calls to vprintf_stderr_common to output a single line of text. This results in strangely formatted output if vprintf_stderr_common is retargeted to an output device that is message-oriented (such as ASL) rather than stream-oriented like stderr. Reviewed by Dan Bernstein. * wtf/Assertions.cpp: (vprintf_stderr_with_prefix): Helper function to prepend a given prefix on to the given format string before handing it off to vprintf_stderr_common. This requires disabling warnings about calling a printf-like function with a non-literal format string for this piece of code. It's safe in this particular case as vprintf_stderr_with_prefix is only ever given a literal prefix. (vprintf_stderr_with_trailing_newline): Helper function to append a trailling newline on to the given format string if one does not already exist. It requires the same treatment with regards to the non-literal format string warning. (WTFReportAssertionFailureWithMessage): Switch to using vprintf_stderr_with_prefix. (WTFReportBacktrace): Switch from calling fprintf directly to using fprintf_stderr_common. (WTFReportFatalError): Switch to using vprintf_stderr_with_prefix. (WTFReportError): Ditto. (WTFLog): Switch to using vprintf_stderr_with_trailing_newline. (WTFLogVerbose): Ditto. 2012-01-04 Gavin Barraclough <barraclough@apple.com> unshift/pop fifo may consume excessive memory https://bugs.webkit.org/show_bug.cgi?id=75588 Reviewed by Sam Weinig. The Array object commonly store data in a vector, consisting of a portion that is in use, a pre-capacity (m_indexBias) and a post-capacity (the delta between m_length and m_vectorLength). Calls to pop with grow the post-capacity, and the current algorithm for increasePrefixVectorLength (used by unshift) will never stink the post-capacity, so a unshift/pop fifo may consume an inordinate amount of memory, whilst having a relatively small active length. * runtime/JSArray.cpp: (JSC::storageSize): - sizeof(JSValue) should be sizeof(WriteBarrier<Unknown>) (JSC::SparseArrayValueMap::put): - sizeof(JSValue) should be sizeof(WriteBarrier<Unknown>) (JSC::JSArray::increaseVectorLength): - sizeof(JSValue) should be sizeof(WriteBarrier<Unknown>) (JSC::JSArray::unshiftCountSlowCase): - renamed from increaseVectorPrefixLength (this was a bad name, since it also moved the ArrayStorage header), rewritten. (JSC::JSArray::shiftCount): - sizeof(JSValue) should be sizeof(WriteBarrier<Unknown>), count should be unsigned (JSC::JSArray::unshiftCount): - sizeof(JSValue) should be sizeof(WriteBarrier<Unknown>), count should be unsigned, increaseVectorPrefixLength renamed to unshiftCountSlowCase (JSC::JSArray::sortNumeric): * runtime/JSArray.h: - Updated function declarations, m_indexBias should be unsigned. 2012-01-04 Mark Rowe <mrowe@apple.com> <http://webkit.org/b/75604> All instances of JSC::ArgumentsData appear to be leaked by JSC::Arguments Since JSC::Arguments has an OwnPtr for a member it needs to override destroy to ensure that the correct destructor is invoked. This is necessary because JSCell subclasses all intentionally have non-virtual destructors. Reviewed by Filip Pizlo. * runtime/Arguments.cpp: (JSC::Arguments::destroy): * runtime/Arguments.h: 2012-01-04 Filip Pizlo <fpizlo@apple.com> Unreviewed, accidentally turned off the JIT in previous commit. Turning it back on. * wtf/Platform.h: 2012-01-04 Filip Pizlo <fpizlo@apple.com> Changed "return" to "break" in some macrology I introduced in http://trac.webkit.org/changeset/104086. This is a benign change, as "return" was technically correct for all uses of the macro. Reviewed by Oliver Hunt. * dfg/DFGGraph.cpp: * wtf/Platform.h: 2012-01-04 Michael Saboff <msaboff@apple.com> StructureStubInfo not reset when corresponding MethodCallLinkInfo is reset https://bugs.webkit.org/show_bug.cgi?id=75583 Reviewed by Filip Pizlo. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::finalizeUnconditionally): Find the corresponding StructureStubInfo and reset the appropriate JIT and the StructureStubInfo itself when reseting a MethodCallLinkInfo. 2012-01-04 Michael Saboff <msaboff@apple.com> Invalid ASSERT() in DFGRepatch.cpp near line 385 https://bugs.webkit.org/show_bug.cgi?id=75584 Reviewed by Filip Pizlo. * dfg/DFGRepatch.cpp: (JSC::DFG::tryBuildGetByIDProtoList): Fixed ASSERT to use ==. 2012-01-04 Filip Pizlo <fpizlo@apple.com> Incorrect use of DFG node reference counts when mutating the graph https://bugs.webkit.org/show_bug.cgi?id=75580 <rdar://problem/10644607> Reviewed by Oliver Hunt. Made deref(node) follow the pattern of ref(node), which it should have to begin with. * dfg/DFGGraph.cpp: (JSC::DFG::Graph::refChildren): (JSC::DFG::Graph::derefChildren): * dfg/DFGGraph.h: (JSC::DFG::Graph::deref): (JSC::DFG::Graph::clearAndDerefChild1): (JSC::DFG::Graph::clearAndDerefChild2): (JSC::DFG::Graph::clearAndDerefChild3): * dfg/DFGNode.h: (JSC::DFG::Node::deref): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::fixupNode): 2012-01-04 Tor Arne Vestbø <tor.arne.vestbo@nokia.com> [Qt] Introduce new qmake variable 'WEBKIT' for signaling dependencies The custom qmake variable 'WEBKIT' is used for signaling that a target depends in some way on other subproject of the WebKit project. For now this is limited to the set of intermediate libraries: wtf, javascriptcore, webcore, and webkit2. This replaces the previous convension of using load(foo) for just include paths, and CONFIG += foo to also link against foo. Adding a dependency results in additional include paths being available, and potentially linking to the library. This is decided by the build system based on conditions such as what kind of target is being built and the general build config. An advantage to his approach is that it simplifies the individual foo.prf files, for example by allowing us to use INCLUDEPATH += and LIBS += as normal instead of prepending. Reviewed by Simon Hausmann. * Target.pri: * jsc.pro: * wtf/wtf.pro: 2012-01-03 Filip Pizlo <fpizlo@apple.com> DFG: The assertion that a double-voted variable cannot become double-unvoted is wrong https://bugs.webkit.org/show_bug.cgi?id=75516 <rdar://problem/10640266> Reviewed by Gavin Barraclough. Removed the offending assertion, since it was wrong. Also hardened the code to make this case less likely by first having the propagator fixpoint converge, and then doing double voting combined with a second fixpoint. This is neutral on benchmarks and fixes the assertion in a fairly low-risk way (i.e. we won't vote a variable double until we've converged to the conclusion that it really is double). * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagatePredictions): * dfg/DFGVariableAccessData.h: (JSC::DFG::VariableAccessData::tallyVotesForShouldUseDoubleFormat): 2012-01-03 Filip Pizlo <fpizlo@apple.com> REGRESSION (r98196-98236): Incorrect layout of iGoogle with RSS feeds https://bugs.webkit.org/show_bug.cgi?id=75303 <rdar://problem/10633533> Reviewed by Gavin Barraclough. The this argument was not being kept alive in some cases during inlining and intrinsic optimizations. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::emitFunctionCheck): (JSC::DFG::ByteCodeParser::handleInlining): 2012-01-03 Gavin Barraclough <barraclough@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-01-03 Gavin Barraclough <barraclough@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-01-03 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=75140 Reviewed by Sam Weinig. Rewrite JSArray::putSlowCase to be much cleaner & simpler. This rewrite only significantly changes behaviour for sparse array, specifically in how sparse arrays are reified back to vector form. This does not affect arrays with less than 10000 entries (since these always use a vector). The more common cases of sparse array behavior (though large sparse arrays are rare) - arrays that always remain sparse, and arrays that are filled in reverse sequential order - should be just as fast or faster (since reification is simpler & no longer requires map lookups) after these changes. Simplifying this code allows all cases of putByIndex that need to grow the vector to do so via increaseVectorLength, which means that this method can encapsulate the policy of determining how the vector should be grown. No performance impact. * runtime/JSArray.cpp: (JSC::isDenseEnoughForVector): - any array of length <= MIN_SPARSE_ARRAY_INDEX is dense enough for a vector. (JSC::JSArray::putByIndex): - simplify & comment. (JSC::JSArray::putByIndexBeyondVectorLength): - Re-written to be much clearer & simpler. (JSC::JSArray::increaseVectorLength): (JSC::JSArray::increaseVectorPrefixLength): - add explicit checks against MAX_STORAGE_VECTOR_LENGTH, so clients do not need do so. (JSC::JSArray::push): - simplify & comment. * runtime/JSArray.h: - removed SparseArrayValueMap::take. 2012-01-03 Gavin Barraclough <barraclough@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2012-01-03 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=75140 Reviewed by Sam Weinig. Simplify JSArray creation - remove ArgsList/JSValue* create methods (this functionality can be implemented in terms of tryCreateUninitialized). * JavaScriptCore.exp: * runtime/ArrayConstructor.cpp: - use constructArray/constructEmptyArray instead of calling JSArray::create directly (JSC::constructArrayWithSizeQuirk): * runtime/JSArray.cpp: * runtime/JSArray.h: - removed ArgsList/JSValue* create methods * runtime/JSGlobalObject.h: (JSC::constructEmptyArray): (JSC::constructArray): - changed to be implemented in terms of JSArray::tryCreateUninitialized 2012-01-03 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=75429 ThrowTypeError should be a singleton object Reviewed by Sam Weinig. Per section 13.2.3 of the spec. We could change setAccessorDescriptor to be able to share the global GetterSetter object, rather than storing the accessor functions and creating a new GetterSetter in defineProperty - but this won't be a small change to PropertyDescriptors (and would probably mean making GetterSetter objects immutable?) - so I'll leave that for another patch. * JavaScriptCore.exp: - don't export setAccessorDescriptor * runtime/Arguments.cpp: (JSC::Arguments::createStrictModeCallerIfNecessary): (JSC::Arguments::createStrictModeCalleeIfNecessary): - call throwTypeErrorGetterSetter instead of createTypeErrorFunction * runtime/Error.cpp: * runtime/Error.h: - remove createTypeErrorFunction * runtime/JSFunction.cpp: * runtime/JSFunction.h: - remove unused createDescriptorForThrowingProperty * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): (JSC::JSGlobalObject::visitChildren): - removed m_strictModeTypeErrorFunctionStructure. * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::internalFunctionStructure): - removed m_strictModeTypeErrorFunctionStructure. * runtime/PropertyDescriptor.cpp: (JSC::PropertyDescriptor::setAccessorDescriptor): - changed to take a GetterSetter * runtime/PropertyDescriptor.h: - changed to take a GetterSetter 2012-01-02 Gavin Barraclough <barraclough@apple.com> Check in fixes for jsc tests following bug #75455. * tests/mozilla/ecma/GlobalObject/15.1.2.2-1.js: * tests/mozilla/ecma/GlobalObject/15.1.2.2-2.js: 2012-01-02 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=75452 If argument to Error is undefined, message is not set Reviewed by Sam Weinig. Per section 15.11.1.1 of the spec. * runtime/ErrorInstance.h: (JSC::ErrorInstance::create): (JSC::ErrorInstance::finishCreation): 2012-01-02 Gavin Barraclough <barraclough@apple.com> ES5 prohibits parseInt from supporting octal https://bugs.webkit.org/show_bug.cgi?id=75455 Reviewed by Sam Weinig. See sections 15.1.2.2 and annex E. * runtime/JSGlobalObjectFunctions.cpp: (JSC::parseInt): 2012-01-02 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=55343 Global JSON should be configurable but isn't Reviewed by Sam Weinig. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): - make JSON configurable 2012-01-01 Filip Pizlo <fpizlo@apple.com> Call instructions should leave room for linking information https://bugs.webkit.org/show_bug.cgi?id=75422 <rdar://problem/10633985> Reviewed by Oliver Hunt. * bytecode/Opcode.h: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitCall): (JSC::BytecodeGenerator::emitConstruct): 2011-12-31 Dan Bernstein <mitz@apple.com> Continue trying to fix the Windows build after r103823. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-12-31 Dan Bernstein <mitz@apple.com> Start trying to fix the Windows build after r103823. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-12-30 Anders Carlsson <andersca@apple.com> Add a ParamStorageTraits specialization for RetainPtr https://bugs.webkit.org/show_bug.cgi?id=75392 Reviewed by Daniel Bates. * wtf/Functional.h: Add a partial specialization of ParamStorageTraits for RetainPtr<T>. * wtf/RetainPtr.h: Bring in the retainPtr function template from WTF. 2011-12-29 Sam Weinig <sam@webkit.org> It should be easier to iterate a Vector backwards https://bugs.webkit.org/show_bug.cgi?id=75359 Reviewed by Anders Carlsson. Adds Vector::rbegin(), Vector::rend(), and Vector::reversed(), a new proxy driven way to access a vector backwards. One can use reversed() in a range-based for loop like so: for (auto val: myVector.reversed()) doSomething(val) * wtf/Vector.h: (WTF::Vector::~Vector): Fix style. (WTF::Vector::rbegin): (WTF::Vector::rend): Added using standard adaptor std::reverse_iterator. (WTF::Vector::reversed): (WTF::Vector::VectorReverseProxy::begin): (WTF::Vector::VectorReverseProxy::end): Add proxy similar to one used in HashMap for keys() and values() which allows access to a Vector backwards for use in range-based for loops. 2011-12-29 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=75140 Reviewed by Oliver Hunt. Start cleaning up JSArray construction. JSArray has a set of create methods, one of which (currently) takes a 'creation mode' enum parameter. Based on that parameter, the constructor does one of two completely different things. If the parameter is 'CreateInitialized' it creates an array, setting the length, but does not eagerly allocate a storage vector of the specified length. A small (BASE_VECTOR_LEN sized) initial vector will be allocated, and cleared, property access to the vector will read the hole value (return undefined). The alternate usage of this method ('CreateCompact') does something very different. It tries to create an array of the requested length, and also allocates a storage vector large enough to hold all properties. It does not clear the storage vector, leaving the memory uninitialized and requiring the user to call a method 'uncheckedSetIndex' to initialize values in the vector. This patch factors out these two behaviours, moving the 'CreateCompact' mode into its own method, 'tryCreateUninitialized' (matching the naming for this functionality in the string classes). 'tryCreateUninitialized' may return 0 if memory allocation fails during construction of the object. The construction pattern changes such that values added during initialization will be marked if a GC is triggered during array allocation. 'CreateInitialized' no longer need be passed to create a normal, fully constructed array with a length, and this method is merged with the version of 'create' that does not take an initial length (length parameter defaults to 0). * JavaScriptCore.exp: * runtime/ArrayConstructor.cpp: (JSC::constructArrayWithSizeQuirk): - removed 'CreateInitialized' argument * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncSplice): - changed to call 'tryCreateUninitialized' * runtime/FunctionPrototype.cpp: (JSC::functionProtoFuncBind): - changed to call 'tryCreateUninitialized' * runtime/JSArray.cpp: (JSC::JSArray::JSArray): - initialize m_storage to null; if construction fails, make destruction safe (JSC::JSArray::finishCreation): - merge versions of this method, takes an initialLength parameter defaulting to zero (JSC::JSArray::tryFinishCreationUninitialized): - version of 'finishCreation' that tries to eagerly allocate storage; may fail & return 0 (JSC::JSArray::~JSArray): - check for null m_storage, in case array construction failed. (JSC::JSArray::increaseVectorPrefixLength): * runtime/JSArray.h: (JSC::JSArray::create): - merge versions of this method, takes an initialLength parameter defaulting to zero (JSC::JSArray::tryCreateUninitialized): - version of 'create' that tries to eagerly allocate storage; may fail & return 0 (JSC::JSArray::initializeIndex): (JSC::JSArray::completeInitialization): - used in conjunction with 'tryCreateUninitialized' to initialize the array * runtime/JSGlobalObject.h: (JSC::constructEmptyArray): - removed 'CreateInitialized' argument * runtime/RegExpConstructor.cpp: (JSC::RegExpMatchesArray::finishCreation): - removed 'CreateInitialized' argument 2011-12-29 Anders Carlsson <andersca@apple.com> Add a retainPtr function template https://bugs.webkit.org/show_bug.cgi?id=75365 Reviewed by Dan Bernstein. This makes it easier to make a RetainPtr using template argument deduction, which is useful when passing RetainPtr objects as function arguments. * wtf/RetainPtr.h: (WTF::retainPtr): 2011-12-28 Yuqiang Xian <yuqiang.xian@intel.com> spill unboxed values in DFG 32_64 https://bugs.webkit.org/show_bug.cgi?id=75291 Reviewed by Filip Pizlo. Currently all the values are spilled as boxed in DFG 32_64, which is not necessary and introduces additional stores/loads. Instead we can spill them as unboxed if feasible. It can be applied to the Integers, Cells and Booleans in DFG 32_64. Doubles are left as is because they don't need to be boxed at all. The modifications to the spill/fill and the OSR exit are required, as well as a bug fix to the "isUnknownJS" logic. * bytecode/ValueRecovery.h: (JSC::ValueRecovery::displacedInRegisterFile): (JSC::ValueRecovery::virtualRegister): (JSC::ValueRecovery::dump): * dfg/DFGGenerationInfo.h: (JSC::DFG::GenerationInfo::isUnknownJS): (JSC::DFG::GenerationInfo::spill): * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::isKnownNotBoolean): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::silentFillGPR): (JSC::DFG::SpeculativeJIT::spill): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillInteger): (JSC::DFG::SpeculativeJIT::fillDouble): (JSC::DFG::SpeculativeJIT::fillJSValue): (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compile): 2011-12-28 Anders Carlsson <andersca@apple.com> Add an implicit block conversion operator to WTF::Function https://bugs.webkit.org/show_bug.cgi?id=75325 Reviewed by Dan Bernstein. * wtf/Compiler.h: Add a define for COMPILER_SUPPORTS(BLOCKS). It's only defined for clang, since the gcc blocks implementation is buggy, especially when it comes to C++. * wtf/Functional.h: Add a block conversion operator that creates and returns an autoreleased block that will call the function when executed. 2011-12-27 Anders Carlsson <andersca@apple.com> Add a new WTF::bind overload that takes 6 parameters https://bugs.webkit.org/show_bug.cgi?id=75287 Reviewed by Sam Weinig. * wtf/Functional.h: 2011-12-27 Sam Weinig <sam@webkit.org> Continue moving compiler feature checks to use the COMPILER_SUPPORTS() macro https://bugs.webkit.org/show_bug.cgi?id=75268 Reviewed by Anders Carlsson. * wtf/Compiler.h: Add support for COMPILER_SUPPORTS(CXX_NULLPTR) and COMPILER_SUPPORTS(CXX_DELETED_FUNCTIONS). * wtf/Noncopyable.h: Use COMPILER_SUPPORTS(CXX_DELETED_FUNCTIONS). * wtf/NullPtr.cpp: * wtf/NullPtr.h: Use COMPILER_SUPPORTS(CXX_NULLPTR). Remove support for HAVE(NULLPTR). * wtf/RefPtr.h: * wtf/RetainPtr.h: Switch from HAVE(NULLPTR) to COMPILER_SUPPORTS(CXX_NULLPTR). 2011-12-27 Anders Carlsson <andersca@apple.com> Misc fixes and cleanups in Functional.h https://bugs.webkit.org/show_bug.cgi?id=75281 Reviewed by Andreas Kling. - Reformat template declarations so that the class begins on a new line. - Change the parameter template parameters to start at P1 instead of P0. - Add function wrappers and bind overloads for 4 and 5 parameter functions. - Change the Function call operator to be const so const functions can be called. * wtf/Functional.h: 2011-12-27 Tony Chang <tony@chromium.org> [chromium] Minor cleanup of gyp files. https://bugs.webkit.org/show_bug.cgi?id=75269 Reviewed by Adam Barth. * JavaScriptCore.gyp/JavaScriptCore.gyp: msvs_guid is no longer needed and vim/emacs specific hooks should be added by the user. 2011-12-27 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=75260 Null name for host function can result in dereference of uninitialize memory Reviewed by Filip Pizlo. This is a recent regression in ToT, if the name passed to finishCreation of a host function is null, we are currently skipping the putDirect, which leaves memory uninitialized. This patch reverts the aspect of the change that introduced the issue. It might be better if functions that don't have a name don't have this property at all, but that's change should be separate from fixing the bug. * runtime/JSFunction.cpp: (JSC::JSFunction::finishCreation): - Always initialize the name property. 2011-12-27 Anders Carlsson <andersca@apple.com> Function should handle wrapping/unwrapping RefPtr and PassRefPtr https://bugs.webkit.org/show_bug.cgi?id=75266 Reviewed by Sam Weinig. Add ParamStorageTraits that can be used for deciding how bound parameters should be stored and peeked at. For RefPtr we want to use the raw pointer when "peeking" to avoid ref-churn. For PassRefPtr, we want to use RefPtr for storage but still use the raw pointer when peeking. * wtf/Functional.h: (WTF::ParamStorageTraits::wrap): (WTF::ParamStorageTraits::unwrap): 2011-12-27 Tony Chang <tony@chromium.org> [chromium] really enable wpo for WebCore libs and for WTF https://bugs.webkit.org/show_bug.cgi?id=75264 Reviewed by Adam Barth. * JavaScriptCore.gyp/JavaScriptCore.gyp: Enable WPO for wtf and yarr. 2011-12-26 Gavin Barraclough <barraclough@apple.com> Errk! OS X build fix. * JavaScriptCore.exp: 2011-12-26 Gavin Barraclough <barraclough@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSObject.h: 2011-12-26 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=75231 Fail to throw in strict mode on assign to read only static properties Reviewed by Filip Pizlo. There are three bugs here: * symbolTablePut should throw for strict mode accesses. * lookupPut should throw for strict mode accesses. * NumberConstructor should override put to call lookupPut, to trap assignment to readonly properties. * runtime/JSActivation.cpp: (JSC::JSActivation::symbolTablePut): (JSC::JSActivation::put): * runtime/JSActivation.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::put): * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::put): * runtime/JSVariableObject.h: (JSC::JSVariableObject::symbolTablePut): * runtime/Lookup.h: (JSC::lookupPut): * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::put): * runtime/NumberConstructor.h: 2011-12-26 Gavin Barraclough <barraclough@apple.com> Fix miss-commit of utf8 change. Reviewed by Filip Pizlo Eeep, patch as landed a while ago had no effect! - acidentally landed modified version of patch used for performance testing. (This isn't covered by layout tests because layour tests don't use jsc, and the tests/mozilla tests use latin1, which was already supported!) Landing changes as intended (and as originally reviewed). * jsc.cpp: (jscSource): 2011-12-26 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix for ARMv7. * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::load16Signed): (JSC::MacroAssemblerARMv7::load8Signed): 2011-12-26 Hajime Morrita <morrita@google.com> Rename WTF_INLINE, JS_INLINE to HIDDEN_INLINE https://bugs.webkit.org/show_bug.cgi?id=74990 Reviewed by Kevin Ollivier. * runtime/JSExportMacros.h: Removed JS_INLINE * wtf/ExportMacros.h: Renamed WTF_INLINE to HIDDEN_INLINE 2011-12-24 Filip Pizlo <fpizlo@apple.com> The ArgumentCount field in the CallFrame should have its tag left blank for other uses https://bugs.webkit.org/show_bug.cgi?id=75199 <rdar://problem/10625105> <rdar://problem/10625106> Reviewed by Oliver Hunt. * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::argumentPayloadSlot): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): * interpreter/CallFrame.h: (JSC::ExecState::argumentCountIncludingThis): (JSC::ExecState::setArgumentCountIncludingThis): * interpreter/Register.h: (JSC::Register::unboxedInt32): (JSC::Register::unboxedBoolean): (JSC::Register::unboxedCell): (JSC::Register::payload): (JSC::Register::tag): * jit/JITCall.cpp: (JSC::JIT::compileOpCall): * jit/JITCall32_64.cpp: (JSC::JIT::compileLoadVarargs): (JSC::JIT::compileOpCall): 2011-12-25 Andreas Kling <awesomekling@apple.com> Yarr: Avoid copying vectors in CharacterClassConstructor. <http://webkit.org/b/75206> Reviewed by Darin Adler. Yarr::CharacterClassConstructor::charClass() was hot when loading twitter feeds (1.2%), replace the usage of Vector::append() by swap() since we're always clearing the source vector afterwards anyway. * yarr/YarrPattern.cpp: (JSC::Yarr::CharacterClassConstructor::charClass): 2011-12-24 Darin Adler <darin@apple.com> Specialize HashTraits for RefPtr to use PassRefPtr as "pass type" to reduce reference count churn https://bugs.webkit.org/show_bug.cgi?id=72476 Reviewed by Sam Weinig. * wtf/HashTraits.h: Defined PassInType and store function in HashTraits<RefPtr>. 2011-12-23 Geoffrey Garen <ggaren@apple.com> Inlined Yarr::execute https://bugs.webkit.org/show_bug.cgi?id=75180 Reviewed reluctantly by Beth Dakin. Tiny speedup on SunSpider string tests. Removes some samples from Instruments. A step toward removing -fomit-frame-pointer. * yarr/YarrJIT.cpp: * yarr/YarrJIT.h: (JSC::Yarr::execute): ONE LINE FUNCTION, Y U NOT INLINED?! 2011-12-23 Filip Pizlo <fpizlo@apple.com> DFG loads from signed 8-bit and 16-bit typed arrays are broken https://bugs.webkit.org/show_bug.cgi?id=75163 Reviewed by Geoffrey Garen. Added 8-bit and 16-bit signed loads. Because doing so on ARM is less trivial, I'm currently disabling Int8Array and Int16Array optimizations on ARM. * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::load8Signed): (JSC::MacroAssemblerX86Common::load16Signed): * assembler/X86Assembler.h: (JSC::X86Assembler::movswl_mr): (JSC::X86Assembler::movsbl_mr): * bytecode/PredictedType.h: (JSC::isActionableMutableArrayPrediction): * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateInt8Array): (JSC::DFG::Node::shouldSpeculateInt16Array): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileGetByValOnIntTypedArray): 2011-12-23 Filip Pizlo <fpizlo@apple.com> DFG does double-to-int conversion incorrectly when storing into int typed arrays https://bugs.webkit.org/show_bug.cgi?id=75164 <rdar://problem/10557547> Reviewed by Geoffrey Garen. * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::branchTruncateDoubleToUint32): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::branchTruncateDoubleToUint32): (JSC::MacroAssemblerX86Common::truncateDoubleToUint32): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): 2011-12-23 Geoffrey Garen <ggaren@apple.com> Refactored String.prototype.replace https://bugs.webkit.org/show_bug.cgi?id=75114 Reviewed by Darin Adler. No performance difference. I think this is a step toward removing -fomit-frame-pointer. * runtime/JSString.cpp: * runtime/JSString.h: Removed the test and special case for a single-character search string because the standard path does this test and special case for us. (As an aside, if we do come up with a unique single-character replace optimization in future, it probably belongs in the replace function, and not in JSString.) * runtime/StringPrototype.cpp: (JSC::stringProtoFuncReplace): Split this mega-sized function into: (JSC::replaceUsingStringSearch): - This reasonably sized function, and (JSC::replaceUsingRegExpSearch): - This still mega-sized function. 2011-12-23 Pierre Rossi <pierre.rossi@gmail.com> [Qt] REGRESSION(r103467): It broke fast/images/animated-gif-restored-from-bfcache.html https://bugs.webkit.org/show_bug.cgi?id=75087 monotonicallyIncreasingTime needs to hava a higher resolution than milliseconds. Reviewed by Darin Adler. * wtf/CurrentTime.cpp: (WTF::monotonicallyIncreasingTime): 2011-12-22 Filip Pizlo <fpizlo@apple.com> DFG should not speculate array even when predictions say that the base is not an array https://bugs.webkit.org/show_bug.cgi?id=75160 <rdar://problem/10622646> <rdar://problem/10622649> Reviewed by Oliver Hunt. Added the ability to call slow path when the base is known to not be an array. Also rationalized the logic for deciding when the index is not an int, and cleaned up the logic for deciding when to speculate typed array. Neutral for the most part, with odd speed-ups and slow-downs. The slow-downs can likely be mitigated by having the notion of a polymorphic array access, where we try, but don't speculate, to access the array one way before either trying some other ways or calling slow path. * bytecode/PredictedType.h: (JSC::isActionableMutableArrayPrediction): (JSC::isActionableArrayPrediction): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateInt8Array): (JSC::DFG::Node::shouldSpeculateInt16Array): (JSC::DFG::Node::shouldSpeculateInt32Array): (JSC::DFG::Node::shouldSpeculateUint8Array): (JSC::DFG::Node::shouldSpeculateUint16Array): (JSC::DFG::Node::shouldSpeculateUint32Array): (JSC::DFG::Node::shouldSpeculateFloat32Array): (JSC::DFG::Node::shouldSpeculateFloat64Array): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::byValIsPure): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileGetIndexedPropertyStorage): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-12-22 Gavin Barraclough <barraclough@apple.com> Unreviewed - fix stylebot issues from last patch. * runtime/JSArray.cpp: (JSC::JSArray::putSlowCase): 2011-12-22 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=75151 Add attributes field to JSArray's SparseMap Reviewed by Sam Weinig. This will be necessary to be able to support non- writable/configurable/enumerable properties, and helpful for getters/setters. Added a concept of being 'inSparseMode' - this indicates the array has a non-standard * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncSort): - JSArray::sort methods not allowed on arrays that are 'inSparseMode'. (must fall back to generic sort alogrithm). * runtime/JSArray.cpp: (JSC::JSArray::finishCreation): - moved reportedMapCapacity into the SparseArrayValueMap object. (JSC::SparseArrayValueMap::find): (JSC::SparseArrayValueMap::put): (JSC::SparseArrayValueMap::visitChildren): - Added. (JSC::JSArray::getOwnPropertySlotByIndex): (JSC::JSArray::getOwnPropertyDescriptor): (JSC::JSArray::putSlowCase): (JSC::JSArray::deletePropertyByIndex): (JSC::JSArray::getOwnPropertyNames): (JSC::JSArray::setLength): (JSC::JSArray::pop): (JSC::JSArray::visitChildren): - Updated for changes in SparseArrayValueMap. (JSC::JSArray::sortNumeric): (JSC::JSArray::sort): (JSC::JSArray::compactForSorting): - Disallow on 'SparseMode' arrays. * runtime/JSArray.h: (JSC::SparseArrayEntry::SparseArrayEntry): - An entry in the sparse array - value (WriteBarrier) + attributes. (JSC::SparseArrayValueMap::SparseArrayValueMap): (JSC::SparseArrayValueMap::sparseMode): (JSC::SparseArrayValueMap::setSparseMode): - Flags to track whether an Array is forced into SparseMode. (JSC::SparseArrayValueMap::remove): (JSC::SparseArrayValueMap::notFound): (JSC::SparseArrayValueMap::isEmpty): (JSC::SparseArrayValueMap::contains): (JSC::SparseArrayValueMap::size): (JSC::SparseArrayValueMap::begin): (JSC::SparseArrayValueMap::end): - accessors to the map (JSC::SparseArrayValueMap::take): - only for use on non-SpareMode arrays. (JSC::JSArray::inSparseMode): - Added. 2011-12-22 Filip Pizlo <fpizlo@apple.com> DFG CFA sometimes generates an incorrect proof that a node is known to be a typed array https://bugs.webkit.org/show_bug.cgi?id=75150 <rdar://problem/10621900> Reviewed by Gavin Barraclough. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): 2011-12-22 Filip Pizlo <fpizlo@apple.com> DFG JIT does exactly the wrong thing when doing strict equality on two known cells https://bugs.webkit.org/show_bug.cgi?id=75138 <rdar://problem/10621526> Reviewed by Oliver Hunt. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeStrictEq): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeStrictEq): 2011-12-22 Balazs Kelemen <kbalazs@webkit.org> Fix debug build with assertions disabled https://bugs.webkit.org/show_bug.cgi?id=75075 Reviewed by Darin Adler. Check whether assertions are disabled instead of NDEBUG where appropriate to avoid "defined but not used" warnings. * wtf/DateMath.cpp: (WTF::initializeDates): 2011-12-22 Mariusz Grzegorczyk <mariusz.g@samsung.com> [EFL] Missing plugins support for efl port https://bugs.webkit.org/show_bug.cgi?id=44505 Reviewed by Anders Carlsson. Add define of ENABLE_PLUGIN_PACKAGE_SIMPLE_HASH for efl port. * wtf/Platform.h: 2011-12-22 Wei Charles <charles.wei@torchmobile.com.cn> Remove un-used data member of LiteralParser::Lex::m_string https://bugs.webkit.org/show_bug.cgi?id=68216 Reviewed by George Staikos. * runtime/LiteralParser.h: 2011-12-21 Dan Bernstein <mitz@apple.com> OS X build fix after r103488. * JavaScriptCore.exp: 2011-12-21 Konrad Piascik <kpiascik@rim.com> Implement the JavaScriptCore bindings for eventListenerHandlerLocation https://bugs.webkit.org/show_bug.cgi?id=74313 Reviewed by Eric Seidel. Updated project files to get Windows and Mac builds working. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.xcodeproj/project.pbxproj: 2011-12-21 Filip Pizlo <fpizlo@apple.com> DFG ConvertThis optimizations do not honor the distinction between the global object and the global this object https://bugs.webkit.org/show_bug.cgi?id=75058 <rdar://problem/10616612> <rdar://problem/10617500> Reviewed by Oliver Hunt. Added a call to toThisObject() in the DFG when planting a direct reference to the global this object. Instead of adding a separate toThisObject() method on JSCell which does not take ExecState*, I reascribed a new contract: if you're calling toThisObject() on JSObject or one of its subtypes, then the ExecState* is optional. * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::globalThisObjectFor): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/JSObject.h: 2011-12-21 Pierre Rossi <pierre.rossi@gmail.com> Implement montonicallyIncreasingClock() on Qt https://bugs.webkit.org/show_bug.cgi?id=62159 Reviewed by Darin Adler. * wtf/CurrentTime.cpp: (WTF::monotonicallyIncreasingTime): 2011-12-20 Filip Pizlo <fpizlo@apple.com> 32_64 baseline JIT should attempt to convert division results to integers, and record when that fails https://bugs.webkit.org/show_bug.cgi?id=74997 <rdar://problem/10612389> Reviewed by Gavin Barraclough. * jit/JITArithmetic32_64.cpp: (JSC::JIT::emit_op_div): 2011-12-20 Filip Pizlo <fpizlo@apple.com> JavaScriptCore should be consistent about how it reads and writes ArgumentCount https://bugs.webkit.org/show_bug.cgi?id=74989 <rdar://problem/10612006> Reviewed by Gavin Barraclough. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JITCall32_64.cpp: (JSC::JIT::compileLoadVarargs): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_get_arguments_length): (JSC::JIT::emit_op_get_argument_by_val): * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::SpecializedThunkJIT): 2011-12-20 Filip Pizlo <fpizlo@apple.com> Value Profiles for arguments should be more easily accessible to the interpreter https://bugs.webkit.org/show_bug.cgi?id=74984 <rdar://problem/10611364> Reviewed by Gavin Barraclough. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::stronglyVisitStrongReferences): (JSC::CodeBlock::shouldOptimizeNow): (JSC::CodeBlock::dumpValueProfiles): * bytecode/CodeBlock.h: (JSC::CodeBlock::setArgumentValueProfileSize): (JSC::CodeBlock::numberOfArgumentValueProfiles): (JSC::CodeBlock::valueProfileForArgument): (JSC::CodeBlock::addValueProfile): (JSC::CodeBlock::valueProfile): (JSC::CodeBlock::valueProfileForBytecodeOffset): (JSC::CodeBlock::totalNumberOfValueProfiles): (JSC::CodeBlock::getFromAllValueProfiles): * bytecode/ValueProfile.h: (JSC::ValueProfile::ValueProfile): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JIT.h: * jit/JITInlineMethods.h: (JSC::JIT::emitValueProfilingSite): 2011-12-20 Gavin Barraclough <barraclough@apple.com> JSC shell should accept utf8 input. Reviewed by Filip Pizlo. * jsc.cpp: (jscSource): (functionRun): (functionLoad): (functionCheckSyntax): (runWithScripts): (runInteractive): 2011-12-20 Gavin Barraclough <barraclough@apple.com> Rubber Stamped by Sam Weinig * runtime/JSGlobalData.cpp: - removed some dead code. 2011-12-19 Geoffrey Garen <ggaren@apple.com> Tightened up Vector<T>::append https://bugs.webkit.org/show_bug.cgi?id=74906 Reviewed by Sam Weinig. Not a measurable speedup, but code inspection shows better code generated, and I believe this is a step toward turning off -fomit-frame-pointer. * wtf/Vector.h: (WTF::::append): (WTF::::appendSlowCase): Split out the slow case into a separate function to keep unnecessary instructions off the hot path. This means the hot path can now be inlined more often. Removed some old MSVC7 cruft. Hopefully, we don't need to hang on to a compiler work-around from 2007. 2011-12-19 Yuqiang Xian <yuqiang.xian@intel.com> Temporary GPR should not be lazily allocated in DFG JIT on X86 https://bugs.webkit.org/show_bug.cgi?id=74908 Reviewed by Filip Pizlo. On X86, we used to allocate a temporary GPR lazily when it's really used rather than defined. This may cause potential issues of allocating registers inside control flow and result in problems in subsequent code generation, for example the DFG JIT may think an operand already being spilled (to satisfy the allocation request) and generate code to read the data from memory, but the allocation and spilling are in a branch which is not taken at runtime, so the generated code is incorrect. Although current DFG JIT code doesn't have this problematic pattern, it's better to cut-off the root to avoid any potential issues in the future. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::GPRTemporary::GPRTemporary): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::GPRTemporary::gpr): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-12-19 Yuqiang Xian <yuqiang.xian@intel.com> Remove unused code for non-speculative Arith operations from DFG JIT https://bugs.webkit.org/show_bug.cgi?id=74905 Reviewed by Filip Pizlo. * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: * dfg/DFGSpeculativeJIT64.cpp: 2011-12-19 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=74903 Exceptions not thrown correctly from DFG JIT on 32bit Reviewed by Oliver Hunt. Arguments for lookupExceptionHandler are not setup correctly. In the case of ARMv7 we rely on lr being preserved over a call, this in invalid. On x86 we don't should be poking the arguments onto the stack! * bytecode/CodeBlock.h: (JSC::CodeBlock::bytecodeOffsetForCallAtIndex): * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::restoreReturnAddressBeforeReturn): * dfg/DFGGPRInfo.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileBody): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::addExceptionCheck): (JSC::DFG::JITCompiler::addFastExceptionCheck): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: 2011-12-19 Filip Pizlo <fpizlo@apple.com> If we detect that we can use the JIT, don't use computed opcode lookups https://bugs.webkit.org/show_bug.cgi?id=74899 <rdar://problem/10604551> Reviewed by Gavin Barraclough. * interpreter/Interpreter.cpp: (JSC::Interpreter::Interpreter): (JSC::Interpreter::initialize): (JSC::Interpreter::privateExecute): * interpreter/Interpreter.h: (JSC::Interpreter::getOpcode): (JSC::Interpreter::getOpcodeID): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): 2011-12-19 Geoffrey Garen <ggaren@apple.com> Try to fix the Qt build. Unreviewed. * wtf/ThreadSpecific.h: #include! 2011-12-18 Filip Pizlo <fpizlo@apple.com> It should be possible to change the value of an Options variable without recompiling the world https://bugs.webkit.org/show_bug.cgi?id=74807 Reviewed by Gavin Barraclough. * runtime/Options.cpp: (JSC::Options::initializeOptions): * runtime/Options.h: 2011-12-19 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r103250. http://trac.webkit.org/changeset/103250 https://bugs.webkit.org/show_bug.cgi?id=74877 it still breaks codegen (Requested by olliej on #webkit). * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateArithNodeFlags): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::byValIsPure): (JSC::DFG::Propagator::clobbersWorld): (JSC::DFG::Propagator::getByValLoadElimination): (JSC::DFG::Propagator::checkStructureLoadElimination): (JSC::DFG::Propagator::getByOffsetLoadElimination): (JSC::DFG::Propagator::getPropertyStorageLoadElimination): (JSC::DFG::Propagator::getIndexedPropertyStorageLoadElimination): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): (JSC::DFG::SpeculativeJIT::compilePutByValForFloatTypedArray): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-12-16 Oliver Hunt <oliver@apple.com> Rolling r103120 back in with merge errors corrected. PutByVal[Alias] unnecessarily reloads the storage buffer https://bugs.webkit.org/show_bug.cgi?id=74747 Reviewed by Gavin Barraclough. Make PutByVal use GetIndexedStorage to load the storage buffer. This required switching PutByVal to a vararg node (which is responsible for most of the noise in this patch). This fixes the remaining portion of the kraken regression caused by the GetByVal storage load elimination, and a 1-5% win on some of the sub tests of the typed array benchmark at: http://stepheneb.github.com/webgl-matrix-benchmarks/matrix_benchmark.html * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateArithNodeFlags): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::byValIndexIsPure): (JSC::DFG::Propagator::clobbersWorld): (JSC::DFG::Propagator::getByValLoadElimination): (JSC::DFG::Propagator::checkStructureLoadElimination): (JSC::DFG::Propagator::getByOffsetLoadElimination): (JSC::DFG::Propagator::getPropertyStorageLoadElimination): (JSC::DFG::Propagator::getIndexedPropertyStorageLoadElimination): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): (JSC::DFG::SpeculativeJIT::compilePutByValForFloatTypedArray): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-12-15 Geoffrey Garen <ggaren@apple.com> Placement new does an unnecessary NULL check https://bugs.webkit.org/show_bug.cgi?id=74676 Reviewed by Sam Weinig. We can define our own version, which skips the NULL check. Not a measurable speedup, but code inspection shows better code generated, and I believe this is a step toward turning off -fomit-frame-pointer. * API/JSCallbackConstructor.h: (JSC::JSCallbackConstructor::create): * API/JSCallbackFunction.h: (JSC::JSCallbackFunction::create): Use the NotNull version of placement new to skip the NULL check. * API/JSCallbackObject.h: Removed a conflicting, unnecessaray placement new. (JSC::JSCallbackObject::create): * debugger/DebuggerActivation.h: (JSC::DebuggerActivation::create): * heap/HandleHeap.cpp: (JSC::HandleHeap::grow): * heap/HandleHeap.h: (JSC::HandleHeap::allocate): * heap/MarkedBlock.cpp: (JSC::MarkedBlock::create): (JSC::MarkedBlock::recycle): * jit/JITCode.h: (JSC::JITCode::clear): * jsc.cpp: (GlobalObject::create): * profiler/CallIdentifier.h: * runtime/Arguments.h: (JSC::Arguments::create): * runtime/ArrayConstructor.h: (JSC::ArrayConstructor::create): * runtime/ArrayPrototype.h: (JSC::ArrayPrototype::create): * runtime/BooleanConstructor.h: (JSC::BooleanConstructor::create): * runtime/BooleanObject.h: (JSC::BooleanObject::create): * runtime/BooleanPrototype.h: (JSC::BooleanPrototype::create): * runtime/DateConstructor.h: (JSC::DateConstructor::create): * runtime/DateInstance.h: (JSC::DateInstance::create): * runtime/DatePrototype.h: (JSC::DatePrototype::create): * runtime/Error.h: (JSC::StrictModeTypeErrorFunction::create): * runtime/ErrorConstructor.h: (JSC::ErrorConstructor::create): * runtime/ErrorInstance.h: (JSC::ErrorInstance::create): * runtime/ErrorPrototype.h: (JSC::ErrorPrototype::create): * runtime/ExceptionHelpers.h: (JSC::InterruptedExecutionError::create): (JSC::TerminatedExecutionError::create): * runtime/Executable.h: (JSC::NativeExecutable::create): (JSC::EvalExecutable::create): (JSC::ProgramExecutable::create): (JSC::FunctionExecutable::create): * runtime/FunctionConstructor.h: (JSC::FunctionConstructor::create): * runtime/FunctionPrototype.h: (JSC::FunctionPrototype::create): * runtime/GetterSetter.h: (JSC::GetterSetter::create): * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::create): * runtime/JSActivation.h: (JSC::JSActivation::create): * runtime/JSArray.h: (JSC::JSArray::create): * runtime/JSBoundFunction.cpp: (JSC::JSBoundFunction::create): * runtime/JSByteArray.h: (JSC::JSByteArray::create): Use the NotNull version of placement new to skip the NULL check. * runtime/JSCell.h: Removed a conflicting, unnecessaray placement new. * runtime/JSFunction.cpp: (JSC::JSFunction::create): * runtime/JSFunction.h: (JSC::JSFunction::create): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::create): * runtime/JSGlobalThis.h: (JSC::JSGlobalThis::create): * runtime/JSNotAnObject.h: (JSC::JSNotAnObject::create): * runtime/JSONObject.h: (JSC::JSONObject::create): * runtime/JSObject.h: (JSC::JSFinalObject::create): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::create): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::create): * runtime/JSStaticScopeObject.h: (JSC::JSStaticScopeObject::create): * runtime/JSString.cpp: (JSC::StringObject::create): * runtime/JSString.h: (JSC::RopeBuilder::createNull): (JSC::RopeBuilder::create): (JSC::RopeBuilder::createHasOtherOwner): * runtime/MathObject.h: (JSC::MathObject::create): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::create): * runtime/NativeErrorPrototype.h: (JSC::NativeErrorPrototype::create): * runtime/NumberConstructor.h: (JSC::NumberConstructor::create): * runtime/NumberObject.h: (JSC::NumberObject::create): * runtime/NumberPrototype.h: (JSC::NumberPrototype::create): * runtime/ObjectConstructor.h: (JSC::ObjectConstructor::create): * runtime/ObjectPrototype.h: (JSC::ObjectPrototype::create): * runtime/RegExp.cpp: (JSC::RegExp::createWithoutCaching): * runtime/RegExpConstructor.h: (JSC::RegExpConstructor::create): * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::create): * runtime/RegExpObject.h: (JSC::RegExpObject::create): * runtime/RegExpPrototype.h: (JSC::RegExpPrototype::create): * runtime/ScopeChain.h: (JSC::ScopeChainNode::create): * runtime/StrictEvalActivation.h: (JSC::StrictEvalActivation::create): * runtime/StringConstructor.h: (JSC::StringConstructor::create): * runtime/StringObject.h: (JSC::StringObject::create): * runtime/StringPrototype.h: (JSC::StringPrototype::create): * runtime/Structure.h: (JSC::Structure::create): (JSC::Structure::createStructure): * runtime/StructureChain.h: (JSC::StructureChain::create): * testRegExp.cpp: (GlobalObject::create): * wtf/BitVector.cpp: (WTF::BitVector::OutOfLineBits::create): Use the NotNull version of placement new to skip the NULL check. * wtf/BumpPointerAllocator.h: (WTF::BumpPointerPool::create): Standardized spacing to make grep easier. * wtf/ByteArray.cpp: (WTF::ByteArray::create): * wtf/Deque.h: (WTF::::append): (WTF::::prepend): Use NotNull, as above. * wtf/FastAllocBase.h: Added a placement new, since this class would otherwise hide the name of the global placement new. (WTF::fastNew): Standardized spacing. Most of these functions don't need NotNull, since they check for NULL, and the optimizer can see that. * wtf/HashTable.h: * wtf/HashTraits.h: (WTF::SimpleClassHashTraits::constructDeletedValue): * wtf/MetaAllocator.cpp: (WTF::MetaAllocator::allocFreeSpaceNode): NotNull, as above. * wtf/StdLibExtras.h: (throw): This is our NotNull placement new. Declaring that we throw is the C++ way to say that operator new will not return NULL. * wtf/ThreadSpecific.h: (WTF::T): * wtf/Vector.h: (WTF::::append): (WTF::::tryAppend): (WTF::::uncheckedAppend): (WTF::::insert): * wtf/text/AtomicStringHash.h: * wtf/text/StringImpl.cpp: (WTF::StringImpl::createUninitialized): (WTF::StringImpl::reallocate): * wtf/text/StringImpl.h: (WTF::StringImpl::tryCreateUninitialized): * wtf/text/StringStatics.cpp: (WTF::AtomicString::init): Use NotNull, as above. * yarr/YarrInterpreter.cpp: (JSC::Yarr::Interpreter::allocDisjunctionContext): (JSC::Yarr::Interpreter::ParenthesesDisjunctionContext::ParenthesesDisjunctionContext): (JSC::Yarr::Interpreter::allocParenthesesDisjunctionContext): Standardized spacing for easy grep. 2011-12-19 Eric Carlson <eric.carlson@apple.com> Enable <track> for Mac build https://bugs.webkit.org/show_bug.cgi?id=74838 Reviewed by Darin Adler. * wtf/Platform.h: 2011-12-18 Filip Pizlo <fpizlo@apple.com> DFG is too sloppy with register allocation https://bugs.webkit.org/show_bug.cgi?id=74835 Reviewed by Gavin Barraclough. Added assertions that at the end of a successfully generated basic block, all use counts should be zero. This revealed a number of bugs: - Array length optimizations were turning a must-generate node into one that is not must-generate, but failing to change the ref count accordingly. - Indexed property storage optimizations were failing to deref their children, or to deref the indexed property storage node itself. Also, they used the Phantom node as a replacement. But the Phantom node is must-generate, which was causing bizarre issues. So this introduces a Nop node, which should be used in cases where you want a node that is skipped and has no children. This does not have any significant performance effect, but it should relieve some register pressure. The main thing this patch adds, though, are the assertions, which should make it easier to do register allocation related changes in the future. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGGenerationInfo.h: (JSC::DFG::GenerationInfo::initConstant): (JSC::DFG::GenerationInfo::initInteger): (JSC::DFG::GenerationInfo::initJSValue): (JSC::DFG::GenerationInfo::initCell): (JSC::DFG::GenerationInfo::initBoolean): (JSC::DFG::GenerationInfo::initDouble): (JSC::DFG::GenerationInfo::initStorage): (JSC::DFG::GenerationInfo::use): * dfg/DFGGraph.h: (JSC::DFG::Graph::clearAndDerefChild1): (JSC::DFG::Graph::clearAndDerefChild2): (JSC::DFG::Graph::clearAndDerefChild3): * dfg/DFGNode.h: (JSC::DFG::Node::deref): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::fixupNode): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-12-18 Benjamin Poulain <bpoulain@apple.com> Remove the duplicated code from ASCIICType.h https://bugs.webkit.org/show_bug.cgi?id=74771 Reviewed by Andreas Kling. Use isASCIIDigit() and isASCIIAlpha() instead of copying the code. * wtf/ASCIICType.h: (WTF::isASCIIDigit): (WTF::isASCIIAlphanumeric): (WTF::isASCIIHexDigit): 2011-12-18 Anders Carlsson <andersca@apple.com> Set the main frame view scroll position asynchronously https://bugs.webkit.org/show_bug.cgi?id=74823 Reviewed by Sam Weinig. * JavaScriptCore.exp: 2011-12-10 Andreas Kling <kling@webkit.org> OpaqueJSClass: Remove RVCT2 workarounds. <http://webkit.org/b/74250> Reviewed by Benjamin Poulain. We no longer need workarounds for the RVCT2 compiler since it was only used for the Symbian port of WebKit which is now defunct. * API/JSClassRef.cpp: (OpaqueJSClass::OpaqueJSClass): (OpaqueJSClassContextData::OpaqueJSClassContextData): 2011-12-16 Benjamin Poulain <bpoulain@apple.com> Remove the duplicated code from ASCIICType.h https://bugs.webkit.org/show_bug.cgi?id=74771 Reviewed by Andreas Kling. The functions were sharing similar code and were defined for the various input types. Use templates instead to avoid code duplication. * wtf/ASCIICType.h: (WTF::isASCII): (WTF::isASCIIAlpha): (WTF::isASCIIAlphanumeric): (WTF::isASCIIDigit): (WTF::isASCIIHexDigit): (WTF::isASCIILower): (WTF::isASCIIOctalDigit): (WTF::isASCIIPrintable): (WTF::isASCIISpace): (WTF::isASCIIUpper): (WTF::toASCIILower): (WTF::toASCIIUpper): (WTF::toASCIIHexValue): (WTF::lowerNibbleToASCIIHexDigit): (WTF::upperNibbleToASCIIHexDigit): 2011-12-16 Filip Pizlo <fpizlo@apple.com> DFG OSR exit may get confused about where in the scratch buffer it stored a value https://bugs.webkit.org/show_bug.cgi?id=74695 Reviewed by Oliver Hunt. The code that reads from the scratch buffer now explicitly knows which locations to read from. No new tests, since this patch covers a case so uncommon that I don't know how to make a test for it. * dfg/DFGOSRExitCompiler.h: (JSC::DFG::OSRExitCompiler::badIndex): (JSC::DFG::OSRExitCompiler::initializePoisoned): (JSC::DFG::OSRExitCompiler::poisonIndex): * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): 2011-12-16 Oliver Hunt <oliver@apple.com> PutByVal[Alias] unnecessarily reloads the storage buffer https://bugs.webkit.org/show_bug.cgi?id=74747 Reviewed by Gavin Barraclough. Make PutByVal use GetIndexedStorage to load the storage buffer. This required switching PutByVal to a vararg node (which is responsible for most of the noise in this patch). This fixes the remaining portion of the kraken regression caused by the GetByVal storage load elimination, and a 1-5% win on some of the sub tests of the typed array benchmark at: http://stepheneb.github.com/webgl-matrix-benchmarks/matrix_benchmark.html * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateArithNodeFlags): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::byValIndexIsPure): (JSC::DFG::Propagator::clobbersWorld): (JSC::DFG::Propagator::getByValLoadElimination): (JSC::DFG::Propagator::checkStructureLoadElimination): (JSC::DFG::Propagator::getByOffsetLoadElimination): (JSC::DFG::Propagator::getPropertyStorageLoadElimination): (JSC::DFG::Propagator::getIndexedPropertyStorageLoadElimination): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): (JSC::DFG::SpeculativeJIT::compilePutByValForFloatTypedArray): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-12-16 Daniel Bates <dbates@rim.com> Include BlackBerryPlatformLog.h instead of BlackBerryPlatformMisc.h Rubber-stamped by Antonio Gomes. BlackBerry::Platform::logV() is declared in BlackBerryPlatformLog.h. That is, it isn't declared in BlackBerryPlatformMisc.h. Hence, we should include BlackBerryPlatformLog.h instead of BlackBerryPlatformMisc.h. * wtf/Assertions.cpp: 2011-12-16 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize destructors https://bugs.webkit.org/show_bug.cgi?id=74331 Reviewed by Geoffrey Garen. This is a megapatch which frees us from the chains of virtual destructors. In order to remove the virtual destructors, which are the last of the virtual functions, from the JSCell hierarchy, we need to add the ClassInfo pointer to the cell rather than to the structure because in order to be able to lazily call the static destroy() functions that will replace the virtual destructors, we need to be able to access the ClassInfo without the danger of the object's Structure being collected before the object itself. After adding the ClassInfo to the cell, we can then begin to remove our use of vptrs for optimizations within the JIT and the GC. When we have removed all of the stored vptrs from JSGlobalData, we can then also remove all of the related VPtrStealingHack code. The replacement for virtual destructors will be to add a static destroy function pointer to the MethodTable stored in ClassInfo. Any subclass of JSCell that has a non-trivial destructor will require its own static destroy function to static call its corresponding destructor, which will now be non-virtual. In future patches we will slowly move away from destructors altogether as we make more and more objects backed by GC memory rather than malloc-ed memory. The GC will now call the static destroy method rather than the virtual destructor. As we go through the hierarchy and add static destroy functions to classes, we will also add a new assert, ASSERT_HAS_TRIVIAL_DESTRUCTOR, to those classes to which it applies. The future goal is to eventually have every class have that assert. * API/JSCallbackConstructor.cpp: (JSC::JSCallbackConstructor::destroy): Add a destroy function to statically call ~JSCallbackConstructor because it has some extra destruction logic. * API/JSCallbackConstructor.h: * API/JSCallbackFunction.cpp: Add trivial destructor assert for JSCallbackFunction. * API/JSCallbackObject.cpp: Add a destroy function to statically call ~JSCallbackObject because it has a member OwnPtr that needs destruction. (JSC::::destroy): * API/JSCallbackObject.h: * JavaScriptCore.exp: Add/remove necessary symbols for JSC. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Same for Windows symbols. * debugger/DebuggerActivation.cpp: DebuggerActivation, for some strange reason, didn't have its own ClassInfo despite the fact that it overrides a number of MethodTable methods. Added the ClassInfo, along with an assertion that its destructor is trivial. * debugger/DebuggerActivation.h: * dfg/DFGOperations.cpp: Remove global data first argument to isJSArray, isJSByteArray, isJSString, as it is no longer necessary. (JSC::DFG::putByVal): * dfg/DFGRepatch.cpp: Ditto. Also remove uses of jsArrayVPtr in favor of using the JSArray ClassInfo pointer. (JSC::DFG::tryCacheGetByID): * dfg/DFGSpeculativeJIT.cpp: Replace uses of the old vptrs with new ClassInfo comparisons since we don't have vptrs anymore. (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality): (JSC::DFG::SpeculativeJIT::compilePeepHoleBranch): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): (JSC::DFG::SpeculativeJIT::compileGetTypedArrayLength): (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): (JSC::DFG::SpeculativeJIT::compilePutByValForFloatTypedArray): (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compileStrictEq): (JSC::DFG::SpeculativeJIT::compileGetIndexedPropertyStorage): * dfg/DFGSpeculativeJIT.h: Ditto. (JSC::DFG::SpeculativeJIT::emitAllocateJSFinalObject): * dfg/DFGSpeculativeJIT32_64.cpp: Ditto. (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: Ditto. (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): * heap/Heap.cpp: Remove all uses of vptrs in GC optimizations and replace them with ClassInfo comparisons. (JSC::Heap::Heap): * heap/MarkStack.cpp: Ditto. (JSC::MarkStackThreadSharedData::markingThreadMain): (JSC::visitChildren): (JSC::SlotVisitor::drain): * heap/MarkStack.h: Ditto. (JSC::MarkStack::MarkStack): * heap/MarkedBlock.cpp: Ditto. (JSC::MarkedBlock::callDestructor): (JSC::MarkedBlock::specializedSweep): * heap/MarkedBlock.h: Ditto. * heap/SlotVisitor.h: Ditto. (JSC::SlotVisitor::SlotVisitor): * heap/VTableSpectrum.cpp: Now that we don't have vptrs, we can't count them. We'll have to rename this class and make it use ClassInfo ptrs in a future patch. (JSC::VTableSpectrum::count): * interpreter/Interpreter.cpp: Remove all global data arguments from isJSArray, etc. functions. (JSC::loadVarargs): (JSC::Interpreter::tryCacheGetByID): (JSC::Interpreter::privateExecute): * jit/JIT.h: Remove vptr argument from emitAllocateBasicJSObject * jit/JITInlineMethods.h: Remove vptr planting, and add ClassInfo planting, remove all vtable related code. (JSC::JIT::emitLoadCharacterString): (JSC::JIT::emitAllocateBasicJSObject): (JSC::JIT::emitAllocateJSFinalObject): (JSC::JIT::emitAllocateJSFunction): * jit/JITOpcodes.cpp: Replace vptr related branch code with corresponding ClassInfo. (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::emit_op_to_primitive): (JSC::JIT::emit_op_convert_this): * jit/JITOpcodes32_64.cpp: Ditto. (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::emit_op_to_primitive): (JSC::JIT::emitSlow_op_eq): (JSC::JIT::emitSlow_op_neq): (JSC::JIT::compileOpStrictEq): (JSC::JIT::emit_op_convert_this): * jit/JITPropertyAccess.cpp: Ditto. (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::emit_op_get_by_val): (JSC::JIT::emitSlow_op_get_by_val): (JSC::JIT::emit_op_put_by_val): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompilePatchGetArrayLength): * jit/JITPropertyAccess32_64.cpp: Ditto. (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::emit_op_get_by_val): (JSC::JIT::emitSlow_op_get_by_val): (JSC::JIT::emit_op_put_by_val): (JSC::JIT::privateCompilePatchGetArrayLength): * jit/JITStubs.cpp: Remove global data argument from isJSString, etc. (JSC::JITThunks::tryCacheGetByID): (JSC::DEFINE_STUB_FUNCTION): * jit/SpecializedThunkJIT.h: Replace vptr related stuff with ClassInfo stuff. (JSC::SpecializedThunkJIT::loadJSStringArgument): * runtime/ArrayConstructor.cpp: Add trivial destructor assert. * runtime/ArrayPrototype.cpp: Remove global data argument from isJSArray. (JSC::arrayProtoFuncToString): (JSC::arrayProtoFuncJoin): (JSC::arrayProtoFuncPop): (JSC::arrayProtoFuncPush): (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): (JSC::arrayProtoFuncFilter): (JSC::arrayProtoFuncMap): (JSC::arrayProtoFuncEvery): (JSC::arrayProtoFuncForEach): (JSC::arrayProtoFuncSome): (JSC::arrayProtoFuncReduce): (JSC::arrayProtoFuncReduceRight): * runtime/BooleanConstructor.cpp: Add trivial destructor assert. * runtime/BooleanObject.cpp: Ditto. * runtime/BooleanPrototype.cpp: Ditto. * runtime/ClassInfo.h: Add destroy function pointer to MethodTable. * runtime/DateConstructor.cpp: Add trivial destructor assert. * runtime/DateInstance.cpp: Add destroy function for DateInstance because it has a RefPtr that needs destruction. (JSC::DateInstance::destroy): * runtime/DateInstance.h: * runtime/Error.cpp: Ditto (because of UString member). (JSC::StrictModeTypeErrorFunction::destroy): * runtime/Error.h: * runtime/ErrorConstructor.cpp: Add trivial destructor assert. * runtime/ErrorInstance.cpp: Ditto. * runtime/ExceptionHelpers.cpp: Ditto. * runtime/Executable.cpp: Add destroy functions for ExecutableBase and subclasses. (JSC::ExecutableBase::destroy): (JSC::NativeExecutable::destroy): (JSC::ScriptExecutable::destroy): (JSC::EvalExecutable::destroy): (JSC::ProgramExecutable::destroy): (JSC::FunctionExecutable::destroy): * runtime/Executable.h: * runtime/FunctionConstructor.cpp: Add trivial destructor assert. * runtime/FunctionPrototype.cpp: Ditto. Also remove global data first arg from isJSArray. (JSC::functionProtoFuncApply): * runtime/GetterSetter.cpp: Ditto. * runtime/InitializeThreading.cpp: Remove call to JSGlobalData::storeVPtrs since it no longer exists. (JSC::initializeThreadingOnce): * runtime/InternalFunction.cpp: Remove vtableAnchor function, add trivial destructor assert, remove first arg from isJSString. (JSC::InternalFunction::displayName): * runtime/InternalFunction.h: Remove VPtrStealingHack. * runtime/JSAPIValueWrapper.cpp: Add trivial destructor assert. * runtime/JSArray.cpp: Add static destroy to call ~JSArray. Replace vptr checks in destructor with ClassInfo checks. (JSC::JSArray::~JSArray): (JSC::JSArray::destroy): * runtime/JSArray.h: Remove VPtrStealingHack. Remove globalData argument from isJSArray and change them to check the ClassInfo rather than the vptrs. (JSC::isJSArray): * runtime/JSBoundFunction.cpp: Add trival destructor assert. Remove first arg from isJSArray. (JSC::boundFunctionCall): (JSC::boundFunctionConstruct): * runtime/JSByteArray.cpp: Add static destroy function, replace vptr checks with ClassInfo checks. (JSC::JSByteArray::~JSByteArray): (JSC::JSByteArray::destroy): * runtime/JSByteArray.h: Remove VPtrStealingHack code. (JSC::isJSByteArray): * runtime/JSCell.cpp: Add trivial destructor assert. Add static destroy function. (JSC::JSCell::destroy): * runtime/JSCell.h: Remove VPtrStealingHack code. Add function for returning the offset of the ClassInfo pointer in the object for use by the JIT. Add the ClassInfo pointer to the JSCell itself, and grab it from the Structure. Remove the vptr and setVPtr functions, as they are no longer used. Add a validatedClassInfo function to JSCell for any clients that want to verify, while in Debug mode, that the ClassInfo contained in the cell is the same one as that contained in the Structure. This isn't used too often, because most of the places where we compare the ClassInfo to things can be called during destruction. Since the Structure is unreliable during the phase when destructors are being called, we can't call validatedClassInfo. (JSC::JSCell::classInfoOffset): (JSC::JSCell::structure): (JSC::JSCell::classInfo): * runtime/JSFunction.cpp: Remove VPtrStealingHack code. Add static destroy, remove vtableAnchor, remove first arg from call to isJSString. (JSC::JSFunction::destroy): (JSC::JSFunction::displayName): * runtime/JSFunction.h: * runtime/JSGlobalData.cpp: Remove all VPtr stealing code and storage, including storeVPtrs, as these vptrs are no longer needed in the codebase. * runtime/JSGlobalData.h: (JSC::TypedArrayDescriptor::TypedArrayDescriptor): Changed the TypedArrayDescriptor to use ClassInfo rather than the vptr. * runtime/JSGlobalObject.cpp: Add static destroy function. (JSC::JSGlobalObject::destroy): * runtime/JSGlobalObject.h: * runtime/JSGlobalThis.cpp: Add trivial destructor assert. * runtime/JSNotAnObject.cpp: Ditto. * runtime/JSONObject.cpp: Ditto. Remove first arg from isJSArray calls. (JSC::Stringifier::Holder::appendNextProperty): (JSC::Walker::walk): * runtime/JSObject.cpp: (JSC::JSFinalObject::destroy): (JSC::JSNonFinalObject::destroy): (JSC::JSObject::destroy): * runtime/JSObject.h: Add trivial destructor assert for JSObject, remove vtableAnchor from JSNonFinalObject and JSFinalObject, add static destroy for JSFinalObject and JSNonFinalObject, add isJSFinalObject utility function similar to isJSArray, remove all VPtrStealingHack code. (JSC::JSObject::finishCreation): (JSC::JSNonFinalObject::finishCreation): (JSC::JSFinalObject::finishCreation): (JSC::isJSFinalObject): * runtime/JSPropertyNameIterator.cpp: Add static destroy. (JSC::JSPropertyNameIterator::destroy): * runtime/JSPropertyNameIterator.h: * runtime/JSStaticScopeObject.cpp: Ditto. (JSC::JSStaticScopeObject::destroy): * runtime/JSStaticScopeObject.h: Ditto. * runtime/JSString.cpp: (JSC::JSString::destroy): * runtime/JSString.h: Ditto. Remove VPtrStealingHack code. Also remove fixupVPtr code, since we no longer need to fixup vptrs. (JSC::jsSingleCharacterString): (JSC::jsSingleCharacterSubstring): (JSC::jsNontrivialString): (JSC::jsString): (JSC::jsSubstring8): (JSC::jsSubstring): (JSC::jsOwnedString): (JSC::jsStringBuilder): (JSC::isJSString): * runtime/JSVariableObject.cpp: (JSC::JSVariableObject::destroy): * runtime/JSVariableObject.h: Ditto. * runtime/JSWrapperObject.cpp: * runtime/JSWrapperObject.h: Add trivial destructor assert. * runtime/MathObject.cpp: Ditto. * runtime/NativeErrorConstructor.cpp: Ditto. * runtime/NumberConstructor.cpp: Ditto. * runtime/NumberObject.cpp: Ditto. * runtime/NumberPrototype.cpp: Ditto. * runtime/ObjectConstructor.cpp: Ditto. * runtime/ObjectPrototype.cpp: Ditto. * runtime/Operations.h: Remove calls to fixupVPtr, remove first arg to isJSString. (JSC::jsString): (JSC::jsLess): (JSC::jsLessEq): * runtime/RegExp.cpp: Add static destroy. (JSC::RegExp::destroy): * runtime/RegExp.h: * runtime/RegExpConstructor.cpp: Add static destroy for RegExpConstructor and RegExpMatchesArray. (JSC::RegExpConstructor::destroy): (JSC::RegExpMatchesArray::destroy): * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: * runtime/RegExpObject.cpp: Add static destroy. (JSC::RegExpObject::destroy): * runtime/RegExpObject.h: * runtime/ScopeChain.cpp: Add trivial destructor assert. * runtime/ScopeChain.h: * runtime/StrictEvalActivation.cpp: Ditto. * runtime/StringConstructor.cpp: * runtime/StringObject.cpp: Ditto. Remove vtableAnchor. * runtime/StringObject.h: * runtime/StringPrototype.cpp: Ditto. * runtime/Structure.cpp: Add static destroy. (JSC::Structure::destroy): * runtime/Structure.h: Move JSCell::finishCreation and JSCell constructor into Structure.h because they need to have the full Structure type to access the ClassInfo to store in the JSCell. (JSC::JSCell::setStructure): (JSC::JSCell::validatedClassInfo): (JSC::JSCell::JSCell): (JSC::JSCell::finishCreation): * runtime/StructureChain.cpp: Add static destroy. (JSC::StructureChain::destroy): * runtime/StructureChain.h: * wtf/Assertions.h: Add new assertion ASSERT_HAS_TRIVIAL_DESTRUCTOR, which uses clangs ability to tell us when a class has a trivial destructor. We will use this assert more in future patches as we move toward having all JSC objects backed by GC memory, which means moving away from using destructors/finalizers. 2011-12-15 Martin Robinson <mrobinson@igalia.com> Fix 'make dist' in preparation for the GTK+ release. * GNUmakefile.list.am: Add missing header. 2011-12-15 Sam Weinig <sam@webkit.org> <rdar://problem/10552550> JavaScriptCore uses obsolete 'cpy' mnemonic in ARM assembly Reviewed by Gavin Barraclough. Original patch by Jim Grosbach. * jit/JITStubs.cpp: (JSC::ctiTrampoline): (JSC::ctiVMThrowTrampoline): Replace uses of the 'cpy' mnemonic with 'mov'. 2011-12-15 Filip Pizlo <fpizlo@apple.com> Value profiling should distinguished between NaN and non-NaN doubles https://bugs.webkit.org/show_bug.cgi?id=74682 Reviewed by Gavin Barraclough. Added PredictDoubleReal and PredictDoubleNaN. PredictDouble is now the union of the two. * bytecode/PredictedType.cpp: (JSC::predictionToString): (JSC::predictionFromValue): * bytecode/PredictedType.h: (JSC::isDoubleRealPrediction): (JSC::isDoublePrediction): 2011-12-15 Anders Carlsson <andersca@apple.com> Regression (r102866): Navigating away from or closing a page with a plugin crashes https://bugs.webkit.org/show_bug.cgi?id=74655 <rdar://problem/10590024> Reviewed by Sam Weinig. Rewrite HasRefAndDeref to work if ref and deref are implemented in base classes, using a modified version of the technique described here: http://groups.google.com/group/comp.lang.c++.moderated/msg/e5fbc9305539f699 * wtf/Functional.h: 2011-12-15 Andy Wingo <wingo@igalia.com> Warnings fixes in Interpreter.cpp and PrivateExecute.cpp https://bugs.webkit.org/show_bug.cgi?id=74624 Reviewed by Darin Adler. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): Fix variables unused in release mode. * wtf/ParallelJobsGeneric.cpp: (WTF::ParallelEnvironment::ParallelEnvironment): Fix signed/unsigned comparison warning, with a cast. 2011-12-15 Andy Wingo <wingo@igalia.com> Use more macrology in JSC::Options https://bugs.webkit.org/show_bug.cgi?id=72938 Reviewed by Filip Pizlo. * runtime/Options.cpp: (JSC::Options::initializeOptions): * runtime/Options.h: Use macros to ensure that all heuristics are declared and have initializers. 2011-12-15 Anders Carlsson <andersca@apple.com> Add ScrollingCoordinator class and ENABLE_THREADED_SCROLLING define https://bugs.webkit.org/show_bug.cgi?id=74639 Reviewed by Andreas Kling. Add ENABLE_THREADED_SCROLLING #define. * wtf/Platform.h: 2011-12-15 Anders Carlsson <andersca@apple.com> EventDispatcher should handle wheel events on the connection queue https://bugs.webkit.org/show_bug.cgi?id=74627 Reviewed by Andreas Kling. Add a BoundFunctionImpl specialization that takes three parameters. * wtf/Functional.h: (WTF::C::): (WTF::R): (WTF::bind): 2011-12-14 Anders Carlsson <andersca@apple.com> Add WTF::Function to wtf/Forward.h https://bugs.webkit.org/show_bug.cgi?id=74576 Reviewed by Adam Roben. * jsc.cpp: Work around a name conflict in the readline library. * wtf/Forward.h: Add Function. 2011-12-15 Igor Oliveira <igor.oliveira@openbossa.org> [Qt] Support requestAnimationFrame API https://bugs.webkit.org/show_bug.cgi?id=74528 Let Qt port use REQUEST_ANIMATION_FRAME_TIMER. Reviewed by Kenneth Rohde Christiansen. * wtf/Platform.h: 2011-12-15 Andy Wingo <wingo@igalia.com> Minor refactor to Parser::parseTryStatement https://bugs.webkit.org/show_bug.cgi?id=74507 Reviewed by Geoffrey Garen. * parser/Parser.cpp (JSC::Parser::parseTryStatement): Use the Parser's declareVariable instead of going directly to the scope. This will facilitate future checks related to harmony block scoping. 2011-12-15 Andy Wingo <wingo@igalia.com> Rename JSC::Heuristics to JSC::Options https://bugs.webkit.org/show_bug.cgi?id=72889 Reviewed by Filip Pizlo. * runtime/Options.cpp: Renamed from Source/JavaScriptCore/runtime/Heuristics.cpp. * runtime/Options.h: Renamed from Source/JavaScriptCore/runtime/Heuristics.h. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::shouldOptimizeNow): * bytecode/CodeBlock.h: (JSC::CodeBlock::likelyToTakeSlowCase): (JSC::CodeBlock::couldTakeSlowCase): (JSC::CodeBlock::likelyToTakeSpecialFastCase): (JSC::CodeBlock::likelyToTakeDeepestSlowCase): (JSC::CodeBlock::likelyToTakeAnySlowCase): (JSC::CodeBlock::reoptimizationRetryCounter): (JSC::CodeBlock::countReoptimization): (JSC::CodeBlock::counterValueForOptimizeAfterWarmUp): (JSC::CodeBlock::counterValueForOptimizeAfterLongWarmUp): (JSC::CodeBlock::optimizeNextInvocation): (JSC::CodeBlock::dontOptimizeAnytimeSoon): (JSC::CodeBlock::optimizeSoon): (JSC::CodeBlock::largeFailCountThreshold): (JSC::CodeBlock::largeFailCountThresholdForLoop): (JSC::CodeBlock::shouldReoptimizeNow): (JSC::CodeBlock::shouldReoptimizeFromLoopNow): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleInlining): * dfg/DFGCapabilities.h: (JSC::DFG::mightCompileEval): (JSC::DFG::mightCompileProgram): (JSC::DFG::mightCompileFunctionForCall): (JSC::DFG::mightCompileFunctionForConstruct): (JSC::DFG::mightInlineFunctionForCall): (JSC::DFG::mightInlineFunctionForConstruct): * dfg/DFGOSRExit.cpp: (JSC::DFG::OSRExit::considerAddingAsFrequentExitSiteSlow): * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGVariableAccessData.h: (JSC::DFG::VariableAccessData::shouldUseDoubleFormatAccordingToVote): * heap/MarkStack.cpp: (JSC::MarkStackSegmentAllocator::allocate): (JSC::MarkStackSegmentAllocator::shrinkReserve): (JSC::MarkStackArray::MarkStackArray): (JSC::MarkStackArray::donateSomeCellsTo): (JSC::MarkStackArray::stealSomeCellsFrom): (JSC::MarkStackThreadSharedData::MarkStackThreadSharedData): (JSC::SlotVisitor::donateSlow): (JSC::SlotVisitor::drain): (JSC::SlotVisitor::drainFromShared): * heap/MarkStack.h: (JSC::MarkStack::mergeOpaqueRootsIfProfitable): (JSC::MarkStack::addOpaqueRoot): (JSC::MarkStackArray::canDonateSomeCells): * heap/SlotVisitor.h: (JSC::SlotVisitor::donate): * jit/JIT.cpp: (JSC::JIT::emitOptimizationCheck): * runtime/InitializeThreading.cpp: (JSC::initializeThreadingOnce): Adapt callers and build systems. * testRegExp.cpp: (CommandLine::CommandLine): * jsc.cpp: (CommandLine::CommandLine): Rename from Options, to avoid name conflict. 2011-12-14 Sam Weinig <sam@webkit.org> Revert unintentional change to JavaScriptCore.def * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-12-14 Sam Weinig <weinig@apple.com> Remove whitespace from InheritedPropertySheets attributes in vsprops files to appease the Visual Studio project migrator. Reviewed by Adam Roben. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreDebug.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreDebugAll.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreDebugCairoCFLite.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedDebug.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedDebugAll.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedDebugCairoCFLite.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedProduction.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedRelease.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedReleaseCairoCFLite.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedReleasePGO.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreProduction.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreRelease.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleaseCairoCFLite.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleasePGO.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleasePGOOptimize.vsprops: * JavaScriptCore.vcproj/WTF/WTFDebug.vsprops: * JavaScriptCore.vcproj/WTF/WTFDebugAll.vsprops: * JavaScriptCore.vcproj/WTF/WTFDebugCairoCFLite.vsprops: * JavaScriptCore.vcproj/WTF/WTFProduction.vsprops: * JavaScriptCore.vcproj/WTF/WTFRelease.vsprops: * JavaScriptCore.vcproj/WTF/WTFReleaseCairoCFLite.vsprops: * JavaScriptCore.vcproj/WTF/WTFReleasePGO.vsprops: * JavaScriptCore.vcproj/jsc/jscDebug.vsprops: * JavaScriptCore.vcproj/jsc/jscDebugAll.vsprops: * JavaScriptCore.vcproj/jsc/jscDebugCairoCFLite.vsprops: * JavaScriptCore.vcproj/jsc/jscProduction.vsprops: * JavaScriptCore.vcproj/jsc/jscRelease.vsprops: * JavaScriptCore.vcproj/jsc/jscReleaseCairoCFLite.vsprops: * JavaScriptCore.vcproj/jsc/jscReleasePGO.vsprops: * JavaScriptCore.vcproj/testRegExp/testRegExpDebug.vsprops: * JavaScriptCore.vcproj/testRegExp/testRegExpDebugAll.vsprops: * JavaScriptCore.vcproj/testRegExp/testRegExpDebugCairoCFLite.vsprops: * JavaScriptCore.vcproj/testRegExp/testRegExpProduction.vsprops: * JavaScriptCore.vcproj/testRegExp/testRegExpRelease.vsprops: * JavaScriptCore.vcproj/testRegExp/testRegExpReleaseCairoCFLite.vsprops: * JavaScriptCore.vcproj/testRegExp/testRegExpReleasePGO.vsprops: * JavaScriptCore.vcproj/testapi/testapiDebug.vsprops: * JavaScriptCore.vcproj/testapi/testapiDebugAll.vsprops: * JavaScriptCore.vcproj/testapi/testapiDebugCairoCFLite.vsprops: * JavaScriptCore.vcproj/testapi/testapiProduction.vsprops: * JavaScriptCore.vcproj/testapi/testapiRelease.vsprops: * JavaScriptCore.vcproj/testapi/testapiReleaseCairoCFLite.vsprops: 2011-12-14 Anders Carlsson <andersca@apple.com> binding a member function should ref/deref the object pointer if needed https://bugs.webkit.org/show_bug.cgi?id=74552 Reviewed by Sam Weinig. Add a HasRefAndDeref helper class template which checks if a given class type has ref and deref member functions which the right type. Use this to determine if we should ref/deref the first parameter. * wtf/Functional.h: (WTF::R): (WTF::C::): (WTF::RefAndDeref::ref): (WTF::RefAndDeref::deref): 2011-12-14 Hajime Morrita <morrita@chromium.org> JS_INLINE and WTF_INLINE should be visible from WebCore https://bugs.webkit.org/show_bug.cgi?id=73191 - Moved Export related macro definitions from config.h to ExportMacros.h and JSExportMacros.h. - Moved WTF_USE_JSC and WTF_USE_V8 from various config.h family to Platform.h. - Replaced JS_EXPORTDATA in wtf moudule with newly introduced WTF_EXPORTDATA. Reviewed by Kevin Ollivier. * JavaScriptCore.xcodeproj/project.pbxproj: * config.h: * runtime/JSExportMacros.h: Added. * wtf/ExportMacros.h: * wtf/Platform.h: * wtf/WTFThreadData.h: * wtf/text/AtomicString.h: * wtf/text/StringStatics.cpp: 2011-12-14 Anders Carlsson <andersca@apple.com> Work around a bug in the MSVC2005 compiler https://bugs.webkit.org/show_bug.cgi?id=74550 Reviewed by Sam Weinig. Add template parameters for the return types of the partial specializations of BoundFunctionImpl. * wtf/Functional.h: (WTF::R): 2011-12-13 Jon Lee <jonlee@apple.com> Enable notifications on Mac. Reviewed by Sam Weinig. * Configurations/FeatureDefines.xcconfig: 2011-12-14 David Kilzer <ddkilzer@apple.com> Remove definition of old ENABLE(YARR) macro <http://webkit.org/b/74532> Reviewed by Darin Adler. * wtf/Platform.h: Removed ENABLE_YARR macros. 2011-12-14 Anders Carlsson <andersca@apple.com> bind should handle member functions https://bugs.webkit.org/show_bug.cgi?id=74529 Reviewed by Sam Weinig. Add FunctionWrapper partial specializations for member function pointers. * wtf/Functional.h: (WTF::C::): 2011-12-14 Gavin Barraclough <barraclough@apple.com> DFG relies on returning a struct in registers https://bugs.webkit.org/show_bug.cgi?id=74527 Reviewed by Geoff Garen. This will not work on all platforms. Returning a uint64_t will more reliably achieve what we want, on 32-bit platforms (on 64-bit, stick with the struct return). * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: (JSC::DFG::DFGHandler::dfgHandlerEncoded): 2011-12-14 Anders Carlsson <andersca@apple.com> Add unary and binary bind overloads https://bugs.webkit.org/show_bug.cgi?id=74524 Reviewed by Sam Weinig. * wtf/Functional.h: (WTF::R): (WTF::FunctionWrapper::ResultType): (WTF::bind): 2011-12-14 Anders Carlsson <andersca@apple.com> Add back the callOnMainThread overload that takes a WTF::Function https://bugs.webkit.org/show_bug.cgi?id=74512 Reviewed by Darin Adler. Add back the overload; the changes to WebCore should hopefully keep Windows building. * wtf/MainThread.cpp: (WTF::callFunctionObject): (WTF::callOnMainThread): * wtf/MainThread.h: 2011-12-13 Filip Pizlo <fpizlo@apple.com> DFG should infer when local variables are doubles https://bugs.webkit.org/show_bug.cgi?id=74480 Reviewed by Oliver Hunt. Introduced the notion that a local variable (though not an argument, yet!) can be stored as a double, and will be guaranteed to always contain a double. This requires more magic in the OSR (conversion in both entry and exit). The inference is quite unorthodox: all uses of a variable vote on whether they think it should be a double or a JSValue, based on how they use it. If they use it in an integer or boxed value context, they vote JSValue. If they use it in a double context, they vote double. This voting is interleaved in the propagator's fixpoint, so that variables voted double then have a double prediction propagated from them. This interleaving is needed because a variable that actually always contains an integer that always gets used in arithmetic that involves doubles may end up being voted double, which then means that all uses of the variable will see a double rather than an integer. This is worth 18% to SunSpider/3d-cube, 7% to Kraken/audio-beat-detection, 7% to Kraken/audio-fft, 6% to Kraken/imaging-darkroom, 20% to Kraken/imaging-gaussian-blur, and just over 1% to Kraken/json-parse-financial. It results in a 1% speed-up on SunSpider and a 4% speed-up in Kraken. Similar results on JSVALUE32_64, though with a bigger win on Kraken (5%) and no overall win on SunSpider. * bytecode/ValueRecovery.h: (JSC::ValueRecovery::alreadyInRegisterFileAsUnboxedDouble): (JSC::ValueRecovery::dump): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::boxDouble): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::noticeOSREntry): * dfg/DFGOSREntry.cpp: (JSC::DFG::prepareOSREntry): * dfg/DFGOSREntry.h: * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::vote): (JSC::DFG::Propagator::doRoundOfDoubleVoting): (JSC::DFG::Propagator::propagatePredictions): (JSC::DFG::Propagator::fixupNode): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::ValueSource::dump): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGVariableAccessData.h: (JSC::DFG::VariableAccessData::VariableAccessData): (JSC::DFG::VariableAccessData::clearVotes): (JSC::DFG::VariableAccessData::vote): (JSC::DFG::VariableAccessData::doubleVoteRatio): (JSC::DFG::VariableAccessData::shouldUseDoubleFormatAccordingToVote): (JSC::DFG::VariableAccessData::shouldUseDoubleFormat): (JSC::DFG::VariableAccessData::tallyVotesForShouldUseDoubleFormat): * runtime/Arguments.cpp: (JSC::Arguments::tearOff): * runtime/Heuristics.cpp: (JSC::Heuristics::initializeHeuristics): * runtime/Heuristics.h: 2011-12-13 Anders Carlsson <andersca@apple.com> Try to fix the Windows build. Remove the callOnMainThread overload that takes a WTF::Function since it's not being used. * wtf/MainThread.cpp: * wtf/MainThread.h: 2011-12-13 Anders Carlsson <andersca@apple.com> Add a very bare-bones implementation of bind and Function to WTF https://bugs.webkit.org/show_bug.cgi?id=74462 Reviewed by Sam Weinig. In order to make it easier to package up function calls and send them across threads, add a (currently very simple) implementation of WTF::bind and WTF::Function to a new wtf/Functional.h header. Currently, all bind can do is bind a nullary function and return a Function object that can be called and copied, but I'll add more as the need arises. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/Functional.h: Added. (WTF::R): (WTF::FunctionImplBase::~FunctionImplBase): (WTF::FunctionWrapper::ResultType): (WTF::FunctionBase::isNull): (WTF::FunctionBase::FunctionBase): (WTF::FunctionBase::impl): (WTF::bind): * wtf/MainThread.cpp: (WTF::callFunctionObject): (WTF::callOnMainThread): * wtf/MainThread.h: * wtf/wtf.pro: 2011-12-13 Geoffrey Garen <ggaren@apple.com> <rdar://problem/10577239> GC Crash introduced in r102545 Reviewed by Gavin Barraclough. MarkedArgumentBuffer was still marking items in forwards order, even though the argument order has been reversed. I fixed this bug, and replaced address calculation code with some helper functions -- mallocBase() and slotFor() -- so it stays fixed everywhere. * runtime/ArgList.cpp: (JSC::MarkedArgumentBuffer::markLists): (JSC::MarkedArgumentBuffer::slowAppend): * runtime/ArgList.h: (JSC::MarkedArgumentBuffer::~MarkedArgumentBuffer): (JSC::MarkedArgumentBuffer::at): (JSC::MarkedArgumentBuffer::append): (JSC::MarkedArgumentBuffer::last): (JSC::MarkedArgumentBuffer::slotFor): (JSC::MarkedArgumentBuffer::mallocBase): 2011-12-13 Filip Pizlo <fpizlo@apple.com> DFG OSR exit for UInt32ToNumber should roll forward, not roll backward https://bugs.webkit.org/show_bug.cgi?id=74463 Reviewed by Gavin Barraclough. Implements roll-forward OSR exit for UInt32ToNumber, which requires ValueRecoveries knowing how to execute the slow path of UInt32ToNumber. * bytecode/CodeBlock.h: (JSC::CodeBlock::lastOSRExit): * bytecode/CodeOrigin.h: (JSC::CodeOrigin::operator!=): * bytecode/ValueRecovery.h: (JSC::ValueRecovery::uint32InGPR): (JSC::ValueRecovery::gpr): (JSC::ValueRecovery::dump): * dfg/DFGAssemblyHelpers.cpp: * dfg/DFGAssemblyHelpers.h: * dfg/DFGOSRExit.h: (JSC::DFG::OSRExit::valueRecoveryForOperand): * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileUInt32ToNumber): (JSC::DFG::SpeculativeJIT::compileGetByValOnIntTypedArray): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativeUInt32ToNumber): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::nonSpeculativeUInt32ToNumber): (JSC::DFG::SpeculativeJIT::compile): 2011-12-13 Oliver Hunt <oliver@apple.com> Arguments object doesn't handle mutation of length property correctly https://bugs.webkit.org/show_bug.cgi?id=74454 Reviewed by Gavin Barraclough. Correct handling of arguments objects with overridden length property * interpreter/Interpreter.cpp: (JSC::loadVarargs): * runtime/Arguments.cpp: (JSC::Arguments::copyToArguments): (JSC::Arguments::fillArgList): 2011-12-13 Filip Pizlo <fpizlo@apple.com> DFG GetByVal CSE rule should match PutByValAlias https://bugs.webkit.org/show_bug.cgi?id=74390 Reviewed by Geoff Garen. Tiny win on some benchmarks. Maybe a 0.2% win on SunSpider. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::getByValLoadElimination): 2011-12-13 Andy Wingo <wingo@igalia.com> Fix interpreter debug build. https://bugs.webkit.org/show_bug.cgi?id=74439 Reviewed by Geoffrey Garen. * bytecode/ValueRecovery.h: Include stdio.h on debug builds. 2011-12-13 Filip Pizlo <fpizlo@apple.com> DFG should know exactly why recompilation was triggered https://bugs.webkit.org/show_bug.cgi?id=74362 Reviewed by Oliver Hunt. Each OSR exit is now individually counted, as well as counting the total number of OSR exits that occurred in a code block. If recompilation is triggered, we check to see if there are OSR exit sites that make up a sufficiently large portion of the total OSR exits that occurred. For any such OSR exit sites, we add a description of the site (bytecode index, kind) to a data structure in the corresponding baseline CodeBlock. Then, when we recompile the code, we immediately know which speculations would be unwise based on the fact that previous such speculations proved to be fruitless. This means 2% win on two of the SunSpider string tests, a 4% win on V8's deltablue, and 5% on Kraken's imaging-darkroom. It is only a minor win in the averages, less than 0.5%. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::tallyFrequentExitSites): * bytecode/CodeBlock.h: (JSC::CodeBlock::addFrequentExitSite): (JSC::CodeBlock::exitProfile): (JSC::CodeBlock::reoptimize): (JSC::CodeBlock::tallyFrequentExitSites): * bytecode/DFGExitProfile.cpp: Added. (JSC::DFG::ExitProfile::ExitProfile): (JSC::DFG::ExitProfile::~ExitProfile): (JSC::DFG::ExitProfile::add): (JSC::DFG::QueryableExitProfile::QueryableExitProfile): (JSC::DFG::QueryableExitProfile::~QueryableExitProfile): * bytecode/DFGExitProfile.h: Added. (JSC::DFG::exitKindToString): (JSC::DFG::exitKindIsCountable): (JSC::DFG::FrequentExitSite::FrequentExitSite): (JSC::DFG::FrequentExitSite::operator!): (JSC::DFG::FrequentExitSite::operator==): (JSC::DFG::FrequentExitSite::hash): (JSC::DFG::FrequentExitSite::bytecodeOffset): (JSC::DFG::FrequentExitSite::kind): (JSC::DFG::FrequentExitSite::isHashTableDeletedValue): (JSC::DFG::FrequentExitSiteHash::hash): (JSC::DFG::FrequentExitSiteHash::equal): (JSC::DFG::QueryableExitProfile::hasExitSite): * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::baselineCodeBlockForOriginAndBaselineCodeBlock): (JSC::DFG::AssemblyHelpers::baselineCodeBlockFor): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::makeSafe): (JSC::DFG::ByteCodeParser::makeDivSafe): (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::handleIntrinsic): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): * dfg/DFGOSRExit.cpp: (JSC::DFG::OSRExit::OSRExit): (JSC::DFG::OSRExit::considerAddingAsFrequentExitSiteSlow): * dfg/DFGOSRExit.h: (JSC::DFG::OSRExit::considerAddingAsFrequentExitSite): * dfg/DFGOSRExitCompiler.cpp: * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::compileGetCharCodeAt): (JSC::DFG::SpeculativeJIT::compileGetByValOnString): (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): (JSC::DFG::SpeculativeJIT::compileGetByValOnByteArray): (JSC::DFG::SpeculativeJIT::compileGetTypedArrayLength): (JSC::DFG::SpeculativeJIT::compileGetByValOnIntTypedArray): (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): (JSC::DFG::SpeculativeJIT::compileGetByValOnFloatTypedArray): (JSC::DFG::SpeculativeJIT::compilePutByValForFloatTypedArray): (JSC::DFG::SpeculativeJIT::compileInstanceOfForObject): (JSC::DFG::SpeculativeJIT::compileSoftModulo): (JSC::DFG::SpeculativeJIT::compileArithMul): (JSC::DFG::SpeculativeJIT::compileGetIndexedPropertyStorage): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::speculationCheck): (JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): * runtime/Heuristics.cpp: (JSC::Heuristics::initializeHeuristics): * runtime/Heuristics.h: 2011-12-13 Michael Saboff <msaboff@apple.com> Cleanup of StringImpl::equal in r102631 post commit https://bugs.webkit.org/show_bug.cgi?id=74421 Reviewed by Darin Adler. * wtf/text/AtomicString.h: (WTF::operator==): Removed cast no longer needed. * wtf/text/StringImpl.h: (WTF::equal): Changed template to several overloaded methods. 2011-12-12 Michael Saboff <msaboff@apple.com> Eliminate Duplicate word at a time equal code in StringImpl.cpp and StringHash.h https://bugs.webkit.org/show_bug.cgi?id=73622 Reviewed by Oliver Hunt. Moved equal(charType1 *, charType2, unsigned) template methods from static StringImpl.cpp to StringImpl.h and then replaced the processor specific character comparison code in StringHash::equal with calls to these methods. This change is worth 3% on SunSpider string-unpack-code as reported by the SunSpider command line harness. No other tests appear to have measurable performance changes. * wtf/text/AtomicString.h: (WTF::operator==): * wtf/text/StringHash.h: (WTF::StringHash::equal): * wtf/text/StringImpl.cpp: * wtf/text/StringImpl.h: (WTF::LChar): (WTF::UChar): (WTF::equal): 2011-12-12 Filip Pizlo <fpizlo@apple.com> ARMv7 version of DFG soft modulo does register allocation inside of control flow https://bugs.webkit.org/show_bug.cgi?id=74354 Reviewed by Gavin Barraclough. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileSoftModulo): 2011-12-12 Andy Wingo <wingo@igalia.com> Simplify autotools configure.ac https://bugs.webkit.org/show_bug.cgi?id=74312 Reviewed by Martin Robinson. * GNUmakefile.am: Add JSC_CPPFLAGS to javascriptcore_cppflags. 2011-12-12 Filip Pizlo <fpizlo@apple.com> DFG GetByVal CSE incorrectly assumes that a non-matching PutByVal cannot clobber https://bugs.webkit.org/show_bug.cgi?id=74329 Reviewed by Gavin Barraclough. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::getByValLoadElimination): 2011-12-09 Alexander Pavlov <apavlov@chromium.org> WebKit does not enumerate over CSS properties in HTMLElement.style https://bugs.webkit.org/show_bug.cgi?id=23946 Reviewed by Darin Adler. Add a few exports to follow the JSCSSStyleDeclaration.cpp changes, introduce an std::sort() comparator function. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/text/WTFString.h: (WTF::codePointCompareLessThan): Used by std::sort() to sort properties. 2011-12-12 Alexander Pavlov <apavlov@chromium.org> Unreviewed, build fix. Revert r102570 which broke SnowLeopard builders. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/text/WTFString.h: 2011-12-09 Alexander Pavlov <apavlov@chromium.org> WebKit does not enumerate over CSS properties in HTMLElement.style https://bugs.webkit.org/show_bug.cgi?id=23946 Reviewed by Darin Adler. Add a few exports to follow the JSCSSStyleDeclaration.cpp changes, introduce an std::sort() comparator function. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/text/WTFString.h: (WTF::codePointCompareLessThan): Used by std::sort() to sort properties. 2011-12-12 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck issues. * GNUmakefile.list.am: 2011-12-11 Sam Weinig <sam@webkit.org> Fix another signed vs. unsigned warning * runtime/ArgList.h: (JSC::MarkedArgumentBuffer::~MarkedArgumentBuffer): 2011-12-11 Sam Weinig <sam@webkit.org> Fix a signed vs. unsigned warning. * runtime/ArgList.cpp: (JSC::MarkedArgumentBuffer::slowAppend): Cast inlineCapacity to an int to appease the warning. This is known OK since inlineCapacity is defined to be 8. 2011-12-11 Geoffrey Garen <ggaren@apple.com> Rolled out *another* debugging change I committed accidentally. Unreviewed. * Configurations/Base.xcconfig: 2011-12-11 Geoffrey Garen <ggaren@apple.com> Rolled out a debug counter I committed accidentally. Unreviewed. * jit/JITStubs.cpp: (JSC::arityCheckFor): 2011-12-10 Geoffrey Garen <ggaren@apple.com> v8 benchmark takes 12-13 million function call slow paths due to extra arguments https://bugs.webkit.org/show_bug.cgi?id=74244 Reviewed by Filip Pizlo. .arguments function of order the Reversed 10% speedup on v8-raytrace, 1.7% speedup on v8 overall, neutral on Kraken and SunSpider. * bytecode/CodeBlock.h: (JSC::CodeBlock::valueProfileForArgument): Clarified that the interface to this function is an argument number. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): (JSC::BytecodeGenerator::emitCall): (JSC::BytecodeGenerator::emitConstruct): (JSC::BytecodeGenerator::isArgumentNumber): Switched to using CallFrame helper functions for computing offsets for arguments, rather than doing the math by hand. Switched to iterating argument offsets backwards (--) instead of forwards (++). * bytecompiler/BytecodeGenerator.h: (JSC::CallArguments::thisRegister): (JSC::CallArguments::argumentRegister): (JSC::CallArguments::registerOffset): Updated for arguments being reversed. * bytecompiler/NodesCodegen.cpp: Allocate arguments in reverse order. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::setArgument): (JSC::DFG::ByteCodeParser::flush): (JSC::DFG::ByteCodeParser::addCall): (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::handleMinMax): (JSC::DFG::ByteCodeParser::handleIntrinsic): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::processPhiStack): Use abstract argument indices that just-in-time convert to bytecode operands (i.e., indexes in the register file) through helper functions. This means only one piece of code needs to know how arguments are laid out in the register file. * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): Ditto. * dfg/DFGGraph.h: (JSC::DFG::Graph::valueProfileFor): Ditto. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): The whole point of this patch: Treat too many arguments as an arity match. * dfg/DFGOSRExit.h: (JSC::DFG::OSRExit::variableForIndex): (JSC::DFG::OSRExit::operandForIndex): Use helper functions, as above. * dfg/DFGOperands.h: (JSC::DFG::operandToArgument): (JSC::DFG::argumentToOperand): These are now the only two lines of code in the DFG compiler that know how arguments are laid out in memory. (JSC::DFG::Operands::operand): (JSC::DFG::Operands::setOperand): Use helper functions, as above. * dfg/DFGOperations.cpp: The whole point of this patch: Treat too many arguments as an arity match. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): Use helper functions, as above. Also, don't tag the caller frame slot as a cell, because it's not a cell. * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): Use helper functions, as above. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): Use helper functions, as above. (JSC::DFG::SpeculativeJIT::checkArgumentTypes): Use already-computed argument virtual register instead of recomputing by hand. * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::callFrameSlot): (JSC::DFG::SpeculativeJIT::argumentSlot): (JSC::DFG::SpeculativeJIT::callFrameTagSlot): (JSC::DFG::SpeculativeJIT::callFramePayloadSlot): (JSC::DFG::SpeculativeJIT::argumentTagSlot): (JSC::DFG::SpeculativeJIT::argumentPayloadSlot): Added a few helper functions for dealing with callee arguments specifically. These still build on top of our other helper functions, and have no direct knowledge of how arguments are laid out in the register file. (JSC::DFG::SpeculativeJIT::resetCallArguments): (JSC::DFG::SpeculativeJIT::addCallArgument): Renamed argumentIndex to argumentOffset to match CallFrame naming. (JSC::DFG::SpeculativeJIT::valueSourceReferenceForOperand): Use helper functions, as above. * interpreter/CallFrame.h: (JSC::ExecState::argumentOffset): (JSC::ExecState::argumentOffsetIncludingThis): (JSC::ExecState::argument): (JSC::ExecState::setArgument): (JSC::ExecState::thisArgumentOffset): (JSC::ExecState::thisValue): (JSC::ExecState::setThisValue): (JSC::ExecState::offsetFor): (JSC::ExecState::hostThisRegister): (JSC::ExecState::hostThisValue): Added a bunch of helper functions for computing where an argument is in the register file. Anything in the runtime that needs to access arguments should use these helpers. * interpreter/CallFrameClosure.h: (JSC::CallFrameClosure::setThis): (JSC::CallFrameClosure::setArgument): (JSC::CallFrameClosure::resetCallFrame): This stuff is a lot simpler, now that too many arguments counts as an arity match and doesn't require preserving two copies of our arguments. * interpreter/Interpreter.cpp: (JSC::Interpreter::slideRegisterWindowForCall): Only need to do something special if the caller provided too few arguments. Key simplification: We never need to maintain two copies of our arguments anymore. (JSC::eval): (JSC::loadVarargs): Use helper functions. (JSC::Interpreter::unwindCallFrame): Updated for new interface. (JSC::Interpreter::execute): (JSC::Interpreter::executeCall): (JSC::Interpreter::executeConstruct): (JSC::Interpreter::prepareForRepeatCall): Seriously, though: use helper functions. (JSC::Interpreter::privateExecute): No need to check for stack overflow when calling host functions because they have zero callee registers. (JSC::Interpreter::retrieveArguments): Explicitly tear off the arguments object, since there's no special constructor for this anymore. * interpreter/Interpreter.h: Reduced the C++ re-entry depth because some workers tests were hitting stack overflow in some of my testing. We should make this test more exact in future. * interpreter/RegisterFile.h: Death to all runtime knowledge of argument location that does not belong to the CallFrame class! * jit/JIT.cpp: (JSC::JIT::privateCompile): I am a broken record and I use helper functions. Also, the whole point of this patch: Treat too many arguments as an arity match. * jit/JITCall32_64.cpp: (JSC::JIT::compileLoadVarargs): * jit/JITCall.cpp: (JSC::JIT::compileLoadVarargs): Updated the argument copying math to use helper functions, for backwards-correctness. Removed the condition pertaining to declared argument count because, now that arguments are always in just one place, this optimization is valid for all functions. Standardized the if predicate for each line of the optimization. This might fix a bug, but I couldn't get the bug to crash in practice. * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_create_arguments): (JSC::JIT::emit_op_get_argument_by_val): (JSC::JIT::emitSlow_op_get_argument_by_val): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_create_arguments): (JSC::JIT::emit_op_get_argument_by_val): (JSC::JIT::emitSlow_op_get_argument_by_val): Removed cti_op_create_arguments_no_params optimization because it's no longer an optimization, now that arguments are always contiguous in a known location. Updated argument access opcode math for backwards-correctness. * jit/JITStubs.cpp: (JSC::arityCheckFor): Updated just like slideRegisterWindowForCall. This function is slightly different because it copies the call frame in addition to the arguments. (In the Interpreter, the call frame is not set up by this point.) (JSC::lazyLinkFor): The whole point of this patch: Treat too many arguments as an arity match. (JSC::DEFINE_STUB_FUNCTION): Updated for new iterface to tearOff(). * jit/JITStubs.h: * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::loadDoubleArgument): (JSC::SpecializedThunkJIT::loadCellArgument): (JSC::SpecializedThunkJIT::loadInt32Argument): Use helper functions! They build strong bones and teeth! * runtime/ArgList.cpp: (JSC::ArgList::getSlice): (JSC::MarkedArgumentBuffer::slowAppend): * runtime/ArgList.h: (JSC::MarkedArgumentBuffer::MarkedArgumentBuffer): (JSC::MarkedArgumentBuffer::~MarkedArgumentBuffer): (JSC::MarkedArgumentBuffer::at): (JSC::MarkedArgumentBuffer::clear): (JSC::MarkedArgumentBuffer::append): (JSC::MarkedArgumentBuffer::removeLast): (JSC::MarkedArgumentBuffer::last): (JSC::ArgList::ArgList): (JSC::ArgList::at): Updated for backwards-correctness. WTF::Vector doesn't play nice with backwards-ness, so I changed to using manual allocation. Fixed a FIXME about not all values being marked in the case of out-of-line arguments. I had to rewrite the loop anyway, and I didn't feel like maintaining fidelity to its old bugs. * runtime/Arguments.cpp: (JSC::Arguments::visitChildren): (JSC::Arguments::copyToArguments): (JSC::Arguments::fillArgList): (JSC::Arguments::getOwnPropertySlotByIndex): (JSC::Arguments::getOwnPropertySlot): (JSC::Arguments::getOwnPropertyDescriptor): (JSC::Arguments::putByIndex): (JSC::Arguments::put): (JSC::Arguments::tearOff): * runtime/Arguments.h: (JSC::Arguments::create): (JSC::Arguments::Arguments): (JSC::Arguments::argument): (JSC::Arguments::finishCreation): Secondary benefit of this patch: deleted lots of tricky code designed to maintain two different copies of function arguments. Now that arguments are always contiguous in one place in memory, this complexity can go away. Reduced down to one create function for the Arguments class, from three. Moved tearOff() into an out-of-line function because it's huge. Moved logic about whether to tear off eagerly into the Arguments class, so we didn't have to duplicate it elsewhere. * runtime/JSActivation.cpp: (JSC::JSActivation::JSActivation): (JSC::JSActivation::visitChildren): Renamed m_numParametersMinusThis to m_numCapturedArgs because if the value really were m_numParametersMinusThis we would be marking too much. (We shouldn't mark 'this' because it can't be captured.) Also, use helper functions. * runtime/JSActivation.h: (JSC::JSActivation::tearOff): Use helper functions. * runtime/JSArray.cpp: (JSC::JSArray::copyToArguments): * runtime/JSArray.h: Use helper functions, as above. 2011-12-10 Mark Hahnenberg <mhahnenberg@apple.com> JSC testapi is crashing on Windows https://bugs.webkit.org/show_bug.cgi?id=74233 Reviewed by Sam Weinig. Same error we've encountered before where we are calling the wrong version of visitChildren and objects that are still reachable aren't getting marked. This problem will go away soon with the removal of vptrs for these sorts of optimizations in favor of using the ClassInfo, but for now we can simply give JSFinalObject a bogus virtual method that Visual Studio can't optimize away to ensure that JSFinalObject will always have a unique vptr. We don't have to worry about JSString or JSArray right now, which are the other two special cases for visitChildren, since they already have their own virtual functions. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSObject.cpp: (JSC::JSFinalObject::vtableAnchor): * runtime/JSObject.h: 2011-12-10 Alexis Menard <alexis.menard@openbossa.org> Unused variable in YarrJIT.cpp. https://bugs.webkit.org/show_bug.cgi?id=74237 Reviewed by Andreas Kling. Variable is set but not used so we can remove it. * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::generatePatternCharacterOnce): 2011-12-09 Filip Pizlo <fpizlo@apple.com> DFG ArithMul power-of-two case does not check for overflow https://bugs.webkit.org/show_bug.cgi?id=74230 Reviewed by Gavin Barraclough. Disabled power-of-2 peephole optimization for multiplication, because it was wrong, and any attempt to fix it would likely introduce code bloat and register pressure. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileArithMul): 2011-12-09 David Levin <levin@chromium.org> REGRESSION(r101863-r102042): Assertion hit: m_verifier.isSafeToUse() in RefCountedBase::ref in FunctionCodeBlock https://bugs.webkit.org/show_bug.cgi?id=73886 Reviewed by Darin Adler. * runtime/SymbolTable.h: (JSC::SharedSymbolTable::SharedSymbolTable): Added deprecatedTurnOffVerifier for another JavaScriptObject, since JavaScriptCore objects allow use on multiple threads. Bug 58091 is about changing these deprecated calls to something else but that something else will still need to be in all of these places. 2011-12-09 Konrad Piascik <kpiascik@rim.com> Remove unnecessary file DissasemblerARM.cpp from build system https://bugs.webkit.org/show_bug.cgi?id=74184 Reviewed by Daniel Bates. * PlatformBlackBerry.cmake: 2011-12-09 Filip Pizlo <fpizlo@apple.com> DFG's interpretation of rare case profiles should be frequency-based not count-based https://bugs.webkit.org/show_bug.cgi?id=74170 Reviewed by Geoff Garen. DFG optimizes for rare cases only when the rare case counter is above some threshold and it also constitutes a large enough fraction of total function executions. Also added some minor debug logic. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::CodeBlock): * bytecode/CodeBlock.h: (JSC::CodeBlock::likelyToTakeSlowCase): (JSC::CodeBlock::couldTakeSlowCase): (JSC::CodeBlock::likelyToTakeSpecialFastCase): (JSC::CodeBlock::likelyToTakeDeepestSlowCase): (JSC::CodeBlock::likelyToTakeAnySlowCase): (JSC::CodeBlock::executionEntryCount): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::makeSafe): (JSC::DFG::ByteCodeParser::makeDivSafe): (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGDriver.cpp: (JSC::DFG::compile): * jit/JIT.cpp: (JSC::JIT::privateCompile): * runtime/Heuristics.cpp: (JSC::Heuristics::initializeHeuristics): * runtime/Heuristics.h: 2011-12-09 Oliver Hunt <oliver@apple.com> PutByValAlias unnecessarily clobbers GetIndexedPropertyStorage https://bugs.webkit.org/show_bug.cgi?id=74223 Reviewed by Geoffrey Garen. Don't clobber GetIndexedPropertyStorage when we see PutByValAlias * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::getIndexedPropertyStorageLoadElimination): 2011-12-09 David Levin <levin@chromium.org> Hash* iterators should allow comparison between const and const versions. https://bugs.webkit.org/show_bug.cgi?id=73370 Reviewed by Darin Adler. * wtf/HashTable.h: Add the operators needed to do this. (WTF::HashTableConstIterator::operator==): (WTF::HashTableConstIterator::operator!=): (WTF::HashTableIterator::operator==): (WTF::HashTableIterator::operator!=): (WTF::operator==): (WTF::operator!=): 2011-12-09 Michael Saboff <msaboff@apple.com> YARR: Multi-character read optimization for 8bit strings https://bugs.webkit.org/show_bug.cgi?id=74191 Reviewed by Oliver Hunt. Changed generatePatternCharacterOnce to generate code for 1 to 4 characters in the 8 bit case. This is worth 29% improvement on SunSpider regexp-dna test. It provides no benefit to v8-regexp. * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::generatePatternCharacterOnce): (JSC::Yarr::YarrGenerator::generate): Spelling fix in comment. 2011-12-09 David Levin <levin@chromium.org> Regression(r53595): Sync xhr requests in workers aren't terminated on worker close. https://bugs.webkit.org/show_bug.cgi?id=71695 Reviewed by Zoltan Herczeg. * wtf/MessageQueue.h: (WTF::MessageQueue::tryGetMessageIgnoringKilled): Added a way to get messages even after the queue has been killed. This is useful when one wants to kill a queue but then go through it to run clean up tasks from it. 2011-12-09 Adrienne Walker <enne@google.com> Fix HashMap<..., OwnPtr<...> >::add compilation errors https://bugs.webkit.org/show_bug.cgi?id=74159 Reviewed by Darin Adler. Add a constructor to OwnPtr that takes the empty value (nullptr_t) from HashTraits so that this function can compile. * wtf/OwnPtr.h: (WTF::OwnPtr::OwnPtr): 2011-12-09 Oliver Hunt <oliver@apple.com> Avoid reloading storage pointer for indexed properties unnecessarily https://bugs.webkit.org/show_bug.cgi?id=74136 Reviewed by Filip Pizlo. Add a node to represent loading property storage for indexed properties. This allows us to reduce code generated for sequential access of arrays, strings, etc. This results in up to 5% improvement in code that is very heavy on indexed reads, such as matrix operations in typed arrays and 20% faster on microbenchmarks. Currently this is only supported by GetByVal and other similar indexed reads. * bytecode/PredictedType.h: (JSC::isFixedIndexedStorageObjectPrediction): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleIntrinsic): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::getIndexedPropertyStorageLoadElimination): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileGetCharCodeAt): (JSC::DFG::SpeculativeJIT::compileGetByValOnString): (JSC::DFG::SpeculativeJIT::compileGetByValOnByteArray): (JSC::DFG::SpeculativeJIT::compileGetByValOnIntTypedArray): (JSC::DFG::SpeculativeJIT::compileGetByValOnFloatTypedArray): (JSC::DFG::SpeculativeJIT::compileGetIndexedPropertyStorage): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-12-08 Fady Samuel <fsamuel@chromium.org> [Chromium] Enable viewport metatag https://bugs.webkit.org/show_bug.cgi?id=73495 Reviewed by Darin Fisher. * wtf/Platform.h: Added ENABLE(VIEWPORT) tag. 2011-12-08 Adam Klein <adamk@chromium.org> Use HashMap<Node*, OwnPtr<...>> in ChildListMutationScope https://bugs.webkit.org/show_bug.cgi?id=73964 Reviewed by Darin Adler. * wtf/HashTraits.h: Add passOut(std::nullptr_t) to allow callers to use HashMap::take on a HashMap of OwnPtrs. 2011-12-08 Thouraya ANDOLSI <thouraya.andolsi@st.com> https://bugs.webkit.org/show_bug.cgi?id=74005 fix unaligned access memory in generatePatternCharacterOnce function for SH4 platforms. Reviewed by Gavin Barraclough. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::load16Unaligned): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::load16Unaligned): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::load16Unaligned): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::lshift32): (JSC::MacroAssemblerSH4::load8): (JSC::MacroAssemblerSH4::load16): (JSC::MacroAssemblerSH4::load16Unaligned): (JSC::MacroAssemblerSH4::branch8): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::load16Unaligned): * jit/JIT.h: * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::generatePatternCharacterOnce): 2011-12-08 Michael Saboff <msaboff@apple.com> Add 8 bit paths for StringTypeAdapter classes https://bugs.webkit.org/show_bug.cgi?id=73882 Reviewed by Darin Adler. Added is8Bit() method and writeTo(LChar*) methods to StringTypeAdapter<> classes. The writeTo(LChar*) method can be used if is8Bit() returns true. The non-native 8 bit classes contain ASSERT(is8Bit()) in their writeTo(LChar*). Updated all of the various versions of tryMakeString() to use 8 bit processing in the updated StringTypeAdapter<> classes. This has slight if any performance improvement on kraken. * runtime/UStringConcatenate.h: * wtf/text/StringConcatenate.h: (WTF::tryMakeString): * wtf/text/StringOperators.h: (WTF::StringAppend::is8Bit): (WTF::StringAppend::writeTo): 2011-12-07 Filip Pizlo <fpizlo@apple.com> DFG CSE should know that CheckFunction is pure https://bugs.webkit.org/show_bug.cgi?id=74044 Reviewed by Oliver Hunt. Possible slight win on V8, no regressions. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::checkFunctionElimination): 2011-12-07 Michael Saboff <msaboff@apple.com> StringBuilderTest.Append and StringBuilderTest.ToStringPreserveCapacity are failing. https://bugs.webkit.org/show_bug.cgi?id=73995 Reviewed by Geoffrey Garen. Problem was that a call to characters on an StringImpl associated with a StringBuilder that is being appended to gets stale. Added a new m_valid16BitShadowlen that keeps the length of the 16 bit shadow that has been upconverted or will be up converted with the first getCharacters(). When StringBuilder::characters or ::reifyString is called, further characters are upconverted if we have a shadow16bit copy and the m_valid16BitShadowlen is updated. * JavaScriptCore.exp: * wtf/text/StringBuilder.cpp: (WTF::StringBuilder::reifyString): * wtf/text/StringBuilder.h: (WTF::StringBuilder::StringBuilder): (WTF::StringBuilder::characters): (WTF::StringBuilder::clear): Cleaned up as part of the change. * wtf/text/StringImpl.cpp: (WTF::StringImpl::getData16SlowCase): (WTF::StringImpl::upconvertCharacters): * wtf/text/StringImpl.h: 2011-12-07 Filip Pizlo <fpizlo@apple.com> Compare and Swap should be enabled on ARMv7 https://bugs.webkit.org/show_bug.cgi?id=74023 Reviewed by Geoff Garen. Implemented weakCompareAndSwap in terms of LDREX/STREX and enabled PARALLEL_GC. It gives the expected speed-up on multi-core ARMv7 devices. * wtf/Atomics.h: (WTF::weakCompareAndSwap): * wtf/Platform.h: 2011-12-07 Filip Pizlo <fpizlo@apple.com> DFG CSE is overzealous with GetByVal https://bugs.webkit.org/show_bug.cgi?id=74042 Reviewed by Oliver Hunt. Made sure that the purity of GetByVal and the limited-clobber-itude of PutByVal is tested in all places that matter. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::byValIsPure): (JSC::DFG::Propagator::clobbersWorld): (JSC::DFG::Propagator::getByValLoadElimination): (JSC::DFG::Propagator::checkStructureLoadElimination): (JSC::DFG::Propagator::getByOffsetLoadElimination): (JSC::DFG::Propagator::getPropertyStorageLoadElimination): (JSC::DFG::Propagator::performNodeCSE): 2011-12-07 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r102267. http://trac.webkit.org/changeset/102267 https://bugs.webkit.org/show_bug.cgi?id=74032 Breaks build on Chromium Mac Debug (Requested by aklein on #webkit). * wtf/HashTraits.h: 2011-12-07 Adam Klein <adamk@chromium.org> Use HashMap<Node*, OwnPtr<...>> in ChildListMutationScope https://bugs.webkit.org/show_bug.cgi?id=73964 Reviewed by Ryosuke Niwa. * wtf/HashTraits.h: Add passOut(std::nullptr_t) to allow callers to use HashMap::take on an entry whose value is null. 2011-12-07 Filip Pizlo <fpizlo@apple.com> Non-Mac devices should benefit from a larger heap https://bugs.webkit.org/show_bug.cgi?id=74015 Reviewed by Geoff Garen. Removed the ENABLE(LARGE_HEAP) option from Platform.h, since it was only used in Heap.cpp, and got in the way of having more granular, per-platform control over what the heap size should be. Bumped the heap size to 8MB on iOS (was 512KB). * heap/Heap.cpp: (JSC::GCTimer::heapSizeForHint): * wtf/Platform.h: 2011-11-30 Simon Hausmann <simon.hausmann@nokia.com> [Qt] V8 build fixes. Reviewed by Tor Arne Vestbø. * yarr/yarr.pri: Don't rely on Source/JavaScriptCore being in VPATH. Prefix SOURCES correctly and make sure that runtime/ is in the include search path when building with v8. 2011-12-06 Filip Pizlo <fpizlo@apple.com> Zapping a block that is Marked leads to dead objects being mistaken for live ones https://bugs.webkit.org/show_bug.cgi?id=73982 Reviewed by Geoff Garen. Changed the zapping code to ignore blocks that are Marked or Zapped. Additionally, the code asserts that: - If we zap a Marked or Zapped block then the free list is empty, because this can only happen if the block was never free-listed. - Zapping can only happen for Marked, Zapped, or FreeListed blocks, since Allocated blocks are those that cannot be referred to by SizeClass::currentBlock (since SizeClass::currentBlock only refers to blocks that are candidates for allocation, and Allocated blocks are those who have been exhausted by allocation and will not be allocated from again), and New blocks cannot be referred to by anything except during a brief window inside the allocation slow-path. * heap/MarkedBlock.cpp: (JSC::MarkedBlock::zapFreeList): 2011-12-06 Filip Pizlo <fpizlo@apple.com> DFG 32_64 call linking does not handle non-cell callees correctly https://bugs.webkit.org/show_bug.cgi?id=73965 Reviewed by Sam Weinig. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitCall): 2011-12-06 Sam Weinig <sam@webkit.org> Remove unintentional type name shadowing in the Interpreter https://bugs.webkit.org/show_bug.cgi?id=73963 Reviewed by Oliver Hunt. * interpreter/Interpreter.cpp: (JSC::Interpreter::prepareForRepeatCall): Replace the parameter name FunctionExecutable, which shadows the FunctionExecutable type name, with functionExecutable. 2011-12-06 Michael Saboff <msaboff@apple.com> r102146 from 73875 broke fast/js/encode-URI-test.html https://bugs.webkit.org/show_bug.cgi?id=73950 Reviewed by Gavin Barraclough. * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncUnescape): Restructured to handle the %uHHHH case to output the resulting character and continue so that a failure in finding 4 hex digits will fall through and output the '%'. Due to style check, changed the temporary character variable to a more descriptive name. 2011-12-06 Filip Pizlo <fpizlo@apple.com> GC zapping logic could benefit from some more assertions https://bugs.webkit.org/show_bug.cgi?id=73947 Reviewed by Gavin Barraclough. - If you're in a zapped block and you're zapped, then your mark bit should never be set. - If you're being marked, then you should never be zapped. * heap/MarkedBlock.h: (JSC::MarkedBlock::isLive): * runtime/Structure.h: (JSC::MarkStack::internalAppend): 2011-12-06 Oliver Hunt <oliver@apple.com> Don't allocate register in typedarray control flow https://bugs.webkit.org/show_bug.cgi?id=73944 Reviewed by Gavin Barraclough. Move a temporary allocation outside of control flow. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileGetByValOnFloatTypedArray): 2011-12-06 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=68328 The generator and intrinsic fields in HashTableValue/HashEntry and associated structures and methods are redundant Reviewed by Geoff Garen. Move the instrinsic enum out of the DFG, into runtime. Add entires for all host functions that have an intrinsic in the form of a generated thunk. Remove the thunk pointer from the hashtable, and make Intrinsic field no longer ifdef on JIT/DFG. In getHostFunction select a thunk genertaor to use based on the Intrinsic. * JavaScriptCore.xcodeproj/project.pbxproj: * create_hash_table: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::handleIntrinsic): * dfg/DFGCapabilities.h: * dfg/DFGIntrinsic.h: Removed. * jit/JITStubs.cpp: (JSC::JITThunks::hostFunctionStub): * jit/JITStubs.h: * runtime/Executable.cpp: (JSC::ExecutableBase::intrinsic): (JSC::NativeExecutable::intrinsic): * runtime/Executable.h: (JSC::ExecutableBase::intrinsicFor): (JSC::NativeExecutable::create): (JSC::NativeExecutable::finishCreation): * runtime/Intrinsic.h: Copied from Source/JavaScriptCore/dfg/DFGIntrinsic.h. * runtime/JSGlobalData.cpp: (JSC::thunkGeneratorForIntrinsic): (JSC::JSGlobalData::getHostFunction): * runtime/JSGlobalData.h: * runtime/Lookup.cpp: (JSC::HashTable::createTable): (JSC::setUpStaticFunctionSlot): * runtime/Lookup.h: (JSC::HashEntry::initialize): (JSC::HashEntry::intrinsic): 2011-12-06 Michael Saboff <msaboff@apple.com> Add 8 bit paths to global object functions https://bugs.webkit.org/show_bug.cgi?id=73875 Added 8 bit paths for converions methods. This is worth 1.5% on kraken audio-oscillator, 1.6% on stanford-crypto-ccm and 2.5% on stanford-crypto-sha256-iterative. See bug for a full report. Reviewed by Oliver Hunt. * runtime/JSGlobalObjectFunctions.cpp: (JSC::decode): Split into a templated helper. (JSC::parseInt): Split into a templated helper. (JSC::parseFloat): Added an 8 bit path (JSC::globalFuncEscape): Added 8 bit path (JSC::globalFuncUnescape): Added 8 bit path * runtime/JSStringBuilder.h: (JSC::JSStringBuilder::append): New append for LChar * wtf/text/StringBuilder.h: (WTF::StringBuilder::append): New append for LChar 2011-11-21 Balazs Kelemen <kbalazs@webkit.org> Enable ParallelJobs by default https://bugs.webkit.org/show_bug.cgi?id=70032 Reviewed by Zoltan Herczeg. According to measurements on Mac and Linux it is a considerable speedup for SVG on multicore. Remove the ENABLE(PARALLEL_JOBS) guard. Fix build on Windows and Chromium. * JavaScriptCore.gypi: Add the files to the build. It was missing for the gyp build system. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Export symbols. * wtf/ParallelJobs.h: * wtf/ParallelJobsGeneric.cpp: (WTF::ParallelEnvironment::ParallelEnvironment): (WTF::ParallelEnvironment::execute): Deinline these to avoid exporting a lot of symbols. These are non-trivial and called only once on a given object so it doesn't seems to be worthwile to inline them. Additionally fix a signed-unsigned comparison in the constructor. * wtf/ParallelJobsGeneric.h: * wtf/Platform.h: 2011-12-06 Simon Hausmann <simon.hausmann@nokia.com> [Qt] build-jsc script doesn't work https://bugs.webkit.org/show_bug.cgi?id=73910 Reviewed by Tor Arne Vestbø. * JavaScriptCore.pro: Build WTF before JavaScriptCore and JSC (moved from top-level WebKit.pro). Also add v8 scopes to only build WTF during v8 builds. 2011-12-05 Anders Carlsson <andersca@apple.com> Add HashMap::keys() and HashMap::values() for easy iteration of hash map keys and values in C++11. Reviewed by Darin Adler. * wtf/HashMap.h: 2011-12-05 Michael Saboff <msaboff@apple.com> Create StringImpl::empty() as an 8 bit string https://bugs.webkit.org/show_bug.cgi?id=73871 Reviewed by Oliver Hunt. * wtf/text/StringStatics.cpp: (WTF::StringImpl::empty): Changed to be an 8 bit string. 2011-12-05 Darin Adler <darin@apple.com> Convert JSClassRef to use HashMap<OwnPtr> https://bugs.webkit.org/show_bug.cgi?id=73780 Reviewed by Andreas Kling. * API/JSCallbackObjectFunctions.h: (JSC::JSCallbackObject::getOwnPropertyNames): Use get() on the hash map entries because the hash map now has an OwnPtr instead of a raw pointer. * API/JSClassRef.cpp: (OpaqueJSClass::OpaqueJSClass): No need to initialize m_staticValues and m_staticFunctions since they are now OwnPtr. Use adoptPtr when allocating. Removed the code that gets and deletes existing entries, and just use set, which now handles deletion automatically due to it being OwnPtr. (OpaqueJSClass::~OpaqueJSClass): Replaced code to do all the deletion with assertion-only NDEBUG-only code. (OpaqueJSClassContextData::OpaqueJSClassContextData): Use adoptPtr when allocating. Use OwnPtr when adding. Removed unneeded code to set staticValues and staticFunctions to 0. Removed unneeded destructor. (OpaqueJSClass::staticValues): Added get call. Also removed unneeded local. (OpaqueJSClass::staticFunctions): Ditto. (OpaqueJSClass::prototype): Added use of adoptPtr. * API/JSClassRef.h: Made the static values and static functions tables use OwnPtr for the entries. Also used OwnPtr for the pointers to the tables themselves. Also removed ~OpaqueJSClassContextData(), letting the compiler generate it. 2011-12-05 Oliver Hunt <oliver@apple.com> Land uncommitted bit of float array support https://bugs.webkit.org/show_bug.cgi?id=73873 Reviewed by Filip Pizlo. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileGetByValOnFloatTypedArray): 2011-12-05 Benjamin Poulain <benjamin@webkit.org> Update String::containsOnlyASCII() to handle 8 bits strings https://bugs.webkit.org/show_bug.cgi?id=73799 Reviewed by Darin Adler. Implement String::containsOnlyASCII() so that it does not call String::characters(). * wtf/text/WTFString.h: (WTF::String::containsOnlyASCII): 2011-12-05 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix for non-DFG platforms. * dfg/DFGRepatch.h: 2011-12-05 Filip Pizlo <fpizlo@apple.com> Old JIT emits 32-bit offsets for put_by_id but sometimes patches them as if they were compact offsets https://bugs.webkit.org/show_bug.cgi?id=73861 Reviewed by Gavin Barraclough. * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::resetPatchPutById): 2011-12-05 Filip Pizlo <fpizlo@apple.com> Unreviewed, build fixes for ARM. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::unreachableForPlatform): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::loadDouble): (JSC::MacroAssemblerARMv7::loadFloat): (JSC::MacroAssemblerARMv7::storeFloat): (JSC::MacroAssemblerARMv7::convertFloatToDouble): (JSC::MacroAssemblerARMv7::convertDoubleToFloat): 2011-12-05 Benjamin Poulain <benjamin@webkit.org> Update String::containsOnlyLatin1() to avoid converting to 16 bits https://bugs.webkit.org/show_bug.cgi?id=73797 Reviewed by Andreas Kling. When the String use 8bits StringImpl, there is no need to iterate over the string. The function charactersAreAllLatin1() is removed because it is not used anywhere. * wtf/text/WTFString.h: (WTF::String::containsOnlyLatin1): 2011-12-05 Michael Saboff <msaboff@apple.com> 8 bit string work slows down Kraken json-stringify-tinderbox https://bugs.webkit.org/show_bug.cgi?id=73457 Added 8 bit path to StringBuilder. StringBuilder starts assuming 8 bit contents and gets converted to 16 bit upon seeing the first 16 bit character or string. Split appendUninitialiezed into an inlined fast and function call slow case. Factored out the processing of the UString argument from Stringifier::appendQuotedString() to a static templated function based on character size. This change eliminates 5% of the 7% slowdown to json-stringify-tinderbox. This change introduces a 4.8% slowdown to json-parse-financial. This slowdown will be addressed in a subsequent patch to StringImpl::equal. Reviewed by Oliver Hunt. * runtime/JSONObject.cpp: (JSC::appendStringToUStringBuilder): (JSC::Stringifier::appendQuotedString): * wtf/text/StringBuilder.cpp: (WTF::StringBuilder::resize): (WTF::StringBuilder::allocateBuffer): (WTF::StringBuilder::allocateBufferUpConvert): (WTF::LChar): (WTF::UChar): (WTF::StringBuilder::reserveCapacity): (WTF::StringBuilder::appendUninitialized): (WTF::StringBuilder::appendUninitializedSlow): (WTF::StringBuilder::append): (WTF::StringBuilder::shrinkToFit): * wtf/text/StringBuilder.h: (WTF::StringBuilder::StringBuilder): (WTF::StringBuilder::append): (WTF::StringBuilder::operator[]): (WTF::StringBuilder::characters8): (WTF::StringBuilder::characters16): (WTF::StringBuilder::charactersBlah): (WTF::LChar): (WTF::UChar): 2011-12-01 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=73624 JIT + INTERPRETER builds are broken Reviewed by Geoff Garen, Sam Weinig. These don't fallback to the interpreter correctly. Thunk creation assumes that is the JIT is compiled in, then it is enabled. * jit/JITStubs.cpp: (JSC::JITThunks::JITThunks): * runtime/Executable.h: (JSC::NativeExecutable::create): (JSC::NativeExecutable::finishCreation): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::getHostFunction): 2011-12-05 Zoltan Herczeg <zherczeg@webkit.org> MacroAssemblerSH4 does not implement readCallTarget https://bugs.webkit.org/show_bug.cgi?id=73434 Reviewed by Csaba Osztrogonác. * assembler/MacroAssemblerSH4.h: Support for SH4. (JSC::MacroAssemblerSH4::readCallTarget): * assembler/SH4Assembler.h: (JSC::SH4Assembler::readCallTarget): 2011-12-04 Filip Pizlo <fpizlo@apple.com> DFG should optimize strict equality https://bugs.webkit.org/show_bug.cgi?id=73764 Reviewed by Oliver Hunt. 1% speed-up on V8. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compileStrictEqForConstant): (JSC::DFG::SpeculativeJIT::compileStrictEq): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compileIntegerCompare): (JSC::DFG::SpeculativeJIT::compileDoubleCompare): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compileIntegerCompare): (JSC::DFG::SpeculativeJIT::compileDoubleCompare): (JSC::DFG::SpeculativeJIT::compile): 2011-12-03 Darin Adler <darin@apple.com> Use HashMap<OwnPtr> for ScriptSampleRecordMap https://bugs.webkit.org/show_bug.cgi?id=73758 Reviewed by Andreas Kling. * bytecode/SamplingTool.cpp: (JSC::SamplingTool::notifyOfScope): Added adoptPtr. (JSC::SamplingTool::dump): Added get. * bytecode/SamplingTool.h: Changed the value type of ScriptSampleRecordMap to be OwnPtr. 2011-12-03 Darin Adler <darin@apple.com> Use HashMap<OwnPtr> for the opaqueJSClassData map https://bugs.webkit.org/show_bug.cgi?id=73759 Reviewed by Andreas Kling. * API/JSClassRef.cpp: (OpaqueJSClass::contextData): Update types. * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::~JSGlobalData): Add an explicit clear of opaqueJSClassData to keep the timing the same. If we didn't care about the order of operations, we could remove this, too. * runtime/JSGlobalData.h: Use OwnPtr instead of raw pointer for the mapped type in the opaqueJSClassData map. 2011-12-03 Darin Adler <darin@apple.com> Change HashMap implementation to use the pass type and peek type from traits for the mapped value https://bugs.webkit.org/show_bug.cgi?id=72474 Reviewed by Anders Carlsson. * wtf/HashMap.h: Added ReferenceTypeMaker struct template. Get PassInType, PassOutType, and PeekType from the traits of the mapped value instead of hard-coding them here. Changed inlineAdd to take a reference to the PassInType instead of the PassInType itself, to accomodate a PassInType that can't be copied. Use the store, peek, and passOut functions from the traits as well. * wtf/HashTraits.h: Updated GenericHashTraits and HashTraits for OwnPtr to include PassInType, PassOutType, PeekType, store, passOut, and peek. Before this, the file had an earlier version that was just PassType, PeekType, pass, and peek. Also commented the HashTraits for RefPtr to foreshadow some work we can do there. * wtf/RefPtrHashMap.h: Same changes as HashMap.h. 2011-12-02 David Levin <levin@chromium.org> Rename WTF class from TemporarilyChange to TemporaryChange. https://bugs.webkit.org/show_bug.cgi?id=73479 Reviewed by Eric Seidel. * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/TemporaryChange.h: Renamed from Source/JavaScriptCore/wtf/TemporarilyChange.h. (WTF::TemporaryChange::TemporaryChange): (WTF::TemporaryChange::~TemporaryChange): 2011-12-02 Mark Hahnenberg <mhahnenberg@apple.com> REGRESSION (r99754): All layout tests crash on Windows https://bugs.webkit.org/show_bug.cgi?id=72305 Reviewed by Geoffrey Garen. Fixes a crash in release builds on Windows. Windows was optimizing the out-of-line virtual destructor in JSFunction away, which left it with no virtual functions. Its vtable ptr was then identical to that of a different class, therefore the optimization in the visitChildren helper function in MarkedStack.cpp was calling an incorrect version of visitChildren on the object, which left its children unmarked, causing them to be collected when they were still reachable. * runtime/JSFunction.cpp: (JSC::JSFunction::vtableAnchor): Add a virtual function to JSFunction that Visual Studio can't optimize away. * runtime/JSFunction.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::storeVPtrs): Add checks to make sure that all virtual pointers that we rely on for optimization purposes are distinct from one another. 2011-12-02 Oliver Hunt <oliver@apple.com> Improve float array support in the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=73722 Reviewed by Gavin Barraclough. Add basic support for float typed arrays in JSC. This is currently less optimal than it could be in the following ways: * float32Array1[0] = float32Array2[0] (eg. an element by element copy) promotes float to double and then back to float. * float64Array[0] will always perform NaN tests in order to prevent signalling NaNs from entering the engine. We also don't support Float32Array on ARMv7 * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::loadDouble): (JSC::MacroAssemblerARMv7::loadFloat): (JSC::MacroAssemblerARMv7::storeDouble): (JSC::MacroAssemblerARMv7::storeFloat): (JSC::MacroAssemblerARMv7::convertFloatToDouble): (JSC::MacroAssemblerARMv7::convertDoubleToFloat): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::loadDouble): (JSC::MacroAssemblerX86Common::loadFloat): (JSC::MacroAssemblerX86Common::storeDouble): (JSC::MacroAssemblerX86Common::storeFloat): (JSC::MacroAssemblerX86Common::convertDoubleToFloat): (JSC::MacroAssemblerX86Common::convertFloatToDouble): * assembler/X86Assembler.h: (JSC::X86Assembler::cvtsd2ss_rr): (JSC::X86Assembler::cvtss2sd_rr): (JSC::X86Assembler::movsd_rm): (JSC::X86Assembler::movss_rm): (JSC::X86Assembler::movsd_mr): (JSC::X86Assembler::movss_mr): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateFloat32Array): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): (JSC::DFG::SpeculativeJIT::compileGetByValOnFloatTypedArray): (JSC::DFG::SpeculativeJIT::compilePutByValForFloatTypedArray): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::compile): 2011-12-02 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r101801. http://trac.webkit.org/changeset/101801 https://bugs.webkit.org/show_bug.cgi?id=73667 Build is still broken (Requested by Ossy on #webkit). * assembler/SH4Assembler.h: 2011-12-01 Darin Adler <darin@apple.com> Prepare to deploy pass and peek types in the HashMap class https://bugs.webkit.org/show_bug.cgi?id=73477 Reviewed by Adam Roben. This patch adds private typedefs inside the HashMap class, and uses them as appropriate. A future patch will actually tie those typedefs to hash traits, which will allow us to make HashMap work with OwnPtr mapped values and to optimize how HashMap works with RefPtr mapped values. Also changed the hash translator and adapter struct templates to use template functions to simplify them and make them more flexible. Also removed some unused template arguments. This goes out of its way to not change behavior. Future patches will change the peek type to be a reference type, which will reduce reference count churn a bit for hash tables with RefPtr mapped values, and then do further optimizations for RefPtr and OwnPtr by getting types from the hash traits. * wtf/HashMap.h: Added MappedPassInType, MappedPassOutType, and MappedPeekType typedefs, and used them for the arguments and return types of the get, set, add, take, and inlineAdd functions. (WTF::HashMapTranslator): Changed this struct template to take fewer arguments, and changed its member functions to be function templates instead. This allows the compiler to determine types more flexibly and also simplifies use of it. (WTF::HashMapTranslatorAdapter): Ditto. (WTF::HashMap::find): Updated to use new HashMapTranslatorAdapter. Also reduced the arguments passed to the HashTable function template. (WTF::HashMap::contains): Ditto. (WTF::HashMap::inlineAdd): Ditto. Also take MappedPassInType. (WTF::HashMap::set): Ditto. (WTF::HashMap::add): Ditto. (WTF::HashMap::inlineGet): Ditto, but return MappedPeekType. (WTF::HashMap::get): Ditto. (WTF::HashMap::take): Ditto, but return MappedPassOutType and use that type in the implementation. (WTF::deleteAllValues): Removed unneeded template arguments from call to deleteAllPairSeconds. (WTF::deleteAllKeys): Removed unneeded template arguments from call to deleteAllPairFirsts. * wtf/HashSet.h: (WTF::IdentityExtractor): Changed this to be a struct rather than a struct template, and replaced the extract function with a function template. This allows the compiler to deduce the type. (WTF::HashSetTranslatorAdapter): Changed this struct template to take fewer arguments, and changed its member functions to be function templates instead. This allows the compiler to determine types more flexibly and also simplifies use of it. (WTF::HashSet::find): Updated to use new HashSetTranslatorAdapter. Also reduced the arguments passed to the HashTable function template. (WTF::HashSet::contains): Ditto. (WTF::HashSet::add): Ditto. * wtf/HashTable.h: (WTF::IdentityHashTranslator): Changed this struct template to take fewer arguments, and changed its member functions to be function templates instead. This allows the compiler to determine types more flexibly and also simplifies use of it. (WTF::HashTable::add): Reduced arguments passed to the function template. (WTF::HashTable::find): Ditto, also reversed the template arguments so the translator comes first so the compiler can deduce the other type. (WTF::HashTable::contains): Ditto. (WTF::HashTable::lookup): Ditto. (WTF::HashTable::lookupForWriting): Ditto. (WTF::HashTable::checkKey): Ditto. (WTF::HashTable::fullLookupForWriting): Ditto. (WTF::HashTable::add): Ditto. (WTF::HashTable::addPassingHashCode): Ditto. (WTF::HashTable::find): Ditto. (WTF::HashTable::contains): Ditto. * wtf/ListHashSet.h: (WTF::ListHashSetNodeHashFunctions): Changed this struct template to take fewer arguments, and changed its member functions to be function templates instead. This allows the compiler to determine types more flexibly and also simplifies use of it. (WTF::ListHashSet::find): Reduced the arguments passed to the HashTable functon template. (WTF::ListHashSetTranslatorAdapter): Changed this struct template in the same way we changed ListHashSetNodeHashFunctions above. (WTF::ListHashSetTranslatorAdapter::equal): (WTF::::contains): (WTF::::add): (WTF::::insertBefore): * wtf/RefPtrHashMap.h: Updated comments. Removed the RefPtrHashMapRawKeyTranslator struct template; we can use the HashMapTranslator struct template from HashMap.h instead now that it is more flexible. Added MappedPassInType, MappedPassOutType, and MappedPeekType typedefs, and used them for the arguments and return types of the get, inlineGet, set, add, take, and inlineAdd functions. Changed the name of the RawKeyTranslator type to Translator since it's now a class that can handle both raw keys and conventional keys. (WTF::HashMap::find): Changed to use Translator instead of RawKeyTranslator. Reduced the arguments passed to the HashTable function template. (WTF::HashMap::contains): Ditto. (WTF::HashMap::inlineAdd): Ditto. Also take MappedPassInType. (WTF::HashMap::set): Ditto. (WTF::HashMap::add): Ditto. (WTF::HashMap::inlineGet): Ditto, but return MappedPeekType. (WTF::HashMap::get): Ditto. (WTF::HashMap::take): Ditto, but return MappedPassOutType and use that type in the implementation. (WTF::deleteAllValues): Removed unneeded template arguments from call to deleteAllPairSeconds. (WTF::deleteAllKeys): Removed unneeded template arguments from call to deleteAllPairFirsts. 2011-12-02 Zoltan Herczeg <zherczeg@webkit.org> MacroAssemblerSH4 does not implement readCallTarget https://bugs.webkit.org/show_bug.cgi?id=73434 Reviewed by Csaba Osztrogonác. * assembler/SH4Assembler.h: (JSC::SH4Assembler::readCallTarget): Support for SH4. 2011-12-02 Hajime Morrita <morrita@chromium.org> Unreviewed, rolling out r101751 and r101775. http://trac.webkit.org/changeset/101751 http://trac.webkit.org/changeset/101775 https://bugs.webkit.org/show_bug.cgi?id=73191 breaks Windows build * JavaScriptCore.xcodeproj/project.pbxproj: * config.h: * runtime/JSExportMacros.h: Removed. * wtf/ExportMacros.h: * wtf/Platform.h: * wtf/WTFThreadData.h: * wtf/text/AtomicString.h: * wtf/text/StringStatics.cpp: 2011-12-01 Hajime Morrita <morrita@chromium.org> JS_INLINE and WTF_INLINE should be visible from WebCore https://bugs.webkit.org/show_bug.cgi?id=73191 - Moved Export related macro definitions from config.h to ExportMacros.h and JSExportMacros.h. - Moved WTF_USE_JSC and WTF_USE_V8 from various config.h family to Platform.h. - Replaced JS_EXPORTDATA in wtf moudule with newly introduced WTF_EXPORTDATA. Reviewed by Kevin Ollivier. * JavaScriptCore.xcodeproj/project.pbxproj: * config.h: * runtime/JSExportMacros.h: Added. * wtf/ExportMacros.h: * wtf/Platform.h: * wtf/WTFThreadData.h: * wtf/text/AtomicString.h: * wtf/text/StringStatics.cpp: 2011-12-01 Michael Saboff <msaboff@apple.com> Changes proposed for 73457 slow down Kraken json-parse-financial https://bugs.webkit.org/show_bug.cgi?id=73584 Restructured StringImpl::equal to take advantage of 8 or 4 bytes at a time when possible. This is worth ~3% on Kraken json-parse-financial. It provides ~2% on SunSpider string-unpack-code. Reviewed by Sam Weinig. * wtf/text/StringImpl.cpp: (WTF::equal): 2011-12-01 Oliver Hunt <oliver@apple.com> Support integer typed arrays in the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=73608 Reviewed by Filip Pizlo. Add support for all the integral typed arrays in the DFG JIT. Currently this loads the contents of Uint32 arrays as doubles, which is clearly not as efficient as it could be, but this is still in the order of 10-20x faster than the existing behaviour. This needed us to add support for writing 16bit values to the macroassembler, and also to support double<->unsigned conversion. * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::strh): (JSC::ARMv7Assembler::vcvt_floatingPointToUnsigned): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::store16): (JSC::MacroAssemblerARMv7::truncateDoubleToUint32): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::store16): (JSC::MacroAssemblerX86Common::truncateDoubleToUint32): * assembler/X86Assembler.h: (JSC::X86Assembler::movw_rm): (JSC::X86Assembler::cvttsd2siq_rr): * bytecode/PredictedType.cpp: (JSC::predictionToString): (JSC::predictionFromClassInfo): * bytecode/PredictedType.h: (JSC::isInt8ArrayPrediction): (JSC::isInt16ArrayPrediction): (JSC::isInt32ArrayPrediction): (JSC::isUint8ArrayPrediction): (JSC::isUint16ArrayPrediction): (JSC::isUint32ArrayPrediction): (JSC::isFloat32ArrayPrediction): (JSC::isFloat64ArrayPrediction): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::initialize): (JSC::DFG::AbstractState::execute): * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateInt8Array): (JSC::DFG::Node::shouldSpeculateInt16Array): (JSC::DFG::Node::shouldSpeculateInt32Array): (JSC::DFG::Node::shouldSpeculateUint8Array): (JSC::DFG::Node::shouldSpeculateUint16Array): (JSC::DFG::Node::shouldSpeculateUint32Array): (JSC::DFG::Node::shouldSpeculateFloat32Array): (JSC::DFG::Node::shouldSpeculateFloat64Array): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::compileGetTypedArrayLength): (JSC::DFG::SpeculativeJIT::compileGetByValOnIntTypedArray): (JSC::DFG::SpeculativeJIT::compilePutByValForIntTypedArray): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/JSGlobalData.h: 2011-12-01 Benjamin Poulain <benjamin@webkit.org> URLs are encoded in UTF-8, then decoded as if they are Latin1 https://bugs.webkit.org/show_bug.cgi?id=71758 Reviewed by Darin Adler. Add the operator == between a String and a Vector of char. The implementation is the same as the comparison of String and char* but adds the length as a parameter for comparing the strings. * JavaScriptCore.exp: * wtf/text/StringImpl.h: (WTF::equal): * wtf/text/WTFString.h: (WTF::operator==): (WTF::operator!=): 2011-12-01 Martin Robinson <mrobinson@igalia.com> [GTK] Read fonts from the jhbuild root https://bugs.webkit.org/show_bug.cgi?id=73487 Reviewed by Gustavo Noronha Silva. Read fonts from the jhbuild root instead of from the system. This will ensure that all testers use the same fonts instead of leaving this up to luck. * wtf/gobject/GlibUtilities.h: Add Assertions.h which was required for the WebKit2TestRunner. 2011-12-01 Martin Robinson <mrobinson@igalia.com> [GTK] Add a helper function to find the current executable's path https://bugs.webkit.org/show_bug.cgi?id=73473 Reviewed by Gustavo Noronha Silva. Add a WTF helper which gets the binary path. This is currently only used in WebKit2. * GNUmakefile.list.am: Add the new file to the source list. * wtf/gobject/GlibUtilities.cpp: Added. (getCurrentExecutablePath): * wtf/gobject/GlibUtilities.h: Added. 2011-12-01 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r101691. http://trac.webkit.org/changeset/101691 https://bugs.webkit.org/show_bug.cgi?id=73588 Tests fail on Chromium bots, early warning system warned committer, please adjust test_expectations in patch (Requested by scheib on #webkit). * JavaScriptCore.exp: * wtf/text/StringImpl.h: * wtf/text/WTFString.h: 2011-12-01 Filip Pizlo <fpizlo@apple.com> ARMv7 only allows for one-shot patching of compact offsets, while the JIT expects to be able to repatch https://bugs.webkit.org/show_bug.cgi?id=73548 Reviewed by Oliver Hunt. * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::setUInt7ForLoad): 2011-11-30 Benjamin Poulain <benjamin@webkit.org> URLs are encoded in UTF-8, then decoded as if they are Latin1 https://bugs.webkit.org/show_bug.cgi?id=71758 Reviewed by Darin Adler. Add the operator == between a String and a Vector of char. The implementation is the same as the comparison of String and char* but adds the length as a parameter for comparing the strings. * JavaScriptCore.exp: * wtf/text/StringImpl.h: (WTF::equal): * wtf/text/WTFString.h: (WTF::operator==): (WTF::operator!=): 2011-11-30 Dmitry Lomov <dslomov@google.com> https://bugs.webkit.org/show_bug.cgi?id=73503 [Chromium][V8] Implement ArrayBuffer transfer in chromium. Portions of this patch come from Luke Zarko. Reviewed by David Levin. * wtf/ArrayBuffer.cpp: (WTF::ArrayBuffer::transfer): Changed prototype from pointers to RefPtr. * wtf/ArrayBuffer.h: (WTF::ArrayBufferContents::transfer): Changed prototype from pointers to RefPtr. (WTF::ArrayBuffer::isNeutered): * wtf/TypedArrayBase.h: (WTF::TypedArrayBase::neuter): 2011-12-01 Chao-ying Fu <fu@mips.com> MacroAssemblerMIPS does not implement readCallTarget https://bugs.webkit.org/show_bug.cgi?id=73432 Reviewed by Zoltan Herczeg. * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::readCallTarget): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::readCallTarget): 2011-12-01 Noel Gordon <noel.gordon@gmail.com> [chromium] Remove wtf/qt/ThreadingQt.cpp from the gyp projects https://bugs.webkit.org/show_bug.cgi?id=73527 Reviewed by Simon Hausmann. wtf/qt/ThreadingQt.cpp was removed in r101477 * JavaScriptCore.gypi: remove wtf/qt/ThreadingQt.cpp 2011-12-01 Filip Pizlo <fpizlo@apple.com> BitVector isInline check could fail https://bugs.webkit.org/show_bug.cgi?id=70691 Reviewed by Gavin Barraclough. Switch back to using the high bit as the inline marker, to make all of the bit indexing operations simpler. Computing the size in words and in bytes of a bitvector, using the number of bits as input is error-prone enough; and with the current approach to solving the X86 bug we end up getting it wrong. Making it right seems hard. So instead, to solve the original problem (the high bit may be meaningful on 32-bit systems), the out-of-line storage pointer is right-shifted by 1. Compared to the original BitVector code, this is a much smaller change (just three lines). This solves a bug where the DFG was corrupting its call frame because BitVector lost track of some bits. * wtf/BitVector.cpp: (WTF::BitVector::setSlow): (WTF::BitVector::resizeOutOfLine): * wtf/BitVector.h: (WTF::BitVector::quickGet): (WTF::BitVector::quickSet): (WTF::BitVector::quickClear): (WTF::BitVector::makeInlineBits): (WTF::BitVector::isInline): (WTF::BitVector::outOfLineBits): 2011-11-30 Filip Pizlo <fpizlo@apple.com> DFG should make it easier to notice node boundaries in disassembly https://bugs.webkit.org/show_bug.cgi?id=73509 Rubber-stamped by Gavin Barraclough If you set XOR_DEBUG_AID to 1 in DFGCommon.h, a pair of xor's will be emitted at node boundaries, where the immediate being xor'd is the node index. * dfg/DFGCommon.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-30 Geoffrey Garen <ggaren@apple.com> Removed ArgList iterators. Reviewed by Gavin Barraclough. Another step toward reversing the argument order. * interpreter/Interpreter.cpp: (JSC::Interpreter::executeCall): (JSC::Interpreter::executeConstruct): Switched from iterator to int. * runtime/ArgList.h: (JSC::ArgList::ArgList): (JSC::ArgList::isEmpty): Removed iterators. * runtime/JSArray.cpp: (JSC::JSArray::finishCreation): Switched from iterator to int. 2011-11-30 Yuqiang Xian <yuqiang.xian@intel.com> 32 bit DFG should handle logicalNot slow case instead of simply bailing out https://bugs.webkit.org/show_bug.cgi?id=73515 Reviewed by Filip Pizlo. This improves Kraken performance by 14%, mainly due to ~3X improvement on imaging-desaturate. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compileLogicalNot): 2011-11-30 Max Vujovic <mvujovic@adobe.com> Some date values not handled consistently with IE/Firefox https://bugs.webkit.org/show_bug.cgi?id=14176 Reviewed by Gavin Barraclough. Changed time zone offset parsing behavior to match IE/Firefox/Opera's in implementation dependent cases like "GMT-4". * wtf/DateMath.cpp: (WTF::parseDateFromNullTerminatedCharacters): 2011-11-30 Mark Hahnenberg <mhahnenberg@apple.com> toStringCallback and valueOfCallback do not check the entire prototype chain for convertToType callback https://bugs.webkit.org/show_bug.cgi?id=73368 Reviewed by Darin Adler. We need to search the entire prototype chain for the convertToType callback, rather than just calling whatever happens to be in the first class of the chain, which potentially could be null. <rdar://problem/10493218> * API/JSCallbackFunction.cpp: (JSC::JSCallbackFunction::toStringCallback): (JSC::JSCallbackFunction::valueOfCallback): 2011-11-29 Sam Weinig <sam@webkit.org> Add adoptCF and adoptNS convenience functions to RetainPtr.h https://bugs.webkit.org/show_bug.cgi?id=73399 Reviewed by Anders Carlsson. * wtf/RetainPtr.h: (WTF::adoptCF): (WTF::adoptNS): These adoption functions match the pattern we use in other smart pointer classes. 2011-11-30 Adam Roben <aroben@apple.com> Fix RetainPtr's move assignment operators Fixes <http://webkit.org/b/73449> RetainPtr's move assignment operators don't modify the pointer being assigned to I didn't write a test for this because we don't have a way of unit testing C++11 code (see <http://webkit.org/b/73448>). Reviewed by Anders Carlsson. * wtf/RetainPtr.h: (WTF::RetainPtr::operator=): Adopt the passed-in RetainPtr's underlying pointer, not our own pointer. 2011-11-30 Csaba Osztrogonác <ossy@webkit.org> Unreviewed rolling out incorrect r101481. * assembler/MIPSAssembler.h: * assembler/MacroAssemblerMIPS.h: 2011-11-30 Simon Hausmann <simon.hausmann@nokia.com> Fix compilation with MingW. Reviewed by Csaba Osztrogonác. * wtf/ThreadingWin.cpp: (WTF::initializeCurrentThreadInternal): MingW doesn't support MSVC exception handling, so for the time being make the thread name setting unimplemented for MingW. 2011-11-30 Simon Hausmann <simon.hausmann@nokia.com> Unreviewed propective build fix for Qt/Windows part 2 after r101477. * wtf/ThreadSpecific.h: Fix the OS(WINDOWS) defines for the friend declaration for ThreadSpecific<T>::Data 2011-11-30 Simon Hausmann <simon.hausmann@nokia.com> Unreviewed propective build fix for Qt/Windows after r101477. * wtf/ThreadSpecific.h: Use OS(WINDOWS) for declaring "destructor", as it's only referenced from within another OS(WINDOWS) section. 2011-11-30 Csaba Osztrogonác <ossy@webkit.org> Unreviewed speculative buildfix after r101457. * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::readCallTarget): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::readCallTarget): 2011-11-30 Andrew Wason <rectalogic@rectalogic.com> Replace Qt QThread threading back-end with pthread/Win32 threading back-ends https://bugs.webkit.org/show_bug.cgi?id=72155 Reviewed by Simon Hausmann. Use ThreadingPthreads and ThreadingWin instead of ThreadingQt. * heap/MachineStackMarker.cpp: * wtf/MainThread.cpp: (WTF::initializeMainThread): * wtf/Platform.h: * wtf/ThreadSpecific.h: Drop QThreadStorage related code. (WTF::::destroy): * wtf/ThreadingPrimitives.h: * wtf/qt/MainThreadQt.cpp: Drop Qt specific isMainThread(). (WTF::initializeMainThreadPlatform): Initialize MainThreadInvoker on main thread to avoid infecting secondary thread with QAdoptedThread. (WTF::scheduleDispatchFunctionsOnMainThread): * wtf/qt/ThreadingQt.cpp: Removed. * wtf/wtf.pro: 2011-11-30 Csaba Osztrogonác <ossy@webkit.org> MacroAssemblerARM does not implement readCallTarget https://bugs.webkit.org/show_bug.cgi?id=73413 Based on Filip Pizlo's patch. Buildfix. Rubber-stamped by Gabor Loki. * assembler/ARMAssembler.h: (JSC::ARMAssembler::readCallTarget): * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::readCallTarget): 2011-11-29 Filip Pizlo <fpizlo@apple.com> Resetting a put_by_id inline cache should preserve the "isDirect" bit https://bugs.webkit.org/show_bug.cgi?id=73375 Reviewed by Gavin Barraclough. For the replace case, we can find out if it was direct by looking at the slow call. For the transition case, we explicitly remember if it was direct. * bytecode/CodeBlock.cpp: (JSC::printStructureStubInfo): * bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::deref): (JSC::StructureStubInfo::visitWeakReferences): * bytecode/StructureStubInfo.h: (JSC::isPutByIdAccess): (JSC::StructureStubInfo::initPutByIdTransition): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGRepatch.cpp: (JSC::DFG::tryCachePutByID): * jit/JIT.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::resetPatchPutById): (JSC::JIT::isDirectPutById): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::resetPatchPutById): * jit/JITStubs.cpp: (JSC::JITThunks::tryCachePutByID): 2011-11-29 Sam Weinig <sam@webkit.org> Remove RetainPtr::releaseRef https://bugs.webkit.org/show_bug.cgi?id=73396 Reviewed by Dan Bernstein. * wtf/RetainPtr.h: Be gone releaseRef! Long live leakRef! 2011-11-29 Sam Weinig <sam@webkit.org> Add move semantics to RetainPtr https://bugs.webkit.org/show_bug.cgi?id=73393 Reviewed by Anders Carlsson. * wtf/RetainPtr.h: (WTF::RetainPtr::RetainPtr): Add a move constructor and move enabled assignment operators to RetainPtr if the compiler being used supports rvalue references. If the compiler does not support it, we fallback to the copy semantics we have always had. 2011-11-29 Yuqiang Xian <yuqiang.xian@intel.com> DFG local CSE may cause incorrect reference counting for a node https://bugs.webkit.org/show_bug.cgi?id=73390 Reviewed by Filip Pizlo. When performing a node substitution, the ref count of the replaced child will be increased, no matter whether the user node is skipped in code generation or not. This will cause the reference count of the replaced child never get the chance to become zero and so the registers occupied by it cannot be reused simply without spilling, if it's used by a "skipped" node. This is a 1% gain on V8 benchmark, tested on IA32 Linux. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::performSubstitution): (JSC::DFG::Propagator::performNodeCSE): 2011-11-29 David Levin <levin@chromium.org> Add a way to revert a variable to its previous value after leaving a scope. https://bugs.webkit.org/show_bug.cgi?id=73371 Reviewed by Adam Barth. In case anyone from Chromium sees this, it is nearly identical to AutoReset but if the same name were used, it causes unnecessary ambiguity. * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/TemporarilyChange.h: Added. (WTF::TemporarilyChange::TemporarilyChange): (WTF::TemporarilyChange::~TemporarilyChange): 2011-11-29 Sam Weinig <sam@webkit.org> Add COMPILER_SUPPORTS macro to allow for compiler feature testing https://bugs.webkit.org/show_bug.cgi?id=73386 Reviewed by Anders Carlsson. * wtf/Compiler.h: Add COMPILER_SUPPORTS and #defines for C++11 variadic templates and rvalue references for Clang. 2011-11-29 Oliver Hunt <oliver@apple.com> Allow WebCore to describe typed arrays to JSC https://bugs.webkit.org/show_bug.cgi?id=73355 Reviewed by Gavin Barraclough. Allow globaldata to track the structure of typed arrays. * runtime/JSGlobalData.h: (JSC::TypedArrayDescriptor::TypedArrayDescriptor): 2011-11-28 Filip Pizlo <fpizlo@apple.com> DFG debugCall() mechanism only works on X86 and X86-64 https://bugs.webkit.org/show_bug.cgi?id=73282 Reviewed by Oliver Hunt. * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::debugCall): 2011-11-28 Filip Pizlo <fpizlo@apple.com> DFG non-X86 ArithDiv does speculation failure after mutating state, without a value recovery https://bugs.webkit.org/show_bug.cgi?id=73286 Reviewed by Gavin Barraclough. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-28 Filip Pizlo <fpizlo@apple.com> Unreviewed build fixes for ARM. * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::readCallTarget): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::setupArgumentsWithExecState): 2011-11-20 Roland Steiner <rolandsteiner@chromium.org> <style scoped>: add ENABLE(STYLE_SCOPED) flag to WebKit https://bugs.webkit.org/show_bug.cgi?id=72848 Add ENABLE_STYLE_SCOPED flag. Reviewed by Dimitri Glazkov. * Configurations/FeatureDefines.xcconfig: 2011-11-28 Jon Lee <jonlee@apple.com> Create skeleton framework for notifications support in WK2 https://bugs.webkit.org/show_bug.cgi?id=73253 <rdar://problem/10356943> * Configurations/FeatureDefines.xcconfig: Split out ENABLE_NOTIFICATIONS based on platform. 2011-11-28 Oliver Hunt <oliver@apple.com> Fix windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-11-28 Oliver Hunt <oliver@apple.com> Fix gyp build * JavaScriptCore.gypi: 2011-11-28 Filip Pizlo <fpizlo@apple.com> GetById should not always speculate cell https://bugs.webkit.org/show_bug.cgi?id=73181 Reviewed by Gavin Barraclough. GetById will now speculate cell if the predictions of the base are cell. Otherwise it will do like the old JIT (and like the old non-speculative DFG JIT): if not cell, go straight to slow-path but otherwise don't OSR out. This is a 1% speed-up on SunSpider. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::setupArgumentsWithExecState): (JSC::DFG::SpeculativeJIT::callOperation): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::cachedGetById): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-28 Oliver Hunt <oliver@apple.com> Move typed array implementations into wtf https://bugs.webkit.org/show_bug.cgi?id=73248 Reviewed by Sam Weinig. Move typed array implementation files from WebCore to wtf. Inline the .cpp files for each of the array views to cut down on unnecessary exports and function call overhead for trivial operations. Added files to all the project files. * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/ArrayBuffer.cpp: Renamed from Source/WebCore/html/canvas/Float32Array.cpp. (WTF::ArrayBuffer::transfer): (WTF::ArrayBuffer::addView): (WTF::ArrayBuffer::removeView): * wtf/ArrayBuffer.h: Renamed from Source/WebCore/html/canvas/ArrayBuffer.cpp. (WTF::ArrayBufferContents::ArrayBufferContents): (WTF::ArrayBufferContents::data): (WTF::ArrayBufferContents::sizeInBytes): (WTF::ArrayBufferContents::transfer): (WTF::ArrayBuffer::~ArrayBuffer): (WTF::ArrayBuffer::clampValue): (WTF::ArrayBuffer::create): (WTF::ArrayBuffer::ArrayBuffer): (WTF::ArrayBuffer::data): (WTF::ArrayBuffer::byteLength): (WTF::ArrayBuffer::slice): (WTF::ArrayBuffer::sliceImpl): (WTF::ArrayBuffer::clampIndex): (WTF::ArrayBufferContents::tryAllocate): (WTF::ArrayBufferContents::~ArrayBufferContents): * wtf/ArrayBufferView.cpp: Copied from Source/WebCore/bindings/js/JSArrayBufferCustom.cpp. (WTF::ArrayBufferView::ArrayBufferView): (WTF::ArrayBufferView::~ArrayBufferView): (WTF::ArrayBufferView::neuter): * wtf/ArrayBufferView.h: Renamed from Source/WebCore/html/canvas/ArrayBufferView.h. (WTF::ArrayBufferView::isByteArray): (WTF::ArrayBufferView::isUnsignedByteArray): (WTF::ArrayBufferView::isShortArray): (WTF::ArrayBufferView::isUnsignedShortArray): (WTF::ArrayBufferView::isIntArray): (WTF::ArrayBufferView::isUnsignedIntArray): (WTF::ArrayBufferView::isFloatArray): (WTF::ArrayBufferView::isDoubleArray): (WTF::ArrayBufferView::isDataView): (WTF::ArrayBufferView::buffer): (WTF::ArrayBufferView::baseAddress): (WTF::ArrayBufferView::byteOffset): (WTF::ArrayBufferView::verifySubRange): (WTF::ArrayBufferView::clampOffsetAndNumElements): (WTF::ArrayBufferView::setImpl): (WTF::ArrayBufferView::setRangeImpl): (WTF::ArrayBufferView::zeroRangeImpl): (WTF::ArrayBufferView::calculateOffsetAndLength): * wtf/CMakeLists.txt: * wtf/Float32Array.h: Renamed from Source/WebCore/html/canvas/Float32Array.h. (WTF::Float32Array::set): (WTF::Float32Array::item): (WTF::Float32Array::isFloatArray): (WTF::Float32Array::create): (WTF::Float32Array::Float32Array): (WTF::Float32Array::subarray): * wtf/Float64Array.h: Renamed from Source/WebCore/html/canvas/Float64Array.h. (WTF::Float64Array::set): (WTF::Float64Array::item): (WTF::Float64Array::isDoubleArray): (WTF::Float64Array::create): (WTF::Float64Array::Float64Array): (WTF::Float64Array::subarray): * wtf/Int16Array.h: Renamed from Source/WebCore/html/canvas/Int16Array.cpp. (WTF::Int16Array::set): (WTF::Int16Array::isShortArray): (WTF::Int16Array::create): (WTF::Int16Array::Int16Array): (WTF::Int16Array::subarray): * wtf/Int32Array.h: Renamed from Source/WebCore/html/canvas/Int32Array.cpp. (WTF::Int32Array::set): (WTF::Int32Array::isIntArray): (WTF::Int32Array::create): (WTF::Int32Array::Int32Array): (WTF::Int32Array::subarray): * wtf/Int8Array.h: Renamed from Source/WebCore/html/canvas/Int8Array.cpp. (WTF::Int8Array::set): (WTF::Int8Array::isByteArray): (WTF::Int8Array::create): (WTF::Int8Array::Int8Array): (WTF::Int8Array::subarray): * wtf/IntegralTypedArrayBase.h: Renamed from Source/WebCore/html/canvas/IntegralTypedArrayBase.h. (WTF::IntegralTypedArrayBase::set): (WTF::IntegralTypedArrayBase::item): (WTF::IntegralTypedArrayBase::IntegralTypedArrayBase): * wtf/TypedArrayBase.h: Renamed from Source/WebCore/html/canvas/TypedArrayBase.h. (WTF::TypedArrayBase::data): (WTF::TypedArrayBase::set): (WTF::TypedArrayBase::setRange): (WTF::TypedArrayBase::zeroRange): (WTF::TypedArrayBase::length): (WTF::TypedArrayBase::byteLength): (WTF::TypedArrayBase::TypedArrayBase): (WTF::TypedArrayBase::create): (WTF::TypedArrayBase::subarrayImpl): * wtf/Uint16Array.h: Renamed from Source/WebCore/html/canvas/Uint16Array.cpp. (WTF::Uint16Array::set): (WTF::Uint16Array::isUnsignedShortArray): (WTF::Uint16Array::create): (WTF::Uint16Array::Uint16Array): (WTF::Uint16Array::subarray): * wtf/Uint32Array.h: Renamed from Source/WebCore/html/canvas/Uint32Array.cpp. (WTF::Uint32Array::set): (WTF::Uint32Array::isUnsignedIntArray): (WTF::Uint32Array::create): (WTF::Uint32Array::Uint32Array): (WTF::Uint32Array::subarray): * wtf/Uint8Array.h: Renamed from Source/WebCore/html/canvas/Uint8Array.h. (WTF::Uint8Array::set): (WTF::Uint8Array::isUnsignedByteArray): (WTF::Uint8Array::create): (WTF::Uint8Array::Uint8Array): (WTF::Uint8Array::subarray): * wtf/wtf.pro: 2011-11-27 Filip Pizlo <fpizlo@apple.com> Don't try to optimize huge code blocks https://bugs.webkit.org/show_bug.cgi?id=73187 Reviewed by Oliver Hunt. This unifies the heuristics used for deciding if a code block is too big to optimize, and sets this heuristic to 1000, which is intuitively better than numeric_limits<unsigned>::max(). It also results in what looks like a speed-up on both SunSpider and V8 (in Tools/Scripts/bencher). * dfg/DFGCapabilities.h: (JSC::DFG::mightCompileEval): (JSC::DFG::mightCompileProgram): (JSC::DFG::mightCompileFunctionForCall): (JSC::DFG::mightCompileFunctionForConstruct): * runtime/Heuristics.cpp: (JSC::Heuristics::initializeHeuristics): * runtime/Heuristics.h: 2011-11-28 Filip Pizlo <fpizlo@apple.com> Either remove the GetMethod node from the DFG backend, or find a use for it https://bugs.webkit.org/show_bug.cgi?id=73178 Reviewed by Gavin Barraclough. More testing seemed to imply that the GetMethod code was indeed not profitable in any major test. So, it's probably best to just remove it. * bytecode/CodeBlock.cpp: (JSC::MethodCallLinkInfo::reset): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: * dfg/DFGNode.h: (JSC::DFG::Node::hasIdentifier): (JSC::DFG::Node::hasHeapPrediction): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGRepatch.cpp: * dfg/DFGRepatch.h: * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::cachedGetById): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::cachedGetById): (JSC::DFG::SpeculativeJIT::compile): 2011-11-28 Michael Saboff <msaboff@apple.com> Change set 101187 from bug 73154 removed already lower case optimization https://bugs.webkit.org/show_bug.cgi?id=73174 Added back the "string is already lower case" optimization. Reviewed by Geoffrey Garen. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncToLowerCase): 2011-11-28 Simon Hausmann <simon.hausmann@nokia.com> Unreviewed prospective build fix. Touch the file to trigger correct rebuild on the Qt mips/sh4/sl bot. * wtf/unicode/qt4/UnicodeQt4.h: 2011-11-28 Tor Arne Vestbø <tor.arne.vestbo@nokia.com> [Qt] Remove cruft from project file Reviewed by Simon Hausmann. * Target.pri: 2011-11-28 Simon Hausmann <simon.hausmann@nokia.com> [Qt] WTF should be built as separate static library https://bugs.webkit.org/show_bug.cgi?id=73201 Reviewed by Tor Arne Vestbø. * Target.pri: Don't claim to build WTF, as that would cause the debug-with-shlibs build to not link in wtf. * jsc.pro: Require wtf. * wtf/wtf.pri: Removed. * wtf/wtf.pro: Added. Pro file to build wtf statically. 2011-11-28 Martin Robinson <mrobinson@igalia.com> [GTK] JavaScriptCore generated sources should build in the DerivedSources directory https://bugs.webkit.org/show_bug.cgi?id=73197 Reviewed by Philippe Normand. Build all JavaScriptCore generated sources in DerivedSources. * GNUmakefile.am: Update generation rules. * GNUmakefile.list.am: Update source lists. 2011-11-27 Filip Pizlo <fpizlo@apple.com> DFG should not emit GetMethod node https://bugs.webkit.org/show_bug.cgi?id=73175 Reviewed by Gavin Barraclough. Replaces all instances of the GetMethod node with GetById. This appears to be a slight win on V8. This patch leaves GetMethod support in the code-base, making this decision easy to reverse, for now. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): 2011-11-26 Hajime Morrita <morrita@chromium.org> Needs WTF_INLINE and JS_INLINE https://bugs.webkit.org/show_bug.cgi?id=72853 Reviewed by Kevin Ollivier. Added WTF_HIDDEN, WTF_INLINE and JS_INLINE which indirect __attribute__((visibility("hidden")) * config.h: * wtf/ExportMacros.h: 2011-11-25 Michael Saboff <msaboff@apple.com> String.prototype.toLower should be optimized for 8 bit strings https://bugs.webkit.org/show_bug.cgi?id=73154 Changed stringProtoFuncToLowerCase to use StringImpl::lower() which has been optimized for 8 bit strings. This is worth ~7% to sunspider string.tagcloud. Reviewed by Filip Pizlo. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncToLowerCase): 2011-11-25 Michael Saboff <msaboff@apple.com> Array.toString always uses StringImpl::characters() https://bugs.webkit.org/show_bug.cgi?id=72969 If all component strings are 8 bit, create an 8 bit result string for toString(). This appears to be performance neutral to sunspider and v8. Reviewed by Filip Pizlo. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncToString): 2011-11-24 Michael Saboff <msaboff@apple.com> UString methods are not character size aware https://bugs.webkit.org/show_bug.cgi?id=72975 Changed the UString number constructors to build 8 bit strings. Modified the other methods to check string bitness and process with 8 bits wherre appropriate. * runtime/UString.cpp: (JSC::UString::number): (JSC::operator==): (JSC::operator<): (JSC::UString::ascii): 2011-11-24 Michael Saboff <msaboff@apple.com> JavaScript string to number conversion functions use characters() https://bugs.webkit.org/show_bug.cgi?id=72974 Change the various JS to number routines to process strings using characters8() or characters16() as appropriate. Implemented using static template methods. Reviewed by Filip Pizlo. * runtime/JSGlobalObjectFunctions.cpp: (JSC::isInfinity): (JSC::jsHexIntegerLiteral): (JSC::jsStrDecimalLiteral): (JSC::toDouble): (JSC::jsToNumber): 2011-11-24 Michael Saboff <msaboff@apple.com> Empty JSStrings are created as 16 bit https://bugs.webkit.org/show_bug.cgi?id=72968 Clear m_is8Bit flag for empty strings. Reviewed by Filip Pizlo. * runtime/JSString.h: (JSC::RopeBuilder::finishCreation): 2011-11-24 Michael Saboff <msaboff@apple.com> Tune JSStringBuilder for 8 bit Strings https://bugs.webkit.org/show_bug.cgi?id=72683 Changed JSStringBuilder to use 8 bit buffers until 16 bit data is added. When 16 bit data is to be added, the 8 bit buffer is converted to 16 bit and building continues with a 16 bit buffer. Reviewed by Filip Pizlo. * runtime/JSStringBuilder.h: (JSC::JSStringBuilder::JSStringBuilder): (JSC::JSStringBuilder::append): (JSC::JSStringBuilder::upConvert): (JSC::JSStringBuilder::build): * runtime/UString.h: (JSC::UString::adopt): * wtf/text/StringImpl.h: (WTF::StringImpl::adopt): 2011-11-24 Zeno Albisser <zeno@webkit.org> [Qt]WK2][Mac] Use Mac port's IPC implementation instead of Unix sockets https://bugs.webkit.org/show_bug.cgi?id=72495 Update defines to not use Unix Domain Sockets for platform Qt on Mac. This enables Qt to reuse existing code for mach ports and Grand Central Dispatch based IPC. Reviewed by Simon Hausmann. * wtf/Platform.h: 2011-11-24 Simon Hausmann <simon.hausmann@nokia.com> [Qt] REGRESSION(r101131): WTF::scheduleDispatchFunctionsOnMainThread() doesn't work reliably Reviewed by Andreas Kling. We must make sure that the MainThreadInvoker object lives in the gui thread. There are a few ways of doing that and this fix seems like the least intrusive one by simply pushing the invoker to the gui thread if it's not there already. * wtf/qt/MainThreadQt.cpp: (WTF::scheduleDispatchFunctionsOnMainThread): 2011-11-24 Patrick Gansterer <paroga@webkit.org> [Qt] Use QEvent for dispatchFunctionsFromMainThread() https://bugs.webkit.org/show_bug.cgi?id=72704 Reviewed by Simon Hausmann. Replace QMetaObject::invokeMethod with QCoreApplication::postEvent. This is the same as what invokeMethod does internally, but reduces the dependency on some internal QThread stuff. * wtf/qt/MainThreadQt.cpp: (WTF::MainThreadInvoker::MainThreadInvoker): (WTF::MainThreadInvoker::event): (WTF::scheduleDispatchFunctionsOnMainThread): 2011-11-23 George Staikos <staikos@webkit.org> Remove BlackBerry OS support from RandomNumberSeed, making QNX=UNIX. https://bugs.webkit.org/show_bug.cgi?id=73028 Reviewed by Daniel Bates. * wtf/RandomNumberSeed.h: (WTF::initializeRandomNumberGenerator): 2011-11-23 Nikolas Zimmermann <nzimmermann@rim.com> Add flags/precision arguments to String::number(double) to allow fine-grained control over the result string https://bugs.webkit.org/show_bug.cgi?id=72793 Reviewed by Zoltan Herczeg. This new code will be used in follow-up patches to replace the String::format("%.2f") usage in platform/text/TextStream.cpp, and String::format("%.6lg") usage in svg/SVGPathStringBuilder.cpp. The String::number(double) currently calls String::format("%.6lg") in trunk. In order to replace this by a variant that properly rounds to six significant figures, JSC code could be refactored. JSCs Number.toPrecision/toFixed uses wtf/dtoa/double-conversion which provides all features we need, except truncating trailing zeros, needed to mimic the "g" format, which is either f or e but with trailing zeros removed, producing shorter results. Changed the default signature to: "static String number(double, unsigned = ShouldRoundSignificantFigures | ShouldTruncateTrailingZeros, unsigned precision = 6);". In WebCore we can now replace String::format() calls like this: String::format("%.2f", f) -> String::number(f, ShouldRoundDecimalPlaces, 2) String::format("%.6lg", f) -> String::number(f) The default parameters for precison & flags exactly match the format of the string produced now, except that the result is rounded according to the rounding mode / formatting mode and precision. This paves the way towards reliable results in the d="" attribute dumps of SVG paths across platforms. The dtoa rounding code enforces a unique zero, resolving all 0.0 vs. -0.0 issues currently seen on Windows, and some Gtk/Qt bots. This patch needs a rebaseline of svg/dom/length-list-parser.html as we don't perfecly mimic the String::format() "lg" mode result for exponentials, we used to return eg. "e-7" and now return "e-07" - the trailing zero truncation hasn't been implemented for exponentials, as this really affects only this test and thus wasn't worth the trouble - in contrary the trailing zero truncation is needed for thousands of other results in "f" notation, and thus needed to match the DRT results. Here's a performance comparision using a JSC release build and some arbitary numbers: Converting 123.456 using old approach took 95.527100ms. avg 0.000955ms/call. Converting 123.456 using new approach took 28.126953ms. avg 0.000281ms/call. Converting 123 using old approach took 85.411133ms. avg 0.000854ms/call. Converting 123 using new approach took 24.190186ms. avg 0.000242ms/call. Converting 0.1 using old approach took 92.622803ms. avg 0.000926ms/call. Converting 0.1 using new approach took 23.317871ms. avg 0.000233ms/call. Converting 1/i using old approach took 106.893066ms. avg 0.001069ms/call. Converting 1/i using new approach took 27.164062ms. avg 0.000272ms/call. For all numbers I've tested in RoundingSignificantFigures mode and 6 digit precision the speedup was at least 250%. * JavaScriptCore.exp: Change String::number(double) signature. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Ditto. * runtime/NumberPrototype.cpp: (JSC::numberProtoFuncToFixed): Refactor this into numberToFixedPrecisionString(), move to wtf/dtoa.cpp. (JSC::numberProtoFuncToPrecision): Ditto, refactor this into numberToFixedWidthString. * wtf/dtoa.cpp: Moved fixedWidth/Precision helpers into dtoa, extend numberToFixedPrecisionString(). Add a mode which allows to truncate trailing zeros/decimal point. to make it possible to use them to generate strings that match the output from String::format("%6.lg"), while using our dtoas rounding facilities. * wtf/dtoa.h: * wtf/dtoa/utils.h: Expose new helper method, which allows us to truncate the result, before generating the output const char*. (WTF::double_conversion::StringBuilder::SetPosition): * wtf/text/WTFString.cpp: (WTF::String::number): Remove String::format("%6.lg") usage! Switch to rounding to six significant figures, while matching the output of String::format. * wtf/text/WTFString.h: 2011-11-23 Hajime Morrita <morrita@chromium.org> WTF::String has extra WTF_EXPORT_PRIVATE https://bugs.webkit.org/show_bug.cgi?id=72858 Reviewed by Kevin Ollivier. * wtf/text/WTFString.h: (WTF::String::String): 2011-11-23 Raphael Kubo da Costa <kubo@profusion.mobi> [CMake] Move the top-level logic to the top-level directory. https://bugs.webkit.org/show_bug.cgi?id=72685 Reviewed by Brent Fulgham. * CMakeLists.txt: Point to the right Source/ directory. * wtf/CMakeLists.txt: Ditto. 2011-11-22 Yuqiang Xian <yuqiang.xian@intel.com> Strength reduction for Mul and Mod operations for known constants in DFG https://bugs.webkit.org/show_bug.cgi?id=72878 Reviewed by Filip Pizlo. Also the code should be commonly shared by both 32_64 and 64. * dfg/DFGNode.h: (JSC::DFG::nodeMayOverflow): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::fmodAsDFGOperation): (JSC::DFG::SpeculativeJIT::compileInstanceOf): (JSC::DFG::isPowerOfTwo): (JSC::DFG::logTwo): (JSC::DFG::SpeculativeJIT::compileSoftModulo): (JSC::DFG::SpeculativeJIT::compileArithMul): (JSC::DFG::SpeculativeJIT::compileArithMod): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-22 Daniel Bates <dbates@rim.com> Add WTF infrastructure for the BlackBerry port https://bugs.webkit.org/show_bug.cgi?id=72970 Reviewed by Antonio Gomes. * wtf/Assertions.cpp: Added BlackBerry-specific logging directive. * wtf/MathExtras.h: (abs): Added; stdlib doesn't contain abs() on QNX. * wtf/Platform.h: Define WTF_PLATFORM_BLACKBERRY and enable some platform features. * wtf/RandomNumberSeed.h: (WTF::initializeRandomNumberGenerator): For the BlackBerry port, we initialize the bad pseudo random number generator using time(3) before initializing the Mersenne Twister random number generator. * wtf/ThreadingPthreads.cpp: (WTF::createThreadInternal): Added. * wtf/blackberry: Added. * wtf/blackberry/MainThreadBlackBerry.cpp: Added. (WTF::initializeMainThreadPlatform): (WTF::scheduleDispatchFunctionsOnMainThread): * wtf/text/WTFString.h: Added constructor and conversion operator for BlackBerry WebString string object. 2011-11-22 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r100988. http://trac.webkit.org/changeset/100988 https://bugs.webkit.org/show_bug.cgi?id=72941 "Broke pixel tests on Chromium-Linux" (Requested by kbalazs on #webkit). * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/ParallelJobs.h: * wtf/ParallelJobsGeneric.cpp: * wtf/ParallelJobsGeneric.h: (WTF::ParallelEnvironment::ParallelEnvironment): (WTF::ParallelEnvironment::execute): * wtf/Platform.h: 2011-11-21 Balazs Kelemen <kbalazs@webkit.org> Enable ParallelJobs by default https://bugs.webkit.org/show_bug.cgi?id=70032 Reviewed by Zoltan Herczeg. According to measurements on Mac and Linux it is a considerable speedup for SVG on multicore. Remove the ENABLE(PARALLEL_JOBS) guard. Fix build on Windows and Chromium. * JavaScriptCore.gypi: Add the files to the build. It was missing for the gyp build system. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Export symbols. * wtf/ParallelJobs.h: * wtf/ParallelJobsGeneric.cpp: (WTF::ParallelEnvironment::ParallelEnvironment): (WTF::ParallelEnvironment::execute): Deinline these to avoid exporting a lot of symbols. These are non-trivial and called only once on a given object so it doesn't seems to be worthwile to inline them. Additionally fix a signed-unsigned comparison in the constructor. * wtf/ParallelJobsGeneric.h: * wtf/Platform.h: 2011-11-21 Filip Pizlo <fpizlo@apple.com> DFG should have richer debug output for CFA and phi processing https://bugs.webkit.org/show_bug.cgi?id=72922 Reviewed by Gavin Barraclough. In the default verbose mode, we now print information about variable state at the bottom of basic blocks in addition to the top, and we also print local variable linking. In the verbose propagation mode, the state of phi processing is dumped more richly and CFA merging (the most subtle part of CFA) is traced as well. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::endBasicBlock): (JSC::DFG::AbstractState::mergeStateAtTail): * dfg/DFGAbstractValue.h: (JSC::DFG::StructureAbstractValue::dump): (JSC::DFG::AbstractValue::dump): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::processPhiStack): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGCommon.h: (JSC::DFG::NodeIndexTraits::dump): * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGNode.h: (JSC::DFG::Node::dumpChildren): * dfg/DFGOSRExitCompiler.cpp: * dfg/DFGOperands.h: (JSC::DFG::OperandValueTraits::dump): (JSC::DFG::dumpOperands): 2011-11-21 Filip Pizlo <fpizlo@apple.com> Showing the data overlay in OpenStreetMap doesn't work, zooming partially broken https://bugs.webkit.org/show_bug.cgi?id=71505 Reviewed by Gavin Barraclough. It turns out that we were corrupting phi nodes in case of overflow. The bug is really obvious, but producing a test case that causes the badness is hard. Even when the phi nodes do get corrupt, there's more that has to happen before it causes incorrect execution - and I wasn't able to reproduce in any kind of sensible reduced case. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::processPhiStack): 2011-11-21 Simon Hausmann <simon.hausmann@nokia.com> [Qt] Speed up debug builds. https://bugs.webkit.org/show_bug.cgi?id=72882 Reviewed by Tor Arne Vestbø. * Target.pri: Make BUILDING_JavaScriptCore available earlier, so it can be used by the build system. 2011-11-21 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r100913. http://trac.webkit.org/changeset/100913 https://bugs.webkit.org/show_bug.cgi?id=72885 "Break Windows build" (Requested by kbalazs on #webkit). * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/ParallelJobs.h: * wtf/ParallelJobsGeneric.cpp: * wtf/ParallelJobsGeneric.h: (WTF::ParallelEnvironment::ParallelEnvironment): (WTF::ParallelEnvironment::execute): * wtf/Platform.h: 2011-11-21 Balazs Kelemen <kbalazs@webkit.org> Enable ParallelJobs by default https://bugs.webkit.org/show_bug.cgi?id=70032 Reviewed by Zoltan Herczeg. According to measurements on Mac and Linux it is a considerable speedup for SVG on multicore. Remove the ENABLE(PARALLEL_JOBS) guard. Fix build on Windows and Chromium. * JavaScriptCore.gypi: Add the files to the build. It was missing for the gyp build system. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Export symbols. * wtf/ParallelJobs.h: * wtf/ParallelJobsGeneric.cpp: (WTF::ParallelEnvironment::ParallelEnvironment): (WTF::ParallelEnvironment::execute): Deinline these to avoid exporting a lot of symbols. These are non-trivial and called only once on a given object so it doesn't seems to be worthwile to inline them. Additionally fix a signed-unsigned comparison in the constructor. * wtf/ParallelJobsGeneric.h: * wtf/Platform.h: 2011-11-21 Andy Wingo <wingo@igalia.com> Add .dir-locals.el file for better Emacs defaults https://bugs.webkit.org/show_bug.cgi?id=72483 Reviewed by Xan Lopez. * .dir-locals.el: Set appropriate directory-local variables for Emacs. 2011-11-21 Filip Pizlo <fpizlo@apple.com> Another attempt at a build fix. * dfg/DFGRepatch.h: (JSC::DFG::dfgResetGetByID): (JSC::DFG::dfgResetPutByID): 2011-11-20 Filip Pizlo <fpizlo@apple.com> Unreviewed interpreter build fix. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::finalizeUnconditionally): * dfg/DFGRepatch.h: 2011-11-20 Yuqiang Xian <yuqiang.xian@intel.com> Improve modulo operation on 32bit platforms https://bugs.webkit.org/show_bug.cgi?id=72501 Reviewed by Filip Pizlo. Extend softModulo to support X86 and MIPS in baseline JIT. Apply the same optimization to 32bit DFG JIT. 1% gain on Kraken, tested on Linux Core i7 Nehalem 32bit. * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compileSoftModulo): (JSC::DFG::SpeculativeJIT::compile): * jit/JITArithmetic32_64.cpp: (JSC::JIT::emit_op_mod): (JSC::JIT::emitSlow_op_mod): * jit/JITOpcodes32_64.cpp: (JSC::JIT::softModulo): * wtf/Platform.h: 2011-11-18 Filip Pizlo <fpizlo@apple.com> Inline caches that refer to otherwise dead objects should be cleared https://bugs.webkit.org/show_bug.cgi?id=72311 Reviewed by Geoff Garen. DFG code blocks now participate in the weak reference harvester fixpoint so that they only consider themselves to be live if either they are currently executing, or their owner is live and all of their weak references are live. If not, the relevant code blocks are jettisoned. Inline caches in both the old JIT and the DFG are now cleared if any of their references are not marked at the end of a GC. This is performance-neutral on SunSpider, V8, and Kraken. With the clear- all-code-on-GC policy that we currently have, it shows a slight reduction in memory usage. If we turn that policy off, it's pretty easy to come up with an example program that will cause ToT to experience linear heap growth, while with this patch, the heap stays small and remains at a constant size. * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::readCallTarget): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::readCallTarget): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::readCallTarget): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::readCallTarget): * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): (JSC::CodeBlock::performTracingFixpointIteration): (JSC::CodeBlock::visitWeakReferences): (JSC::CodeBlock::finalizeUnconditionally): (JSC::CodeBlock::stronglyVisitStrongReferences): (JSC::MethodCallLinkInfo::reset): (JSC::ProgramCodeBlock::jettison): (JSC::EvalCodeBlock::jettison): (JSC::FunctionCodeBlock::jettison): * bytecode/CodeBlock.h: (JSC::CodeBlock::reoptimize): (JSC::CodeBlock::shouldImmediatelyAssumeLivenessDuringScan): * bytecode/Instruction.h: (JSC::PolymorphicAccessStructureList::visitWeak): * bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::visitWeakReferences): * bytecode/StructureStubInfo.h: (JSC::isGetByIdAccess): (JSC::isPutByIdAccess): (JSC::StructureStubInfo::reset): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGOperations.cpp: * dfg/DFGRepatch.cpp: (JSC::DFG::dfgRepatchByIdSelfAccess): (JSC::DFG::dfgResetGetByID): (JSC::DFG::dfgResetPutByID): * dfg/DFGRepatch.h: (JSC::DFG::dfgResetGetByID): (JSC::DFG::dfgResetPutByID): * jit/JIT.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::resetPatchGetById): (JSC::JIT::resetPatchPutById): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::resetPatchGetById): (JSC::JIT::resetPatchPutById): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jit/JITWriteBarrier.h: (JSC::JITWriteBarrierBase::clearToMaxUnsigned): 2011-11-20 Filip Pizlo <fpizlo@apple.com> Showing the data overlay in OpenStreetMap doesn't work, zooming partially broken https://bugs.webkit.org/show_bug.cgi?id=71505 Reviewed by Oliver Hunt. The bytecode generator was assuming that call_varargs never reuses the base register (i.e. the function being called) for the result. This is no longer true. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitCallVarargs): * bytecompiler/BytecodeGenerator.h: * bytecompiler/NodesCodegen.cpp: (JSC::ApplyFunctionCallDotNode::emitBytecode): 2011-11-20 Filip Pizlo <fpizlo@apple.com> DFG 32_64 should directly store double virtual registers on SetLocal https://bugs.webkit.org/show_bug.cgi?id=72845 Reviewed by Oliver Hunt. 2% win on Kraken. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-20 Noel Gordon <noel.gordon@gmail.com> [chromium] Remove DFG::JITCodeGenerator from the gyp projects https://bugs.webkit.org/show_bug.cgi?id=72842 Reviewed by Filip Pizlo. dfg/DFGJITCodeGenerator.{h,cpp} were removed in r100244 * JavaScriptCore.gypi: remove dfg/DFGJITCodeGenerator.{h,cpp} 2011-11-18 Daniel Bates <dbates@rim.com> Add CMake build infrastructure for the BlackBerry port https://bugs.webkit.org/show_bug.cgi?id=72768 Reviewed by Antonio Gomes. * PlatformBlackBerry.cmake: Added. * shell/PlatformBlackBerry.cmake: Added. * wtf/PlatformBlackBerry.cmake: Added. 2011-11-18 Filip Pizlo <fpizlo@apple.com> DFG JIT fails speculation on InstanceOf if the base is not an object https://bugs.webkit.org/show_bug.cgi?id=72709 Reviewed by Geoff Garen. InstanceOf already leverages the fact that we only allow the default hasInstance implementation. So, if the base is predicted to possibly be not an object and the CFA has not yet proven otherwise, InstanceOf will abstain from speculating cell and instead return false if the base is not a cell. This appears to be a 1% speed-up on V8 on the V8 harness. 3-4% or so speed-up in earley-boyer. Neutral according to bencher on SunSpider, V8, and Kraken. In 32-bit, it's a 0.5% win on SunSpider and a 1.9% win on V8 even on my harness, due to a 12.5% win on earley-boyer. I also took this opportunity to make the code for InstanceOf common between the two JITs. This was partially successful, in that the "common code" has a bunch of #if's, but overall it seems like a code size reduction. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileInstanceOfForObject): (JSC::DFG::SpeculativeJIT::compileInstanceOf): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-18 Mark Hahnenberg <mhahnenberg@apple.com> Forgot to completely de-virtualize isDynamicScope https://bugs.webkit.org/show_bug.cgi?id=72763 Reviewed by Darin Adler. * runtime/JSActivation.h: Removed virtual keyword. 2011-11-18 Filip Pizlo <fpizlo@apple.com> Crash in JSC::DFG::OSRExitCompiler::compileExit(JSC::DFG::OSRExit const&, JSC::DFG::SpeculationRecovery*) https://bugs.webkit.org/show_bug.cgi?id=72292 Reviewed by Darin Adler. Fix this for 32_64. * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): 2011-11-18 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize ExecutableBase::intrinsic https://bugs.webkit.org/show_bug.cgi?id=72548 Reviewed by Oliver Hunt. * runtime/Executable.cpp: (JSC::ExecutableBase::intrinsic): Dynamic cast to NativeExecutable. If successful, call intrinsic, otherwise return default value. * runtime/Executable.h: * runtime/JSCell.h: (JSC::jsDynamicCast): Add jsDynamicCast that duplicates the functionality of dynamic_cast in C++ but uses ClassInfo rather than requiring C++ RTTI. 2011-11-18 Patrick Gansterer <paroga@webkit.org> [CMake] Remove duplicate dtoa files from CMakeLists.txt https://bugs.webkit.org/show_bug.cgi?id=72711 Reviewed by Brent Fulgham. * wtf/CMakeLists.txt: 2011-11-17 Michael Saboff <msaboff@apple.com> [Qt] REGRESSION(r100510): Enable 8 Bit Strings in JavaScriptCore https://bugs.webkit.org/show_bug.cgi?id=72602 Fixed StringImpl::foldCase by adding return in the case we need to handle folding of 8 bit strings with Latin-1 characters. Fixed case where StringImpl::replace was using a char temp instead of an LChar temp. Because of the second change, I changed other uses of char or unsigned char to LChar. Reviewed by Zoltan Herczeg. * wtf/text/StringImpl.cpp: (WTF::StringImpl::upper): (WTF::StringImpl::foldCase): (WTF::equal): (WTF::equalIgnoringCase): (WTF::StringImpl::replace): 2011-11-17 Patrick Gansterer <paroga@webkit.org> [CMake] Move FAST_MALLOC specific lines from Platform*.cmake to CMakeLists.txt https://bugs.webkit.org/show_bug.cgi?id=72644 Reviewed by Brent Fulgham. All ports need to do the same determination about fast malloc. Move the CMake code from platform specific files into the generic one, so that additional ports can reuse it. * wtf/CMakeLists.txt: * wtf/PlatformEfl.cmake: * wtf/PlatformWinCE.cmake: 2011-11-17 Mark Hahnenberg <mhahnenberg@apple.com> Add finalizer to JSActivation https://bugs.webkit.org/show_bug.cgi?id=72575 Reviewed by Geoffrey Garen. * runtime/JSActivation.cpp: (JSC::JSActivation::finishCreation): Attach finalize function to objects during creation. (JSC::JSActivation::finalize): * runtime/JSActivation.h: Replaced virtual destructor with static finalize function. 2011-11-15 Filip Pizlo <fpizlo@apple.com> Code block jettisoning should be part of the GC's transitive closure https://bugs.webkit.org/show_bug.cgi?id=72467 Reviewed by Geoff Garen. Replaced JettisonedCodeBlocks with DFGCodeBlocks. The latter knows about all DFG code blocks (i.e. those that may be jettisoned, and may have inlined weak references) and helps track what state each of those code blocks is in during GC. The state consists of two flags; mayBeExecuting, which tells if the code block is live from call frames; and isJettisoned, which tells if the code block is not owned by any executable and thus should be deleted as soon as it is not mayBeExecuting. - Not executing, Not jettisoned: The code block may or may not be reachable from any executables, but it is owned by an executable, and hence should be kept alive if its executable is live and if all of its weak references are live. Otherwise it should be deleted during the current GC cycle, and its outgoing references should not be scanned. - Not executing but jettisoned: The code block should be deleted as soon as possible and none of its outgoing references should be scanned. - Executing but not jettisoned: The code block should be kept alive during this GC cycle, and all of its outgoing references (including the weak ones) should be scanned and marked strongly. The mayBeExecuting bit will be cleared at the end of the GC cycle. - Executing and jettisoned: The code block should be kept alive during this GC cycle, and all of its outgoing references (including the weak ones) should be scanned and marked strongly. However, on the next GC cycle, it will have its mayBeExecuting bit cleared and hence it will become a candidate for immediate deletion provided it is not executing again. This is performance-neutral. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::~CodeBlock): * bytecode/CodeBlock.h: (JSC::CodeBlock::setJITCode): (JSC::CodeBlock::DFGData::DFGData): (JSC::DFGCodeBlocks::mark): * heap/ConservativeRoots.cpp: (JSC::ConservativeRoots::add): * heap/ConservativeRoots.h: * heap/DFGCodeBlocks.cpp: Added. (JSC::DFGCodeBlocks::DFGCodeBlocks): (JSC::DFGCodeBlocks::~DFGCodeBlocks): (JSC::DFGCodeBlocks::jettison): (JSC::DFGCodeBlocks::clearMarks): (JSC::DFGCodeBlocks::deleteUnmarkedJettisonedCodeBlocks): (JSC::DFGCodeBlocks::traceMarkedCodeBlocks): * heap/DFGCodeBlocks.h: Added. * heap/Heap.cpp: (JSC::Heap::jettisonDFGCodeBlock): (JSC::Heap::markRoots): (JSC::Heap::collect): * heap/Heap.h: * heap/JettisonedCodeBlocks.cpp: Removed. * heap/JettisonedCodeBlocks.h: Removed. * interpreter/RegisterFile.cpp: (JSC::RegisterFile::gatherConservativeRoots): * interpreter/RegisterFile.h: * runtime/Executable.cpp: (JSC::jettisonCodeBlock): 2011-11-16 Filip Pizlo <fpizlo@apple.com> Unreviewed, build fix for 32-bit. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-16 Geoffrey Garen <ggaren@apple.com> Some CachedCall cleanup, in preparation for reversing argument order. Reviewed by Gavin Barraclough. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::stronglyVisitWeakReferences): A build fix for the interpreter, so I can test it. * interpreter/CachedCall.h: (JSC::CachedCall::CachedCall): Renamed argCount to argumentCount because we are not that desperate for character saving. (JSC::CachedCall::setThis): (JSC::CachedCall::setArgument): Adopted new 0-based argument indexing for CallFrameClosure. * interpreter/CallFrameClosure.h: (JSC::CallFrameClosure::setThis): (JSC::CallFrameClosure::setArgument): (JSC::CallFrameClosure::resetCallFrame): Provide 0-based argument indexing, with an explicit setter for 'this', since that's how most clients think. * interpreter/Interpreter.cpp: (JSC::Interpreter::prepareForRepeatCall): * interpreter/Interpreter.h: Change argCount to argumentCountIncludingThis, for clarity. 2011-11-16 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize ScriptExecutable::unlinkCalls https://bugs.webkit.org/show_bug.cgi?id=72546 Reviewed by Geoffrey Garen. * runtime/Executable.cpp: (JSC::FunctionExecutable::~FunctionExecutable): Added an empty explicit virtual destructor to prevent a very odd compilation error due to the fact that the compiler was trying to generate the implicit inline destructor in every translation unit, some of which didn't have complete type information on the things that needed to be destructed in the implicit destructor. * runtime/Executable.h: (JSC::EvalExecutable::createStructure): Used new type value from JSType (JSC::ProgramExecutable::createStructure): Ditto (JSC::FunctionExecutable::createStructure): Ditto (JSC::ScriptExecutable::unlinkCalls): Condition upon the type value, cast and call the corresponding unlinkCalls implementation. * runtime/JSType.h: Added new values for EvalExecutable, ProgramExecutable, and FunctionExecutable. Remove explicit numbers, since that just adds noise to patches and they currently have no significance. 2011-11-16 Filip Pizlo <fpizlo@apple.com> JSC::CodeBlock should know which references generated by the DFG are weak https://bugs.webkit.org/show_bug.cgi?id=72563 Reviewed by Geoff Garen. CodeBlock::m_dfgData now tracks weak references and weak reference transitions (like ephemerons) generated by the DFG. The DFG makes sure to notify the CodeBlock of all uses of weak references and weak reference transitions. CodeBlock currently marks them strongly, since the weak marking logic is not in place, yet. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): (JSC::CodeBlock::stronglyVisitWeakReferences): * bytecode/CodeBlock.h: (JSC::CodeBlock::appendWeakReference): (JSC::CodeBlock::shrinkWeakReferencesToFit): (JSC::CodeBlock::appendWeakReferenceTransition): (JSC::CodeBlock::shrinkWeakReferenceTransitionsToFit): (JSC::CodeBlock::WeakReferenceTransition::WeakReferenceTransition): * bytecode/CodeOrigin.h: (JSC::CodeOrigin::codeOriginOwner): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::addWeakReference): (JSC::DFG::JITCompiler::addWeakReferenceTransition): (JSC::DFG::JITCompiler::branchWeakPtr): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillJSValue): (JSC::DFG::SpeculativeJIT::compile): 2011-11-16 Michael Saboff <msaboff@apple.com> LayoutTests for Debug Builds Crashes in JavaScriptCore/yarr/YarrInterpreter.cpp(185) https://bugs.webkit.org/show_bug.cgi?id=72561 Removed #if USE(JSC) and therefore the ASSERT_NOT_REACHED(). Simplified the code in the process. Reviewed by James Robinson. * yarr/YarrInterpreter.cpp: (JSC::Yarr::Interpreter::CharAccess::CharAccess): (JSC::Yarr::Interpreter::CharAccess::~CharAccess): 2011-11-16 Geoffrey Garen <ggaren@apple.com> Interpreter build fixes. * bytecode/CodeBlock.h: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): 2011-11-16 Patrick Gansterer <paroga@webkit.org> Unreviewed. Build fix for !ENABLE(JIT) after r100363. * bytecode/CodeBlock.h: 2011-11-16 Geoffrey Garen <ggaren@apple.com> Rolled back in r100375 and r100385 with 32-bit build fixed. * dfg/DFGOperations.cpp: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ArgList.cpp: (JSC::ArgList::getSlice): * runtime/ArgList.h: * runtime/JSArray.cpp: (JSC::JSArray::finishCreation): * runtime/JSArray.h: (JSC::JSArray::create): * runtime/JSGlobalObject.h: (JSC::constructArray): 2011-11-16 Filip Pizlo <fpizlo@apple.com> DFG global variable CSE mishandles the cross-global-object inlining corner case https://bugs.webkit.org/show_bug.cgi?id=72542 Reviewed by Geoff Garen. Moved code to get the global object for a code origin into CodeBlock, so it is more broadly accessible. Fixed CSE to compare both the variable number, and the global object, before deciding to perform elimination. * bytecode/CodeBlock.h: (JSC::CodeBlock::globalObjectFor): * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::globalObjectFor): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::globalVarLoadElimination): (JSC::DFG::Propagator::performNodeCSE): 2011-11-16 Michael Saboff <msaboff@apple.com> Enable 8 Bit Strings in JavaScriptCore https://bugs.webkit.org/show_bug.cgi?id=71337 This patch turns on 8 bit strings in StringImpl and enables their use in JavaScriptCore. Some of the changes are to turn on code that had been staged (Lexer.cpp, Identifier.cpp, SmallStrings.cpp and some of StringImpl.{h,cpp}). Other changes are minor fixes to make 8 bit strings work (UString.h, StringImpl::getData16SlowCase()). Changed StringBuffer to be a templated class based on character type. This change rippled into WebCore code as well. Reviewed by Geoffrey Garen. * JavaScriptCore.exp: * parser/Lexer.cpp: (JSC::::append8): Changed to use 8 bit buffers. (JSC::::parseIdentifier): Changed to use 8 bit buffers. (JSC::::parseString): Changed to use 8 bit buffers. * runtime/Identifier.cpp: (JSC::IdentifierCStringTranslator::translate): 8 bit version keeps data 8 bit (JSC::Identifier::toUInt32FromCharacters): Templated helper. (JSC::Identifier::toUInt32): Added 8 bit optimized path. * runtime/SmallStrings.cpp: (JSC::SmallStringsStorage::SmallStringsStorage): Changed to be 8 bit strings * runtime/UString.h: (JSC::UString::characters): Now calls StringImpl::characters() * wtf/Forward.h: * wtf/text/StringBuffer.h: Made StringBuffer a template base on character type. (WTF::StringBuffer::StringBuffer): (WTF::StringBuffer::characters): (WTF::StringBuffer::release): * wtf/text/StringImpl.cpp: (WTF::StringImpl::create): (WTF::StringImpl::getData16SlowCase): Fixed null terminated case. (WTF::StringImpl::removeCharacters): Added 8 bit path. (WTF::StringImpl::simplifyMatchedCharactersToSpace): (WTF::StringImpl::simplifyWhiteSpace): (WTF::equal): Removed bug from code copied from null terminated version. (WTF::StringImpl::adopt): Added 8 bit path. (WTF::StringImpl::createWithTerminatingNullCharacter): Fixed 8 bi flag propagation. * wtf/text/StringImpl.h: (WTF::StringImpl::StringImpl): Added new 8 bit constructor. (WTF::StringImpl::characters8): Removed ASSERT_NOT_REACHED(). (WTF::getCharacters<LChar>): Added templated accessor for 8 bit strings. (WTF::getCharacters<UChar>): Added templated accessor for 16 bit strings. * wtf/text/WTFString.h: (WTF::String::adopt): Changed to use StringBuffer template. 2011-11-16 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize ExecutableBase::clearCodeVirtual https://bugs.webkit.org/show_bug.cgi?id=72337 Reviewed by Darin Adler. Added static finalize functions to the subclasses of ExecutableBase that provide an implementation of clearCodeVirtual, changed all of the clearCodeVirtual methods to non-virtual clearCode method, and had the finalize functions call the corresponding clearCode methods. * runtime/Executable.cpp: (JSC::ExecutableBase::clearCode): (JSC::NativeExecutable::finalize): (JSC::EvalExecutable::finalize): (JSC::EvalExecutable::clearCode): (JSC::ProgramExecutable::finalize): (JSC::ProgramExecutable::clearCode): (JSC::FunctionExecutable::discardCode): (JSC::FunctionExecutable::finalize): (JSC::FunctionExecutable::clearCode): * runtime/Executable.h: (JSC::ExecutableBase::finishCreation): (JSC::NativeExecutable::create): (JSC::EvalExecutable::create): (JSC::ProgramExecutable::create): (JSC::FunctionExecutable::create): 2011-11-16 Yusuke Suzuki <utatane.tea@gmail.com> String new RegExp('\n').toString() returns is invalid RegularExpressionLiteral https://bugs.webkit.org/show_bug.cgi?id=71572 Reviewed by Gavin Barraclough and Darin Adler. * runtime/RegExpObject.cpp: (JSC::regExpObjectSource): 2011-11-16 Darin Adler <darin@apple.com> Specialize HashTraits for OwnPtr to use PassOwnPtr and raw pointer https://bugs.webkit.org/show_bug.cgi?id=72475 Reviewed by Adam Roben. * wtf/HashTraits.h: Specialize HashTraits for OwnPtr. Do overloads so we can pass a nullptr and also be sure to get the raw pointer type from the OwnPtr template so we handle both forms of OwnPtr: OwnPtr<T> and OwnPtr<T*>. 2011-11-16 Simon Hausmann <simon.hausmann@nokia.com> [Qt] Centralize hide_symbols and ensure all libs are built with symbol visibility & bsymbolic_functions Reviewed by Tor Arne Vestbø. * Target.pri: Eliminate duplicated symbol stuff that lives now in default_post.prf. 2011-11-16 Simon Hausmann <simon.hausmann@nokia.com> Unreviewed, rolling out r100266. http://trac.webkit.org/changeset/100266 Broke WTR. * Target.pri: 2011-11-16 Darin Adler <darin@apple.com> Add a "pass type" and "peek type" concept to HashTraits https://bugs.webkit.org/show_bug.cgi?id=72473 Reviewed by Filip Pizlo. * wtf/HashTraits.h: Added the pass type and peek type. For OwnPtr, the pass type will be PassOwnPtr and the peek type will be a raw pointer. 2011-11-16 Darin Adler <darin@apple.com> Fix some hash traits that don't derive from the base hash traits https://bugs.webkit.org/show_bug.cgi?id=72470 Reviewed by Filip Pizlo. Hash traits structures need to derive from the base hash traits in HashTraits.h, but some were not. This is needed for compatibility with some additional traits we will be adding to make OwnPtr work with HashMap. * runtime/Identifier.h: Make IdentifierMapIndexHashTraits derive from HashTraits<int>. This enabled removal of all the members except for the ones that control the empty value, because this is otherwise the same as the standard int hash. * runtime/SymbolTable.h: Changed SymbolTableIndexHashTraits to derive from HashTraits<SymbolTableEntry> and removed redundant members. 2011-11-15 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r100375 and r100385. http://trac.webkit.org/changeset/100375 http://trac.webkit.org/changeset/100385 https://bugs.webkit.org/show_bug.cgi?id=72465 They broke 32 bit builds on Qt (Requested by ossy on #webkit). * dfg/DFGOperations.cpp: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ArgList.cpp: (JSC::ArgList::getSlice): * runtime/ArgList.h: (JSC::ArgList::ArgList): * runtime/JSArray.cpp: * runtime/JSArray.h: * runtime/JSGlobalObject.h: 2011-11-15 George Staikos <staikos@webkit.org> Remove the guard page from the addressable stack region on QNX. https://bugs.webkit.org/show_bug.cgi?id=72455 Reviewed by Daniel Bates. * wtf/StackBounds.cpp: (WTF::StackBounds::initialize): 2011-11-15 Michael Saboff <msaboff@apple.com> Towards 8 bit Strings - Update utf8() and ascii() methods for 8 bit strings https://bugs.webkit.org/show_bug.cgi?id=72323 Added 8 bit optimized paths for String and UString ascii() and utf8() methods. Added String::characters8(), characters16() and is8Bit() helper methods. Added an new Unicode::convertLatin1ToUTF8() method that works on LChar (8 bit) strings that is a stripped down version of convertUTF16ToUTF8(). Reviewed by Geoff Garen. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/UString.cpp: (JSC::UString::utf8): * wtf/text/WTFString.cpp: (WTF::String::ascii): (WTF::String::utf8): * wtf/text/WTFString.h: (WTF::String::characters8): (WTF::String::characters16): (WTF::String::is8Bit): (WTF::LChar): (WTF::UChar): * wtf/unicode/UTF8.cpp: (WTF::Unicode::convertLatin1ToUTF8): * wtf/unicode/UTF8.h: * wtf/unicode/Unicode.h: 2011-11-15 Darin Adler <darin@apple.com> REGRESSION (r98887): ParserArena and Keywords leaking https://bugs.webkit.org/show_bug.cgi?id=72428 Reviewed by Sam Weinig. * parser/Lexer.h: Made Keywords destructor public since OwnPtr and PassOwnPtr need to be able to destroy it. * parser/Parser.cpp: (JSC::Parser::Parser): Use get now that parserArena is an OwnPtr. * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): Use adoptPtr to initialize OwnPtr members. * runtime/JSGlobalData.h: Make parserArena and keywords be OwnPtr. 2011-11-15 Geoffrey Garen <ggaren@apple.com> Removed another use of ArgList that baked in the assumption that arguments are forward in the regiter file. Reviewed by Sam Weinig. * dfg/DFGOperations.cpp: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): Use our new array creation API, instead of working through ArgList. * runtime/ArgList.h: Removed! 2011-11-15 Geoffrey Garen <ggaren@apple.com> Removed a use of ArgList that baked in the assumption that arguments are forward in the regiter file. Reviewed by Sam Weinig. * dfg/DFGOperations.cpp: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): Use new API. * runtime/ArgList.cpp: (JSC::ArgList::getSlice): No need to provide an arbitrary constructor -- getSlice can do the right thing by using its rights to private data. * runtime/ArgList.h: Removed constructor that took a forward-contiguous set of arguments. * runtime/JSArray.cpp: (JSC::JSArray::finishCreation): * runtime/JSArray.h: (JSC::JSArray::create): * runtime/JSGlobalObject.h: (JSC::constructArray): Added explicit support for creating an array from a pre-allocated set of values, so we could stop relying on the ArgList API we want to remove. 2011-11-15 Filip Pizlo <fpizlo@apple.com> Crash in JSC::DFG::OSRExitCompiler::compileExit(JSC::DFG::OSRExit const&, JSC::DFG::SpeculationRecovery*) https://bugs.webkit.org/show_bug.cgi?id=72292 Reviewed by Geoff Garen. We need to be careful about how we look for the baseline CodeBlock if we're lazy-compiling an OSR exit after our CodeBlock has been jettisoned. In short, use CodeBlock::baselineVersion() instead of CodeBlock::alternative(). No performance effect. No tests because all of our heuristics work very hard to make sure that this never happens in the first place. OSR exits are rare by design, and jettisoning of CodeBlocks (i.e. recompilation) is even rarer. Furthermore, OSR exits after a CodeBlock has been jettisoned is rarer still because the whole point of jettisoning is to bring the probability of future OSR exits to as close to zero as possible. But even that isn't enough to trigger this bug; it requires the OSR exit after a jettison to be the first of its kind; our whole design tries to ensure that CodeBlocks tend to OSR exit at a handful (i.e. 1 in most cases) of points, and since jettisoning is triggered by OSR, in most sane cases the OSR exits after jettison will not require lazy OSR compilation. So this is a truly evil case, and any test for it would be quite fragile. * bytecode/CodeBlock.h: (JSC::CodeBlock::specializationKind): (JSC::CodeBlock::largeFailCountThreshold): (JSC::CodeBlock::largeFailCountThresholdForLoop): * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::AssemblyHelpers): (JSC::DFG::AssemblyHelpers::baselineCodeBlockFor): (JSC::DFG::AssemblyHelpers::baselineCodeBlock): * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGOSRExitCompiler.cpp: * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): 2011-11-15 Geoffrey Garen <ggaren@apple.com> Use MarkedArgumentBuffer to avoid making assumptions about argument order https://bugs.webkit.org/show_bug.cgi?id=72418 Reviewed by Sam Weinig. A step toward reversing the argument order. * runtime/JSONObject.cpp: (JSC::Stringifier::toJSON): (JSC::Stringifier::appendStringifiedValue): (JSC::Walker::callReviver): Don't assume that ArgList wants to point at arguments in forward order. Instead, use MarkedArgumentBuffer, which will make the decision for us. 2011-11-15 Filip Pizlo <fpizlo@apple.com> DFG should distinguish between constants in the constant pool and weak constants added as artifacts of code generation https://bugs.webkit.org/show_bug.cgi?id=72367 Reviewed by Geoff Garen. Added the notion of a WeakJSConstant, which is like a JSConstant except that it can only refer to JSCell*. Currently all WeakJSConstants are also backed by constants in the constant pool, since weak references originated from machine code are not yet properly handled. Replaced CheckMethod, and MethodCheckData, with a combination of WeakJSConstant and CheckStructure. This results in improved CSE, leading to a 1% win on V8. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::cellConstant): (JSC::DFG::ByteCodeParser::prepareToParseBlock): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::Graph::getJSConstantPrediction): (JSC::DFG::Graph::valueOfJSConstant): (JSC::DFG::Graph::valueOfInt32Constant): (JSC::DFG::Graph::valueOfNumberConstant): (JSC::DFG::Graph::valueOfBooleanConstant): * dfg/DFGNode.h: (JSC::DFG::Node::isWeakConstant): (JSC::DFG::Node::hasConstant): (JSC::DFG::Node::weakConstant): (JSC::DFG::Node::valueOfJSConstant): (JSC::DFG::Node::isInt32Constant): (JSC::DFG::Node::isDoubleConstant): (JSC::DFG::Node::isNumberConstant): (JSC::DFG::Node::isBooleanConstant): (JSC::DFG::Node::hasIdentifier): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-15 Michael Saboff <msaboff@apple.com> Towards 8 bit Strings - Initial JS String Tuning https://bugs.webkit.org/show_bug.cgi?id=72326 Added 8 bit optimized paths for the methods below. Reviewed by Geoffrey Garen. * runtime/JSString.h: (JSC::jsSubstring8): * runtime/StringPrototype.cpp: (JSC::jsSpliceSubstrings): (JSC::jsSpliceSubstringsWithSeparators): (JSC::stringProtoFuncReplace): (JSC::stringProtoFuncCharCodeAt): 2011-11-15 Gavin Barraclough <barraclough@apple.com> Result of Error.prototype.toString not ES5 conformant https://bugs.webkit.org/show_bug.cgi?id=70889 Reviewed by Oliver Hunt. * runtime/ErrorPrototype.cpp: (JSC::errorProtoFuncToString): 2011-11-15 Simon Hausmann <simon.hausmann@nokia.com> [Qt] Centralize hide_symbols and ensure all libs are built with symbol visibility & bsymbolic_functions Reviewed by Tor Arne Vestbø. * Target.pri: Eliminate duplicated symbol stuff that lives now in default_post.prf. 2011-11-15 Yuqiang Xian <yuqiang.xian@intel.com> Remove DFGJITCompilerInlineMethods https://bugs.webkit.org/show_bug.cgi?id=72366 Reviewed by Filip Pizlo. Those methods are actually seldom used. Modify the few such places and remove DFGJITCompilerInlineMethods stuffs totally. * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::addressOfDoubleConstant): * dfg/DFGJITCompilerInlineMethods.h: Removed. * dfg/DFGSpeculativeJIT.cpp: * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::silentFillFPR): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillJSValue): (JSC::DFG::SpeculativeJIT::cachedGetMethod): 2011-11-14 Filip Pizlo <fpizlo@apple.com> DFG::SpeculativeJIT and DFG::JITCodeGenerator should be combined https://bugs.webkit.org/show_bug.cgi?id=72348 Reviewed by Gavin Barraclough. Moved all of JITCodeGenerator into SpeculativeJIT. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * dfg/DFGJITCodeGenerator.cpp: Removed. * dfg/DFGJITCodeGenerator.h: Removed. * dfg/DFGJITCodeGenerator32_64.cpp: Removed. * dfg/DFGJITCodeGenerator64.cpp: Removed. * dfg/DFGJITCompiler.cpp: * dfg/DFGRepatch.cpp: (JSC::DFG::generateProtoChainAccessStub): (JSC::DFG::tryCacheGetByID): (JSC::DFG::tryCachePutByID): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::clearGenerationInfo): (JSC::DFG::SpeculativeJIT::fillStorage): (JSC::DFG::SpeculativeJIT::useChildren): (JSC::DFG::SpeculativeJIT::isStrictInt32): (JSC::DFG::SpeculativeJIT::isKnownInteger): (JSC::DFG::SpeculativeJIT::isKnownNumeric): (JSC::DFG::SpeculativeJIT::isKnownCell): (JSC::DFG::SpeculativeJIT::isKnownNotCell): (JSC::DFG::SpeculativeJIT::isKnownNotInteger): (JSC::DFG::SpeculativeJIT::isKnownNotNumber): (JSC::DFG::SpeculativeJIT::isKnownBoolean): (JSC::DFG::SpeculativeJIT::writeBarrier): (JSC::DFG::SpeculativeJIT::markCellCard): (JSC::DFG::SpeculativeJIT::nonSpeculativeCompare): (JSC::DFG::SpeculativeJIT::nonSpeculativeStrictEq): (JSC::DFG::dataFormatString): (JSC::DFG::SpeculativeJIT::dump): (JSC::DFG::SpeculativeJIT::checkConsistency): (JSC::DFG::GPRTemporary::GPRTemporary): (JSC::DFG::GPRTemporary::adopt): (JSC::DFG::FPRTemporary::FPRTemporary): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::at): (JSC::DFG::SpeculativeJIT::lock): (JSC::DFG::SpeculativeJIT::unlock): (JSC::DFG::SpeculativeJIT::canReuse): (JSC::DFG::SpeculativeJIT::reuse): (JSC::DFG::SpeculativeJIT::allocate): (JSC::DFG::SpeculativeJIT::tryAllocate): (JSC::DFG::SpeculativeJIT::fprAllocate): (JSC::DFG::SpeculativeJIT::isFilled): (JSC::DFG::SpeculativeJIT::isFilledDouble): (JSC::DFG::SpeculativeJIT::use): (JSC::DFG::SpeculativeJIT::selectScratchGPR): (JSC::DFG::SpeculativeJIT::silentSpillGPR): (JSC::DFG::SpeculativeJIT::silentSpillFPR): (JSC::DFG::SpeculativeJIT::silentFillGPR): (JSC::DFG::SpeculativeJIT::silentFillFPR): (JSC::DFG::SpeculativeJIT::silentSpillAllRegisters): (JSC::DFG::SpeculativeJIT::silentFillAllRegisters): (JSC::DFG::SpeculativeJIT::boxDouble): (JSC::DFG::SpeculativeJIT::unboxDouble): (JSC::DFG::SpeculativeJIT::spill): (JSC::DFG::SpeculativeJIT::isConstant): (JSC::DFG::SpeculativeJIT::isJSConstant): (JSC::DFG::SpeculativeJIT::isInt32Constant): (JSC::DFG::SpeculativeJIT::isDoubleConstant): (JSC::DFG::SpeculativeJIT::isNumberConstant): (JSC::DFG::SpeculativeJIT::isBooleanConstant): (JSC::DFG::SpeculativeJIT::isFunctionConstant): (JSC::DFG::SpeculativeJIT::valueOfInt32Constant): (JSC::DFG::SpeculativeJIT::valueOfNumberConstant): (JSC::DFG::SpeculativeJIT::addressOfDoubleConstant): (JSC::DFG::SpeculativeJIT::valueOfJSConstant): (JSC::DFG::SpeculativeJIT::valueOfBooleanConstant): (JSC::DFG::SpeculativeJIT::valueOfFunctionConstant): (JSC::DFG::SpeculativeJIT::isNullConstant): (JSC::DFG::SpeculativeJIT::identifier): (JSC::DFG::SpeculativeJIT::flushRegisters): (JSC::DFG::SpeculativeJIT::isFlushed): (JSC::DFG::SpeculativeJIT::valueOfJSConstantAsImmPtr): (JSC::DFG::SpeculativeJIT::bitOp): (JSC::DFG::SpeculativeJIT::shiftOp): (JSC::DFG::SpeculativeJIT::detectPeepHoleBranch): (JSC::DFG::SpeculativeJIT::addressOfCallData): (JSC::DFG::SpeculativeJIT::tagOfCallData): (JSC::DFG::SpeculativeJIT::payloadOfCallData): (JSC::DFG::SpeculativeJIT::integerResult): (JSC::DFG::SpeculativeJIT::noResult): (JSC::DFG::SpeculativeJIT::cellResult): (JSC::DFG::SpeculativeJIT::booleanResult): (JSC::DFG::SpeculativeJIT::jsValueResult): (JSC::DFG::SpeculativeJIT::storageResult): (JSC::DFG::SpeculativeJIT::doubleResult): (JSC::DFG::SpeculativeJIT::initConstantInfo): (JSC::DFG::SpeculativeJIT::resetCallArguments): (JSC::DFG::SpeculativeJIT::addCallArgument): (JSC::DFG::SpeculativeJIT::setupArguments): (JSC::DFG::SpeculativeJIT::setupArgumentsExecState): (JSC::DFG::SpeculativeJIT::setupArgumentsWithExecState): (JSC::DFG::SpeculativeJIT::setupTwoStubArgs): (JSC::DFG::SpeculativeJIT::setupStubArguments): (JSC::DFG::SpeculativeJIT::callOperation): (JSC::DFG::SpeculativeJIT::appendCallWithExceptionCheck): (JSC::DFG::SpeculativeJIT::appendCallWithExceptionCheckSetResult): (JSC::DFG::SpeculativeJIT::setupResults): (JSC::DFG::SpeculativeJIT::appendCallSetResult): (JSC::DFG::SpeculativeJIT::addBranch): (JSC::DFG::SpeculativeJIT::linkBranches): (JSC::DFG::SpeculativeJIT::block): (JSC::DFG::SpeculativeJIT::checkConsistency): (JSC::DFG::SpeculativeJIT::BranchRecord::BranchRecord): (JSC::DFG::IntegerOperand::IntegerOperand): (JSC::DFG::IntegerOperand::~IntegerOperand): (JSC::DFG::IntegerOperand::index): (JSC::DFG::IntegerOperand::format): (JSC::DFG::IntegerOperand::gpr): (JSC::DFG::IntegerOperand::use): (JSC::DFG::DoubleOperand::DoubleOperand): (JSC::DFG::DoubleOperand::~DoubleOperand): (JSC::DFG::DoubleOperand::index): (JSC::DFG::DoubleOperand::fpr): (JSC::DFG::DoubleOperand::use): (JSC::DFG::JSValueOperand::JSValueOperand): (JSC::DFG::JSValueOperand::~JSValueOperand): (JSC::DFG::JSValueOperand::index): (JSC::DFG::JSValueOperand::gpr): (JSC::DFG::JSValueOperand::jsValueRegs): (JSC::DFG::JSValueOperand::isDouble): (JSC::DFG::JSValueOperand::fill): (JSC::DFG::JSValueOperand::tagGPR): (JSC::DFG::JSValueOperand::payloadGPR): (JSC::DFG::JSValueOperand::fpr): (JSC::DFG::JSValueOperand::use): (JSC::DFG::StorageOperand::StorageOperand): (JSC::DFG::StorageOperand::~StorageOperand): (JSC::DFG::StorageOperand::index): (JSC::DFG::StorageOperand::gpr): (JSC::DFG::StorageOperand::use): (JSC::DFG::GPRTemporary::~GPRTemporary): (JSC::DFG::GPRTemporary::gpr): (JSC::DFG::FPRTemporary::~FPRTemporary): (JSC::DFG::FPRTemporary::fpr): (JSC::DFG::FPRTemporary::FPRTemporary): (JSC::DFG::GPRResult::GPRResult): (JSC::DFG::GPRResult2::GPRResult2): (JSC::DFG::FPRResult::FPRResult): (JSC::DFG::FPRResult::lockedResult): (JSC::DFG::SpeculativeJIT::SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillInteger): (JSC::DFG::SpeculativeJIT::fillDouble): (JSC::DFG::SpeculativeJIT::fillJSValue): (JSC::DFG::SpeculativeJIT::nonSpeculativeValueToNumber): (JSC::DFG::SpeculativeJIT::nonSpeculativeValueToInt32): (JSC::DFG::SpeculativeJIT::nonSpeculativeUInt32ToNumber): (JSC::DFG::SpeculativeJIT::nonSpeculativeKnownConstantArithOp): (JSC::DFG::SpeculativeJIT::nonSpeculativeBasicArithOp): (JSC::DFG::SpeculativeJIT::cachedGetById): (JSC::DFG::SpeculativeJIT::cachedPutById): (JSC::DFG::SpeculativeJIT::cachedGetMethod): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompareNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull): (JSC::DFG::SpeculativeJIT::nonSpeculativeCompareNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranch): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompare): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeStrictEq): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeStrictEq): (JSC::DFG::SpeculativeJIT::emitCall): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillInteger): (JSC::DFG::SpeculativeJIT::fillDouble): (JSC::DFG::SpeculativeJIT::fillJSValue): (JSC::DFG::SpeculativeJIT::nonSpeculativeValueToNumber): (JSC::DFG::SpeculativeJIT::nonSpeculativeValueToInt32): (JSC::DFG::SpeculativeJIT::nonSpeculativeUInt32ToNumber): (JSC::DFG::SpeculativeJIT::nonSpeculativeKnownConstantArithOp): (JSC::DFG::SpeculativeJIT::nonSpeculativeBasicArithOp): (JSC::DFG::SpeculativeJIT::cachedGetById): (JSC::DFG::SpeculativeJIT::cachedPutById): (JSC::DFG::SpeculativeJIT::cachedGetMethod): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompareNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranchNull): (JSC::DFG::SpeculativeJIT::nonSpeculativeCompareNull): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeBranch): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompare): (JSC::DFG::SpeculativeJIT::nonSpeculativePeepholeStrictEq): (JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeStrictEq): (JSC::DFG::SpeculativeJIT::emitCall): * runtime/JSFunction.h: 2011-11-14 Filip Pizlo <fpizlo@apple.com> Weak reference harvesters should run to fixpoint https://bugs.webkit.org/show_bug.cgi?id=72346 Reviewed by Oliver Hunt. * heap/Heap.cpp: (JSC::Heap::markRoots): * heap/ListableHandler.h: (JSC::ListableHandler::next): (JSC::ListableHandler::List::head): (JSC::ListableHandler::List::removeNext): (JSC::ListableHandler::List::removeAll): * heap/MarkStack.cpp: (JSC::MarkStackThreadSharedData::reset): (JSC::SlotVisitor::harvestWeakReferences): * heap/MarkStack.h: (JSC::MarkStack::isEmpty): 2011-11-14 Oliver Hunt <oliver@apple.com> Start migrating typed array impl types to WTF https://bugs.webkit.org/show_bug.cgi?id=72336 Reviewed by Geoffrey Garen. Add typed array impls to WTF forwarding header. * wtf/Forward.h: 2011-11-14 Julien Chaffraix <jchaffraix@webkit.org> Add --css-grid-layout to build-webkit and the build systems https://bugs.webkit.org/show_bug.cgi?id=72320 Reviewed by Ojan Vafai. * Configurations/FeatureDefines.xcconfig: 2011-11-14 Geoffrey Garen <ggaren@apple.com> A little bit of arguments / activation cleanup https://bugs.webkit.org/show_bug.cgi?id=72339 Reviewed by Gavin Barraclough. Renamed copyRegisters => tearOff to match bytecode and other terminology. Renamed setActivation => didTearOffActivation to indicate that this is a notification the object may choose to ignore. Moved "Should I ignore?" code into the arguments object to avoid duplication elsewhere. * interpreter/Interpreter.cpp: (JSC::Interpreter::unwindCallFrame): (JSC::Interpreter::privateExecute): (JSC::Interpreter::retrieveArguments): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Arguments.h: (JSC::Arguments::createAndTearOff): (JSC::Arguments::didTearOffActivation): (JSC::Arguments::finishCreationButDontTearOff): (JSC::Arguments::finishCreation): (JSC::Arguments::finishCreationAndTearOff): (JSC::Arguments::tearOff): * runtime/JSActivation.h: (JSC::JSActivation::tearOff): Moved Activation's code into its own header because that's where it belongs. 2011-11-14 Gavin Barraclough <barraclough@apple.com> Should sign the jsc binary https://bugs.webkit.org/show_bug.cgi?id=72332 Reviewed by David Kilzer. * Configurations/JSC.xcconfig: * entitlements.plist: Added. 2011-11-14 Filip Pizlo <fpizlo@apple.com> DFG's inline references to objects should be tracked https://bugs.webkit.org/show_bug.cgi?id=72313 Reviewed by Gavin Barraclough. Added a pinCell() method in the parser that currently creates a dummy constant in CodeBlock. Added calls to pinCell() wherever the DFG would inline a constant reference that the original code would not have referred to. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getCellConstantIndex): (JSC::DFG::ByteCodeParser::pinCell): (JSC::DFG::ByteCodeParser::cellConstant): (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::parseBlock): 2011-11-14 Filip Pizlo <fpizlo@apple.com> DFG put_by_id transition optimizations test the wrong structures https://bugs.webkit.org/show_bug.cgi?id=72324 Reviewed by Gavin Barraclough. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::structureChainIsStillValid): (JSC::DFG::ByteCodeParser::parseBlock): * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompilePutByIdTransition): 2011-11-14 Michael Saboff <msaboff@apple.com> Further changes and cleanup to JSString.h and cpp. Reviewed by Darin Adler. * runtime/JSString.cpp: (JSC::JSString::resolveRope): Change PassRefPtr to RefPtr. Eliminated exec in slow case calls. (JSC::JSString::resolveRopeSlowCase8): Darin and I agreed that this should have 8 in name. (JSC::JSString::resolveRopeSlowCase): Removed exec parameter. * runtime/JSString.h: 2011-11-14 Adam Barth <abarth@webkit.org> DateMath.cpp should not depend on JavaScriptCore https://bugs.webkit.org/show_bug.cgi?id=71747 Reviewed by Darin Adler. This patch moves the JSC-specific parts of DateMath into JSDateMath in JavaScriptCore. There shouldn't be any behavior change. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * runtime/DateConstructor.cpp: * runtime/DateConversion.cpp: * runtime/DateInstance.cpp: * runtime/DateInstanceCache.h: * runtime/DatePrototype.cpp: * runtime/InitializeThreading.cpp: * runtime/JSDateMath.cpp: Copied from Source/JavaScriptCore/wtf/DateMath.cpp. (JSC::timeToMS): (JSC::msToSeconds): * runtime/JSDateMath.h: Copied from Source/JavaScriptCore/wtf/DateMath.h. * wtf/DateMath.cpp: (WTF::isLeapYear): (WTF::msToDays): (WTF::msToMinutes): (WTF::msToHours): (WTF::parseDateFromNullTerminatedCharacters): (WTF::makeRFC2822DateString): * wtf/DateMath.h: 2011-11-14 Michael Saboff <msaboff@apple.com> Towards 8 bit strings - Add 8 bit handling to JSString Ropes https://bugs.webkit.org/show_bug.cgi?id=72317 Added bit to track that a rope is made up of all 8 bit fibers. Created an 8 bit path (fast and slow cases) to handle 8 bit only ropes. Reviewed by Oliver Hunt. * runtime/JSString.cpp: (JSC::JSString::resolveRope): (JSC::JSString::resolveRopeSlowCase8): (JSC::JSString::resolveRopeSlowCase16): * runtime/JSString.h: (JSC::RopeBuilder::finishCreation): (JSC::RopeBuilder::is8Bit): (JSC::jsSubstring8): 2011-11-14 Geoffrey Garen <ggaren@apple.com> A little bit of function call cleanup https://bugs.webkit.org/show_bug.cgi?id=72314 Reviewed by Oliver Hunt. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitCall): Renamed callFrame to registerOffset because this value doesn't give you the offset of the callee's call frame. (JSC::BytecodeGenerator::emitReturn): Tightened to use equality instead of greater-than. Removed comment since its reasoning was wrong. (JSC::BytecodeGenerator::emitConstruct): Updated for rename mentioned above. (JSC::BytecodeGenerator::isArgumentNumber): Provided a more precise way to ask this question, giving the bytecode generator more freedom to change internal implementation details. * bytecompiler/BytecodeGenerator.h: Reduced default vector capacity because 16 was overkill. (JSC::CallArguments::registerOffset): Updated for rename mentioned above. * bytecompiler/NodesCodegen.cpp: (JSC::CallArguments::CallArguments): (JSC::CallArguments::newArgument): Factored out argument allocation into a helper function, so I can change it later. (JSC::CallFunctionCallDotNode::emitBytecode): (JSC::FunctionBodyNode::emitBytecode): Use helper function mentioned above. 2011-11-14 Tony Chang <tony@chromium.org> Remove the CSS3_FLEXBOX compile time flag and enable on all ports https://bugs.webkit.org/show_bug.cgi?id=72196 Reviewed by Ojan Vafai. * Configurations/FeatureDefines.xcconfig: 2011-11-14 Mark Rowe <mrowe@apple.com> <rdar://problem/10424154> testRegExp should not be installed as part of JavaScriptCore testRegExp and testapi.js were being installed in the JavaScriptCore framework. As test-only tools they shouldn't be installed there by default, only when FORCE_TOOL_INSTALL is set to YES. This patch incorprorates a few related changes: 1) Make the jsc and testRegExp targets be configured via .xcconfig files. 2) Sets up testRegExp so that SKIP_INSTALL is YES by default, and only NO when FORCE_TOOL_INSTALL is YES. 3) Switches the testapi target to using a script build phase to install testapi.js so that the installation will be skipped when SKIP_INSTALL is YES. I'm not sure why this isn't the built-in behavior when a Copy Files build phase has "Copy only when installing" checked, but it doesn't seem to be. 4) Other random cleanup such as removing a bogus group that refers to files that do not exist, moving testRegExp.cpp in to the tests group, etc. Reviewed by Geoff Garen. * Configurations/JSC.xcconfig: Added. * Configurations/TestRegExp.xcconfig: Added. * JavaScriptCore.xcodeproj/project.pbxproj: 2011-11-14 Michael Saboff <msaboff@apple.com> Towards 8 bit strings - Add 8 bit paths to StringImpl methods https://bugs.webkit.org/show_bug.cgi?id=72290 Added 8 bit patchs to StringImpl to number and find methods. Reviewed by Oliver Hunt. * wtf/text/StringImpl.cpp: (WTF::StringImpl::toIntStrict): (WTF::StringImpl::toUIntStrict): (WTF::StringImpl::toInt64Strict): (WTF::StringImpl::toUInt64Strict): (WTF::StringImpl::toIntPtrStrict): (WTF::StringImpl::toInt): (WTF::StringImpl::toUInt): (WTF::StringImpl::toInt64): (WTF::StringImpl::toUInt64): (WTF::StringImpl::toIntPtr): (WTF::StringImpl::toDouble): (WTF::StringImpl::toFloat): (WTF::StringImpl::find): (WTF::StringImpl::reverseFind): * wtf/text/WTFString.cpp: (WTF::toIntegralType): (WTF::lengthOfCharactersAsInteger): (WTF::charactersToIntStrict): (WTF::charactersToUIntStrict): (WTF::charactersToInt64Strict): (WTF::charactersToUInt64Strict): (WTF::charactersToIntPtrStrict): (WTF::charactersToInt): (WTF::charactersToUInt): (WTF::charactersToInt64): (WTF::charactersToUInt64): (WTF::charactersToIntPtr): (WTF::toDoubleType): (WTF::charactersToDouble): (WTF::charactersToFloat): * wtf/text/WTFString.h: (WTF::find): (WTF::reverseFind): 2011-11-14 Vincent Scheib <scheib@chromium.org> Mouse Lock: Renaming to 'Pointer Lock': ENABLE Flags https://bugs.webkit.org/show_bug.cgi?id=72286 Reviewed by Adam Barth. * wtf/Platform.h: 2011-11-14 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=72280 Rubber stamped by Geoff Garen. Fix 32-bit Lion. * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveAndCommit): 2011-11-14 Geoffrey Garen <ggaren@apple.com> 32-bit Build fix: declare virtual register indices to be int rather than unsigned, since they can be positive or negative. For better clarity, explicitly use ReturnPC instead of -1 as the "invalid" state, since we'll never load and operate on the ReturnPC as a JS value. * jit/JIT.cpp: (JSC::JIT::JIT): * jit/JIT.h: * jit/JITInlineMethods.h: (JSC::JIT::emitLoadTag): (JSC::JIT::emitLoadPayload): (JSC::JIT::emitLoad): (JSC::JIT::emitLoad2): (JSC::JIT::emitLoadDouble): (JSC::JIT::emitLoadInt32ToDouble): (JSC::JIT::emitStore): (JSC::JIT::emitStoreInt32): (JSC::JIT::emitStoreAndMapInt32): (JSC::JIT::emitStoreCell): (JSC::JIT::emitStoreBool): (JSC::JIT::emitStoreDouble): (JSC::JIT::map): (JSC::JIT::unmap): (JSC::JIT::isMapped): (JSC::JIT::getMappedPayload): (JSC::JIT::getMappedTag): (JSC::JIT::emitJumpSlowCaseIfNotJSCell): 2011-11-14 Michael Saboff <msaboff@apple.com> Remove unused m_data member from UStringSourceProvider https://bugs.webkit.org/show_bug.cgi?id=72289 Removed unused m_data member from UStringSourceProvider. Reviewed by Oliver Hunt. * parser/SourceProvider.h: (JSC::UStringSourceProvider::UStringSourceProvider): 2011-11-14 Michael Saboff <msaboff@apple.com> Towards 8 Bit Strings: Templatize YARR Parser https://bugs.webkit.org/show_bug.cgi?id=72288 Changed Yarr::Parser to be a template based on character type. Reviewed by Oliver Hunt. * yarr/YarrParser.h: (JSC::Yarr::Parser::Parser): (JSC::Yarr::parse): 2011-11-14 Geoffrey Garen <ggaren@apple.com> 32-bit build fix: Removed unused declaration. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::emitCall): 2011-11-12 Geoffrey Garen <ggaren@apple.com> Standardized the JS calling convention https://bugs.webkit.org/show_bug.cgi?id=72221 Reviewed by Oliver Hunt. This patch standardizes the calling convention so that the caller always sets up the callee's CallFrame. Adjustments for call type, callee type, argument count, etc. now always take place after that initial setup. This is a step toward reversing the argument order, but also has these immediate benefits (measured on x64): (1) 1% benchmark speedup across the board. (2) 50% code size reduction in baseline JIT function calls. (3) 1.5x speedup for single-dispatch .apply forwarding. (4) 1.1x speedup for multi-dispatch .apply forwarding. This change affected the baseline JIT most, since the baseline JIT had lots of ad hoc calling conventions for different caller / callee types. * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::branchPtr): (JSC::MacroAssemblerX86_64::branchAddPtr): Optimize compare to 0 into a test, like other assemblers do. (I added some compares to 0, and didn't want them to be slow.) * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): Merged op_load_varargs into op_call_varargs so op_call_varargs could share code generation with other forms of op_call. This is also a small optimization, since op_*varargs no longer have to pass arguments to each other through the register file. (JSC::CallLinkInfo::unlink): * bytecode/CodeBlock.h: Added a new call type: CallVarargs. This allows us to link functions called through .apply syntax. We need to distinguish CallVarargs from Call because CallVarargs changes its argument count on each inovcation, so we must always link to the argument count checking version of the callee. * bytecode/Opcode.h: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitCallVarargs): * bytecompiler/BytecodeGenerator.h: Merged op_load_varargs into op_call_varargs. * bytecompiler/NodesCodegen.cpp: (JSC::ApplyFunctionCallDotNode::emitBytecode): Ditto. Also, simplified some of this bytecode generation to remove redundant copies. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::emitCall): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::emitCall): Added a new call type: CallVarargs. DFG doesn't support this type, but its code needs to change slightly to accomodate a 3-state variable. Stopped passing the argument count in regT1 because this is non-standard. (The argument count goes in the CallFrame. This trades speed on the slow path for speed and code size on the fast path, and simplicity on all paths. A good trade, in my opinion.) * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileEntry): (JSC::DFG::JITCompiler::link): (JSC::DFG::JITCompiler::compile): (JSC::DFG::JITCompiler::compileFunction): Tweaked code to make CallFrame setup more obvious when single-stepping. Also, updated for argument count not being in regT1. * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::addJSCall): (JSC::DFG::JITCompiler::JSCallRecord::JSCallRecord): Added a new call type: CallVarargs. * dfg/DFGOperations.cpp: Do finish CallFrame setup in one place before doing anything else. Don't check for stack overflow because we have no callee registers, and our caller has already checked for its own registers. * dfg/DFGRepatch.cpp: (JSC::DFG::dfgLinkFor): We can link to our callee even if our argument count doesn't match -- we just need to link to the argument count checking version. * interpreter/CallFrameClosure.h: (JSC::CallFrameClosure::setArgument): BUG FIX: When supplying too many arguments from C++, we need to supply a full copy of the arguments prior to the subset copy that matches our callee's argument count. (That is what the standard calling convention would have produced in JS.) I would have split this into its own patch, but I couldn't find a way to get the JIT to fail a regression test in this area without my patch applied. * interpreter/Interpreter.cpp: Let the true code bomb begin! (JSC::eval): Fixed up this helper function to operate on eval()'s CallFrame, and not eval()'s caller frame. We no longer leave the CallFrame pointing to eval()'s caller during a call to eval(), since that is not standard. (JSC::loadVarargs): Factored out a shared helper function for use by JIT and interpreter because half the code means one quarter the bugs -- in my programming, at least. (JSC::Interpreter::execute): Removed a now-unused way to invoke eval. (JSC::Interpreter::privateExecute): Removed an invalid ASSERT following putDirect, because it got in the way of my testing. (When putting a function, the cached base of a PutPropertySlot can be 0 to signify "do not optimize".) op_call_eval: Updated for new, standard eval calling convention. op_load_varargs: Merged op_load_varargs into op_call_varargs. op_call_varags: Updated for new, standard eval calling convention. Don't check for stack overflow because the loadVarargs helper function already checked. * interpreter/Interpreter.h: (JSC::Interpreter::execute): Headers are fun and educational! * interpreter/RegisterFile.cpp: (JSC::RegisterFile::growSlowCase): * interpreter/RegisterFile.h: (JSC::RegisterFile::grow): Factored out the slow case into a slow case because it was cramping the style of my fast case. * jit/JIT.cpp: (JSC::JIT::privateCompile): Moved initialization of RegisterFile::CodeBlock to make it more obvious when debugging. Removed assumption that argument count is in regT1, as above. Removed call to restoreArgumentReference() because the JITStubCall abstraction does this for us. (JSC::JIT::linkFor): Link even if we miss on argument count, as above. * jit/JIT.h: * jit/JITCall32_64.cpp: (JSC::JIT::emitSlow_op_call): (JSC::JIT::emitSlow_op_call_eval): (JSC::JIT::emitSlow_op_call_varargs): (JSC::JIT::emitSlow_op_construct): (JSC::JIT::emit_op_call_eval): (JSC::JIT::emit_op_call_varargs): Share all function call code generation. Don't count call_eval when accounting for linkable function calls because eval doesn't link. (Its fast path is to perform the eval.) (JSC::JIT::compileLoadVarargs): Ported this inline copying optimization to our new calling convention. The key to this optimization is the observation that, in a function that declares no arguments, if any arguments are passed, they all end up right behind 'this'. (JSC::JIT::compileCallEval): (JSC::JIT::compileCallEvalSlowCase): Factored out eval for a little clarity. (JSC::JIT::compileOpCall): (JSC::JIT::compileOpCallSlowCase): If you are still with me, dear reader, this is the whole point of my patch. The caller now unconditionally moves the CallFrame forward and fills in the data it knows before taking any branches to deal with weird caller/callee pairs. This also means that there is almost no slow path for calls -- it all gets folded into the shared virtual call stub. The only things remaining in the slow path are the rare case counter and a call to the stub. * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::privateCompileCTINativeCall): Updated for values being in different registers or in memory, based on our new standard calling convention. Added a shared path for calling out to CTI helper functions for non-JS calls. * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_method_check): method_check emits its own code and the following get_by_id's code, so it needs to add both when informing result chaining of its result. This is important because the standard calling convention can now take advantage of this chaining. * jit/JITCall.cpp: (JSC::JIT::compileLoadVarargs): (JSC::JIT::compileCallEval): (JSC::JIT::compileCallEvalSlowCase): (JSC::JIT::compileOpCall): (JSC::JIT::compileOpCallSlowCase): * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::emit_op_call_eval): (JSC::JIT::emit_op_call_varargs): (JSC::JIT::emitSlow_op_call): (JSC::JIT::emitSlow_op_call_eval): (JSC::JIT::emitSlow_op_call_varargs): (JSC::JIT::emitSlow_op_construct): Observe, as I write all of my code a second time, now with 64 bits. * jit/JITStubs.cpp: (JSC::throwExceptionFromOpCall): (JSC::jitCompileFor): (JSC::arityCheckFor): (JSC::lazyLinkFor): A lot of mechanical changes here for one purpose: Exceptions thrown in the middle of a function call now use a shared helper function (throwExceptionFromOpCall). This function understands that the CallFrame currently points to the callEE, and the exception must be thrown by the callER. (The old calling convention would often still have the CallFrame pointing at the callER at the point of an exception. That is not the way of our new, standard calling convention.) (JSC::op_call_eval): Finish standard CallFrame setup before calling our eval helper function, which now depends on that setup. * runtime/Arguments.h: (JSC::Arguments::length): Renamed numProvidedArguments() to length() because that's what other objects call it, and the difference made our new loadVarargs helper function hard to read. * runtime/Executable.cpp: (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): Interpreter build fixes. * runtime/FunctionPrototype.cpp: (JSC::functionProtoFuncApply): Honor Arguments::MaxArguments even when the .apply call_varargs optimization fails. (This bug appears on layout tests when you disable the optimization.) 2011-11-11 Jer Noble <jer.noble@apple.com> Implement MediaController. https://bugs.webkit.org/show_bug.cgi?id=71408 Reviewed by Eric Carlson. Change the definition of WTF_USE_COREAUDIO to exclude Windows completely, as CoreAudioClock.h is not available there. * wtf/Platform.h: 2011-11-14 Patrick Gansterer <paroga@webkit.org> [WIN] Remove dependency on pthread from FastMalloc https://bugs.webkit.org/show_bug.cgi?id=72098 Reviewed by Adam Roben. All pthread calls are already ported to native Windows calls. Use the native version for all OS(WINDOWS) to remove the runtime dependency on the pthread dll. * wtf/FastMalloc.cpp: 2011-11-14 Simon Hausmann <simon.hausmann@nokia.com> [Qt] Replace use of QApplication with QGuiApplication. Reviewed by Tor Arne Vestbø. * wtf/qt/compat/qguiapplication.h: (QGuiApplication::styleHints): Introduce styleHints wrapper hack. 2011-11-14 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck build. * GNUmakefile.list.am: Add missing files. 2011-11-11 Yury Semikhatsky <yurys@chromium.org> Web Inspector: function remote objetct should provide access to function position in the script https://bugs.webkit.org/show_bug.cgi?id=71808 Exposed accessor for function source code. Reviewed by Pavel Feldman. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSFunction.cpp: (JSC::JSFunction::sourceCode): * runtime/JSFunction.h: 2011-11-13 Yuqiang Xian <yuqiang.xian@intel.com> Fix silent spilling/filling GPRs in DFG 32_64 https://bugs.webkit.org/show_bug.cgi?id=72201 Reviewed by Gavin Barraclough. Current silentSpillGPR/silentFillGPR may not work as expected for some cases in 32_64. If there's a JSValue which was retained by two GPRs, we may end up failing to spill/fill some GPRs or redundantly spilling/filling some GPRs. For example, if we tend to exclude "eax" from spilling while a JSValue is retained by both "eax" and "edx", then "edx" won't be spilled as well (wrong). And if another JSValue is retained by "ecx" and "ebx", both "ecx" and "ebx" will be spilled twice. The similar problem applies to silentFillGPR. The fix is to make silentSpillGPR/silentFillGPR more straightforward, i.e., spilling/filling based on the GPR instead of the virtual register. FPR spilling/filling is also modified accordingly to make it consistent with GPR spilling/filling. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentSpillGPR): (JSC::DFG::JITCodeGenerator::silentSpillFPR): (JSC::DFG::JITCodeGenerator::silentFillGPR): (JSC::DFG::JITCodeGenerator::silentFillFPR): (JSC::DFG::JITCodeGenerator::silentSpillAllRegisters): (JSC::DFG::JITCodeGenerator::silentFillAllRegisters): 2011-11-12 Laszlo Gombos <laszlo.1.gombos@nokia.com> [Qt][Symbian] Remove support for WINSCW compiler https://bugs.webkit.org/show_bug.cgi?id=70178 Reviewed by Chang Shu. * API/JSStringRef.h: * create_hash_table: Revert r45553. * runtime/JSGlobalData.cpp: Revert r45553. * runtime/LiteralParser.cpp: Remove WINSCW comment. (JSC::LiteralParser::Lexer::lexString): * runtime/Lookup.h: Revert r45553. * runtime/Structure.h: Revert r48461. * wtf/Alignment.h: * wtf/Assertions.h: Revert r52337. * wtf/Compiler.h: * wtf/ListRefPtr.h: Revert r48988. (WTF::ListRefPtr::~ListRefPtr): * wtf/OwnArrayPtr.h: Revert r45911. (WTF::OwnArrayPtr::operator UnspecifiedBoolType): * wtf/PassOwnArrayPtr.h: (WTF::PassOwnArrayPtr::operator UnspecifiedBoolType): * wtf/PassRefPtr.h: * wtf/StaticConstructors.h: * wtf/unicode/qt4/UnicodeQt4.h: 2011-11-12 Patrick Gansterer <paroga@webkit.org> Unreviewed. Add ENABLE(DFG_JIT) around DFGCorrectableJumpPoint code. * dfg/DFGCorrectableJumpPoint.cpp: * dfg/DFGCorrectableJumpPoint.h: 2011-11-12 Patrick Gansterer <paroga@webkit.org> [CMake] Move list of DFG source files into correct file https://bugs.webkit.org/show_bug.cgi?id=72212 Reviewed by Daniel Bates. The DFG files are platform independent. So move them from the EFL specific file into the general CMakeLists.txt. * CMakeLists.txt: * PlatformEfl.cmake: 2011-11-12 Patrick Gansterer <paroga@webkit.org> Fix "unused variable" warning in JSLock https://bugs.webkit.org/show_bug.cgi?id=72213 Reviewed by Anders Carlsson. Use ASSERT_UNUSED() instead of ASSERT() to make sure that the variable is also used in the release build. * runtime/JSLock.cpp: (JSC::JSLock::lock): (JSC::JSLock::unlock): 2011-11-11 Gavin Barraclough <barraclough@apple.com> Update iOS compiler version. Reviewed by David Kilzer. * Configurations/CompilerVersion.xcconfig: - Update compiler version. 2011-11-11 Gavin Barraclough <barraclough@apple.com> Update iOS port's configuration setting, particularly in Platform.h https://bugs.webkit.org/show_bug.cgi?id=72187 Reviewed by David Kilzer. * interpreter/Interpreter.h: - Lower the reentry depth. * runtime/DatePrototype.cpp: - iOS also uses CF. * wtf/FastMalloc.cpp: (WTF::TCMalloc_PageHeap::IncrementalScavenge): - Update fastmalloc configuration for iOS. * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveAndCommit): - Added flag. * wtf/Platform.h: - Update platform configuration for iOS. 2011-11-11 David Kilzer <ddkilzer@apple.com> Only define BUILDING_ON_* and TARGETING_* macros when building for Mac OS X <http://webkit.org/b/72175> Reviewed by Joseph Pecoraro. * wtf/Platform.h: Move the definition of the BUILDING_ON_* and TARGETING_* macros to where the WTF_OS_MAC_OS_X macro is defined so that they're only defined on Mac OS X builds. Also include Availability.h, which is needed on iOS builds. 2011-11-11 Darin Adler <darin@apple.com> Remove all releaseRef implementations except for RetainPtr https://bugs.webkit.org/show_bug.cgi?id=71423 Reviewed by Julien Chaffraix. * API/JSRetainPtr.h: Removed releaseRef. * wtf/PassRefPtr.h: Removed releaseRef. 2011-11-11 Darin Adler <darin@apple.com> * JavaScriptCore.xcodeproj/project.pbxproj: Let a newer Xcode update this file. If an older Xcode downgrades this file and we have a risk of some kind of oscillating commit situation, please contact me so I know not to do this again. 2011-11-11 Mark Hahnenberg <mhahnenberg@apple.com> Add jsCast to replace static_cast https://bugs.webkit.org/show_bug.cgi?id=72071 Reviewed by Geoffrey Garen. Added new jsCast and changed all of the static_cast sites in functions that are in the MethodTable to use jsCast instead. * API/JSCallbackFunction.cpp: (JSC::JSCallbackFunction::toStringCallback): (JSC::JSCallbackFunction::valueOfCallback): * API/JSCallbackObject.h: (JSC::JSCallbackObject::visitChildren): * API/JSCallbackObjectFunctions.h: (JSC::::className): (JSC::::getOwnPropertySlot): (JSC::::getOwnPropertyDescriptor): (JSC::::put): (JSC::::deleteProperty): (JSC::::deletePropertyByIndex): (JSC::::getConstructData): (JSC::::hasInstance): (JSC::::getCallData): (JSC::::getOwnPropertyNames): * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::visitChildren): (JSC::DebuggerActivation::className): (JSC::DebuggerActivation::getOwnPropertySlot): (JSC::DebuggerActivation::put): (JSC::DebuggerActivation::putWithAttributes): (JSC::DebuggerActivation::deleteProperty): (JSC::DebuggerActivation::getOwnPropertyNames): (JSC::DebuggerActivation::getOwnPropertyDescriptor): (JSC::DebuggerActivation::defineGetter): (JSC::DebuggerActivation::defineSetter): * runtime/Arguments.cpp: (JSC::Arguments::visitChildren): (JSC::Arguments::getOwnPropertySlotByIndex): (JSC::Arguments::getOwnPropertySlot): (JSC::Arguments::getOwnPropertyDescriptor): (JSC::Arguments::getOwnPropertyNames): (JSC::Arguments::putByIndex): (JSC::Arguments::put): (JSC::Arguments::deletePropertyByIndex): (JSC::Arguments::deleteProperty): * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::getOwnPropertySlot): (JSC::ArrayConstructor::getOwnPropertyDescriptor): * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::getOwnPropertySlot): (JSC::ArrayPrototype::getOwnPropertyDescriptor): * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::getOwnPropertySlot): (JSC::BooleanPrototype::getOwnPropertyDescriptor): * runtime/DateConstructor.cpp: (JSC::DateConstructor::getOwnPropertySlot): (JSC::DateConstructor::getOwnPropertyDescriptor): * runtime/DatePrototype.cpp: (JSC::DatePrototype::getOwnPropertySlot): (JSC::DatePrototype::getOwnPropertyDescriptor): * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::getOwnPropertySlot): (JSC::ErrorPrototype::getOwnPropertyDescriptor): * runtime/Executable.cpp: (JSC::ExecutableBase::clearCode): (JSC::EvalExecutable::visitChildren): (JSC::ProgramExecutable::visitChildren): (JSC::FunctionExecutable::visitChildren): * runtime/GetterSetter.cpp: (JSC::GetterSetter::visitChildren): * runtime/JSActivation.cpp: (JSC::JSActivation::visitChildren): (JSC::JSActivation::getOwnPropertyNames): (JSC::JSActivation::getOwnPropertySlot): (JSC::JSActivation::put): (JSC::JSActivation::putWithAttributes): * runtime/JSArray.cpp: (JSC::JSArray::getOwnPropertySlotByIndex): (JSC::JSArray::getOwnPropertySlot): (JSC::JSArray::getOwnPropertyDescriptor): (JSC::JSArray::put): (JSC::JSArray::putByIndex): (JSC::JSArray::deleteProperty): (JSC::JSArray::deletePropertyByIndex): (JSC::JSArray::getOwnPropertyNames): (JSC::JSArray::visitChildren): * runtime/JSBoundFunction.cpp: (JSC::JSBoundFunction::hasInstance): (JSC::JSBoundFunction::visitChildren): * runtime/JSByteArray.cpp: (JSC::JSByteArray::getOwnPropertySlot): (JSC::JSByteArray::getOwnPropertyDescriptor): (JSC::JSByteArray::getOwnPropertySlotByIndex): (JSC::JSByteArray::put): (JSC::JSByteArray::putByIndex): (JSC::JSByteArray::getOwnPropertyNames): * runtime/JSCell.h: (JSC::JSCell::visitChildren): (JSC::jsCast): * runtime/JSFunction.cpp: (JSC::JSFunction::visitChildren): (JSC::JSFunction::getCallData): (JSC::JSFunction::getOwnPropertySlot): (JSC::JSFunction::getOwnPropertyDescriptor): (JSC::JSFunction::getOwnPropertyNames): (JSC::JSFunction::put): (JSC::JSFunction::deleteProperty): (JSC::JSFunction::getConstructData): * runtime/JSGlobalData.cpp: (JSC::StackPreservingRecompiler::operator()): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::put): (JSC::JSGlobalObject::putWithAttributes): (JSC::JSGlobalObject::defineGetter): (JSC::JSGlobalObject::defineSetter): (JSC::JSGlobalObject::visitChildren): (JSC::JSGlobalObject::getOwnPropertySlot): (JSC::JSGlobalObject::getOwnPropertyDescriptor): (JSC::JSGlobalObject::clearRareData): * runtime/JSGlobalThis.cpp: (JSC::JSGlobalThis::visitChildren): * runtime/JSONObject.cpp: (JSC::JSONObject::getOwnPropertySlot): (JSC::JSONObject::getOwnPropertyDescriptor): * runtime/JSObject.cpp: (JSC::JSObject::finalize): (JSC::JSObject::visitChildren): (JSC::JSObject::getOwnPropertySlotByIndex): (JSC::JSObject::put): (JSC::JSObject::putByIndex): (JSC::JSObject::deleteProperty): (JSC::JSObject::deletePropertyByIndex): * runtime/JSObject.h: (JSC::JSObject::getOwnPropertySlot): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::visitChildren): * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::visitChildren): (JSC::JSStaticScopeObject::put): (JSC::JSStaticScopeObject::putWithAttributes): (JSC::JSStaticScopeObject::getOwnPropertySlot): * runtime/JSString.cpp: (JSC::JSString::visitChildren): (JSC::JSString::toThisObject): (JSC::JSString::getOwnPropertySlot): (JSC::JSString::getOwnPropertySlotByIndex): * runtime/JSVariableObject.cpp: (JSC::JSVariableObject::deleteProperty): (JSC::JSVariableObject::getOwnPropertyNames): * runtime/JSWrapperObject.cpp: (JSC::JSWrapperObject::visitChildren): * runtime/MathObject.cpp: (JSC::MathObject::getOwnPropertySlot): (JSC::MathObject::getOwnPropertyDescriptor): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::visitChildren): * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::getOwnPropertySlot): (JSC::NumberConstructor::getOwnPropertyDescriptor): * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::getOwnPropertySlot): (JSC::NumberPrototype::getOwnPropertyDescriptor): * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::getOwnPropertySlot): (JSC::ObjectConstructor::getOwnPropertyDescriptor): * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::put): (JSC::ObjectPrototype::getOwnPropertySlotByIndex): (JSC::ObjectPrototype::getOwnPropertySlot): (JSC::ObjectPrototype::getOwnPropertyDescriptor): * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::getOwnPropertySlot): (JSC::RegExpConstructor::getOwnPropertyDescriptor): (JSC::RegExpConstructor::put): * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::getOwnPropertySlot): (JSC::RegExpMatchesArray::getOwnPropertySlotByIndex): (JSC::RegExpMatchesArray::getOwnPropertyDescriptor): (JSC::RegExpMatchesArray::put): (JSC::RegExpMatchesArray::putByIndex): (JSC::RegExpMatchesArray::deleteProperty): (JSC::RegExpMatchesArray::deletePropertyByIndex): (JSC::RegExpMatchesArray::getOwnPropertyNames): * runtime/RegExpObject.cpp: (JSC::RegExpObject::visitChildren): (JSC::RegExpObject::getOwnPropertySlot): (JSC::RegExpObject::getOwnPropertyDescriptor): (JSC::RegExpObject::put): * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::getOwnPropertySlot): (JSC::RegExpPrototype::getOwnPropertyDescriptor): * runtime/ScopeChain.cpp: (JSC::ScopeChainNode::visitChildren): * runtime/StringConstructor.cpp: (JSC::StringConstructor::getOwnPropertySlot): (JSC::StringConstructor::getOwnPropertyDescriptor): * runtime/StringObject.cpp: (JSC::StringObject::getOwnPropertySlot): (JSC::StringObject::getOwnPropertySlotByIndex): (JSC::StringObject::getOwnPropertyDescriptor): (JSC::StringObject::deleteProperty): (JSC::StringObject::getOwnPropertyNames): * runtime/StringPrototype.cpp: (JSC::StringPrototype::getOwnPropertySlot): (JSC::StringPrototype::getOwnPropertyDescriptor): * runtime/Structure.cpp: (JSC::Structure::visitChildren): * runtime/StructureChain.cpp: (JSC::StructureChain::visitChildren): 2011-11-11 Gavin Barraclough <barraclough@apple.com> Enable DFG JIT for ARMv7/iOS. Rubber stamped by Oliver Hunt. * wtf/Platform.h: - enable DFG JIT for ARMv7/iOS. 2011-11-11 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize supportsProfiling, supportsRichSourceInfo, shouldInterruptScript in JSGlobalObject https://bugs.webkit.org/show_bug.cgi?id=72035 Reviewed by Geoffrey Garen. De-virtualized the methods through the use of a new method table just for JSGlobalObject and subclasses. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * bytecompiler/BytecodeGenerator.cpp: Changed call sites to use the new GlobalObjectMethodTable. (JSC::BytecodeGenerator::BytecodeGenerator): * interpreter/Interpreter.cpp: Ditto. (JSC::Interpreter::execute): * runtime/JSGlobalObject.cpp: Added a static const GlobalObjectMethodTable with the correct function pointers. * runtime/JSGlobalObject.h: Added a field in JSGlobalObject to keep track of the current method table. (JSC::JSGlobalObject::JSGlobalObject): (JSC::JSGlobalObject::globalObjectMethodTable): The new struct to contain the function pointers. (JSC::JSGlobalObject::supportsProfiling): Made static to put in the method table. (JSC::JSGlobalObject::supportsRichSourceInfo): Ditto. (JSC::JSGlobalObject::shouldInterruptScript): Ditto. * runtime/TimeoutChecker.cpp: Changed call sites to use the new GlobalObjectMethodTable for lookup. (JSC::TimeoutChecker::didTimeOut): 2011-11-11 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSGlobalObject::allowsAccessFrom https://bugs.webkit.org/show_bug.cgi?id=71969 Reviewed by Darin Adler. * runtime/JSGlobalObject.h: Removed allowsAccessFrom from JSGlobalObject since it is exclusive to JSDOMWindowBase and WebScriptObject. 2011-11-11 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r99950. http://trac.webkit.org/changeset/99950 https://bugs.webkit.org/show_bug.cgi?id=72117 "Landed wrong patch by mistake" (Requested by yurys on #webkit). * JavaScriptCore.exp: * runtime/JSFunction.cpp: * runtime/JSFunction.h: 2011-11-11 Patrick Gansterer <paroga@webkit.org> Unreviewed. Build fix for !ENABLE(JIT) after r99898. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::CodeBlock): 2011-11-10 Dan Bernstein <mitz@apple.com> Disabling assertions breaks the debug build https://bugs.webkit.org/show_bug.cgi?id=72091 Reviewed by Geoff Garen. * dfg/DFGNode.h: Made hasIdentifier() available when assertions are disabled. It is used in Graph::dump(). * runtime/JSObject.cpp: (JSC::JSObject::visitChildren): Update m_isCheckingForDefaultMarkViolation only if assertions are enabled. * wtf/Deque.h: (WTF::::checkIndexValidity): Changed ASSERT to ASSERT_UNUSED. * wtf/ThreadRestrictionVerifier.h: (WTF::ThreadRestrictionVerifier::setShared): Guarded the definition of a local variable that is only used in an assertion. 2011-11-10 Filip Pizlo <fpizlo@apple.com> JSString forgets to clear m_fibers when resolving ropes https://bugs.webkit.org/show_bug.cgi?id=72089 Reviewed by Geoff Garen. * runtime/JSString.cpp: (JSC::JSString::resolveRopeSlowCase): 2011-11-09 Filip Pizlo <fpizlo@apple.com> DFG byte array support sometimes clamps values incorrectly https://bugs.webkit.org/show_bug.cgi?id=71975 Reviewed by Oliver Hunt. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): 2011-11-10 Filip Pizlo <fpizlo@apple.com> ValueProfile/PredictedType contains dead code, and doesn't recognize functions https://bugs.webkit.org/show_bug.cgi?id=72065 Reviewed by Gavin Barraclough and Geoff Garen. Added PredictFunction support, and did some cleaning up along the way. ValueProfile no longer has statistics machinery, because we never used it. Rearranged some bits in PredictedType to more easily make room for one more object type. Changed some debug code to use more consistent conventions (ByteArray becomes Bytearray so that if we ever have a "Byte" prediction we don't get confused between a prediction that is the union of Byte and Array and a prediction that indicates precisely a ByteArray). * bytecode/PredictedType.cpp: (JSC::predictionToString): (JSC::predictionFromClassInfo): * bytecode/PredictedType.h: (JSC::isFunctionPrediction): * bytecode/ValueProfile.cpp: * bytecode/ValueProfile.h: (JSC::ValueProfile::dump): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): 2011-11-10 David Kilzer <ddkilzer@apple.com> <http://webkit.org/b/72049> Specify testapi.js install path using JAVASCRIPTCORE_FRAMEWORKS_DIR Reviewed by Joseph Pecoraro. * JavaScriptCore.xcodeproj/project.pbxproj: The testapi.js script should use JAVASCRIPTCORE_FRAMEWORKS_DIR in its dstPath for installation. Also removed "Versions/A/" from the path since this is unneeded due the default symlinks present in the framework. 2011-11-10 Gavin Barraclough <barraclough@apple.com> Add ARMv7 support to the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=72061 Reviewed by Geoff Garen. * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::preserveReturnAddressAfterCall): (JSC::DFG::AssemblyHelpers::restoreReturnAddressBeforeReturn): (JSC::DFG::AssemblyHelpers::emitPutImmediateToCallFrameHeader): (JSC::DFG::AssemblyHelpers::boxDouble): (JSC::DFG::AssemblyHelpers::unboxDouble): - Add CPU(ARM) copies of these functions. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::spill): - Fix matching of '}' re #if blocks, makes some tools happy. (JSC::DFG::JITCodeGenerator::setupArguments): (JSC::DFG::JITCodeGenerator::setupArgumentsWithExecState): (JSC::DFG::JITCodeGenerator::appendCallWithExceptionCheckSetResult): (JSC::DFG::JITCodeGenerator::appendCallSetResult): - Add CPU(ARM) / 4 argument register copies of these functions. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): - Should use callOperation to plant a call to a DFG_OPERATION. (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::cachedPutById): - These methods need to plant a relinkable jump; we currently do so using beginUninterruptedSequence() / endUninterruptedSequence(). * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): - Should use callOperation to plant a call to a DFG_OPERATION. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::linkOSRExits): - This method needs to plant a relinkable jump; we currently do so using beginUninterruptedSequence() / endUninterruptedSequence(). (JSC::DFG::JITCompiler::compileBody): - Add abstraction to retrieve the pc after a call. * dfg/DFGOSRExitCompiler.cpp: - Fix a bug - CodeLocationLabel needs a data address rather than an executable one, but can just take a MacroAssemblerCodePtr instead! * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::compileClampDoubleToByte): - Add FIXME comment to come back to! - bug#72054. * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::speculationCheck): - Add missing method (ooops, required by bug#72047) * dfg/DFGSpeculativeJIT32_64.cpp: - Need to wrap fmod on ARMv7. 2011-11-10 Filip Pizlo <fpizlo@apple.com> DFG should not reparse code that was just parsed https://bugs.webkit.org/show_bug.cgi?id=71977 Reviewed by Geoff Garen. The instruction stream of a code block is now kept around until the next GC. When doing either an optimizing compilation of an executable, or inlining of an executable, we now try to find the already preexisting bytecode. If we find it, we don't have to parse. If we don't find it, we parse as before. Inlining takes the extra step of caching code blocks, so if the same executable gets inlined multiple times into the same caller, then we parse it at most once even if prior to inlining that executable did not have any code blocks with an instruction stream. Also fixed a silly bug where the strict mode for various operations was being determined by looking at the machine code block rather than the inlinee. To enable the delete-on-next-GC policy, I introduced the notion of an ultra weak finalizer, which anyone can register during tracing. This is thread-safe (for parallel GC) and stop-the-world-safe (so calls to free() are postponed until the world is resumed). This required reusing some facilities previously created for WeakReferenceHarvester, so I created a common utility class. I also retweaked the handling of WeakReferenceHarvesters, since they should be executed during stop-the-world since in the future we may want to allow them to call drain(). 2% win on SunSpider. 2% win on V8, when run in my harness. Neutral elsewhere. * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::CodeBlock): (JSC::CodeBlock::visitAggregate): (JSC::CodeBlock::copyPostParseDataFrom): (JSC::CodeBlock::copyPostParseDataFromAlternative): (JSC::CodeBlock::finalizeUnconditionally): * bytecode/CodeBlock.h: (JSC::CodeBlock::canProduceCopyWithBytecode): (JSC::CodeBlock::discardBytecodeLater): (JSC::CodeBlock::handleBytecodeDiscardingOpportunity): (JSC::GlobalCodeBlock::GlobalCodeBlock): (JSC::ProgramCodeBlock::ProgramCodeBlock): (JSC::EvalCodeBlock::EvalCodeBlock): (JSC::FunctionCodeBlock::FunctionCodeBlock): (JSC::BytecodeDestructionBlocker::BytecodeDestructionBlocker): (JSC::BytecodeDestructionBlocker::~BytecodeDestructionBlocker): * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::strictModeFor): * dfg/DFGByteCodeCache.h: Added. (JSC::DFG::CodeBlockKey::CodeBlockKey): (JSC::DFG::CodeBlockKey::operator==): (JSC::DFG::CodeBlockKey::hash): (JSC::DFG::CodeBlockKey::executable): (JSC::DFG::CodeBlockKey::kind): (JSC::DFG::CodeBlockKey::isHashTableDeletedValue): (JSC::DFG::CodeBlockKeyHash::hash): (JSC::DFG::CodeBlockKeyHash::equal): (JSC::DFG::ByteCodeCache::ByteCodeCache): (JSC::DFG::ByteCodeCache::~ByteCodeCache): (JSC::DFG::ByteCodeCache::get): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleInlining): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * heap/Heap.cpp: (JSC::Heap::finalizeUnconditionally): (JSC::Heap::markRoots): (JSC::Heap::collect): * heap/Heap.h: * heap/ListableHandler.h: Added. (JSC::ListableHandler::ListableHandler): (JSC::ListableHandler::~ListableHandler): (JSC::ListableHandler::List::List): (JSC::ListableHandler::List::addNotThreadSafe): (JSC::ListableHandler::List::addThreadSafe): (JSC::ListableHandler::List::hasNext): (JSC::ListableHandler::List::removeNext): * heap/MarkStack.cpp: (JSC::MarkStackThreadSharedData::MarkStackThreadSharedData): (JSC::SlotVisitor::harvestWeakReferences): (JSC::SlotVisitor::finalizeUnconditionally): * heap/MarkStack.h: (JSC::MarkStack::addWeakReferenceHarvester): (JSC::MarkStack::addUnconditionalFinalizer): * heap/SlotVisitor.h: * heap/UnconditionalFinalizer.h: Added. (JSC::UnconditionalFinalizer::~UnconditionalFinalizer): * heap/WeakReferenceHarvester.h: (JSC::WeakReferenceHarvester::WeakReferenceHarvester): (JSC::WeakReferenceHarvester::~WeakReferenceHarvester): * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::baselineCodeBlockFor): (JSC::FunctionExecutable::codeBlockWithBytecodeFor): (JSC::FunctionExecutable::produceCodeBlockFor): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): * runtime/Executable.h: (JSC::FunctionExecutable::profiledCodeBlockFor): 2011-11-10 Gavin Barraclough <barraclough@apple.com> Add ARMv7 register info for the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=72050 Reviewed by Geoff Garen. * dfg/DFGFPRInfo.h: (JSC::DFG::FPRInfo::toRegister): (JSC::DFG::FPRInfo::toIndex): (JSC::DFG::FPRInfo::debugName): * dfg/DFGGPRInfo.h: (JSC::DFG::GPRInfo::toRegister): (JSC::DFG::GPRInfo::toIndex): (JSC::DFG::GPRInfo::debugName): 2011-11-10 Gavin Barraclough <barraclough@apple.com> #ifdef CPU(X86) specific div/mod code in DFGSpeculativeJIT32_64 https://bugs.webkit.org/show_bug.cgi?id=72047 Reviewed by Geoff Garen. We currently don't attempt to abstract divide through the macro assembler, due to these instructions commonly having specific requirements. This means there is architecture specific code in the JIT - #ifdef it, and provide a common implementation. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::fmodAsDFGOperation): (JSC::DFG::SpeculativeJIT::compile): 2011-11-10 Gavin Barraclough <barraclough@apple.com> Add ENABLE_VALUE_PROFILER support for ARMv7 https://bugs.webkit.org/show_bug.cgi?id=72043 Reviewed by Geoff Garen. This requires us to make a bucketCounterRegister available; to do so we'll need to spill more registers on entry to JIT code. * jit/JITArithmetic32_64.cpp: (JSC::JIT::emitSlow_op_mod): - cleanup location of UNUSED_PARAM * jit/JITStubs.cpp: (JSC::ctiTrampoline): (JSC::ctiVMThrowTrampoline): (JSC::ctiOpThrowNotCaught): (JSC::JITThunks::JITThunks): * jit/JITStubs.h: - Update JITStackFrame structure & asm code to spill more registers. * jit/JSInterfaceJIT.h: - Assign a bucketCounterRegister. 2011-11-10 Gavin Barraclough <barraclough@apple.com> Fix sampling counters on ARMv7, move add64 functionality to macro assembler https://bugs.webkit.org/show_bug.cgi?id=72040 Reviewed by Geoff Garen. The ability to add an integer to a uint64_t in memory is poorly copied in multiple places & ifdef'ed on architecture, addWithCarry32 is also a badly designed interface since add32 is not required to set flags (we have no concept of flags in the macro assembler interface). * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::add64): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::add64): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::add64): * dfg/DFGAssemblyHelpers.cpp: * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::emitCount): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::writeBarrier): * jit/JIT.h: * jit/JITInlineMethods.h: (JSC::JIT::emitCount): 011-11-10 Ryuan Choi <ryuan.choi@samsung.com> [CMAKE] Refactoring CMakeLists${PORT}.txt to Platform${PORT}.cmake https://bugs.webkit.org/show_bug.cgi?id=56705 Reviewed by Adam Roben. * CMakeLists.txt: * PlatformEfl.cmake: Renamed from Source/JavaScriptCore/CMakeListsEfl.txt. * PlatformWinCE.cmake: Renamed from Source/JavaScriptCore/CMakeListsWinCE.txt. * shell/CMakeLists.txt: * shell/PlatformEfl.cmake: Renamed from Source/JavaScriptCore/shell/CMakeListsEfl.txt. * shell/PlatformWinCE.cmake: Renamed from Source/JavaScriptCore/shell/CMakeListsWinCE.txt. * wtf/CMakeLists.txt: * wtf/PlatformEfl.cmake: Renamed from Source/JavaScriptCore/wtf/CMakeListsEfl.txt. * wtf/PlatformWinCE.cmake: Renamed from Source/JavaScriptCore/wtf/CMakeListsWinCE.txt. 2011-11-10 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix make distcheck build. * GNUmakefile.list.am: Add missing files. 2011-11-09 Michael Saboff <msaboff@apple.com> Towards 8 Bit Strings: Templatize JSC::LiteralParser class by character type https://bugs.webkit.org/show_bug.cgi?id=71862 Changed LiteralParser to be templatized of character type. Moved five enums out of class definition to work around a clang compiler defect. Added lexIdentifier templated method to break out character specific versions. Added static setParserTokenString templated method to handle setting approriately sized string pointer. To keep code in LiteralParser.cpp and keep LiteralParser.h small, the two flavors of LiteralParser are explicitly instantiated at the end of LiteralParser.cpp. Reviewed by Oliver Hunt. * API/JSValueRef.cpp: (JSValueMakeFromJSONString): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::Interpreter::callEval): (JSC::Interpreter::execute): * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncEval): * runtime/JSONObject.cpp: (JSC::JSONProtoFuncParse): * runtime/LiteralParser.cpp: (JSC::isJSONWhiteSpace): (JSC::::tryJSONPParse): (JSC::::makeIdentifier): (JSC::::Lexer::lex): (JSC::::Lexer::lexIdentifier): (JSC::::Lexer::next): (JSC::LChar): (JSC::UChar): (JSC::isSafeStringCharacter): (JSC::::Lexer::lexString): (JSC::::Lexer::lexNumber): (JSC::::parse): * runtime/LiteralParser.h: (JSC::LiteralParser::LiteralParser): (JSC::LiteralParser::getErrorMessage): (JSC::LiteralParser::tryLiteralParse): (JSC::LiteralParser::Lexer::Lexer): (JSC::LiteralParser::Lexer::currentToken): (JSC::LiteralParser::Lexer::getErrorMessage): * runtime/UString.h: (JSC::LChar): (JSC::UChar): * wtf/text/StringBuilder.cpp: (WTF::StringBuilder::append): * wtf/text/StringBuilder.h: (WTF::StringBuilder::append): 2011-11-09 Filip Pizlo <fpizlo@apple.com> Multiple CodeBlock should be able to share the same instruction stream without copying https://bugs.webkit.org/show_bug.cgi?id=71978 Reviewed by Oliver Hunt. This refactors CodeBlock::m_instructions to be a Vector boxed in a ref-counted object, but otherwise does not take advantage of this. This is performance neutral. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::printStructure): (JSC::CodeBlock::printStructures): (JSC::CodeBlock::dump): (JSC::CodeBlock::CodeBlock): (JSC::CodeBlock::visitAggregate): (JSC::CodeBlock::shrinkToFit): * bytecode/CodeBlock.h: (JSC::CodeBlock::hasInstructions): (JSC::CodeBlock::numberOfInstructions): (JSC::CodeBlock::instructions): * jit/JIT.cpp: (JSC::JIT::JIT): 2011-11-09 Gavin Barraclough <barraclough@apple.com> Renovate ARMv7 assembler/macro-assembler https://bugs.webkit.org/show_bug.cgi?id=71982 Reviewed by Geoff Garen. ARMv7Assembler: * add support for strb (byte stores) * rename the VMOV_CtoS opcodes (there are currently backwards!) * add support for adc (add with carry) * add support for vsqrt, vabs * add support for vmov (between FPRs, and to/from GPR pairs). * remove '_F64' postfixes from instructions (these aren't helpful, functions can already be distinguished by their signatures). * rename vcvt_F64_S32 to vcvt_signedToFloatingPoint, the prior postfix was unhelpful in failing to distinguish the types (S32 indicates a single precision register, but the type could be float, int32, or uint32). * rename vcvtr_S32_F64 to vcvt_floatingPointToSigned, as for previous, also vcvtr was the incorrect name for the operation (the emitted instruction truncates). MacroAssemblerARMv7: * add 3-operand versions of and32, lshift32, or32, rshift32, urshift32, sub32, xor32, * add store8, and store32 imm to base-index. * fix load32WithCompactAddressOffsetPatch to work for all gprs (the fix is a little kludgy but functional; to do better we'll have to also fix the repatching code). * Update supportsFloating* flags (all features now supported). * add moveDouble, storeDouble to absolute address, addDouble to absolute address * add 3-operand double operations. * implement sqrtDouble/absDouble * add branchTruncateDoubleToInt32, implement truncateDoubleToInt32 * move should do nothing if src == dest * branchTest8-on-memory can be implemented in terms of branchTest32-on-register (branchTest8-on-register has been removed). * add 3-operand branchAdd32, branchSub32, also branchAdd32 absolute address. 2011-11-09 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=71873 Reviewed by Geoff Garen. Incrementally re-landing these changes, trying to determine what went wrong. (The whole patch failed tests on the build bot but worked locally. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleIntrinsic): 2011-11-09 Filip Pizlo <fpizlo@apple.com> DFG OSR exit code should be lazily generated https://bugs.webkit.org/show_bug.cgi?id=71744 Reviewed by Gavin Barraclough. The OSR exit code is now generated the first time it is executed, rather than right after speculative compilation. Because most OSR exits are never taken, this should greatly reduce both code size and compilation time. This is a 1% win on SunSpider, and a 1% win on V8 when running in my harness. No change in V8 in V8's harness (due to the long runs, so compile time is not an issue) and no change in Kraken (again, long runs of small code so compile time has no measurable effect). * CMakeListsEfl.txt: * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * assembler/AbstractMacroAssembler.h: * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::jump): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::jump): * assembler/X86Assembler.h: (JSC::X86Assembler::jmp_m): * bytecode/CodeBlock.h: (JSC::CodeBlock::createDFGDataIfNecessary): (JSC::CodeBlock::appendDFGOSREntryData): (JSC::CodeBlock::numberOfDFGOSREntries): (JSC::CodeBlock::dfgOSREntryData): (JSC::CodeBlock::dfgOSREntryDataForBytecodeIndex): (JSC::CodeBlock::appendOSRExit): (JSC::CodeBlock::appendSpeculationRecovery): (JSC::CodeBlock::numberOfOSRExits): (JSC::CodeBlock::numberOfSpeculationRecoveries): (JSC::CodeBlock::osrExit): (JSC::CodeBlock::speculationRecovery): * dfg/DFGAssemblyHelpers.h: (JSC::DFG::AssemblyHelpers::debugCall): * dfg/DFGCorrectableJumpPoint.cpp: Added. (JSC::DFG::CorrectableJumpPoint::codeLocationForRepatch): * dfg/DFGCorrectableJumpPoint.h: Added. (JSC::DFG::CorrectableJumpPoint::CorrectableJumpPoint): (JSC::DFG::CorrectableJumpPoint::switchToLateJump): (JSC::DFG::CorrectableJumpPoint::correctInitialJump): (JSC::DFG::CorrectableJumpPoint::correctLateJump): (JSC::DFG::CorrectableJumpPoint::initialJump): (JSC::DFG::CorrectableJumpPoint::lateJump): (JSC::DFG::CorrectableJumpPoint::correctJump): (JSC::DFG::CorrectableJumpPoint::getJump): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::linkOSRExits): (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: * dfg/DFGOSRExit.cpp: Added. (JSC::DFG::OSRExit::OSRExit): (JSC::DFG::OSRExit::dump): * dfg/DFGOSRExit.h: * dfg/DFGOSRExitCompiler.cpp: Added. * dfg/DFGOSRExitCompiler.h: * dfg/DFGOSRExitCompiler32_64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT.cpp: * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::speculationCheck): * dfg/DFGThunks.cpp: Added. (JSC::DFG::osrExitGenerationThunkGenerator): * dfg/DFGThunks.h: Added. * jit/JITCode.h: (JSC::JITCode::dataAddressAtOffset): * runtime/JSGlobalData.h: 2011-11-09 Mark Hahnenberg <mhahnenberg@apple.com> Fixing build breakage Unreviewed build fix * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-11-09 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSVariableObject::isDynamicScope https://bugs.webkit.org/show_bug.cgi?id=71933 Reviewed by Geoffrey Garen. * runtime/JSActivation.cpp: * runtime/JSActivation.h: Inlined and de-virtualized isDynamicScope (JSC::JSActivation::isDynamicScope): * runtime/JSGlobalObject.cpp: * runtime/JSGlobalObject.h: Inlined and de-virtualized isDynamicScope (JSC::JSGlobalObject::isDynamicScope): * runtime/JSStaticScopeObject.cpp: * runtime/JSStaticScopeObject.h: Inlined and de-virtualized isDynamicScope (JSC::JSStaticScopeObject::createStructure): Changed createStructure to use new JSType (JSC::JSStaticScopeObject::isDynamicScope): * runtime/JSType.h: Added new type for JSStaticScopeObject * runtime/JSVariableObject.cpp: De-virtualized and added an implementation that checks the object's type and calls the corresponding implementation. (JSC::JSVariableObject::isDynamicScope): * runtime/JSVariableObject.h: 2011-11-09 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSGlobalObject::hasOwnPropertyForWrite https://bugs.webkit.org/show_bug.cgi?id=71934 Reviewed by Geoffrey Garen. * runtime/JSGlobalObject.h: Removed the virtual-ness of hasOwnPropertyForWrite since nobody overrides it. 2011-11-09 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=71873 Reviewed by Geoff Garen. Incrementally re-landing these changes, trying to determine what went wrong. (The whole patch failed tests on the build bot but worked locally. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::absDouble): * assembler/MacroAssemblerARMv7.h: * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::absDouble): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::absDouble): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::absDouble): * assembler/MacroAssemblerX86Common.h: * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::absDouble): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/ThunkGenerators.cpp: (JSC::absThunkGenerator): 2011-11-09 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSObject::getOwnPropertyDescriptor https://bugs.webkit.org/show_bug.cgi?id=71523 Reviewed by Sam Weinig. Added getOwnPropertyDescriptor to the MethodTable, changed all of the virtual versions of getOwnPropertyDescriptor to static ones, and changed all of the call sites to the corresponding lookup in the MethodTable. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::getOwnPropertyDescriptor): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::getOwnPropertyDescriptor): * debugger/DebuggerActivation.h: * runtime/Arguments.cpp: (JSC::Arguments::getOwnPropertyDescriptor): * runtime/Arguments.h: * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::getOwnPropertyDescriptor): * runtime/ArrayConstructor.h: * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::getOwnPropertyDescriptor): * runtime/ArrayPrototype.h: * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::getOwnPropertyDescriptor): * runtime/BooleanPrototype.h: * runtime/ClassInfo.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::getOwnPropertyDescriptor): * runtime/DateConstructor.h: * runtime/DatePrototype.cpp: (JSC::DatePrototype::getOwnPropertyDescriptor): * runtime/DatePrototype.h: * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::getOwnPropertyDescriptor): * runtime/ErrorPrototype.h: * runtime/JSArray.cpp: (JSC::JSArray::getOwnPropertyDescriptor): * runtime/JSArray.h: * runtime/JSByteArray.cpp: (JSC::JSByteArray::getOwnPropertyDescriptor): * runtime/JSByteArray.h: * runtime/JSCell.cpp: (JSC::JSCell::getOwnPropertyDescriptor): * runtime/JSCell.h: * runtime/JSFunction.cpp: (JSC::JSFunction::getOwnPropertyDescriptor): * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::getOwnPropertyDescriptor): * runtime/JSGlobalObject.h: * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::getOwnPropertyDescriptor): * runtime/JSNotAnObject.h: * runtime/JSONObject.cpp: (JSC::JSONObject::getOwnPropertyDescriptor): * runtime/JSONObject.h: * runtime/JSObject.cpp: (JSC::JSObject::vtableAnchor): (JSC::JSObject::propertyIsEnumerable): (JSC::JSObject::getOwnPropertyDescriptor): (JSC::JSObject::getPropertyDescriptor): (JSC::JSObject::defineOwnProperty): * runtime/JSObject.h: * runtime/JSString.cpp: Removed getOwnPropertyDescriptor, since this seems to be a relic from a bygone era when getOwnPropertyDescriptor was rooted in JSCell rather than JSObject. There were no call sites for this version of getOwnPropertyDescriptor in the entire project. * runtime/JSString.h: * runtime/Lookup.h: (JSC::getStaticPropertyDescriptor): (JSC::getStaticFunctionDescriptor): (JSC::getStaticValueDescriptor): * runtime/MathObject.cpp: (JSC::MathObject::getOwnPropertyDescriptor): * runtime/MathObject.h: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::getOwnPropertyDescriptor): * runtime/NumberConstructor.h: * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::getOwnPropertyDescriptor): * runtime/NumberPrototype.h: * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::getOwnPropertyDescriptor): (JSC::objectConstructorGetOwnPropertyDescriptor): * runtime/ObjectConstructor.h: * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::getOwnPropertyDescriptor): * runtime/ObjectPrototype.h: * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::getOwnPropertyDescriptor): * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::getOwnPropertyDescriptor): * runtime/RegExpObject.cpp: (JSC::RegExpObject::getOwnPropertyDescriptor): * runtime/RegExpObject.h: * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::getOwnPropertyDescriptor): * runtime/RegExpPrototype.h: * runtime/StringConstructor.cpp: (JSC::StringConstructor::getOwnPropertyDescriptor): * runtime/StringConstructor.h: * runtime/StringObject.cpp: (JSC::StringObject::vtableAnchor): Added to prevent a weak vtable. (JSC::StringObject::getOwnPropertyDescriptor): * runtime/StringObject.h: * runtime/StringPrototype.cpp: (JSC::StringPrototype::getOwnPropertyDescriptor): * runtime/StringPrototype.h: 2011-11-09 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=71873 Reviewed by Geoff Garen. Incrementally re-landing these changes, trying to determine what went wrong. (The whole patch failed tests on the build bot but worked locally. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::supportsFloatingPoint): (JSC::MacroAssemblerARM::supportsFloatingPointTruncate): (JSC::MacroAssemblerARM::supportsFloatingPointSqrt): (JSC::MacroAssemblerARM::supportsFloatingPointAbs): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::supportsFloatingPoint): (JSC::MacroAssemblerARMv7::supportsFloatingPointTruncate): (JSC::MacroAssemblerARMv7::supportsFloatingPointSqrt): (JSC::MacroAssemblerARMv7::supportsFloatingPointAbs): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::supportsFloatingPoint): (JSC::MacroAssemblerMIPS::supportsFloatingPointTruncate): (JSC::MacroAssemblerMIPS::supportsFloatingPointSqrt): (JSC::MacroAssemblerMIPS::supportsFloatingPointAbs): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::supportsFloatingPoint): (JSC::MacroAssemblerSH4::supportsFloatingPointTruncate): (JSC::MacroAssemblerSH4::supportsFloatingPointSqrt): (JSC::MacroAssemblerSH4::supportsFloatingPointAbs): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::supportsFloatingPoint): (JSC::MacroAssemblerX86::supportsFloatingPointTruncate): (JSC::MacroAssemblerX86::supportsFloatingPointSqrt): (JSC::MacroAssemblerX86::supportsFloatingPointAbs): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::supportsFloatingPoint): (JSC::MacroAssemblerX86_64::supportsFloatingPointTruncate): (JSC::MacroAssemblerX86_64::supportsFloatingPointSqrt): (JSC::MacroAssemblerX86_64::supportsFloatingPointAbs): * jit/ThunkGenerators.cpp: (JSC::absThunkGenerator): 2011-11-08 Darin Adler <darin@apple.com> Add code path in HashTable for emptyValueIsZero that does not require copying the empty value https://bugs.webkit.org/show_bug.cgi?id=71875 Reviewed by Anders Carlsson. This is a step along the path of making OwnPtr work as HashMap value types. * wtf/Alignment.h: Moved the AlignedBufferChar and AlignedBuffer types from Vector.h here. Also fixed include style. To include other WTF headers inside WTF, we use "" includes. I did not change the code to fix style checker complaints. * wtf/HashTable.h: Added includes as needed and fixed include style. (WTF::doubleHash): Removed the uneeeded and inappropriate "static" in this function, which gave it internal linkage for no good reason. (WTF::HashTable::checkKey): Made this use AlignedBuffer for the deleted value check to avoid construction/destruction problems instead of doing the trick where we construct and destroy an empty value twice. It's cleaner and simpler and avoids copying the empty value. (WTF::HashTable::initializeBucket): Specialized initializeBucket to use memset when the empty value is zero rather than copying an empty value. * wtf/Vector.h: Moved the AlignedBufferChar and AlignedBuffer types into Alignment.h. 2011-11-09 Gabor Rapcsanyi <rgabor@webkit.org> Buildfix for 32bit debug mode. Reviewed by Csaba Osztrogonác. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::dump): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): 2011-11-09 Andy Wingo <wingo@igalia.com> Enable the DFG JIT on X86-64 Linux platforms https://bugs.webkit.org/show_bug.cgi?id=71373 Reviewed by Csaba Osztrogonác. * wtf/Platform.h (ENABLE_DFG_JIT): Enable the DFG JIT on the x86-64 GNU/Linux platform. * CMakeListsEfl.txt: Add JSValue64 implementations to EFL build. 2011-11-09 Csaba Osztrogonác <ossy@webkit.org> Enable the DFG JIT on x86-64 Linux platforms https://bugs.webkit.org/show_bug.cgi?id=71373 Enable DFG JIT by default on X86 Linux and Mac platforms https://bugs.webkit.org/show_bug.cgi?id=71686 Buildfix for stricter compilers: -Werror=unused-but-set-variable Reviewed by Zoltan Herczeg. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-09 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r99678. http://trac.webkit.org/changeset/99678 https://bugs.webkit.org/show_bug.cgi?id=71882 broke the build with -Werror=unused-but-set-variable (Requested by tronical_ on #webkit). * CMakeListsEfl.txt: * wtf/Platform.h: 2011-11-09 Andy Wingo <wingo@igalia.com> Enable the DFG JIT on X86-64 Linux platforms https://bugs.webkit.org/show_bug.cgi?id=71373 Reviewed by Filip Pizlo. * wtf/Platform.h (ENABLE_DFG_JIT): Enable the DFG JIT on the x86-64 GNU/Linux platform. * CMakeListsEfl.txt: Add JSValue64 implementations to EFL build. 2011-11-09 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSObject::defineOwnProperty https://bugs.webkit.org/show_bug.cgi?id=71429 Reviewed by Geoffrey Garen. Added defineOwnProperty to the MethodTable, changed all the virtual implementations of defineOwnProperty to static ones, and replaced all call sites with corresponding lookups in the MethodTable. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/Arguments.cpp: (JSC::Arguments::createStrictModeCallerIfNecessary): (JSC::Arguments::createStrictModeCalleeIfNecessary): * runtime/ClassInfo.h: * runtime/JSCell.cpp: (JSC::JSCell::defineOwnProperty): * runtime/JSCell.h: * runtime/JSObject.cpp: (JSC::JSObject::defineOwnProperty): * runtime/JSObject.h: * runtime/ObjectConstructor.cpp: (JSC::objectConstructorDefineProperty): (JSC::defineProperties): 2011-11-09 Simon Hausmann <simon.hausmann@nokia.com> [Qt] Build system cleanup https://bugs.webkit.org/show_bug.cgi?id=71815 Reviewed by Kenneth Rohde Christiansen. * wtf/wtf.pri: Moved the glib dependency to javascriptcore.prf. 2011-11-08 Simon Hausmann <simon.hausmann@nokia.com> [Qt] Replace use of QApplication with QGuiApplication https://bugs.webkit.org/show_bug.cgi?id=71794 Reviewed by Andreas Kling. Add compat headers for use when building with Qt 4: QGuiApplication is typedef'ed to QApplication. * wtf/qt/compat/QGuiApplication: Added. * wtf/qt/compat/qguiapplication.h: Added. 2011-11-08 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r99647. http://trac.webkit.org/changeset/99647 https://bugs.webkit.org/show_bug.cgi?id=71876 It broke jsc and layout tests on all bot (Requested by Ossy_night on #webkit). * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::supportsFloatingPoint): (JSC::MacroAssemblerARM::supportsFloatingPointTruncate): (JSC::MacroAssemblerARM::supportsFloatingPointSqrt): (JSC::MacroAssemblerARM::supportsDoubleBitops): (JSC::MacroAssemblerARM::andnotDouble): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::supportsFloatingPoint): (JSC::MacroAssemblerARMv7::supportsFloatingPointTruncate): (JSC::MacroAssemblerARMv7::supportsFloatingPointSqrt): (JSC::MacroAssemblerARMv7::supportsDoubleBitops): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::andnotDouble): (JSC::MacroAssemblerMIPS::supportsFloatingPoint): (JSC::MacroAssemblerMIPS::supportsFloatingPointTruncate): (JSC::MacroAssemblerMIPS::supportsFloatingPointSqrt): (JSC::MacroAssemblerMIPS::supportsDoubleBitops): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::supportsFloatingPoint): (JSC::MacroAssemblerSH4::supportsFloatingPointTruncate): (JSC::MacroAssemblerSH4::supportsFloatingPointSqrt): (JSC::MacroAssemblerSH4::supportsDoubleBitops): (JSC::MacroAssemblerSH4::andnotDouble): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::MacroAssemblerX86): (JSC::MacroAssemblerX86::supportsFloatingPoint): (JSC::MacroAssemblerX86::supportsFloatingPointTruncate): (JSC::MacroAssemblerX86::supportsFloatingPointSqrt): (JSC::MacroAssemblerX86::supportsDoubleBitops): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::andnotDouble): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::supportsFloatingPoint): (JSC::MacroAssemblerX86_64::supportsFloatingPointTruncate): (JSC::MacroAssemblerX86_64::supportsFloatingPointSqrt): (JSC::MacroAssemblerX86_64::supportsDoubleBitops): * assembler/X86Assembler.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleIntrinsic): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/ThunkGenerators.cpp: (JSC::absThunkGenerator): * runtime/JSGlobalData.cpp: 2011-11-08 Gavin Barraclough <barraclough@apple.com> Better abstract 'abs' operation through the MacroAssembler. https://bugs.webkit.org/show_bug.cgi?id=71873 Reviewed by Geoff Garen. Currently the x86 specific instruction sequence to perform a double abs is duplicated throughout the JITs / thunk generators. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::supportsFloatingPoint): (JSC::MacroAssemblerARM::supportsFloatingPointTruncate): (JSC::MacroAssemblerARM::supportsFloatingPointSqrt): (JSC::MacroAssemblerARM::supportsFloatingPointAbs): (JSC::MacroAssemblerARM::absDouble): - Renamed supportsFloatingPointAbs, make these methods static so that we can check the JIT's capabilites before we begin compilation. * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::supportsFloatingPoint): (JSC::MacroAssemblerARMv7::supportsFloatingPointTruncate): (JSC::MacroAssemblerARMv7::supportsFloatingPointSqrt): (JSC::MacroAssemblerARMv7::supportsFloatingPointAbs): - Renamed supportsFloatingPointAbs, make these methods static so that we can check the JIT's capabilites before we begin compilation. * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::absDouble): (JSC::MacroAssemblerMIPS::supportsFloatingPoint): (JSC::MacroAssemblerMIPS::supportsFloatingPointTruncate): (JSC::MacroAssemblerMIPS::supportsFloatingPointSqrt): (JSC::MacroAssemblerMIPS::supportsFloatingPointAbs): - Renamed supportsFloatingPointAbs, make these methods static so that we can check the JIT's capabilites before we begin compilation. * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::supportsFloatingPoint): (JSC::MacroAssemblerSH4::supportsFloatingPointTruncate): (JSC::MacroAssemblerSH4::supportsFloatingPointSqrt): (JSC::MacroAssemblerSH4::supportsFloatingPointAbs): (JSC::MacroAssemblerSH4::absDouble): - Renamed supportsFloatingPointAbs, make these methods static so that we can check the JIT's capabilites before we begin compilation. * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::absDouble): (JSC::MacroAssemblerX86::supportsFloatingPoint): (JSC::MacroAssemblerX86::supportsFloatingPointTruncate): (JSC::MacroAssemblerX86::supportsFloatingPointSqrt): (JSC::MacroAssemblerX86::supportsFloatingPointAbs): - Made supports* methods static so that we can check the JIT's capabilites before we begin compilation. Added absDouble. * assembler/MacroAssemblerX86Common.h: - Removed andnotDouble, added s_maskSignBit. * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::absDouble): (JSC::MacroAssemblerX86_64::supportsFloatingPoint): (JSC::MacroAssemblerX86_64::supportsFloatingPointTruncate): (JSC::MacroAssemblerX86_64::supportsFloatingPointSqrt): (JSC::MacroAssemblerX86_64::supportsFloatingPointAbs): - Made supports* methods static so that we can check the JIT's capabilites before we begin compilation. Added absDouble. * assembler/X86Assembler.h: (JSC::X86Assembler::andpd_rr): (JSC::X86Assembler::andpd_mr): - Added support for andpd instruction. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleIntrinsic): - Added checks for supportsFloatingPointAbs, supportsFloatingPointSqrt. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): - Switched to use doubleAbs, we can now also reuse the operand register for the result. * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): - Switched to use doubleAbs, we can now also reuse the operand register for the result. * jit/ThunkGenerators.cpp: - Switched to use doubleAbs. (JSC::absThunkGenerator): * runtime/JSGlobalData.cpp: - Declared MacroAssemblerX86Common::s_maskSignBit here. This is a little ugly, but it doesn't seem worth adding a whole extra .cpp to the compile for just one constant. 2011-11-08 Gavin Barraclough <barraclough@apple.com> Move duplicates of SYMBOL_STRING* macros to the single location https://bugs.webkit.org/show_bug.cgi?id=71456 Reviewed by Sam Weinig. * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGOperations.cpp: * jit/JITStubs.cpp: * wtf/InlineASM.h: Added. - Moved asm related macros. 2011-11-08 Gavin Barraclough <barraclough@apple.com> Move code to handle 8bit regs from X86Assembler to MacroAssembler https://bugs.webkit.org/show_bug.cgi?id=71867 Reviewed by Oliver Hunt. This code is fine, but is in the wrong place really. X86 assembler should basically just format up exactly the instruction you request - not expand out to a set of instructions (that is what the macro assembler layer is for!). For other 8-bit ops, on X86 we don't guard against clients accessing the XH registers. * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::store8): * assembler/X86Assembler.h: (JSC::X86Assembler::movb_rm): - moved some code. 2011-11-08 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix for GTK. * GNUmakefile.list.am: 2011-11-08 Gavin Barraclough <barraclough@apple.com> Build fix. * assembler/X86Assembler.h: 2011-11-08 Gavin Barraclough <barraclough@apple.com> Errrk, failed to commit this in last change. * assembler/X86Assembler.h: 2011-11-08 Gavin Barraclough <barraclough@apple.com> Remove an unused method. Rubber stamped by Geoff Garen. * assembler/AbstractMacroAssembler.h: * assembler/AssemblerBuffer.h: - removed rewindToLabel. 2011-11-08 Gavin Barraclough <barraclough@apple.com> Fix OSR entry points to calculate offsets correctly WRT to branch compaction. https://bugs.webkit.org/show_bug.cgi?id=71864 Reviewed by Filip Pizlo. * assembler/LinkBuffer.h: (JSC::LinkBuffer::offsetOf): - We use this to return the offsets into the code of the entry points. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileEntry): (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::compile): (JSC::DFG::JITCompiler::compileFunction): - Move the construction of the speculative JIT outside of compileBody, such that it is still available to link the OSR entry points at the point we are linking. * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::noticeOSREntry): - Pass the label of the block & linkbuffer into noticeOSREntry. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::linkOSREntries): - Moved call to noticeOSREntry until we we linking. * dfg/DFGSpeculativeJIT.h: * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompileSlowCases): (JSC::JIT::privateCompile): - Moved calculation of entries until we we linking. * jit/JIT.h: - Removed some members. 2011-11-08 Filip Pizlo <fpizlo@apple.com> DFG OSR exit code should be generated by a separate compiler, not related to DFG::JITCompiler https://bugs.webkit.org/show_bug.cgi?id=71787 Reviewed by Gavin Barraclough. Moves the exitSpeculativeWithOSR() method from JITCompiler to OSRExitCompiler::compileExit(). * CMakeListsEfl.txt: * JavaScriptCore.xcodeproj/project.pbxproj: * Target.pri: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::linkOSRExits): * dfg/DFGJITCompiler32_64.cpp: Removed. * dfg/DFGOSRExitCompiler.h: Added. (JSC::DFG::OSRExitCompiler::OSRExitCompiler): * dfg/DFGOSRExitCompiler32_64.cpp: Added. (JSC::DFG::OSRExitCompiler::compileExit): * dfg/DFGOSRExitCompiler64.cpp: Added. (JSC::DFG::OSRExitCompiler::compileExit): * runtime/JSValue.h: 2011-11-08 Filip Pizlo <fpizlo@apple.com> Basic DFG definitions should be moved out of DFGNode.h https://bugs.webkit.org/show_bug.cgi?id=71861 Rubber-stamped by Gavin Barraclough. * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGCommon.h: Added. (JSC::DFG::NodeIndexTraits::defaultValue): * dfg/DFGNode.h: * dfg/DFGOSRExit.h: * dfg/DFGRegisterBank.h: 2011-11-08 Michael Saboff <msaboff@apple.com> Towards 8 Bit Strings: Templatize JSC::Parser class by Lexer type https://bugs.webkit.org/show_bug.cgi?id=71761 Templatized Parser based on Lexer<T>. Moved two enums, SourceElementsMode and FunctionRequirements out of Parser definition to work around a clang compiler defect. Cleaned up SourceCode data() to return StringImpl* and eliminated the recently added stringData() virtual method. To keep code in Parser.cpp and keep Parser.h small, the two flavors of Parser are explicitly instantiated at the end of Parser.cpp. Reviewed by Gavin Barraclough. * interpreter/Interpreter.cpp: (JSC::appendSourceToError): * parser/Lexer.cpp: (JSC::::setCode): (JSC::::sourceCode): * parser/Parser.cpp: (JSC::::Parser): (JSC::::~Parser): (JSC::::parseInner): (JSC::::didFinishParsing): (JSC::::allowAutomaticSemicolon): (JSC::::parseSourceElements): (JSC::::parseVarDeclaration): (JSC::::parseConstDeclaration): (JSC::::parseDoWhileStatement): (JSC::::parseWhileStatement): (JSC::::parseVarDeclarationList): (JSC::::parseConstDeclarationList): (JSC::::parseForStatement): (JSC::::parseBreakStatement): (JSC::::parseContinueStatement): (JSC::::parseReturnStatement): (JSC::::parseThrowStatement): (JSC::::parseWithStatement): (JSC::::parseSwitchStatement): (JSC::::parseSwitchClauses): (JSC::::parseSwitchDefaultClause): (JSC::::parseTryStatement): (JSC::::parseDebuggerStatement): (JSC::::parseBlockStatement): (JSC::::parseStatement): (JSC::::parseFormalParameters): (JSC::::parseFunctionBody): (JSC::::parseFunctionInfo): (JSC::::parseFunctionDeclaration): (JSC::::parseExpressionOrLabelStatement): (JSC::::parseExpressionStatement): (JSC::::parseIfStatement): (JSC::::parseExpression): (JSC::::parseAssignmentExpression): (JSC::::parseConditionalExpression): (JSC::::isBinaryOperator): (JSC::::parseBinaryExpression): (JSC::::parseProperty): (JSC::::parseObjectLiteral): (JSC::::parseStrictObjectLiteral): (JSC::::parseArrayLiteral): (JSC::::parsePrimaryExpression): (JSC::::parseArguments): (JSC::::parseMemberExpression): (JSC::::parseUnaryExpression): * parser/Parser.h: (JSC::::parse): (JSC::parse): * parser/SourceCode.h: (JSC::SourceCode::data): (JSC::SourceCode::subExpression): * parser/SourceProvider.h: (JSC::UStringSourceProvider::data): 2011-11-08 Gavin Barraclough <barraclough@apple.com> Fix PropertyAccessRecords in DFG JIT to take account of branch compaction. https://bugs.webkit.org/show_bug.cgi?id=71855 Reviewed by Filip Pizlo. The DFG JIT presently calculates a set of offsets early, before branches have been compacted. This won't work on ARMv7. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::differenceBetweenCodePtr): * assembler/LinkBuffer.h: (JSC::LinkBuffer::locationOf): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: (JSC::DFG::PropertyAccessRecord::PropertyAccessRecord): (JSC::DFG::JITCompiler::addPropertyAccess): 2011-11-08 Gavin Barraclough <barraclough@apple.com> DFG JIT calculation of OSR entry points is not THUMB2 safe https://bugs.webkit.org/show_bug.cgi?id=71852 Reviewed by Oliver Hunt. Executable addresses are tagged with a low bit set to distinguish between THUMB2 and traditional ARM. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGOSREntry.cpp: (JSC::DFG::prepareOSREntry): * jit/JITCode.h: (JSC::JITCode::executableAddressAtOffset): (JSC::JITCode::start): (JSC::JITCode::size): 2011-11-08 Michael Saboff <msaboff@apple.com> JSC::Parser::Parser leaks Lexer member https://bugs.webkit.org/show_bug.cgi?id=71847 Changed m_lexer member of Parser to be OwnPtr to fix a memory leak. Reviewed by Oliver Hunt. * parser/Parser.cpp: (JSC::Parser::Parser): (JSC::Parser::parseFunctionBody): * parser/Parser.h: 2011-11-08 Yuqiang Xian <yuqiang.xian@intel.com> Enable DFG JIT by default on X86 Linux and Mac platforms https://bugs.webkit.org/show_bug.cgi?id=71686 Reviewed by Filip Pizlo. We can get 9% on SunSpider, 89% on Kraken and 37% on V8, on Linux X86. * wtf/Platform.h: 2011-11-08 Yuqiang Xian <yuqiang.xian@intel.com> DFG 32_64 - update make lists for efl, gtk, and Qt ports with DFG change r99519 https://bugs.webkit.org/show_bug.cgi?id=71768 Reviewed by Geoffrey Garen. Also includes a fix to make the newly introduced AssemblyHelpers friend of JSValue as we need the Tag definitions. * CMakeListsEfl.txt: * GNUmakefile.list.am: * Target.pri: * runtime/JSValue.h: 2011-11-07 Yuqiang Xian <yuqiang.xian@intel.com> Fix gcc 4.4 compilation warnings in DFG 32_64 https://bugs.webkit.org/show_bug.cgi?id=71762 Reviewed by Filip Pizlo. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::registersMatched): 2011-11-07 Filip Pizlo <fpizlo@apple.com> DFG code base should allow for classes not related to DFG::JITCompiler to use DFG idioms https://bugs.webkit.org/show_bug.cgi?id=71746 Reviewed by Gavin Barraclough. * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGAssemblyHelpers.cpp: Added. (JSC::DFG::AssemblyHelpers::decodedCodeMapFor): (JSC::DFG::AssemblyHelpers::emitCount): (JSC::DFG::AssemblyHelpers::setSamplingFlag): (JSC::DFG::AssemblyHelpers::clearSamplingFlag): (JSC::DFG::AssemblyHelpers::jitAssertIsInt32): (JSC::DFG::AssemblyHelpers::jitAssertIsJSInt32): (JSC::DFG::AssemblyHelpers::jitAssertIsJSNumber): (JSC::DFG::AssemblyHelpers::jitAssertIsJSDouble): (JSC::DFG::AssemblyHelpers::jitAssertIsCell): * dfg/DFGAssemblyHelpers.h: Added. * dfg/DFGJITCompiler.cpp: * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::JITCompiler): (JSC::DFG::JITCompiler::graph): * dfg/DFGJITCompiler32_64.cpp: * dfg/DFGOSRExit.h: Added. (JSC::DFG::SpeculationRecovery::SpeculationRecovery): (JSC::DFG::SpeculationRecovery::type): (JSC::DFG::SpeculationRecovery::dest): (JSC::DFG::SpeculationRecovery::src): (JSC::DFG::OSRExit::numberOfRecoveries): (JSC::DFG::OSRExit::valueRecovery): (JSC::DFG::OSRExit::isArgument): (JSC::DFG::OSRExit::isVariable): (JSC::DFG::OSRExit::argumentForIndex): (JSC::DFG::OSRExit::variableForIndex): (JSC::DFG::OSRExit::operandForArgument): (JSC::DFG::OSRExit::operandForIndex): * dfg/DFGSpeculativeJIT.h: 2011-11-07 Filip Pizlo <fpizlo@apple.com> Switch back to 1+1 value profiling buckets, since it didn't help on arewefastyet, but it appears to help on other benchmarks. Rubber stamped by Oliver Hunt. * bytecode/ValueProfile.h: 2011-11-07 Ariya Hidayat <ariya@sencha.com> "use strict" can not contain escape sequences or line continuation https://bugs.webkit.org/show_bug.cgi?id=71532 Reviewed by Darin Adler. Store the actual literal length (before the escapes and line continuation are encoded) while parsing the directive and use it for the directive comparison. * parser/Parser.cpp: (JSC::Parser::parseSourceElements): (JSC::Parser::parseStatement): * parser/Parser.h: 2011-11-06 Filip Pizlo <fpizlo@apple.com> DFG operationCreateThis slow path may get the wrong callee in case of inlining https://bugs.webkit.org/show_bug.cgi?id=71647 Reviewed by Oliver Hunt. No new tests because I only saw this manifest itself when I had other bugs leading to spurious slow path executions. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-07 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSObject::putWithAttributes https://bugs.webkit.org/show_bug.cgi?id=71716 Reviewed by Darin Adler. Added putWithAttributes to the MethodTable, changed all the virtual implementations of putWithAttributes to static ones, and replaced all call sites with corresponding lookups in the MethodTable. * API/JSObjectRef.cpp: (JSObjectSetProperty): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::putWithAttributes): * debugger/DebuggerActivation.h: * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): * runtime/ClassInfo.h: * runtime/JSActivation.cpp: (JSC::JSActivation::putWithAttributes): * runtime/JSActivation.h: * runtime/JSCell.cpp: (JSC::JSCell::putWithAttributes): * runtime/JSCell.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::putWithAttributes): * runtime/JSGlobalObject.h: * runtime/JSObject.cpp: (JSC::JSObject::putWithAttributes): (JSC::putDescriptor): * runtime/JSObject.h: * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::putWithAttributes): * runtime/JSStaticScopeObject.h: * runtime/JSVariableObject.cpp: (JSC::JSVariableObject::putWithAttributes): * runtime/JSVariableObject.h: 2011-11-07 Dmitry Lomov <dslomov@google.com> Unreviewed. Release build fix. * parser/Lexer.cpp: (JSC::assertCharIsIn8BitRange): 2011-11-07 Filip Pizlo <fpizlo@apple.com> Switch the value profiler back to 8 buckets, because we suspect that while this is more expensive it's also more stable. Rubber stamped by Geoff Garen. * bytecode/ValueProfile.h: 2011-11-07 Andrew Wason <rectalogic@rectalogic.com> Uninitialized Heap member var https://bugs.webkit.org/show_bug.cgi?id=71722 Reviewed by Filip Pizlo. * heap/Heap.cpp: (JSC::Heap::Heap): Initialize m_blockFreeingThreadShouldQuit 2011-11-07 Yuqiang Xian <yuqiang.xian@intel.com> DFG 32_64 - registers cannot be reused arbitrarily if speculation failures are possible https://bugs.webkit.org/show_bug.cgi?id=71684 Reviewed by Filip Pizlo. Currently in DFG JIT, we try to reuse the physical register of an operand for temporary usage if the current use of the operand is the last use. But sometimes this can be wrong, for example if there are possible speculation failures and we need to fallback to baseline JIT, the value of the operand which is supposed to be hold in the physical register can be modified by register reusing. The fixes the last inspector failures in layout test on Mac 32-bit if switching on DFG. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::compile): 2011-11-07 Ryosuke Niwa <rniwa@webkit.org> REGRESSION(r99436): Broke Snow Leopard debug build https://bugs.webkit.org/show_bug.cgi?id=71713 Reviewed by Darin Adler. Put the assertion in a template and use template specialization to avoid warning when instantiated with UChar or LChar. In the long term, we should have traits for unsigned integral types and use that to specialize template instead of specializing it for UChar and LChar. * parser/Lexer.cpp: (JSC::assertCharIsIn8BitRange): (JSC::::append8): 2011-11-07 ChangSeok Oh <shivamidow@gmail.com> [EFL] Support requestAnimationFrame API https://bugs.webkit.org/show_bug.cgi?id=67112 Reviewed by Andreas Kling. Let EFL port use REQUEST_ANIMATION_FRAME_TIMER. * wtf/Platform.h: 2011-11-07 Michael Saboff <msaboff@apple.com> Towards 8 Bit Strings: Templatize JSC::Lexer class by character type https://bugs.webkit.org/show_bug.cgi?id=71331 Change the Lexer class to be a template class based on the character type of the source. In the process updated the parseIdentifier() and parseString() methods to create 8 bit strings where possible. Also added some helper methods for accumulating temporary string data in the 8 and 16 bit vectors. Changed the SourceProvider::data() virtual method to return a StringImpl* instead of a UChar*. Updated the KeywordLookup generator to create code to match keywords for both 8 and 16 bit source strings. Due to a compiler bug (<rdar://problem/10194295>) moved enum definition outside of Lexer class declaration. Remove second enum no longer needed. Reviewed by Darin Adler. * KeywordLookupGenerator.py: * interpreter/Interpreter.cpp: (JSC::Interpreter::callEval): * parser/Lexer.cpp: (JSC::::Lexer): (JSC::::~Lexer): (JSC::::getInvalidCharMessage): (JSC::::currentCharacter): (JSC::::setCode): (JSC::::internalShift): (JSC::::shift): (JSC::::peek): (JSC::::getUnicodeCharacter): (JSC::::shiftLineTerminator): (JSC::::lastTokenWasRestrKeyword): (JSC::::record8): (JSC::::append8): (JSC::::append16): (JSC::::record16): (JSC::::parseIdentifier): (JSC::::parseIdentifierSlowCase): (JSC::::parseString): (JSC::::parseStringSlowCase): (JSC::::parseHex): (JSC::::parseOctal): (JSC::::parseDecimal): (JSC::::parseNumberAfterDecimalPoint): (JSC::::parseNumberAfterExponentIndicator): (JSC::::parseMultilineComment): (JSC::::nextTokenIsColon): (JSC::::lex): (JSC::::scanRegExp): (JSC::::skipRegExp): (JSC::::clear): (JSC::::sourceCode): * parser/Lexer.h: (JSC::Lexer::append16): (JSC::Lexer::currentOffset): (JSC::Lexer::setOffsetFromCharOffset): (JSC::::isWhiteSpace): (JSC::::isLineTerminator): (JSC::::convertHex): (JSC::::convertUnicode): (JSC::::makeIdentifier): (JSC::::setCodeStart): (JSC::::makeIdentifierLCharFromUChar): (JSC::::lexExpectIdentifier): * parser/Parser.cpp: (JSC::Parser::Parser): (JSC::Parser::parseProperty): (JSC::Parser::parseMemberExpression): * parser/Parser.h: (JSC::Parser::next): (JSC::Parser::nextExpectIdentifier): * parser/ParserArena.h: (JSC::IdentifierArena::makeIdentifier): (JSC::IdentifierArena::makeIdentifierLCharFromUChar): * parser/SourceCode.h: (JSC::SourceCode::subExpression): * parser/SourceProvider.h: (JSC::UStringSourceProvider::stringData): * parser/SourceProviderCache.h: * parser/SyntaxChecker.h: * runtime/FunctionPrototype.cpp: (JSC::insertSemicolonIfNeeded): * runtime/Identifier.cpp: (JSC::IdentifierTable::add): (JSC::IdentifierLCharFromUCharTranslator::hash): (JSC::IdentifierLCharFromUCharTranslator::equal): (JSC::IdentifierLCharFromUCharTranslator::translate): (JSC::Identifier::add8): * runtime/Identifier.h: (JSC::Identifier::Identifier): (JSC::Identifier::createLCharFromUChar): (JSC::Identifier::canUseSingleCharacterString): (JSC::IdentifierCharBufferTranslator::hash): (JSC::IdentifierCharBufferTranslator::equal): (JSC::IdentifierCharBufferTranslator::translate): (JSC::Identifier::add): (JSC::Identifier::equal): (JSC::IdentifierTable::add): * runtime/JSGlobalObjectFunctions.cpp: (JSC::decode): (JSC::parseIntOverflow): (JSC::globalFuncUnescape): * runtime/JSGlobalObjectFunctions.h: (JSC::parseIntOverflow): * runtime/LiteralParser.cpp: (JSC::LiteralParser::tryJSONPParse): (JSC::LiteralParser::Lexer::lexString): * wtf/text/StringImpl.h: 2011-11-07 Tor Arne Vestbø <tor.arne.vestbo@nokia.com> [Qt] Put the jsc binary in 'bin' instead of leaving it deep in the build tree Allows us to not package up the whole Source/JavaScriptCore directory for the buildbots. Reviewed-by Simon Hausmann. * jsc.pro: 2011-11-06 Filip Pizlo <fpizlo@apple.com> REGRESSION(r99374): GTK+ build of the jsc binary doesn't like the call to initializeMainThread, and crashes https://bugs.webkit.org/show_bug.cgi?id=71643 Reviewed by Sam Weinig. * jsc.cpp: (main): 2011-11-06 Sam Weinig <sam@webkit.org> Add space missing from some class declarations https://bugs.webkit.org/show_bug.cgi?id=71632 Reviewed by Anders Carlsson. * assembler/AssemblerBufferWithConstantPool.h: * bytecode/CodeBlock.h: * dfg/DFGVariableAccessData.h: * heap/VTableSpectrum.h: * jit/ExecutableAllocator.cpp: * jit/ExecutableAllocatorFixedVMPool.cpp: * wtf/MetaAllocatorHandle.h: * wtf/UnionFind.h: 2011-11-06 Sam Weinig <sam@webkit.org> Allow use of FINAL in JavaScriptCore https://bugs.webkit.org/show_bug.cgi?id=71630 Reviewed by Anders Carlsson. * Configurations/Base.xcconfig: Don't warn about C++11 extensions used in C++98 mode. 2011-11-05 Filip Pizlo <fpizlo@apple.com> Value profiling should just use two buckets https://bugs.webkit.org/show_bug.cgi?id=71619 Reviewed by Gavin Barraclough. Added one more configuration options (like Heuristics::minimumOptimizationDelay), improved debugging in JIT optimization support, changed the number of buckets in the value profile from 9 to 2, and wrote a more optimal value profiling path in the old JIT to take advantage of this. It's still possible to play around with larger numbers of buckets, and we should probably keep this for a little while until we convince ourselves that using just two buckets is the right call. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::shouldOptimizeNow): * bytecode/ValueProfile.h: * jit/JITInlineMethods.h: (JSC::JIT::emitValueProfilingSite): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Heuristics.cpp: (JSC::Heuristics::initializeHeuristics): * runtime/Heuristics.h: 2011-11-03 Filip Pizlo <fpizlo@apple.com> JSC should be able to sample itself in a more flexible way than just sampling flags https://bugs.webkit.org/show_bug.cgi?id=71522 Reviewed by Gavin Barraclough. Added a construct that looks like SamplingRegion samplingRegion("name"). * JavaScriptCore.exp: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/SamplingTool.cpp: (JSC::SamplingRegion::Locker::Locker): (JSC::SamplingRegion::Locker::~Locker): (JSC::SamplingRegion::sample): (JSC::SamplingRegion::dump): (JSC::SamplingRegion::dumpInternal): (JSC::SamplingThread::threadStartFunc): * bytecode/SamplingTool.h: (JSC::SamplingRegion::SamplingRegion): (JSC::SamplingRegion::~SamplingRegion): (JSC::SamplingRegion::exchangeCurrent): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::generate): * dfg/DFGDriver.cpp: (JSC::DFG::compile): * heap/Heap.cpp: (JSC::Heap::markRoots): (JSC::Heap::collect): * heap/VTableSpectrum.cpp: (JSC::VTableSpectrum::countVPtr): (JSC::VTableSpectrum::dump): * heap/VTableSpectrum.h: * jsc.cpp: (main): (runWithScripts): * parser/Parser.h: (JSC::parse): * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): * wtf/Atomics.h: (WTF::weakCompareAndSwap): * wtf/Platform.h: * wtf/Spectrum.h: Added. (WTF::Spectrum::Spectrum): (WTF::Spectrum::add): (WTF::Spectrum::get): (WTF::Spectrum::begin): (WTF::Spectrum::end): (WTF::Spectrum::KeyAndCount::KeyAndCount): (WTF::Spectrum::KeyAndCount::operator<): (WTF::Spectrum::buildList): * wtf/wtf.pri: 2011-11-05 Sam Weinig <sam@webkit.org> Fix windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-11-04 Sam Weinig <sam@webkit.org> Reduce the number of putWithAttributes https://bugs.webkit.org/show_bug.cgi?id=71597 Reviewed by Adam Roben. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Remove exports of removed functions. * runtime/JSActivation.cpp: (JSC::JSActivation::putWithAttributes): Calling the overload without the extra parameters does the same thing. * runtime/JSObject.cpp: (JSC::JSObject::putWithAttributes): * runtime/JSObject.h: Remove four unused JSObject::putWithAttributes overloads and make one of the remaining two overloads not virtual, since no one overrides it. 2011-11-04 Pratik Solanki <psolanki@apple.com> sqrtDouble and andnotDouble should be declared noreturn https://bugs.webkit.org/show_bug.cgi?id=71592 Reviewed by Sam Weinig. * assembler/MacroAssemblerARMv7.h: 2011-11-04 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSObject::hasInstance https://bugs.webkit.org/show_bug.cgi?id=71430 Reviewed by Darin Adler. Added hasInstance to the MethodTable, changed all the virtual implementations of hasInstance to static ones, and replaced all call sites with corresponding lookups in the MethodTable. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::hasInstance): * API/JSValueRef.cpp: (JSValueIsInstanceOfConstructor): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ClassInfo.h: * runtime/JSBoundFunction.cpp: (JSC::JSBoundFunction::hasInstance): * runtime/JSBoundFunction.h: * runtime/JSCell.cpp: (JSC::JSCell::hasInstance): * runtime/JSCell.h: * runtime/JSObject.cpp: (JSC::JSObject::hasInstance): * runtime/JSObject.h: 2011-11-04 Tor Arne Vestbø <tor.arne.vestbo@nokia.com> [Qt] Refactor and clean up the qmake build system The qmake build system has accumulated a bit of cruft and redundancy over time. There's also a fairly tight coupling between how to build the various targets, and _what_ to build, making it harder to add new rules or sources. This patch aims to elevate these issues somewhat. This is a short-list of the changes: * The rules for how to build targets are now mostly contained as prf-files in Tools/qmake/mkspecs/features. Using mkspecs also allows us to do pre- and post-processing of each project file, which helps to clean up the actual project files. * Derived sources are no longer generated as a separate make-step but is part of each target's project file as a subdir. Makefile rules are used to ensure that we run make on the derived sources before running qmake on the actual target makefile. This makes it easier to keep a proper dependency between derived sources and the target. * We use GNU make and the compiler to generate dependencies on UNIX-based systems running Qt 5. This allows us to lessen the need to run qmake, which should reduce compile time. * WebKit2 is now build by default if building with Qt 5. It can be disabled by passing --no-webkit2 to build-webkit. The result of these changes are hopefully a cleaner and easier build system to modify, and faster build times due to no longer running qmake on every single build. It's also a first step towards possibly generating the list of sources using another build system. https://bugs.webkit.org/show_bug.cgi?id=71222 Reviewed by Simon Hausmann. * DerivedSources.pri: Added. * DerivedSources.pro: Removed. * JavaScriptCore.pro: * Target.pri: Copied from Source/JavaScriptCore/JavaScriptCore.pro. * headers.pri: Removed. * jsc.pro: * wtf/wtf.pri: * yarr/yarr.pri: 2011-11-04 Yuqiang Xian <yuqiang.xian@intel.com> More code clean-up in DFG 32_64 https://bugs.webkit.org/show_bug.cgi?id=71540 Remove unnecessary code duplications, and fix compilation warnings. Reviewed by Gavin Barraclough. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::emitCount): (JSC::DFG::JITCompiler::setSamplingFlag): (JSC::DFG::JITCompiler::clearSamplingFlag): (JSC::DFG::JITCompiler::jitAssertIsCell): * dfg/DFGJITCompiler32_64.cpp: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-04 Csaba Osztrogonác <ossy@webkit.org> De-virtualize JSObject::hasInstance https://bugs.webkit.org/show_bug.cgi?id=71430 Unreviewed rolling out r99238, because it made a test crash on all platform. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::hasInstance): * API/JSValueRef.cpp: (JSValueIsInstanceOfConstructor): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ClassInfo.h: * runtime/JSBoundFunction.cpp: (JSC::JSBoundFunction::hasInstance): * runtime/JSBoundFunction.h: * runtime/JSCell.cpp: * runtime/JSCell.h: * runtime/JSObject.cpp: (JSC::JSObject::hasInstance): * runtime/JSObject.h: 2011-11-03 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSObject::getPropertyNames https://bugs.webkit.org/show_bug.cgi?id=71306 Reviewed by Darin Adler. Added getPropertyNames to the MethodTable, changed all the virtual implementations of getPropertyNames to static ones, and replaced all call sites with corresponding lookups in the MethodTable. * API/JSObjectRef.cpp: (JSObjectCopyPropertyNames): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::getOwnPropertyNames): * runtime/ClassInfo.h: * runtime/JSCell.cpp: (JSC::JSCell::getPropertyNames): * runtime/JSCell.h: * runtime/JSObject.cpp: (JSC::JSObject::getPropertyNames): (JSC::JSObject::getOwnPropertyNames): * runtime/JSObject.h: * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::create): * runtime/ScopeChain.cpp: (JSC::ScopeChainNode::print): * runtime/Structure.cpp: (JSC::Structure::getPropertyNamesFromStructure): * runtime/Structure.h: 2011-11-03 Darin Adler <darin@apple.com> Change remaining callers of releaseRef to call leakRef https://bugs.webkit.org/show_bug.cgi?id=71422 * wtf/text/AtomicString.cpp: (WTF::HashAndUTF8CharactersTranslator::translate): Use leakRef. 2011-11-02 Darin Adler <darin@apple.com> Change remaining callers of releaseRef to call leakRef https://bugs.webkit.org/show_bug.cgi?id=71422 * wtf/text/AtomicString.cpp: (WTF::HashAndUTF8CharactersTranslator::translate): Use leakRef. 2011-11-03 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSObject::hasInstance https://bugs.webkit.org/show_bug.cgi?id=71430 Reviewed by Darin Adler. Added hasInstance to the MethodTable, changed all the virtual implementations of hasInstance to static ones, and replaced all call sites with corresponding lookups in the MethodTable. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::hasInstance): * API/JSValueRef.cpp: (JSValueIsInstanceOfConstructor): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ClassInfo.h: * runtime/JSBoundFunction.cpp: (JSC::JSBoundFunction::hasInstance): * runtime/JSBoundFunction.h: * runtime/JSCell.cpp: (JSC::JSCell::hasInstance): * runtime/JSCell.h: * runtime/JSObject.cpp: (JSC::JSObject::hasInstance): * runtime/JSObject.h: 2011-11-03 Filip Pizlo <fpizlo@apple.com> JIT-specific code should be able to refer to register types even on JIT-disabled builds https://bugs.webkit.org/show_bug.cgi?id=71498 Reviewed by Gavin Barraclough. * assembler/MacroAssembler.h: (MacroAssembler::MacroAssembler): 2011-11-03 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSObject::className https://bugs.webkit.org/show_bug.cgi?id=71428 Reviewed by Sam Weinig. Added className to the MethodTable, changed all the virtual implementations of className to static ones, and replaced all call sites with corresponding lookups in the MethodTable. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::className): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::className): * debugger/DebuggerActivation.h: * jsc.cpp: (GlobalObject::createStructure): * profiler/Profiler.cpp: (JSC::Profiler::createCallIdentifier): * runtime/ClassInfo.h: * runtime/JSCell.cpp: (JSC::JSCell::className): * runtime/JSCell.h: * runtime/JSObject.cpp: (JSC::JSObject::className): * runtime/JSObject.h: * runtime/ObjectPrototype.cpp: (JSC::objectProtoFuncToString): * testRegExp.cpp: (GlobalObject::createStructure): 2011-11-02 Jer Noble <jer.noble@apple.com> Add Clock class and platform-specific implementations. https://bugs.webkit.org/show_bug.cgi?id=71341 Reviewed by Sam Weinig. Add WTF_USE_COREAUDIO macro for use by PlatformClockCA. * wtf/Platform.h: 2011-11-03 Pavel Feldman <pfeldman@chromium.org> Not reviewed: fixing win build. step2. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-11-03 Pavel Feldman <pfeldman@chromium.org> Not reviewed: fix windows build, step1 * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-11-03 Pavel Feldman <pfeldman@google.com> Web Inspector: preserve script location for inline handlers. https://bugs.webkit.org/show_bug.cgi?id=71367 Makes SourceCode factories receive TextPosition instead of the line number; Stores consistent position values in SourceCode and SourceProvider; Reviewed by Yury Semikhatsky. * API/JSBase.cpp: (JSEvaluateScript): (JSCheckScriptSyntax): * API/JSObjectRef.cpp: (JSObjectMakeFunction): * parser/SourceCode.h: (JSC::makeSource): * parser/SourceProvider.h: (JSC::SourceProvider::SourceProvider): (JSC::SourceProvider::startPosition): (JSC::UStringSourceProvider::create): (JSC::UStringSourceProvider::UStringSourceProvider): * runtime/FunctionConstructor.cpp: (JSC::constructFunction): (JSC::constructFunctionSkippingEvalEnabledCheck): * runtime/FunctionConstructor.h: 2011-11-03 Kentaro Hara <haraken@chromium.org> Fixed wrong implementation of doubleValue % 2^{64}. https://bugs.webkit.org/show_bug.cgi?id=67980 Reviewed by Hajime Morita. fast/events/constructors/progress-event-constructor.html was failing because of the wrong implementation of conversion from an ECMAScript value to an IDL unsigned long long value (Spec: http://www.w3.org/TR/WebIDL/#es-unsigned-long-long). In particular, the calculation of doubleValue % 2^{64} was wrong. This patch implemented it correctly in doubleToInteger() in wtf/MathExtras.h. * wtf/MathExtras.h: (doubleToInteger): Implemented the spec correctly. 2011-11-03 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r99089. http://trac.webkit.org/changeset/99089 https://bugs.webkit.org/show_bug.cgi?id=71448 @plt postfix for math functions cause crash on Linux 32 (the symbol is defined but it points to NULL) (Requested by zherczeg on #webkit). * dfg/DFGOperations.cpp: * jit/JITStubs.cpp: * jit/ThunkGenerators.cpp: 2011-11-02 Filip Pizlo <fpizlo@apple.com> DFG inlining breaks function.arguments[something] if the argument being retrieved was subjected to DFG's unboxing optimizations https://bugs.webkit.org/show_bug.cgi?id=71436 Reviewed by Oliver Hunt. This makes inlined arguments retrieval use some of the same machinery as OSR to determine where from, and how, to retrieve a value that the DFG might have somehow squirreled away while the old JIT would put it in its obvious location, using an obvious format. To that end, previously DFG-internal notions such as DataFormat, VirtualRegister, and ValueRecovery are now in bytecode/ since they are stored as part of InlineCallFrames. * bytecode/CodeOrigin.h: * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * interpreter/CallFrame.cpp: (JSC::CallFrame::trueCallerFrame): * interpreter/CallFrame.h: (JSC::ExecState::inlineCallFrame): * interpreter/Register.h: (JSC::Register::asInlineCallFrame): (JSC::Register::unboxedInt32): (JSC::Register::unboxedBoolean): (JSC::Register::unboxedCell): * runtime/Arguments.h: (JSC::Arguments::finishCreationAndCopyRegisters): 2011-11-02 Filip Pizlo <fpizlo@apple.com> ValueRecovery should be moved out of the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=71439 Reviewed by Oliver Hunt. * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/DataFormat.h: Added. (JSC::dataFormatToString): (JSC::needDataFormatConversion): (JSC::isJSFormat): (JSC::isJSInteger): (JSC::isJSDouble): (JSC::isJSCell): (JSC::isJSBoolean): * bytecode/ValueRecovery.h: Added. (JSC::ValueRecovery::ValueRecovery): (JSC::ValueRecovery::alreadyInRegisterFile): (JSC::ValueRecovery::alreadyInRegisterFileAsUnboxedInt32): (JSC::ValueRecovery::alreadyInRegisterFileAsUnboxedCell): (JSC::ValueRecovery::alreadyInRegisterFileAsUnboxedBoolean): (JSC::ValueRecovery::inGPR): (JSC::ValueRecovery::inPair): (JSC::ValueRecovery::inFPR): (JSC::ValueRecovery::displacedInRegisterFile): (JSC::ValueRecovery::constant): (JSC::ValueRecovery::technique): (JSC::ValueRecovery::isInRegisters): (JSC::ValueRecovery::gpr): (JSC::ValueRecovery::tagGPR): (JSC::ValueRecovery::payloadGPR): (JSC::ValueRecovery::fpr): (JSC::ValueRecovery::virtualRegister): (JSC::ValueRecovery::dump): * bytecode/VirtualRegister.h: Added. * dfg/DFGGenerationInfo.h: (JSC::DFG::GenerationInfo::isJSFormat): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::ValueSource::dump): * dfg/DFGSpeculativeJIT.h: * dfg/DFGVariableAccessData.h: 2011-11-02 Sam Weinig <sam@webkit.org> Object.getOwnPropertyDescriptor() does not retrieve the getter/setter from a property on the window that has been overridden with a getter/setter https://bugs.webkit.org/show_bug.cgi?id=71333 Reviewed by Gavin Barraclough. Tested by fast/dom/getter-on-window-object2.html * runtime/PropertyDescriptor.cpp: (JSC::PropertyDescriptor::setDescriptor): The attributes returned from Structure::get do not include Getter or Setter, so instead check if the value is a GetterSetter like we do elsewhere. If it is, update the descriptor's attributes accordingly. 2011-11-02 Yuqiang Xian <yuqiang.xian@intel.com> FunctionPtr should accept FASTCALL functions on X86 https://bugs.webkit.org/show_bug.cgi?id=71434 Reviewed by Filip Pizlo. On X86 we sometimes use FASTCALL convention functions, for example the cti functions, and we may need the pointers to such functions, e.g., in current DFG register file check and arity check, though long term we may avoid such usage of cti calls in DFG. * assembler/MacroAssemblerCodeRef.h: (JSC::FunctionPtr::FunctionPtr): 2011-11-02 Filip Pizlo <fpizlo@apple.com> Inlined uses of the global object should use the right global object https://bugs.webkit.org/show_bug.cgi?id=71427 Reviewed by Oliver Hunt. * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::globalObjectFor): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-11-02 Yuqiang Xian <yuqiang.xian@intel.com> Remove some unnecessary loads/stores in DFG JIT 32_64 https://bugs.webkit.org/show_bug.cgi?id=71090 Reviewed by Filip Pizlo. In fillSpeculateCell and OSR exit, some unnecessary loads/stores can be eliminated. * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateCell): 2011-11-02 Adam Klein <adamk@chromium.org> Replace usage of StringImpl with String where possible in CharacterData and Text https://bugs.webkit.org/show_bug.cgi?id=71383 Reviewed by Darin Adler. * wtf/text/WTFString.h: (WTF::String::containsOnlyWhitespace): Added new method. 2011-11-02 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSObject::getOwnPropertyNames https://bugs.webkit.org/show_bug.cgi?id=71307 Reviewed by Darin Adler. Added getOwnPropertyNames to the MethodTable, changed all the virtual implementations of getOwnPropertyNames to static ones, and replaced all call sites with corresponding lookups in the MethodTable. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::getOwnPropertyNames): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::getOwnPropertyNames): * debugger/DebuggerActivation.h: * runtime/Arguments.cpp: (JSC::Arguments::getOwnPropertyNames): * runtime/Arguments.h: * runtime/ClassInfo.h: * runtime/JSActivation.cpp: (JSC::JSActivation::getOwnPropertyNames): * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::getOwnPropertyNames): * runtime/JSArray.h: * runtime/JSByteArray.cpp: (JSC::JSByteArray::getOwnPropertyNames): * runtime/JSByteArray.h: * runtime/JSCell.cpp: (JSC::JSCell::getOwnPropertyNames): * runtime/JSCell.h: * runtime/JSFunction.cpp: (JSC::JSFunction::getOwnPropertyNames): * runtime/JSFunction.h: * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::getOwnPropertyNames): * runtime/JSNotAnObject.h: * runtime/JSONObject.cpp: (JSC::Stringifier::Holder::appendNextProperty): (JSC::Walker::walk): * runtime/JSObject.cpp: (JSC::JSObject::getPropertyNames): (JSC::JSObject::getOwnPropertyNames): * runtime/JSObject.h: * runtime/JSVariableObject.cpp: (JSC::JSVariableObject::~JSVariableObject): (JSC::JSVariableObject::getOwnPropertyNames): * runtime/JSVariableObject.h: * runtime/ObjectConstructor.cpp: (JSC::objectConstructorGetOwnPropertyNames): (JSC::objectConstructorKeys): (JSC::defineProperties): * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::getOwnPropertyNames): * runtime/StringObject.cpp: (JSC::StringObject::getOwnPropertyNames): * runtime/StringObject.h: * runtime/Structure.h: 2011-11-02 Dean Jackson <dino@apple.com> Add ENABLE_CSS_SHADERS flag https://bugs.webkit.org/show_bug.cgi?id=71394 Reviewed by Sam Weinig. * Configurations/FeatureDefines.xcconfig: 2011-11-02 Alexey Shabalin <a.shabalin@gmail.com> TEXTREL in libjavascriptcoregtk-1.0.so.0.11.0 on x86 (or i586) https://bugs.webkit.org/show_bug.cgi?id=70610 Reviewed by Martin Robinson. Properly annotate ASM on BSD and Linux x86 systems. * dfg/DFGOperations.cpp: Add annotation for X86. * jit/JITStubs.cpp: Ditto. * jit/ThunkGenerators.cpp: Ditto. 2011-11-02 Xianzhu Wang <wangxianzhu@chromium.org> Missing Force8BitConstructor in 8-bit version of StringImpl::reallocate() https://bugs.webkit.org/show_bug.cgi?id=71347 Reviewed by Geoffrey Garen. * wtf/text/StringImpl.cpp: (WTF::StringImpl::reallocate): 2011-11-01 Darin Adler <darin@apple.com> Cut down on malloc/free a bit in the parser arena https://bugs.webkit.org/show_bug.cgi?id=71343 Reviewed by Oliver Hunt. * parser/ParserArena.cpp: (JSC::ParserArena::deallocateObjects): Call the destructors of the deletable objects before freeing the pools. Don't call fastFree on the deletable objects any more. * parser/ParserArena.h: (JSC::ParserArena::allocateDeletable): Use allocateFreeable instead of fastMalloc here. 2011-11-01 Sam Weinig <sam@webkit.org> Implement __lookupGetter__/__lookupSetter__ in terms of getPropertyDescriptor https://bugs.webkit.org/show_bug.cgi?id=71336 Reviewed by Darin Adler. * debugger/DebuggerActivation.cpp: * debugger/DebuggerActivation.h: Remove overrides of lookupGetter/lookupSetter, which are no longer needed due to implementing getPropertyDescriptor. * runtime/JSObject.cpp: (JSC::JSObject::lookupGetter): (JSC::JSObject::lookupSetter): * runtime/JSObject.h: De-virtualize lookupGetter/lookupSetter, and implement them in terms of getPropertyDescriptor. 2011-11-01 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSObject::defineSetter https://bugs.webkit.org/show_bug.cgi?id=71303 Reviewed by Darin Adler. Added defineSetter to the MethodTable, changed all the virtual implementations of defineSetter to static ones, and replaced all call sites with corresponding lookups in the MethodTable. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::defineSetter): * debugger/DebuggerActivation.h: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ClassInfo.h: * runtime/JSCell.cpp: (JSC::JSCell::defineSetter): * runtime/JSCell.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::defineSetter): * runtime/JSGlobalObject.h: * runtime/JSObject.cpp: (JSC::JSObject::defineSetter): (JSC::putDescriptor): * runtime/JSObject.h: * runtime/ObjectPrototype.cpp: (JSC::objectProtoFuncDefineSetter): 2011-11-01 Filip Pizlo <fpizlo@apple.com> DFG inlining breaks function.arguments https://bugs.webkit.org/show_bug.cgi?id=71329 Reviewed by Oliver Hunt. The DFG was forgetting to store code origin mappings for inlined call sites. Some of the fast-path optimizations for CallFrame::trueCallerFrame() were wrong. An assertion in Arguments was wrong. I also took the opportunity to decrease code duplication between DFG64 and DFG32_64, because I didn't feel like writing the same code twice. * bytecode/CodeBlock.h: (JSC::ExecState::isInlineCallFrame): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileEntry): (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::link): (JSC::DFG::JITCompiler::compile): (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGJITCompiler32_64.cpp: * dfg/DFGNode.h: * interpreter/CallFrame.cpp: (JSC::CallFrame::trueCallerFrame): * interpreter/CallFrame.h: * runtime/Arguments.h: (JSC::Arguments::getArgumentsData): 2011-11-01 Xianzhu Wang <wangxianzhu@chromium.org> StringImpl::reallocate() should have a 8-bit version https://bugs.webkit.org/show_bug.cgi?id=71210 Reviewed by Geoffrey Garen. * wtf/text/StringImpl.cpp: (WTF::StringImpl::reallocate): * wtf/text/StringImpl.h: 2011-10-31 Filip Pizlo <fpizlo@apple.com> The GC should be parallel https://bugs.webkit.org/show_bug.cgi?id=70995 Reviewed by Geoff Garen. Added parallel tracing to the GC. This works by having local mark stacks per thread, and a global shared one. Threads sometimes donate cells from the mark stack to the global one if the heuristics tell them that it's affordable to do so. Threads that have depleted their local mark stacks try to steal some from the shared one. Marking is now done using an atomic weak relaxed CAS (compare-and-swap). This is a 23% speed-up on V8-splay when I use 4 marking threads, leading to a 3.5% speed-up on V8. It also appears that this reduces GC pause times on real websites by more than half. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::~Heap): (JSC::Heap::markRoots): * heap/Heap.h: * heap/MarkStack.cpp: (JSC::MarkStackSegmentAllocator::MarkStackSegmentAllocator): (JSC::MarkStackSegmentAllocator::~MarkStackSegmentAllocator): (JSC::MarkStackSegmentAllocator::allocate): (JSC::MarkStackSegmentAllocator::release): (JSC::MarkStackSegmentAllocator::shrinkReserve): (JSC::MarkStackArray::MarkStackArray): (JSC::MarkStackArray::~MarkStackArray): (JSC::MarkStackArray::expand): (JSC::MarkStackArray::refill): (JSC::MarkStackArray::donateSomeCellsTo): (JSC::MarkStackArray::stealSomeCellsFrom): (JSC::MarkStackThreadSharedData::markingThreadMain): (JSC::MarkStackThreadSharedData::markingThreadStartFunc): (JSC::MarkStackThreadSharedData::MarkStackThreadSharedData): (JSC::MarkStackThreadSharedData::~MarkStackThreadSharedData): (JSC::MarkStackThreadSharedData::reset): (JSC::MarkStack::reset): (JSC::SlotVisitor::donateSlow): (JSC::SlotVisitor::drain): (JSC::SlotVisitor::drainFromShared): (JSC::MarkStack::mergeOpaqueRoots): (JSC::SlotVisitor::harvestWeakReferences): * heap/MarkStack.h: (JSC::MarkStackSegment::data): (JSC::MarkStackSegment::capacityFromSize): (JSC::MarkStackSegment::sizeFromCapacity): (JSC::MarkStackArray::postIncTop): (JSC::MarkStackArray::preDecTop): (JSC::MarkStackArray::setTopForFullSegment): (JSC::MarkStackArray::setTopForEmptySegment): (JSC::MarkStackArray::top): (JSC::MarkStackArray::validatePrevious): (JSC::MarkStack::addWeakReferenceHarvester): (JSC::MarkStack::mergeOpaqueRootsIfNecessary): (JSC::MarkStack::mergeOpaqueRootsIfProfitable): (JSC::MarkStack::MarkStack): (JSC::MarkStack::addOpaqueRoot): (JSC::MarkStack::containsOpaqueRoot): (JSC::MarkStack::opaqueRootCount): (JSC::MarkStackArray::append): (JSC::MarkStackArray::canRemoveLast): (JSC::MarkStackArray::removeLast): (JSC::MarkStackArray::isEmpty): (JSC::MarkStackArray::canDonateSomeCells): (JSC::MarkStackArray::size): (JSC::ParallelModeEnabler::ParallelModeEnabler): (JSC::ParallelModeEnabler::~ParallelModeEnabler): * heap/MarkedBlock.h: (JSC::MarkedBlock::testAndSetMarked): * heap/SlotVisitor.h: (JSC::SlotVisitor::donate): (JSC::SlotVisitor::donateAndDrain): (JSC::SlotVisitor::donateKnownParallel): (JSC::SlotVisitor::SlotVisitor): * heap/WeakReferenceHarvester.h: * runtime/Heuristics.cpp: (JSC::Heuristics::initializeHeuristics): * runtime/Heuristics.h: * wtf/Atomics.h: (WTF::weakCompareAndSwap): * wtf/Bitmap.h: (WTF::::Bitmap): (WTF::::get): (WTF::::set): (WTF::::testAndSet): (WTF::::testAndClear): (WTF::::concurrentTestAndSet): (WTF::::concurrentTestAndClear): (WTF::::clear): (WTF::::clearAll): (WTF::::nextPossiblyUnset): (WTF::::findRunOfZeros): (WTF::::count): (WTF::::isEmpty): (WTF::::isFull): * wtf/MainThread.h: (WTF::isMainThreadOrGCThread): * wtf/Platform.h: * wtf/ThreadSpecific.h: (WTF::::isSet): * wtf/mac/MainThreadMac.mm: (WTF::initializeGCThreads): (WTF::initializeMainThreadPlatform): (WTF::initializeMainThreadToProcessMainThreadPlatform): (WTF::registerGCThread): (WTF::isMainThreadOrGCThread): 2011-10-31 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSObject::defaultValue https://bugs.webkit.org/show_bug.cgi?id=71146 Reviewed by Sam Weinig. Added defaultValue to the MethodTable. Replaced all virtual versions of defaultValue with static versions. Replaced all call sites with lookups in the MethodTable. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/ClassInfo.h: * runtime/ExceptionHelpers.cpp: (JSC::InterruptedExecutionError::defaultValue): (JSC::TerminatedExecutionError::defaultValue): * runtime/ExceptionHelpers.h: * runtime/JSCell.cpp: (JSC::JSCell::defaultValue): * runtime/JSCell.h: * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::defaultValue): * runtime/JSNotAnObject.h: * runtime/JSObject.cpp: (JSC::JSObject::getPrimitiveNumber): (JSC::JSObject::defaultValue): * runtime/JSObject.h: (JSC::JSObject::toPrimitive): 2011-10-31 Mark Hahnenberg <mhahnenberg@apple.com> Interpreter build fix Unreviewed build fix * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * runtime/Executable.cpp: (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): 2011-10-31 Filip Pizlo <fpizlo@apple.com> DFG OSR exits should add to value profiles https://bugs.webkit.org/show_bug.cgi?id=71202 Reviewed by Oliver Hunt. Value profiles now have an extra special slot not used by the old JIT's profiling, which is reserved for OSR exits. The DFG's OSR exit code now knows which register, node index, and value profiling site was responsible for the (possibly flawed) information that led to the OSR failure. This is somewhat opportunistic and imperfect; if there's a lot of control flow between the value profiling site and the OSR failure point, then this mechanism simply gives up. It also gives up if the OSR failure is caused by either known deficiencies in the DFG (like that we always assume that the index in a strict charCodeAt access is within bounds) or where the OSR failure would be catalogues and profiled through other means (like slow case counters). This patch also adds the notion of a JSValueRegs, which is either a single register in JSVALUE64 or a pair in JSVALUE32_64. We should probably move the 32_64 DFG towards using this, since it often makes it easier to share code between 64 and 32_64. Also fixed a number of pathologies that this uncovered. op_method_check didn't have a value profiling site on the slow path. GetById should not always force OSR exit if it never executed in the old JIT; we may be able to infer its type if it's a array or string length get. Finally, these changes benefit from a slight tweak to optimization delay heuristics (profile fullness is now 0.35 instead of 0.25). 3.8% speed-up on Kraken, mostly due to ~35% on both stanford-crypto-aes and imaging-darkroom. * bytecode/ValueProfile.cpp: (JSC::ValueProfile::computeStatistics): (JSC::ValueProfile::computeUpdatedPrediction): * bytecode/ValueProfile.h: (JSC::ValueProfile::ValueProfile): (JSC::ValueProfile::specFailBucket): (JSC::ValueProfile::numberOfSamples): (JSC::ValueProfile::isLive): (JSC::ValueProfile::numberOfInt32s): (JSC::ValueProfile::numberOfDoubles): (JSC::ValueProfile::numberOfCells): (JSC::ValueProfile::numberOfObjects): (JSC::ValueProfile::numberOfFinalObjects): (JSC::ValueProfile::numberOfStrings): (JSC::ValueProfile::numberOfArrays): (JSC::ValueProfile::numberOfBooleans): (JSC::ValueProfile::dump): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getPredictionWithoutOSRExit): (JSC::DFG::ByteCodeParser::getPrediction): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGPRInfo.h: (JSC::DFG::JSValueRegs::JSValueRegs): (JSC::DFG::JSValueRegs::operator!): (JSC::DFG::JSValueRegs::gpr): (JSC::DFG::JSValueSource::JSValueSource): (JSC::DFG::JSValueSource::unboxedCell): (JSC::DFG::JSValueSource::operator!): (JSC::DFG::JSValueSource::isAddress): (JSC::DFG::JSValueSource::offset): (JSC::DFG::JSValueSource::base): (JSC::DFG::JSValueSource::gpr): (JSC::DFG::JSValueSource::asAddress): (JSC::DFG::JSValueSource::notAddress): (JSC::DFG::JSValueRegs::tagGPR): (JSC::DFG::JSValueRegs::payloadGPR): (JSC::DFG::JSValueSource::tagGPR): (JSC::DFG::JSValueSource::payloadGPR): (JSC::DFG::JSValueSource::hasKnownTag): (JSC::DFG::JSValueSource::tag): * dfg/DFGGenerationInfo.h: (JSC::DFG::GenerationInfo::jsValueRegs): * dfg/DFGGraph.h: (JSC::DFG::Graph::valueProfileFor): * dfg/DFGJITCodeGenerator.h: (JSC::JSValueOperand::jsValueRegs): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::valueProfileFor): * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::OSRExit::OSRExit): (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::compileGetCharCodeAt): (JSC::DFG::SpeculativeJIT::compileGetByValOnString): (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): (JSC::DFG::SpeculativeJIT::compileGetByValOnByteArray): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::speculationCheck): (JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): * jit/JITPropertyAccess.cpp: (JSC::JIT::emitSlow_op_method_check): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emitSlow_op_method_check): * runtime/Heuristics.cpp: (JSC::Heuristics::initializeHeuristics): * runtime/JSValue.h: 2011-10-31 Sam Weinig <sam@webkit.org> Remove need for virtual JSObject::unwrappedObject https://bugs.webkit.org/show_bug.cgi?id=71034 Reviewed by Geoffrey Garen. * JavaScriptCore.exp: Update exports. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: Add JSGlobalThis.cpp. * runtime/JSGlobalThis.cpp: Added. (JSC::JSGlobalThis::visitChildren): (JSC::JSGlobalThis::unwrappedObject): * runtime/JSGlobalThis.h: (JSC::JSGlobalThis::createStructure): Move underlying object from JSDOMWindowShell down to JSGlobalThis and corresponding visitChildren method. * runtime/JSObject.cpp: (JSC::JSObject::unwrappedObject): Change unwrappedObject from virtual, to just needing an if check. * runtime/JSObject.h: (JSC::JSObject::isGlobalThis): * runtime/JSType.h: Add isGlobalThis predicate and type. 2011-10-31 Xianzhu Wang <wangxianzhu@chromium.org> WTF::StringImpl::create(const char*, unsigned) calls itself https://bugs.webkit.org/show_bug.cgi?id=71206 The original implementation just calls itself, causing infinite recursion. Cast the first parameter to const LChar* to fix that. Reviewed by Ryosuke Niwa. * wtf/text/StringImpl.h: (WTF::StringImpl::create): 2011-10-31 Andy Wingo <wingo@igalia.com> Fix DFG JIT compilation on Linux targets. https://bugs.webkit.org/show_bug.cgi?id=70904 Reviewed by Darin Adler. * jit/JITStubs.cpp (SYMBOL_STRING_RELOCATION): Simplify this macro. * dfg/DFGOperations.cpp (SYMBOL_STRING_RELOCATION): Copy the simplified definition from jit/JITStubs.cpp. (FUNCTION_WRAPPER_WITH_RETURN_ADDRESS, getHostCallReturnValue): Use the macro to access trampoline targets through the PLT on PIC systems, instead of introducing a text relocation. Otherwise, the library fails to link. 2011-10-31 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSObject::defineGetter https://bugs.webkit.org/show_bug.cgi?id=71134 Reviewed by Darin Adler. Added defineGetter to the MethodTable. Replaced all virtual versions of defineGetter with static versions. Replaced all call sites with lookups in the MethodTable. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::defineGetter): * debugger/DebuggerActivation.h: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ClassInfo.h: * runtime/JSCell.cpp: (JSC::JSCell::defineGetter): * runtime/JSCell.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::defineGetter): * runtime/JSGlobalObject.h: * runtime/JSObject.cpp: (JSC::JSObject::defineGetter): (JSC::putDescriptor): * runtime/JSObject.h: * runtime/ObjectPrototype.cpp: (JSC::objectProtoFuncDefineGetter): 2011-10-31 Michael Saboff <msaboff@apple.com> Towards 8-bit Strings: Move Lexer and Parser Objects out of JSGlobalData https://bugs.webkit.org/show_bug.cgi?id=71138 Restructure and movement of Lexer and Parser code. Moved Lexer and Parser objects out of JSGlobalData. Added a new ParserTokens class and instance to JSGlobalData that have JavaScript token related definitions. Replaced JSGlobalData arguments to Node classes with lineNumber, as that was the only use of the JSGlobalData. Combined JSParser and Parser classes into one class, eliminating JSParser.h and .cpp. Various supporting #include changes. These mostly mechanical changes are done in preparation to making the Lexer and Parser template classes. Reviewed by Darin Adler. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecompiler/NodesCodegen.cpp: (JSC::ArrayNode::toArgumentList): (JSC::ApplyFunctionCallDotNode::emitBytecode): * parser/ASTBuilder.h: (JSC::ASTBuilder::ASTBuilder): (JSC::ASTBuilder::createSourceElements): (JSC::ASTBuilder::createCommaExpr): (JSC::ASTBuilder::createLogicalNot): (JSC::ASTBuilder::createUnaryPlus): (JSC::ASTBuilder::createVoid): (JSC::ASTBuilder::thisExpr): (JSC::ASTBuilder::createResolve): (JSC::ASTBuilder::createObjectLiteral): (JSC::ASTBuilder::createArray): (JSC::ASTBuilder::createNumberExpr): (JSC::ASTBuilder::createString): (JSC::ASTBuilder::createBoolean): (JSC::ASTBuilder::createNull): (JSC::ASTBuilder::createBracketAccess): (JSC::ASTBuilder::createDotAccess): (JSC::ASTBuilder::createRegExp): (JSC::ASTBuilder::createNewExpr): (JSC::ASTBuilder::createConditionalExpr): (JSC::ASTBuilder::createAssignResolve): (JSC::ASTBuilder::createFunctionExpr): (JSC::ASTBuilder::createFunctionBody): (JSC::ASTBuilder::createGetterOrSetterProperty): (JSC::ASTBuilder::createArguments): (JSC::ASTBuilder::createArgumentsList): (JSC::ASTBuilder::createPropertyList): (JSC::ASTBuilder::createElementList): (JSC::ASTBuilder::createFormalParameterList): (JSC::ASTBuilder::createClause): (JSC::ASTBuilder::createClauseList): (JSC::ASTBuilder::createFuncDeclStatement): (JSC::ASTBuilder::createBlockStatement): (JSC::ASTBuilder::createExprStatement): (JSC::ASTBuilder::createIfStatement): (JSC::ASTBuilder::createForLoop): (JSC::ASTBuilder::createForInLoop): (JSC::ASTBuilder::createEmptyStatement): (JSC::ASTBuilder::createVarStatement): (JSC::ASTBuilder::createReturnStatement): (JSC::ASTBuilder::createBreakStatement): (JSC::ASTBuilder::createContinueStatement): (JSC::ASTBuilder::createTryStatement): (JSC::ASTBuilder::createSwitchStatement): (JSC::ASTBuilder::createWhileStatement): (JSC::ASTBuilder::createDoWhileStatement): (JSC::ASTBuilder::createLabelStatement): (JSC::ASTBuilder::createWithStatement): (JSC::ASTBuilder::createThrowStatement): (JSC::ASTBuilder::createDebugger): (JSC::ASTBuilder::createConstStatement): (JSC::ASTBuilder::appendConstDecl): (JSC::ASTBuilder::combineCommaNodes): (JSC::ASTBuilder::appendBinaryOperation): (JSC::ASTBuilder::createAssignment): (JSC::ASTBuilder::createNumber): (JSC::ASTBuilder::makeTypeOfNode): (JSC::ASTBuilder::makeDeleteNode): (JSC::ASTBuilder::makeNegateNode): (JSC::ASTBuilder::makeBitwiseNotNode): (JSC::ASTBuilder::makeMultNode): (JSC::ASTBuilder::makeDivNode): (JSC::ASTBuilder::makeModNode): (JSC::ASTBuilder::makeAddNode): (JSC::ASTBuilder::makeSubNode): (JSC::ASTBuilder::makeLeftShiftNode): (JSC::ASTBuilder::makeRightShiftNode): (JSC::ASTBuilder::makeURightShiftNode): (JSC::ASTBuilder::makeBitOrNode): (JSC::ASTBuilder::makeBitAndNode): (JSC::ASTBuilder::makeBitXOrNode): (JSC::ASTBuilder::makeFunctionCallNode): (JSC::ASTBuilder::makeBinaryNode): (JSC::ASTBuilder::makeAssignNode): (JSC::ASTBuilder::makePrefixNode): (JSC::ASTBuilder::makePostfixNode): * parser/JSParser.cpp: Removed. * parser/JSParser.h: Removed. * parser/Lexer.cpp: (JSC::Keywords::Keywords): (JSC::Lexer::Lexer): (JSC::Lexer::~Lexer): (JSC::Lexer::setCode): (JSC::Lexer::parseIdentifier): * parser/Lexer.h: (JSC::Keywords::isKeyword): (JSC::Keywords::getKeyword): (JSC::Keywords::~Keywords): (JSC::Lexer::setIsReparsing): (JSC::Lexer::isReparsing): (JSC::Lexer::lineNumber): (JSC::Lexer::setLastLineNumber): (JSC::Lexer::lastLineNumber): (JSC::Lexer::prevTerminator): (JSC::Lexer::sawError): (JSC::Lexer::getErrorMessage): (JSC::Lexer::currentOffset): (JSC::Lexer::setOffset): (JSC::Lexer::setLineNumber): (JSC::Lexer::sourceProvider): (JSC::Lexer::isWhiteSpace): (JSC::Lexer::isLineTerminator): (JSC::Lexer::convertHex): (JSC::Lexer::convertUnicode): (JSC::Lexer::makeIdentifier): (JSC::Lexer::lexExpectIdentifier): * parser/NodeConstructors.h: (JSC::ParserArenaFreeable::operator new): (JSC::ParserArenaDeletable::operator new): (JSC::ParserArenaRefCounted::ParserArenaRefCounted): (JSC::Node::Node): (JSC::ExpressionNode::ExpressionNode): (JSC::StatementNode::StatementNode): (JSC::NullNode::NullNode): (JSC::BooleanNode::BooleanNode): (JSC::NumberNode::NumberNode): (JSC::StringNode::StringNode): (JSC::RegExpNode::RegExpNode): (JSC::ThisNode::ThisNode): (JSC::ResolveNode::ResolveNode): (JSC::ElementNode::ElementNode): (JSC::ArrayNode::ArrayNode): (JSC::PropertyNode::PropertyNode): (JSC::PropertyListNode::PropertyListNode): (JSC::ObjectLiteralNode::ObjectLiteralNode): (JSC::BracketAccessorNode::BracketAccessorNode): (JSC::DotAccessorNode::DotAccessorNode): (JSC::ArgumentListNode::ArgumentListNode): (JSC::ArgumentsNode::ArgumentsNode): (JSC::NewExprNode::NewExprNode): (JSC::EvalFunctionCallNode::EvalFunctionCallNode): (JSC::FunctionCallValueNode::FunctionCallValueNode): (JSC::FunctionCallResolveNode::FunctionCallResolveNode): (JSC::FunctionCallBracketNode::FunctionCallBracketNode): (JSC::FunctionCallDotNode::FunctionCallDotNode): (JSC::CallFunctionCallDotNode::CallFunctionCallDotNode): (JSC::ApplyFunctionCallDotNode::ApplyFunctionCallDotNode): (JSC::PrePostResolveNode::PrePostResolveNode): (JSC::PostfixResolveNode::PostfixResolveNode): (JSC::PostfixBracketNode::PostfixBracketNode): (JSC::PostfixDotNode::PostfixDotNode): (JSC::PostfixErrorNode::PostfixErrorNode): (JSC::DeleteResolveNode::DeleteResolveNode): (JSC::DeleteBracketNode::DeleteBracketNode): (JSC::DeleteDotNode::DeleteDotNode): (JSC::DeleteValueNode::DeleteValueNode): (JSC::VoidNode::VoidNode): (JSC::TypeOfResolveNode::TypeOfResolveNode): (JSC::TypeOfValueNode::TypeOfValueNode): (JSC::PrefixResolveNode::PrefixResolveNode): (JSC::PrefixBracketNode::PrefixBracketNode): (JSC::PrefixDotNode::PrefixDotNode): (JSC::PrefixErrorNode::PrefixErrorNode): (JSC::UnaryOpNode::UnaryOpNode): (JSC::UnaryPlusNode::UnaryPlusNode): (JSC::NegateNode::NegateNode): (JSC::BitwiseNotNode::BitwiseNotNode): (JSC::LogicalNotNode::LogicalNotNode): (JSC::BinaryOpNode::BinaryOpNode): (JSC::MultNode::MultNode): (JSC::DivNode::DivNode): (JSC::ModNode::ModNode): (JSC::AddNode::AddNode): (JSC::SubNode::SubNode): (JSC::LeftShiftNode::LeftShiftNode): (JSC::RightShiftNode::RightShiftNode): (JSC::UnsignedRightShiftNode::UnsignedRightShiftNode): (JSC::LessNode::LessNode): (JSC::GreaterNode::GreaterNode): (JSC::LessEqNode::LessEqNode): (JSC::GreaterEqNode::GreaterEqNode): (JSC::ThrowableBinaryOpNode::ThrowableBinaryOpNode): (JSC::InstanceOfNode::InstanceOfNode): (JSC::InNode::InNode): (JSC::EqualNode::EqualNode): (JSC::NotEqualNode::NotEqualNode): (JSC::StrictEqualNode::StrictEqualNode): (JSC::NotStrictEqualNode::NotStrictEqualNode): (JSC::BitAndNode::BitAndNode): (JSC::BitOrNode::BitOrNode): (JSC::BitXOrNode::BitXOrNode): (JSC::LogicalOpNode::LogicalOpNode): (JSC::ConditionalNode::ConditionalNode): (JSC::ReadModifyResolveNode::ReadModifyResolveNode): (JSC::AssignResolveNode::AssignResolveNode): (JSC::ReadModifyBracketNode::ReadModifyBracketNode): (JSC::AssignBracketNode::AssignBracketNode): (JSC::AssignDotNode::AssignDotNode): (JSC::ReadModifyDotNode::ReadModifyDotNode): (JSC::AssignErrorNode::AssignErrorNode): (JSC::CommaNode::CommaNode): (JSC::ConstStatementNode::ConstStatementNode): (JSC::SourceElements::SourceElements): (JSC::EmptyStatementNode::EmptyStatementNode): (JSC::DebuggerStatementNode::DebuggerStatementNode): (JSC::ExprStatementNode::ExprStatementNode): (JSC::VarStatementNode::VarStatementNode): (JSC::IfNode::IfNode): (JSC::IfElseNode::IfElseNode): (JSC::DoWhileNode::DoWhileNode): (JSC::WhileNode::WhileNode): (JSC::ForNode::ForNode): (JSC::ContinueNode::ContinueNode): (JSC::BreakNode::BreakNode): (JSC::ReturnNode::ReturnNode): (JSC::WithNode::WithNode): (JSC::LabelNode::LabelNode): (JSC::ThrowNode::ThrowNode): (JSC::TryNode::TryNode): (JSC::ParameterNode::ParameterNode): (JSC::FuncExprNode::FuncExprNode): (JSC::FuncDeclNode::FuncDeclNode): (JSC::CaseClauseNode::CaseClauseNode): (JSC::ClauseListNode::ClauseListNode): (JSC::CaseBlockNode::CaseBlockNode): (JSC::SwitchNode::SwitchNode): (JSC::ConstDeclNode::ConstDeclNode): (JSC::BlockNode::BlockNode): (JSC::ForInNode::ForInNode): * parser/NodeInfo.h: * parser/Nodes.cpp: (JSC::StatementNode::setLoc): (JSC::ScopeNode::ScopeNode): (JSC::ProgramNode::ProgramNode): (JSC::ProgramNode::create): (JSC::EvalNode::EvalNode): (JSC::EvalNode::create): (JSC::FunctionBodyNode::FunctionBodyNode): (JSC::FunctionBodyNode::create): * parser/Nodes.h: (JSC::Node::lineNo): * parser/Parser.cpp: (JSC::Parser::Parser): (JSC::Parser::~Parser): (JSC::Parser::parseInner): (JSC::Parser::allowAutomaticSemicolon): (JSC::Parser::parseSourceElements): (JSC::Parser::parseVarDeclaration): (JSC::Parser::parseConstDeclaration): (JSC::Parser::parseDoWhileStatement): (JSC::Parser::parseWhileStatement): (JSC::Parser::parseVarDeclarationList): (JSC::Parser::parseConstDeclarationList): (JSC::Parser::parseForStatement): (JSC::Parser::parseBreakStatement): (JSC::Parser::parseContinueStatement): (JSC::Parser::parseReturnStatement): (JSC::Parser::parseThrowStatement): (JSC::Parser::parseWithStatement): (JSC::Parser::parseSwitchStatement): (JSC::Parser::parseSwitchClauses): (JSC::Parser::parseSwitchDefaultClause): (JSC::Parser::parseTryStatement): (JSC::Parser::parseDebuggerStatement): (JSC::Parser::parseBlockStatement): (JSC::Parser::parseStatement): (JSC::Parser::parseFormalParameters): (JSC::Parser::parseFunctionBody): (JSC::Parser::parseFunctionInfo): (JSC::Parser::parseFunctionDeclaration): (JSC::LabelInfo::LabelInfo): (JSC::Parser::parseExpressionOrLabelStatement): (JSC::Parser::parseExpressionStatement): (JSC::Parser::parseIfStatement): (JSC::Parser::parseExpression): (JSC::Parser::parseAssignmentExpression): (JSC::Parser::parseConditionalExpression): (JSC::isUnaryOp): (JSC::Parser::isBinaryOperator): (JSC::Parser::parseBinaryExpression): (JSC::Parser::parseProperty): (JSC::Parser::parseObjectLiteral): (JSC::Parser::parseStrictObjectLiteral): (JSC::Parser::parseArrayLiteral): (JSC::Parser::parsePrimaryExpression): (JSC::Parser::parseArguments): (JSC::Parser::parseMemberExpression): (JSC::Parser::parseUnaryExpression): * parser/Parser.h: (JSC::isEvalNode): (JSC::EvalNode): (JSC::DepthManager::DepthManager): (JSC::DepthManager::~DepthManager): (JSC::ScopeLabelInfo::ScopeLabelInfo): (JSC::Scope::Scope): (JSC::Scope::startSwitch): (JSC::Scope::endSwitch): (JSC::Scope::startLoop): (JSC::Scope::endLoop): (JSC::Scope::inLoop): (JSC::Scope::breakIsValid): (JSC::Scope::continueIsValid): (JSC::Scope::pushLabel): (JSC::Scope::popLabel): (JSC::Scope::getLabel): (JSC::Scope::setIsFunction): (JSC::Scope::isFunction): (JSC::Scope::isFunctionBoundary): (JSC::Scope::declareVariable): (JSC::Scope::declareWrite): (JSC::Scope::preventNewDecls): (JSC::Scope::allowsNewDecls): (JSC::Scope::declareParameter): (JSC::Scope::useVariable): (JSC::Scope::setNeedsFullActivation): (JSC::Scope::collectFreeVariables): (JSC::Scope::getUncapturedWrittenVariables): (JSC::Scope::getCapturedVariables): (JSC::Scope::setStrictMode): (JSC::Scope::strictMode): (JSC::Scope::isValidStrictMode): (JSC::Scope::shadowsArguments): (JSC::Scope::copyCapturedVariablesToVector): (JSC::Scope::saveFunctionInfo): (JSC::Scope::restoreFunctionInfo): (JSC::ScopeRef::ScopeRef): (JSC::ScopeRef::operator->): (JSC::ScopeRef::index): (JSC::ScopeRef::hasContainingScope): (JSC::ScopeRef::containingScope): (JSC::Parser::AllowInOverride::AllowInOverride): (JSC::Parser::AllowInOverride::~AllowInOverride): (JSC::Parser::AutoPopScopeRef::AutoPopScopeRef): (JSC::Parser::AutoPopScopeRef::~AutoPopScopeRef): (JSC::Parser::AutoPopScopeRef::setPopped): (JSC::Parser::currentScope): (JSC::Parser::pushScope): (JSC::Parser::popScopeInternal): (JSC::Parser::popScope): (JSC::Parser::declareVariable): (JSC::Parser::declareWrite): (JSC::Parser::findCachedFunctionInfo): (JSC::Parser::isFunctionBodyNode): (JSC::Parser::next): (JSC::Parser::nextExpectIdentifier): (JSC::Parser::nextTokenIsColon): (JSC::Parser::consume): (JSC::Parser::getToken): (JSC::Parser::match): (JSC::Parser::tokenStart): (JSC::Parser::tokenLine): (JSC::Parser::tokenEnd): (JSC::Parser::getTokenName): (JSC::Parser::updateErrorMessageSpecialCase): (JSC::Parser::updateErrorMessage): (JSC::Parser::updateErrorWithNameAndMessage): (JSC::Parser::startLoop): (JSC::Parser::endLoop): (JSC::Parser::startSwitch): (JSC::Parser::endSwitch): (JSC::Parser::setStrictMode): (JSC::Parser::strictMode): (JSC::Parser::isValidStrictMode): (JSC::Parser::declareParameter): (JSC::Parser::breakIsValid): (JSC::Parser::continueIsValid): (JSC::Parser::pushLabel): (JSC::Parser::popLabel): (JSC::Parser::getLabel): (JSC::Parser::autoSemiColon): (JSC::Parser::canRecurse): (JSC::Parser::lastTokenEnd): (JSC::Parser::DepthManager::DepthManager): (JSC::Parser::DepthManager::~DepthManager): (JSC::Parser::parse): (JSC::parse): * parser/ParserTokens.h: Added. (JSC::JSTokenInfo::JSTokenInfo): * parser/SourceCode.h: (JSC::SourceCode::subExpression): * parser/SourceProviderCacheItem.h: * parser/SyntaxChecker.h: (JSC::SyntaxChecker::SyntaxChecker): (JSC::SyntaxChecker::makeFunctionCallNode): (JSC::SyntaxChecker::createCommaExpr): (JSC::SyntaxChecker::makeAssignNode): (JSC::SyntaxChecker::makePrefixNode): (JSC::SyntaxChecker::makePostfixNode): (JSC::SyntaxChecker::makeTypeOfNode): (JSC::SyntaxChecker::makeDeleteNode): (JSC::SyntaxChecker::makeNegateNode): (JSC::SyntaxChecker::makeBitwiseNotNode): (JSC::SyntaxChecker::createLogicalNot): (JSC::SyntaxChecker::createUnaryPlus): (JSC::SyntaxChecker::createVoid): (JSC::SyntaxChecker::thisExpr): (JSC::SyntaxChecker::createResolve): (JSC::SyntaxChecker::createObjectLiteral): (JSC::SyntaxChecker::createArray): (JSC::SyntaxChecker::createNumberExpr): (JSC::SyntaxChecker::createString): (JSC::SyntaxChecker::createBoolean): (JSC::SyntaxChecker::createNull): (JSC::SyntaxChecker::createBracketAccess): (JSC::SyntaxChecker::createDotAccess): (JSC::SyntaxChecker::createRegExp): (JSC::SyntaxChecker::createNewExpr): (JSC::SyntaxChecker::createConditionalExpr): (JSC::SyntaxChecker::createAssignResolve): (JSC::SyntaxChecker::createFunctionExpr): (JSC::SyntaxChecker::createFunctionBody): (JSC::SyntaxChecker::createArguments): (JSC::SyntaxChecker::createArgumentsList): (JSC::SyntaxChecker::createProperty): (JSC::SyntaxChecker::createPropertyList): (JSC::SyntaxChecker::createFuncDeclStatement): (JSC::SyntaxChecker::createBlockStatement): (JSC::SyntaxChecker::createExprStatement): (JSC::SyntaxChecker::createIfStatement): (JSC::SyntaxChecker::createForLoop): (JSC::SyntaxChecker::createForInLoop): (JSC::SyntaxChecker::createEmptyStatement): (JSC::SyntaxChecker::createVarStatement): (JSC::SyntaxChecker::createReturnStatement): (JSC::SyntaxChecker::createBreakStatement): (JSC::SyntaxChecker::createContinueStatement): (JSC::SyntaxChecker::createTryStatement): (JSC::SyntaxChecker::createSwitchStatement): (JSC::SyntaxChecker::createWhileStatement): (JSC::SyntaxChecker::createWithStatement): (JSC::SyntaxChecker::createDoWhileStatement): (JSC::SyntaxChecker::createLabelStatement): (JSC::SyntaxChecker::createThrowStatement): (JSC::SyntaxChecker::createDebugger): (JSC::SyntaxChecker::createConstStatement): (JSC::SyntaxChecker::appendConstDecl): (JSC::SyntaxChecker::createGetterOrSetterProperty): (JSC::SyntaxChecker::combineCommaNodes): (JSC::SyntaxChecker::operatorStackPop): * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::checkSyntax): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::produceCodeBlockFor): (JSC::FunctionExecutable::fromGlobalCode): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::~JSGlobalData): * runtime/JSGlobalData.h: * runtime/LiteralParser.cpp: (JSC::LiteralParser::tryJSONPParse): 2011-10-31 Filip Pizlo <fpizlo@apple.com> REGRESSION (r97118): Reproducible crash in JSCell::toPrimitive when adding https://bugs.webkit.org/show_bug.cgi?id=71227 Reviewed by Oliver Hunt. No new tests, since while I can see exactly where the DFG went wrong on the site in question from looking at the generated machine code, and while I can certainly believe that such a scenario would happen, I cannot visualize how to make it happen reproducibly. It requires an odd combination of double values getting spilled and then refilled, but then reboxed at just the right time so that the spilled value is an unboxed double while the in-register value is a boxed double. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentFillGPR): 2011-10-30 Filip Pizlo <fpizlo@apple.com> JSParser::parsePrimaryExpression should have an overflow check https://bugs.webkit.org/show_bug.cgi?id=71197 Reviewed by Geoff Garen. * parser/JSParser.cpp: (JSC::JSParser::parsePrimaryExpression): 2011-10-30 Filip Pizlo <fpizlo@apple.com> DFG ValueAdd(string, int) should not fail speculation https://bugs.webkit.org/show_bug.cgi?id=71195 Reviewed by Geoff Garen. 1% speed-up on V8. * dfg/DFGNode.h: (JSC::DFG::Node::shouldNotSpeculateInteger): (JSC::DFG::Node::shouldSpeculateInteger): 2011-10-30 Filip Pizlo <fpizlo@apple.com> The DFG inliner should not flush the callee https://bugs.webkit.org/show_bug.cgi?id=71191 Reviewed by Oliver Hunt. 0.6% speed-up on V8. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): * bytecode/CodeOrigin.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::flush): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * interpreter/CallFrame.cpp: (JSC::CallFrame::trueCallerFrameSlow): 2011-10-28 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize isGlobalObject, isVariableObject, isActivationObject, and isErrorInstance in JSObject https://bugs.webkit.org/show_bug.cgi?id=70968 Reviewed by Geoffrey Garen. * API/JSCallbackObject.cpp: Added two specializations for createStructure that use different JSTypes in their TypeInfo. Had to also create a specialization for JSNonFinalObject, even JSGlobalObject was the only that needed it because Windows wouldn't build without it. (JSC::::createStructure): * API/JSCallbackObject.h: * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/ErrorInstance.h: Removed virtual function and changed JSType provided to TypeInfo in createStructure. (JSC::ErrorInstance::createStructure): * runtime/ErrorPrototype.h: Ditto (JSC::ErrorPrototype::createStructure): * runtime/JSActivation.h: Ditto (JSC::JSActivation::createStructure): * runtime/JSGlobalObject.h: Ditto (JSC::JSGlobalObject::createStructure): * runtime/JSObject.h: De-virtualized functions. They now check the JSType of the object for the corresponding type. (JSC::JSObject::isGlobalObject): (JSC::JSObject::isVariableObject): (JSC::JSObject::isActivationObject): (JSC::JSObject::isErrorInstance): * runtime/JSType.h: Added new types for GlobalObject, VariableObject, ActivationObject, and ErrorInstance. * runtime/JSVariableObject.cpp: Removed virtual function. * runtime/JSVariableObject.h: Changed JSType provided to TypeInfo in createStructure. (JSC::JSVariableObject::createStructure): 2011-10-28 Pavel Feldman <pfeldman@google.com> Reset line numbers for scripts generated with document.write. https://bugs.webkit.org/show_bug.cgi?id=71099 Reviewed by Yury Semikhatsky. * wtf/text/TextPosition.h: (WTF::OrdinalNumber::OrdinalNumber): 2011-10-27 Daniel Bates <dbates@rim.com> CMake: Add support to optionally install the built JavaScript shell https://bugs.webkit.org/show_bug.cgi?id=71062 Reviewed by Antonio Gomes. Generate an installation rule for installing the JavaScript shell in /bin (with respect to the prefix path) when SHOULD_INSTALL_JS_SHELL is defined. * shell/CMakeLists.txt: 2011-10-27 Kentaro Hara <haraken@chromium.org> Generate WebKitCSSMatrix constructor for JSC by [Constructor] IDL https://bugs.webkit.org/show_bug.cgi?id=70215 Reviewed by Adam Barth. Added a method that judges if a given JSValue is empty. Tests: transforms/svg-vs-css.xhtml transforms/cssmatrix-2d-interface.xhtml transforms/cssmatrix-3d-interface.xhtml * runtime/JSValue.h: * runtime/JSValueInlineMethods.h: (JSC::JSValue::isEmpty): 2011-10-27 Michael Saboff <msaboff@apple.com> ENH: Add 8 bit string support to JSC JIT https://bugs.webkit.org/show_bug.cgi?id=71073 Changed the JIT String character access generation to create code to check the character size and load8() or load16() as approriate. Reviewed by Gavin Barraclough. * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::load8): * assembler/X86Assembler.h: (JSC::X86Assembler::movzbl_mr): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileGetCharCodeAt): (JSC::DFG::SpeculativeJIT::compileGetByValOnString): * jit/JITInlineMethods.h: (JSC::JIT::emitLoadCharacterString): * jit/JITPropertyAccess.cpp: (JSC::JIT::stringGetByValStubGenerator): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::stringGetByValStubGenerator): * jit/JSInterfaceJIT.h: (JSC::ThunkHelpers::stringImplFlagsOffset): (JSC::ThunkHelpers::stringImpl8BitFlag): * jit/ThunkGenerators.cpp: (JSC::stringCharLoad): 2011-10-27 Filip Pizlo <fpizlo@apple.com> If the bytecode generator emits code after the return in the first basic block, DFG's inliner crashes https://bugs.webkit.org/show_bug.cgi?id=71071 Reviewed by Gavin Barraclough. Removed some cruft dealing with parsing failures due to unsupported functionality (that's never reached anymore due to it being caught in DFGCapabilities). This allowed me to repurpose the bool return from parseBlock() to mean: true if we should continue to parse, or false if we've already parsed all live code. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::parseCodeBlock): 2011-10-27 Joseph Pecoraro <pecoraro@apple.com> Reviewed by David Kilzer. Make FeatureDefines Identical Across OS X Projects https://bugs.webkit.org/show_bug.cgi?id=71051 * Configurations/FeatureDefines.xcconfig: 2011-10-27 Filip Pizlo <fpizlo@apple.com> Crash in JSC::Structure::materializePropertyMap when viewing Garden-O-Matic https://bugs.webkit.org/show_bug.cgi?id=71045 Reviewed by Geoff Garen. Make sure that if a structure is pinned, it also has a property map. * runtime/Structure.cpp: (JSC::Structure::changePrototypeTransition): (JSC::Structure::despecifyFunctionTransition): (JSC::Structure::getterSetterTransition): (JSC::Structure::toDictionaryTransition): (JSC::Structure::preventExtensionsTransition): (JSC::Structure::addPropertyWithoutTransition): (JSC::Structure::removePropertyWithoutTransition): (JSC::Structure::pin): (JSC::Structure::copyPropertyTableForPinning): * runtime/Structure.h: (JSC::Structure::materializePropertyMapIfNecessaryForPinning): 2011-10-27 Michael Saboff <msaboff@apple.com> 32bit build failure after r98624 https://bugs.webkit.org/show_bug.cgi?id=71064 Disambiguated operator overload with unsigned index (0u). Reviewed by Sam Weinig. * runtime/UString.h: (JSC::operator==): 2011-10-27 Gustavo Noronha Silva <gns@gnome.org> Fix building on GNU/kFreeBSD https://bugs.webkit.org/show_bug.cgi?id=71005 Reviewed by Darin Adler. * config.h: * wtf/Platform.h: 2011-10-27 Michael Saboff <msaboff@apple.com> Investigate storing strings in 8-bit buffers when possible https://bugs.webkit.org/show_bug.cgi?id=66161 Investigate storing strings in 8-bit buffers when possible https://bugs.webkit.org/show_bug.cgi?id=66161 Added support for 8 bit string data in StringImpl. Changed (UChar*) m_data to m_data16. Added char* m_data8 as a union with m_data16. Added UChar* m_copyData16 to the other union to store a 16 bit copy of an 8 bit string when needed. Added characters8() and characters16() accessor methods that assume the caller has checked the underlying string type via the new is8Bit() method. The characters() method will return a UChar* of the string, materializing a 16 bit copy if the string is an 8 bit string. Added two flags, one for 8 bit buffer and a second for a 16 bit copy for an 8 bit string. Fixed method name typo (StringHasher::defaultCoverter()). Over time the goal is to eliminate calls to characters() and us the character8() and characters16() accessors. This patch does not include changes that actually create 8 bit strings. This is the first of at least 8 patches. Subsequent patches will be submitted for JIT changes, making the JSC lexer, parser and literal parser, JavaScript string changes and then changes in webcore to take advantage of the 8 bit strings. This change is performance neutral for SunSpider and V8 when run from the command line with "jsc". Reviewed by Geoffrey Garen. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def * interpreter/Interpreter.cpp: (JSC::Interpreter::callEval): * parser/SourceProvider.h: (JSC::UStringSourceProvider::data): (JSC::UStringSourceProvider::UStringSourceProvider): * runtime/Identifier.cpp: (JSC::IdentifierCStringTranslator::hash): (JSC::IdentifierCStringTranslator::equal): (JSC::IdentifierCStringTranslator::translate): (JSC::Identifier::add): (JSC::Identifier::toUInt32): * runtime/Identifier.h: (JSC::Identifier::equal): (JSC::operator==): (JSC::operator!=): * runtime/JSString.cpp: (JSC::JSString::resolveRope): (JSC::JSString::resolveRopeSlowCase): * runtime/RegExp.cpp: (JSC::RegExp::match): * runtime/StringPrototype.cpp: (JSC::jsSpliceSubstringsWithSeparators): * runtime/UString.cpp: (JSC::UString::UString): (JSC::equalSlowCase): (JSC::UString::utf8): * runtime/UString.h: (JSC::UString::characters): (JSC::UString::characters8): (JSC::UString::characters16): (JSC::UString::is8Bit): (JSC::UString::operator[]): (JSC::UString::find): (JSC::operator==): * wtf/StringHasher.h: (WTF::StringHasher::computeHash): (WTF::StringHasher::defaultConverter): * wtf/text/AtomicString.cpp: (WTF::CStringTranslator::hash): (WTF::CStringTranslator::equal): (WTF::CStringTranslator::translate): (WTF::AtomicString::add): * wtf/text/AtomicString.h: (WTF::AtomicString::AtomicString): (WTF::AtomicString::contains): (WTF::AtomicString::find): (WTF::AtomicString::add): (WTF::operator==): (WTF::operator!=): (WTF::equalIgnoringCase): * wtf/text/StringConcatenate.h: * wtf/text/StringHash.h: (WTF::StringHash::equal): (WTF::CaseFoldingHash::hash): * wtf/text/StringImpl.cpp: (WTF::StringImpl::~StringImpl): (WTF::StringImpl::createUninitialized): (WTF::StringImpl::create): (WTF::StringImpl::getData16SlowCase): (WTF::StringImpl::containsOnlyWhitespace): (WTF::StringImpl::substring): (WTF::StringImpl::characterStartingAt): (WTF::StringImpl::lower): (WTF::StringImpl::upper): (WTF::StringImpl::fill): (WTF::StringImpl::foldCase): (WTF::StringImpl::stripMatchedCharacters): (WTF::StringImpl::removeCharacters): (WTF::StringImpl::simplifyMatchedCharactersToSpace): (WTF::StringImpl::toIntStrict): (WTF::StringImpl::toUIntStrict): (WTF::StringImpl::toInt64Strict): (WTF::StringImpl::toUInt64Strict): (WTF::StringImpl::toIntPtrStrict): (WTF::StringImpl::toInt): (WTF::StringImpl::toUInt): (WTF::StringImpl::toInt64): (WTF::StringImpl::toUInt64): (WTF::StringImpl::toIntPtr): (WTF::StringImpl::toDouble): (WTF::StringImpl::toFloat): (WTF::equal): (WTF::equalIgnoringCase): (WTF::StringImpl::find): (WTF::StringImpl::findIgnoringCase): (WTF::StringImpl::reverseFind): (WTF::StringImpl::replace): (WTF::StringImpl::defaultWritingDirection): (WTF::StringImpl::adopt): (WTF::StringImpl::createWithTerminatingNullCharacter): * wtf/text/StringImpl.h: (WTF::StringImpl::StringImpl): (WTF::StringImpl::create): (WTF::StringImpl::create8): (WTF::StringImpl::tryCreateUninitialized): (WTF::StringImpl::flagsOffset): (WTF::StringImpl::flagIs8Bit): (WTF::StringImpl::dataOffset): (WTF::StringImpl::is8Bit): (WTF::StringImpl::characters8): (WTF::StringImpl::characters16): (WTF::StringImpl::characters): (WTF::StringImpl::has16BitShadow): (WTF::StringImpl::setHash): (WTF::StringImpl::hash): (WTF::StringImpl::copyChars): (WTF::StringImpl::operator[]): (WTF::StringImpl::find): (WTF::StringImpl::findIgnoringCase): (WTF::equal): (WTF::equalIgnoringCase): (WTF::StringImpl::isolatedCopy): * wtf/text/WTFString.cpp: (WTF::String::String): (WTF::String::append): (WTF::String::format): (WTF::String::fromUTF8): (WTF::String::fromUTF8WithLatin1Fallback): * wtf/text/WTFString.h: (WTF::String::find): (WTF::String::findIgnoringCase): (WTF::String::contains): (WTF::String::append): (WTF::String::fromUTF8): (WTF::String::fromUTF8WithLatin1Fallback): (WTF::operator==): (WTF::operator!=): (WTF::equalIgnoringCase): * wtf/unicode/Unicode.h: * yarr/YarrJIT.cpp: (JSC::Yarr::execute): * yarr/YarrJIT.h: (JSC::Yarr::YarrCodeBlock::execute): * yarr/YarrParser.h: (JSC::Yarr::Parser::Parser): 2011-10-27 Mark Hahnenberg <mhahnenberg@apple.com> Fixing windows build Unreviewed build fix * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-10-27 Mark Hahnenberg <mhahnenberg@apple.com> Add ability to check for presence of static members at compile time https://bugs.webkit.org/show_bug.cgi?id=70986 Reviewed by Geoffrey Garen. Added new CREATE_MEMBER_CHECKER macro to instantiate the template and the HAS_MEMBER_NAMED macro to use that template to check if the specified class does indeed have a method with that name. This mechanism is not currently used anywhere, but will be in the future when adding virtual methods from JSObject to the MethodTable. * runtime/ClassInfo.h: 2011-10-27 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSCell::toThisObject https://bugs.webkit.org/show_bug.cgi?id=70958 Reviewed by Geoffrey Garen. Converted all instances of toThisObject to static functions, added toThisObject to the MethodTable, and replaced all call sites with a corresponding lookup in the MethodTable. * API/JSContextRef.cpp: * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/ClassInfo.h: * runtime/JSActivation.cpp: (JSC::JSActivation::toThisObject): * runtime/JSActivation.h: * runtime/JSCell.cpp: (JSC::JSCell::toThisObject): * runtime/JSCell.h: * runtime/JSObject.cpp: (JSC::JSObject::put): (JSC::JSObject::toThisObject): * runtime/JSObject.h: (JSC::JSValue::toThisObject): * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::toThisObject): * runtime/JSStaticScopeObject.h: * runtime/JSString.cpp: (JSC::JSString::toThisObject): * runtime/JSString.h: * runtime/StrictEvalActivation.cpp: (JSC::StrictEvalActivation::toThisObject): * runtime/StrictEvalActivation.h: 2011-10-27 Yuqiang Xian <yuqiang.xian@intel.com> Fix a small bug in callOperation after r98431 https://bugs.webkit.org/show_bug.cgi?id=70984 Reviewed by Geoffrey Garen. TrustedImmPtr is not expecting "int" type parameters. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): 2011-10-26 Oliver Hunt <oliver@apple.com> Restore structure-clearing behaviour of allocateCell<> https://bugs.webkit.org/show_bug.cgi?id=70976 Reviewed by Geoffrey Garen. This restores the logic that allows the markstack to filter live objects that have not yet been initialised. * runtime/JSCell.h: (JSC::JSCell::clearStructure): Validation-safe method to clear a cell's structure. (JSC::allocateCell): Call the above method. * runtime/Structure.h: (JSC::MarkStack::internalAppend): Don't visit cells that haven't been initialised. 2011-10-26 Filip Pizlo <fpizlo@apple.com> REGRESSION (r97030): Cannot log in to progressive.com https://bugs.webkit.org/show_bug.cgi?id=70094 Reviewed by Oliver Hunt. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleCall): 2011-10-26 Mark Hahnenberg <mhahnenberg@apple.com> Remove getOwnPropertySlotVirtual https://bugs.webkit.org/show_bug.cgi?id=70741 Reviewed by Geoffrey Garen. Removed all declarations and definitions of getOwnPropertySlotVirtual. Also replaced all call sites to getOwnPropertyVirtualVirtual with a corresponding lookup in the MethodTable. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::getOwnPropertyDescriptor): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::getOwnPropertySlot): * debugger/DebuggerActivation.h: * runtime/Arguments.cpp: * runtime/Arguments.h: * runtime/ArrayConstructor.cpp: * runtime/ArrayConstructor.h: * runtime/ArrayPrototype.cpp: * runtime/ArrayPrototype.h: * runtime/BooleanPrototype.cpp: * runtime/BooleanPrototype.h: * runtime/DateConstructor.cpp: * runtime/DateConstructor.h: * runtime/DatePrototype.cpp: * runtime/DatePrototype.h: (JSC::DatePrototype::create): * runtime/ErrorPrototype.cpp: * runtime/ErrorPrototype.h: * runtime/JSActivation.cpp: * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::getOwnPropertySlotByIndex): * runtime/JSArray.h: * runtime/JSByteArray.cpp: * runtime/JSByteArray.h: * runtime/JSCell.cpp: * runtime/JSCell.h: * runtime/JSFunction.cpp: (JSC::JSFunction::getOwnPropertyDescriptor): (JSC::JSFunction::getOwnPropertyNames): (JSC::JSFunction::put): * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: * runtime/JSGlobalObject.h: * runtime/JSNotAnObject.cpp: * runtime/JSNotAnObject.h: * runtime/JSONObject.cpp: (JSC::Stringifier::Holder::appendNextProperty): (JSC::Walker::walk): * runtime/JSONObject.h: * runtime/JSObject.cpp: (JSC::JSObject::getOwnPropertySlotByIndex): (JSC::JSObject::hasOwnProperty): * runtime/JSObject.h: (JSC::JSCell::fastGetOwnPropertySlot): (JSC::JSObject::getPropertySlot): (JSC::JSValue::get): * runtime/JSStaticScopeObject.cpp: * runtime/JSStaticScopeObject.h: * runtime/JSString.cpp: (JSC::JSString::getOwnPropertySlot): * runtime/JSString.h: * runtime/MathObject.cpp: * runtime/MathObject.h: (JSC::MathObject::create): * runtime/NumberConstructor.cpp: * runtime/NumberConstructor.h: * runtime/NumberPrototype.cpp: * runtime/NumberPrototype.h: * runtime/ObjectConstructor.cpp: * runtime/ObjectConstructor.h: * runtime/ObjectPrototype.cpp: * runtime/ObjectPrototype.h: * runtime/RegExpConstructor.cpp: * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::createStructure): * runtime/RegExpObject.cpp: * runtime/RegExpObject.h: * runtime/RegExpPrototype.cpp: * runtime/RegExpPrototype.h: * runtime/StringConstructor.cpp: * runtime/StringConstructor.h: * runtime/StringObject.cpp: * runtime/StringObject.h: * runtime/StringPrototype.cpp: * runtime/StringPrototype.h: 2011-10-26 Alejandro G. Castro <alex@igalia.com> [GTK] [WK2] Add WebKit2 distcheck support https://bugs.webkit.org/show_bug.cgi?id=70933 Reviewed by Martin Robinson. * GNUmakefile.list.am: Add MemoryStatistics.h to the sources list. 2011-10-26 Michael Saboff <msaboff@apple.com> Increase StringImpl Flag Bits for 8 bit Strings https://bugs.webkit.org/show_bug.cgi?id=70937 Increased the number of bits used for flags in StringImpl from 6 to 8 bits. This frees up 2 flag bits that will be used for 8-bit string support. Updated hash methods accordingly. Changed hash value masking from the low bits to the high bits. Reviewed by Darin Adler. * create_hash_table: * wtf/StringHasher.h: (WTF::StringHasher::hash): * wtf/text/StringImpl.h: 2011-10-26 Dan Bernstein <mitz@apple.com> Build fix. Reverted r98488, which caused the scripts’ status messages to be included in the generated files. * create_hash_table: * create_jit_stubs: 2011-10-26 Tor Arne Vestbø <tor.arne.vestbo@nokia.com> Don't print regular output to STDERR when generating hashtables and JIT stubs Reviewed by Simon Hausmann. * create_hash_table: * create_jit_stubs: 2011-10-25 Gavin Barraclough <barraclough@apple.com> Split DFGJITCodeGenerator::callOperation methods https://bugs.webkit.org/show_bug.cgi?id=70870 Reviewed by Filip Pizlo. The DFGJITCodeGenerator currently contains two sets of callOperation methods. One set works with the JSVALUE64 value representation and passes arguments in registers (suitable for use on x86-64), and one set works with the JSVALUE32_64 value representation and passes arguments in memory (suitable for use on x86). By refactoring out the representation and calling convention specific aspects of the code we can also configure the DFG JIT to operator on platforms that use the JSVALUE32_64 value representation but pass arguments in registers. On platforms supported by the JIT, the payload precedes the tag of a value in argument/result ordering, as such, in order to make the setupResults method generally applicable to return the results of a function that are returned in two registers, the ordering of arguments to this function has been reversed - as is the ordering of augments passed to setupArguments methods, with respect to the ordering with which they are passed in to callOperation. This inconsistency will be resolved in a later change when we combine the pairs of arguments passed into callOperation, such that the function signatures can be made consistent across the two value representations (the callOperation methods will be passed a reference to a struct representing the JSValue temporary, this will consist of two gprs on 32_64 and one on 64). * dfg/DFGJITCodeGenerator.h: (JSC::DFG::resetCallArguments): (JSC::DFG::addCallArgument): - moved, removed tag,payload version of this method. (JSC::DFG::setupArguments): (JSC::DFG::setupArgumentsExecState): (JSC::DFG::setupArgumentsWithExecState): - Calling convention specific portion of callOperation refactored out into these methods. (JSC::DFG::callOperation): - updated these methods to use setupArguments* methods. (JSC::DFG::setupResults): - setupResults is now passed payload,tag. (JSC::DFG::appendCallWithExceptionCheckSetResult): - Added fpr versions of this function. (JSC::DFG::appendCallSetResult): - Added versions of this function without exception check. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::emitCall): - setupResults is now passed payload,tag. 2011-10-25 Mark Hahnenberg <mhahnenberg@apple.com> Remove deletePropertyVirtual https://bugs.webkit.org/show_bug.cgi?id=70738 Reviewed by Geoffrey Garen. Removed all declarations and definitions of deletePropertyVirtual. Also replaced all call sites to deletePropertyVirtual with a corresponding lookup in the MethodTable. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::deletePropertyByIndex): * API/JSObjectRef.cpp: (JSObjectDeleteProperty): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::deleteProperty): * debugger/DebuggerActivation.h: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Arguments.cpp: * runtime/Arguments.h: * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncPop): (JSC::arrayProtoFuncReverse): (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): * runtime/JSActivation.cpp: * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::deleteProperty): (JSC::JSArray::deletePropertyByIndex): * runtime/JSArray.h: * runtime/JSCell.cpp: (JSC::JSCell::deleteProperty): (JSC::JSCell::deletePropertyByIndex): * runtime/JSCell.h: * runtime/JSFunction.cpp: * runtime/JSFunction.h: * runtime/JSNotAnObject.cpp: * runtime/JSNotAnObject.h: * runtime/JSONObject.cpp: (JSC::Walker::walk): * runtime/JSObject.cpp: (JSC::JSObject::deletePropertyByIndex): (JSC::JSObject::defineOwnProperty): * runtime/JSObject.h: * runtime/JSVariableObject.cpp: * runtime/JSVariableObject.h: * runtime/RegExpMatchesArray.h: * runtime/StrictEvalActivation.cpp: * runtime/StrictEvalActivation.h: * runtime/StringObject.cpp: * runtime/StringObject.h: 2011-10-25 Mark Hahnenberg <mhahnenberg@apple.com> Remove putVirtual https://bugs.webkit.org/show_bug.cgi?id=70740 Reviewed by Geoffrey Garen. Removed all declarations and definitions of putVirtual. Also replaced all call sites to putVirtual with a corresponding lookup in the MethodTable. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: * API/JSObjectRef.cpp: (JSObjectSetProperty): (JSObjectSetPropertyAtIndex): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::put): * debugger/DebuggerActivation.h: * dfg/DFGOperations.cpp: * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): (JSC::Interpreter::privateExecute): * jsc.cpp: (GlobalObject::finishCreation): * runtime/Arguments.cpp: * runtime/Arguments.h: * runtime/ArrayPrototype.cpp: (JSC::putProperty): (JSC::arrayProtoFuncConcat): (JSC::arrayProtoFuncPush): (JSC::arrayProtoFuncReverse): (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSlice): (JSC::arrayProtoFuncSort): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): (JSC::arrayProtoFuncFilter): (JSC::arrayProtoFuncMap): * runtime/JSActivation.cpp: * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::putSlowCase): (JSC::JSArray::push): (JSC::JSArray::shiftCount): (JSC::JSArray::unshiftCount): * runtime/JSArray.h: * runtime/JSByteArray.cpp: * runtime/JSByteArray.h: * runtime/JSCell.cpp: (JSC::JSCell::put): (JSC::JSCell::putByIndex): * runtime/JSCell.h: * runtime/JSFunction.cpp: * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: * runtime/JSGlobalObject.h: * runtime/JSNotAnObject.cpp: * runtime/JSNotAnObject.h: * runtime/JSONObject.cpp: (JSC::Walker::walk): * runtime/JSObject.cpp: (JSC::JSObject::putByIndex): (JSC::JSObject::defineOwnProperty): * runtime/JSObject.h: (JSC::JSValue::put): * runtime/JSStaticScopeObject.cpp: * runtime/JSStaticScopeObject.h: * runtime/ObjectPrototype.cpp: * runtime/ObjectPrototype.h: * runtime/RegExpConstructor.cpp: * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: * runtime/RegExpObject.cpp: * runtime/RegExpObject.h: * runtime/StringObject.cpp: * runtime/StringObject.h: * runtime/StringPrototype.cpp: (JSC::stringProtoFuncSplit): 2011-10-25 Gavin Barraclough <barraclough@apple.com> Separate out function linking & exception check data structures. https://bugs.webkit.org/show_bug.cgi?id=70858 Reviewed by Oliver Hunt. This will make it easier to refactor the callOperation methods to spilt the value representation specific handling from the cpu/calling-convention implementation. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::appendCallWithExceptionCheck): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::emitCall): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::emitCall): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: (JSC::DFG::CallLinkRecord::CallLinkRecord): (JSC::DFG::CallExceptionRecord::CallExceptionRecord): (JSC::DFG::JITCompiler::JITCompiler): (JSC::DFG::JITCompiler::notifyCall): (JSC::DFG::JITCompiler::appendCall): (JSC::DFG::JITCompiler::addExceptionCheck): (JSC::DFG::JITCompiler::addFastExceptionCheck): * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::link): 2011-10-25 Filip Pizlo <fpizlo@apple.com> Tiered compilation may introduce dangling pointers in constant buffers https://bugs.webkit.org/show_bug.cgi?id=70854 Reviewed by Oliver Hunt. Tiered compilation now copies constant buffers, which fixes the regression in https://bugs.webkit.org/show_bug.cgi?id=70246. No new tests because this regression relies on a subtle interleaving of optimized compilation and garbage collection, and cannot be reproduced in a simple test. This also adds some new debug support, which was used to fix this bug and is likely to be useful in the future. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::copyDataFrom): (JSC::CodeBlock::usesOpcode): * bytecode/CodeBlock.h: * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): 2011-10-25 Mark Hahnenberg <mhahnenberg@apple.com> Fixing Windows build after r98367 Unreviewed build fix * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-10-25 Yuqiang Xian <yuqiang.xian@intel.com> Add missing DFG file entries to the make lists for GTK and Qt ports https://bugs.webkit.org/show_bug.cgi?id=70806 Reviewed by Darin Adler. * GNUmakefile.list.am: * JavaScriptCore.pro: 2011-10-25 Mark Hahnenberg <mhahnenberg@apple.com> Add getOwnPropertySlot to MethodTable https://bugs.webkit.org/show_bug.cgi?id=69807 Reviewed by Oliver Hunt. * JavaScriptCore.exp: * runtime/ClassInfo.h: Added both versions of getOwnPropertySlot to the MethodTable. * runtime/JSCell.h: Changed getOwnPropertySlot to be protected so other classes can reference it in their MethodTables. 2011-10-25 Oliver Hunt <oliver@apple.com> Need to support marking of multiple nested codeblocks when compiling https://bugs.webkit.org/show_bug.cgi?id=70832 Reviewed by Gavin Barraclough. When inlining a function we end up with multiple codeblocks being compiled at the same time, so we need to support a list of live codeblocks. * heap/Heap.cpp: (JSC::Heap::markRoots): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: (JSC::JSGlobalData::startedCompiling): (JSC::JSGlobalData::finishedCompiling): 2011-10-24 Yuqiang Xian <yuqiang.xian@intel.com> DFG JIT 32_64 - fillInteger should accept DataFormatJSInteger https://bugs.webkit.org/show_bug.cgi?id=70798 Reviewed by Filip Pizlo. When filling an integer for a known integer node (not speculated), it should accept DataFormatJSInteger as well. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::fillInteger): 2011-10-24 Geoffrey Garen <ggaren@apple.com> Build fix: removed some cases of threadsafeCopy() that I missed in my previous patch. * JavaScriptCore.order: 2011-10-24 Geoffrey Garen <ggaren@apple.com> Removed SharedUChar and tightened language around its previous uses https://bugs.webkit.org/show_bug.cgi?id=70698 Reviewed by David Levin. - Removed SharedUChar because most of its functionality has moved into other abstraction layers, and we want remaining clients to choose their abstractions explicitly instead of relying on StringImpl to provide this behavior implicitly, since we think they can sometimes make more efficient choices. - Renamed "threadSafeCopy" and "crossThreadCopy" to "isolatedCopy" because the former names could give the impression that the resulting object was thread-safe, but actually it's just an isolated copy, which is not thread-safe by itself, but can be used to implement a thread-safe algorithm through isolation. * wtf/CrossThreadRefCounted.h: Removed. * JavaScriptCore.exp: Export! * wtf/text/StringImpl.cpp: (WTF::StringImpl::~StringImpl): Removed the stuff mentioned above. * wtf/text/StringImpl.h: (WTF::StringImpl::length): Ditto. (WTF::StringImpl::isolatedCopy): Inlined this, since it's now trivial. * wtf/text/WTFString.cpp: (WTF::String::isolatedCopy): * wtf/text/WTFString.h: Updated for StringImpl changes. * API/OpaqueJSString.h: * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.order: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/CMakeLists.txt: * wtf/OwnFastMallocPtr.h: * wtf/RefCounted.h: * wtf/SizeLimits.cpp: * wtf/ThreadSafeRefCounted.h: * wtf/wtf.pri: * yarr/YarrPattern.h: Updated these files to accomodate removal of CrossThreadRefCounted.h. 2011-10-24 Oliver Hunt <oliver@apple.com> Crash in void JSC::validateCell<JSC::RegExp*>(JSC::RegExp*) https://bugs.webkit.org/show_bug.cgi?id=70689 Reviewed by Filip Pizlo. While performing codegen we need to make the GlobalData explicitly aware of the codeblock being compiled, as compilation may trigger GC and CodeBlock holds GC values, but has not yet been assigned to its owner executable. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): (JSC::BytecodeGenerator::~BytecodeGenerator): * bytecompiler/BytecodeGenerator.h: * heap/AllocationSpace.cpp: (JSC::AllocationSpace::allocateSlowCase): * heap/Heap.cpp: (JSC::Heap::markRoots): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: (JSC::JSGlobalData::startedCompiling): (JSC::JSGlobalData::finishedCompiling): 2011-10-24 Filip Pizlo <fpizlo@apple.com> Object-or-other branch speculation may corrupt the state for OSR if the child of the branch is an integer https://bugs.webkit.org/show_bug.cgi?id=70777 Reviewed by Oliver Hunt. * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): 2011-10-24 Filip Pizlo <fpizlo@apple.com> op_new_array_buffer is not inlined correctly https://bugs.webkit.org/show_bug.cgi?id=70770 Reviewed by Oliver Hunt. Disabled inlining of op_new_array_buffer, for now. * dfg/DFGCapabilities.h: (JSC::DFG::canInlineOpcode): 2011-10-24 Yuqiang Xian <yuqiang.xian@intel.com> Add boolean speculations to DFG JIT 32_64 https://bugs.webkit.org/show_bug.cgi?id=70706 Reviewed by Filip Pizlo. Different from the boolean speculations in DFG 64, the boolean speculations in DFG 32_64 will use a 32bit GPR to hold the primitive boolean instead of a JSBoolean. This choice is not only for performance, but also to save a register as we're short of registers on X86. To accomplish this we make use of DataFormatBoolean, allow a value to be represented as a primitive boolean and converted from/to a JSBoolean. This patch also fixes SpillOrder in 32_64, which should be different from 64, and fixes needDataFormatConversion logic in 32_64. * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::branchTest32): We don't expect byte test actually as it doesn't work for registers esp..edi on X86. * dfg/DFGGenerationInfo.h: (JSC::DFG::needDataFormatConversion): (JSC::DFG::GenerationInfo::initBoolean): (JSC::DFG::GenerationInfo::gpr): (JSC::DFG::GenerationInfo::fillInteger): (JSC::DFG::GenerationInfo::fillBoolean): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::checkConsistency): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::use): (JSC::DFG::JITCodeGenerator::silentSpillGPR): (JSC::DFG::JITCodeGenerator::silentFillGPR): (JSC::DFG::JITCodeGenerator::spill): (JSC::DFG::cellResult): (JSC::DFG::booleanResult): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompareNull): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeStrictEq): * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::ValueSource::dump): (JSC::DFG::ValueRecovery::dump): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::ValueSource::forPrediction): (JSC::DFG::ValueRecovery::alreadyInRegisterFileAsUnboxedBoolean): (JSC::DFG::ValueRecovery::inGPR): (JSC::DFG::ValueRecovery::gpr): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): 2011-10-24 Mark Hahnenberg <mhahnenberg@apple.com> Fixing Windows build Unreviewed build fix * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-10-24 Yuqiang Xian <yuqiang.xian@intel.com> BitVector isInline check could fail https://bugs.webkit.org/show_bug.cgi?id=70691 Reviewed by Geoffrey Garen. Current BitVector uses the highest bit of m_bitsOrPointer to indicate whether it's an inlined bit set or a pointer to an outOfLine bit set. This check may fail in case the pointer also has the highest bit set, which is surely possible on IA32 (Linux). In this case the check failure can result in unexpected behaviors, for example if the BitVector is incorrectly determined as having an inlined bit set, then setting a bit exceeding maxInlineBits will wrongly modify the memory adjacent to the BitVector object. This fix is to use the lowest bit of m_bitsOrPointer to indicate inline or outofline, based on the assumption that the pointer to OutOfLineBits should be 4 or 8 byte aligned. We could mark the lowest bit (bit 0) with 1 for inlined bit set, and bits 1~bitsInPointer are used for bit set/test. In this case we need do one bit more shift for bit set/test. * wtf/BitVector.cpp: (WTF::BitVector::resizeOutOfLine): * wtf/BitVector.h: (WTF::BitVector::quickGet): (WTF::BitVector::quickSet): (WTF::BitVector::quickClear): (WTF::BitVector::makeInlineBits): (WTF::BitVector::isInline): 2011-10-24 Mark Hahnenberg <mhahnenberg@apple.com> Rename static getOwnPropertySlot to getOwnPropertySlotByIndex https://bugs.webkit.org/show_bug.cgi?id=70271 Reviewed by Darin Adler. Renaming versions of getOwnPropertySlot that use an unsigned as the property name to "getOwnPropertySlotByIndex" in preparation for adding them to the MethodTable, which requires unique names for each method. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/Arguments.cpp: (JSC::Arguments::getOwnPropertySlotVirtual): (JSC::Arguments::getOwnPropertySlotByIndex): * runtime/Arguments.h: * runtime/JSArray.cpp: (JSC::JSArray::getOwnPropertySlotVirtual): (JSC::JSArray::getOwnPropertySlotByIndex): (JSC::JSArray::getOwnPropertySlot): * runtime/JSArray.h: * runtime/JSByteArray.cpp: (JSC::JSByteArray::getOwnPropertySlotVirtual): (JSC::JSByteArray::getOwnPropertySlotByIndex): * runtime/JSByteArray.h: * runtime/JSCell.cpp: (JSC::JSCell::getOwnPropertySlotVirtual): (JSC::JSCell::getOwnPropertySlotByIndex): * runtime/JSCell.h: * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::getOwnPropertySlotVirtual): (JSC::JSNotAnObject::getOwnPropertySlotByIndex): * runtime/JSNotAnObject.h: * runtime/JSObject.cpp: (JSC::JSObject::getOwnPropertySlotVirtual): (JSC::JSObject::getOwnPropertySlotByIndex): * runtime/JSObject.h: * runtime/JSString.cpp: (JSC::JSString::getOwnPropertySlotVirtual): (JSC::JSString::getOwnPropertySlotByIndex): * runtime/JSString.h: * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::getOwnPropertySlotVirtual): (JSC::ObjectPrototype::getOwnPropertySlotByIndex): * runtime/ObjectPrototype.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::getOwnPropertySlotVirtual): (JSC::RegExpMatchesArray::getOwnPropertySlotByIndex): * runtime/StringObject.cpp: (JSC::StringObject::getOwnPropertySlotVirtual): (JSC::StringObject::getOwnPropertySlotByIndex): * runtime/StringObject.h: 2011-10-24 Patrick Gansterer <paroga@webkit.org> Interpreter build fix after r98179. * bytecode/CodeBlock.h: Moved CodeBlock::baselineVersion() into ENABLE(JIT) block, since it is only used there. 2011-10-23 Geoffrey Garen <ggaren@apple.com> Fixed a typo Darin spotted. * wtf/StringHasher.h: (WTF::StringHasher::hash): Expelliarmus! 2011-10-23 Geoffrey Garen <ggaren@apple.com> Removed StringImpl::createStrippingNullCharacters https://bugs.webkit.org/show_bug.cgi?id=70700 Reviewed by David Levin. It was unused. * JavaScriptCore.exp: * wtf/text/StringImpl.cpp: * wtf/text/StringImpl.h: 2011-10-22 Filip Pizlo <fpizlo@apple.com> DFG should inline constructors https://bugs.webkit.org/show_bug.cgi?id=70675 Reviewed by Oliver Hunt. Adds support for inlining constructors. Also fixes two pathologies uncovered along the way: CheckMethod claimed that it never returned a result (causing CheckMethod -> SetLocal -> GetLocal sequences to result in the GetLocal doing OSR exit), and get_by_id parsing never checked if it was hot in slow path. Also fiddled with inlining heuristics; it appears that for now, the more inlining, the happier V8 is. Finally, a bug was uncovered where a silent spill of a boxed integer that had previously been spilled unboxed causes the silent fill to forget to unbox. This appears to be a 4% speed-up on V8 in their harness, or a 1% speed-up in my harness. The difference is due to warm-up: in my harness we see significant amounts of time spent in compilation, but in V8's harness compilation gets amortizes. Profiling indicates that we have the potential for a 5% win from basic optimizations like generating OSR exits lazily and holding onto bytecode longer. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleCall): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::handleMinMax): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGCapabilities.h: (JSC::DFG::mightInlineFunctionForConstruct): (JSC::DFG::canInlineOpcode): (JSC::DFG::mightInlineFunctionFor): (JSC::DFG::canInlineFunctionFor): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentFillGPR): * runtime/Executable.h: (JSC::isCall): (JSC::ExecutableBase::intrinsicFor): * runtime/Heuristics.cpp: (JSC::Heuristics::initializeHeuristics): * runtime/Heuristics.h: 2011-10-23 Noel Gordon <noel.gordon@gmail.com> [chromium] Remove RopeImpl.{h,cpp} from the gyp projects https://bugs.webkit.org/show_bug.cgi?id=70703 Reviewed by Kent Tamura. runtime/RopeImpl.{h,cpp} were removed in r97872, remove references to these files from the gyp project files. * JavaScriptCore.gypi: 2011-10-23 Mark Hahnenberg <mhahnenberg@apple.com> Add deleteProperty to the MethodTable https://bugs.webkit.org/show_bug.cgi?id=70162 Reviewed by Sam Weinig. * JavaScriptCore.exp: * runtime/ClassInfo.h: Added both versions of deleteProperty to the MethodTable. * runtime/JSFunction.h: Changed JSFunction::deleteProperty to be protected rather than private for subclasses who don't provide their own implementation. 2011-10-23 Mark Hahnenberg <mhahnenberg@apple.com> Remove getConstructDataVirtual https://bugs.webkit.org/show_bug.cgi?id=70638 Reviewed by Darin Adler. Removed all declarations and definitions of getConstructDataVirtual. Also replaced all call sites to getConstructDataVirtual with a corresponding lookup in the MethodTable. * API/JSCallbackConstructor.cpp: * API/JSCallbackConstructor.h: * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: * API/JSObjectRef.cpp: (JSObjectIsConstructor): (JSObjectCallAsConstructor): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * dfg/DFGOperations.cpp: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ArrayConstructor.cpp: * runtime/ArrayConstructor.h: * runtime/BooleanConstructor.cpp: * runtime/BooleanConstructor.h: * runtime/DateConstructor.cpp: * runtime/DateConstructor.h: * runtime/Error.h: (JSC::StrictModeTypeErrorFunction::getConstructData): * runtime/ErrorConstructor.cpp: * runtime/ErrorConstructor.h: * runtime/FunctionConstructor.cpp: * runtime/FunctionConstructor.h: * runtime/JSCell.cpp: * runtime/JSCell.h: * runtime/JSFunction.cpp: * runtime/JSFunction.h: * runtime/JSObject.h: (JSC::getConstructData): * runtime/NativeErrorConstructor.cpp: * runtime/NativeErrorConstructor.h: * runtime/NumberConstructor.cpp: * runtime/NumberConstructor.h: * runtime/ObjectConstructor.cpp: * runtime/ObjectConstructor.h: * runtime/RegExpConstructor.cpp: * runtime/RegExpConstructor.h: * runtime/StringConstructor.cpp: * runtime/StringConstructor.h: 2011-10-23 Geoffrey Garen <ggaren@apple.com> Try to fix the SL build. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): Cast away int vs unisgned warning. 2011-10-21 Geoffrey Garen <ggaren@apple.com> Separated string lifetime bits from character buffer state bits https://bugs.webkit.org/show_bug.cgi?id=70673 Reviewed by Anders Carlsson. Moved the static/immortal bit into the bottom bit of the refcount, and moved all other bits into the high bits of the hash code. This is the first step toward a new Characters/PassString class, and it makes ref/deref slightly more efficient. * create_hash_table: * wtf/StringHasher.h: (WTF::StringHasher::hash): Tweaked the string hashing function to leave the top bits clear, so they can be used as flags. Fixed some small differences between the PERL copy of this function and the C++ copy of this function, which could have in theory caused subtle crashes. * wtf/text/StringImpl.cpp: (WTF::StringImpl::sharedBuffer): (WTF::StringImpl::createWithTerminatingNullCharacter): * wtf/text/StringImpl.h: (WTF::StringImpl::StringImpl): (WTF::StringImpl::cost): Renamed s_refCountFlagShouldReportedCost to s_didReportExtraCost, since the original name was both self-contradictory and used as a double-negative. (WTF::StringImpl::isIdentifier): (WTF::StringImpl::setIsIdentifier): (WTF::StringImpl::hasTerminatingNullCharacter): (WTF::StringImpl::isAtomic): (WTF::StringImpl::setIsAtomic): (WTF::StringImpl::setHash): (WTF::StringImpl::rawHash): (WTF::StringImpl::hasHash): (WTF::StringImpl::existingHash): (WTF::StringImpl::hash): (WTF::StringImpl::hasOneRef): (WTF::StringImpl::ref): (WTF::StringImpl::deref): (WTF::StringImpl::bufferOwnership): (WTF::StringImpl::isStatic): Moved the static/immortal bit into the bottom bit of the refcount. Now, all lifetime information lives in the refcount field. Moved the other bits into the hash code field. 2011-10-21 Filip Pizlo <fpizlo@apple.com> DFG inlining sometimes fails to reset constant references https://bugs.webkit.org/show_bug.cgi?id=70668 Reviewed by Anders Carlsson. Reset constant references when we need to (new block created) and not when we don't (change of inlining depth). * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::prepareToParseBlock): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::parseCodeBlock): 2011-10-21 Filip Pizlo <fpizlo@apple.com> DFG should have inlining https://bugs.webkit.org/show_bug.cgi?id=69996 Reviewed by Oliver Hunt. Implements inlining that's hooked into the bytecode parser. Only works for calls, for now, though nothing fundamentally prevents us from inlining constructor calls. 2% overall speed-up on all benchmarks. 7% speed-up on V8 (around 34% and 27% on deltablue and richards respectively), neutral on Kraken and SunSpider. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): * bytecode/CodeBlock.h: (JSC::CodeBlock::baselineVersion): (JSC::CodeBlock::setInstructionCount): (JSC::CodeBlock::likelyToTakeSlowCase): (JSC::CodeBlock::couldTakeSlowCase): (JSC::CodeBlock::likelyToTakeSpecialFastCase): (JSC::CodeBlock::likelyToTakeDeepestSlowCase): (JSC::CodeBlock::likelyToTakeAnySlowCase): * bytecode/CodeOrigin.h: (JSC::CodeOrigin::inlineDepthForCallFrame): (JSC::CodeOrigin::inlineDepth): (JSC::CodeOrigin::operator==): (JSC::CodeOrigin::inlineStack): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::generate): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::beginBasicBlock): (JSC::DFG::AbstractState::execute): (JSC::DFG::AbstractState::mergeStateAtTail): * dfg/DFGBasicBlock.h: (JSC::DFG::BasicBlock::BasicBlock): (JSC::DFG::BasicBlock::ensureLocals): (JSC::DFG::UnlinkedBlock::UnlinkedBlock): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::getDirect): (JSC::DFG::ByteCodeParser::get): (JSC::DFG::ByteCodeParser::setDirect): (JSC::DFG::ByteCodeParser::set): (JSC::DFG::ByteCodeParser::getLocal): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::flush): (JSC::DFG::ByteCodeParser::InlineStackEntry::~InlineStackEntry): (JSC::DFG::ByteCodeParser::InlineStackEntry::remapOperand): (JSC::DFG::ByteCodeParser::handleInlining): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::processPhiStack): (JSC::DFG::ByteCodeParser::linkBlock): (JSC::DFG::ByteCodeParser::linkBlocks): (JSC::DFG::ByteCodeParser::handleSuccessor): (JSC::DFG::ByteCodeParser::determineReachability): (JSC::DFG::ByteCodeParser::buildOperandMapsIfNecessary): (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): (JSC::DFG::ByteCodeParser::parseCodeBlock): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGCapabilities.cpp: (JSC::DFG::canHandleOpcodes): (JSC::DFG::canCompileOpcodes): (JSC::DFG::canInlineOpcodes): * dfg/DFGCapabilities.h: (JSC::DFG::mightCompileEval): (JSC::DFG::mightCompileProgram): (JSC::DFG::mightCompileFunctionForCall): (JSC::DFG::mightCompileFunctionForConstruct): (JSC::DFG::mightInlineFunctionForCall): (JSC::DFG::mightInlineFunctionForConstruct): (JSC::DFG::canInlineOpcode): (JSC::DFG::canInlineOpcodes): (JSC::DFG::canInlineFunctionForCall): (JSC::DFG::canInlineFunctionForConstruct): * dfg/DFGGraph.cpp: (JSC::DFG::printWhiteSpace): (JSC::DFG::Graph::dumpCodeOrigin): (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::GetBytecodeBeginForBlock::operator()): (JSC::DFG::Graph::blockIndexForBytecodeOffset): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::decodedCodeMapFor): (JSC::DFG::JITCompiler::linkOSRExits): (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::debugCall): (JSC::DFG::JITCompiler::baselineCodeBlockFor): * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGNode.h: (JSC::DFG::Node::hasVariableAccessData): (JSC::DFG::Node::shouldGenerate): * dfg/DFGOperands.h: (JSC::DFG::Operands::ensureLocals): (JSC::DFG::Operands::setLocal): (JSC::DFG::Operands::getLocal): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::OSRExit::OSRExit): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * interpreter/CallFrame.cpp: (JSC::CallFrame::trueCallerFrameSlow): * jit/JITCall.cpp: (JSC::JIT::compileOpCallSlowCase): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Executable.cpp: (JSC::FunctionExecutable::baselineCodeBlockFor): (JSC::FunctionExecutable::produceCodeBlockFor): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): * runtime/Executable.h: (JSC::FunctionExecutable::profiledCodeBlockFor): (JSC::FunctionExecutable::parameterCount): * runtime/Heuristics.cpp: (JSC::Heuristics::initializeHeuristics): * runtime/Heuristics.h: * runtime/JSFunction.h: 2011-10-21 Mark Hahnenberg <mhahnenberg@apple.com> Add put to the MethodTable https://bugs.webkit.org/show_bug.cgi?id=70439 Reviewed by Oliver Hunt. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/ClassInfo.h: Added put and putByIndex to the MethodTable. * runtime/JSFunction.h: Changed access modifier for put to protected since some subclasses of JSFunction need to reference it in their MethodTables. 2011-10-21 Mark Hahnenberg <mhahnenberg@apple.com> Add finalizer to JSObject https://bugs.webkit.org/show_bug.cgi?id=70336 Reviewed by Darin Adler. * heap/MarkedBlock.cpp: (JSC::MarkedBlock::callDestructor): Skip the call to the destructor if we're a JSFinalObject, since the finalizer takes care of things. * runtime/JSCell.h: (JSC::JSCell::~JSCell): Remove the GC validation due to a conflict with future changes and the fact that we no longer always call the destructor, making the information provided less useful. * runtime/JSObject.cpp: (JSC::JSObject::finalize): Add finalizer for JSObject. (JSC::JSObject::allocatePropertyStorage): The first time we need to allocate out-of-line property storage, we add a finalizer to ourself. * runtime/JSObject.h: 2011-10-21 Simon Hausmann <simon.hausmann@nokia.com> Remove QtScript source code from WebKit. https://bugs.webkit.org/show_bug.cgi?id=64088 Reviewed by Tor Arne Vestbø. Removed dead code that isn't developed anymore. * JavaScriptCore.gypi: * JavaScriptCore.pri: * qt/api/QtScript.pro: Removed. * qt/api/qscriptconverter_p.h: Removed. * qt/api/qscriptengine.cpp: Removed. * qt/api/qscriptengine.h: Removed. * qt/api/qscriptengine_p.cpp: Removed. * qt/api/qscriptengine_p.h: Removed. * qt/api/qscriptfunction.cpp: Removed. * qt/api/qscriptfunction_p.h: Removed. * qt/api/qscriptoriginalglobalobject_p.h: Removed. * qt/api/qscriptprogram.cpp: Removed. * qt/api/qscriptprogram.h: Removed. * qt/api/qscriptprogram_p.h: Removed. * qt/api/qscriptstring.cpp: Removed. * qt/api/qscriptstring.h: Removed. * qt/api/qscriptstring_p.h: Removed. * qt/api/qscriptsyntaxcheckresult.cpp: Removed. * qt/api/qscriptsyntaxcheckresult.h: Removed. * qt/api/qscriptsyntaxcheckresult_p.h: Removed. * qt/api/qscriptvalue.cpp: Removed. * qt/api/qscriptvalue.h: Removed. * qt/api/qscriptvalue_p.h: Removed. * qt/api/qscriptvalueiterator.cpp: Removed. * qt/api/qscriptvalueiterator.h: Removed. * qt/api/qscriptvalueiterator_p.h: Removed. * qt/api/qtscriptglobal.h: Removed. * qt/benchmarks/benchmarks.pri: Removed. * qt/benchmarks/benchmarks.pro: Removed. * qt/benchmarks/qscriptengine/qscriptengine.pro: Removed. * qt/benchmarks/qscriptengine/tst_qscriptengine.cpp: Removed. * qt/benchmarks/qscriptvalue/qscriptvalue.pro: Removed. * qt/benchmarks/qscriptvalue/tst_qscriptvalue.cpp: Removed. * qt/tests/qscriptengine/qscriptengine.pro: Removed. * qt/tests/qscriptengine/tst_qscriptengine.cpp: Removed. * qt/tests/qscriptstring/qscriptstring.pro: Removed. * qt/tests/qscriptstring/tst_qscriptstring.cpp: Removed. * qt/tests/qscriptvalue/qscriptvalue.pro: Removed. * qt/tests/qscriptvalue/tst_qscriptvalue.cpp: Removed. * qt/tests/qscriptvalue/tst_qscriptvalue.h: Removed. * qt/tests/qscriptvalue/tst_qscriptvalue_generated_comparison.cpp: Removed. * qt/tests/qscriptvalue/tst_qscriptvalue_generated_init.cpp: Removed. * qt/tests/qscriptvalue/tst_qscriptvalue_generated_istype.cpp: Removed. * qt/tests/qscriptvalue/tst_qscriptvalue_generated_totype.cpp: Removed. * qt/tests/qscriptvalueiterator/qscriptvalueiterator.pro: Removed. * qt/tests/qscriptvalueiterator/tst_qscriptvalueiterator.cpp: Removed. * qt/tests/tests.pri: Removed. * qt/tests/tests.pro: Removed. 2011-10-21 Zheng Liu <zheng.z.liu@intel.com> bytecompiler sometimes generates incorrect bytecode for put_by_id https://bugs.webkit.org/show_bug.cgi?id=70403 Reviewed by Filip Pizlo. * bytecompiler/NodesCodegen.cpp: (JSC::AssignDotNode::emitBytecode): (JSC::AssignBracketNode::emitBytecode): 2011-10-20 Filip Pizlo <fpizlo@apple.com> DFG should not try to predict argument types by looking at the values of argument registers at the time of compilation https://bugs.webkit.org/show_bug.cgi?id=70578 Reviewed by Oliver Hunt. * bytecode/CodeBlock.cpp: * dfg/DFGDriver.cpp: (JSC::DFG::compile): (JSC::DFG::tryCompile): (JSC::DFG::tryCompileFunction): * dfg/DFGDriver.h: (JSC::DFG::tryCompileFunction): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGGraph.h: * runtime/Executable.cpp: (JSC::FunctionExecutable::compileOptimizedForCall): (JSC::FunctionExecutable::compileOptimizedForConstruct): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): * runtime/Executable.h: (JSC::FunctionExecutable::compileForCall): (JSC::FunctionExecutable::compileForConstruct): (JSC::FunctionExecutable::compileFor): (JSC::FunctionExecutable::compileOptimizedFor): 2011-10-20 Filip Pizlo <fpizlo@apple.com> DFG call optimization handling will fail if the call had been unlinked due to the callee being optimized https://bugs.webkit.org/show_bug.cgi?id=70468 Reviewed by Geoff Garen. If a call had ever been linked, we remember this fact as well as the function to which it was linked even if unlinkIncomingCalls() or unlinkCalls() are called. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): * bytecode/CodeBlock.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGRepatch.cpp: (JSC::DFG::dfgLinkFor): * jit/JIT.cpp: (JSC::JIT::linkFor): 2011-10-20 Yuqiang Xian <yuqiang.xian@intel.com> DFG JIT 32_64 - Fix ByteArray speculation https://bugs.webkit.org/show_bug.cgi?id=70571 Reviewed by Filip Pizlo. * dfg/DFGSpeculativeJIT.h: (JSC::DFG::ValueSource::forPrediction): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-20 Vincent Scheib <scheib@chromium.org> MouseLock compile and run time flags. https://bugs.webkit.org/show_bug.cgi?id=70530 Reviewed by Darin Fisher. * wtf/Platform.h: 2011-10-20 Mark Hahnenberg <mhahnenberg@apple.com> Rename static deleteProperty to deletePropertyByIndex https://bugs.webkit.org/show_bug.cgi?id=70257 Reviewed by Geoffrey Garen. Renaming versions of deleteProperty that use an unsigned as the property name to "deletePropertyByIndex" in preparation for adding them to the MethodTable, which requires unique names for each method. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::deletePropertyVirtual): (JSC::::deletePropertyByIndex): * runtime/Arguments.cpp: (JSC::Arguments::deletePropertyVirtual): (JSC::Arguments::deletePropertyByIndex): * runtime/Arguments.h: * runtime/JSArray.cpp: (JSC::JSArray::deletePropertyVirtual): (JSC::JSArray::deletePropertyByIndex): * runtime/JSArray.h: * runtime/JSCell.cpp: (JSC::JSCell::deletePropertyVirtual): (JSC::JSCell::deletePropertyByIndex): * runtime/JSCell.h: * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::deletePropertyVirtual): (JSC::JSNotAnObject::deletePropertyByIndex): * runtime/JSNotAnObject.h: * runtime/JSObject.cpp: (JSC::JSObject::deletePropertyVirtual): (JSC::JSObject::deletePropertyByIndex): * runtime/JSObject.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::deletePropertyVirtual): (JSC::RegExpMatchesArray::deletePropertyByIndex): 2011-10-20 Filip Pizlo <fpizlo@apple.com> https://bugs.webkit.org/show_bug.cgi?id=70482 DFG-related stubs in the old JIT should not be built if the DFG is disabled Reviewed by Zoltan Herczeg. Aiming for a slight code size/build time reduction if the DFG is not in play. This should also make further DFG development slightly easier since the bodies of these JIT stubs can now safely refer to things that are only declared when the DFG is enabled. * jit/JITStubs.cpp: * jit/JITStubs.h: 2011-10-19 Filip Pizlo <fpizlo@apple.com> DFG ConvertThis emits slow code when the source node is known to be, but not predicted to be, a final object https://bugs.webkit.org/show_bug.cgi?id=70466 Reviewed by Oliver Hunt. Added a new case in ConvertThis compilation. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-19 Filip Pizlo <fpizlo@apple.com> Optimization triggers in the old JIT may sometimes fire repeatedly even though there is no optimization to be done https://bugs.webkit.org/show_bug.cgi?id=70467 Reviewed by Oliver Hunt. If optimize_from_ret does nothing, it delays the next optimization trigger. This is performance-neutral. * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Heuristics.cpp: (JSC::Heuristics::initializeHeuristics): 2011-10-19 Yuqiang Xian <yuqiang.xian@intel.com> DFG JIT 32_64 - remove unnecessary double unboxings in fillDouble/fillSpeculateDouble https://bugs.webkit.org/show_bug.cgi?id=70460 Reviewed by Filip Pizlo. As pointed out by Gavin in bug #70418, when a value is already in memory we can avoid loading it to two GPRs at first and then unboxing them to a FPR. This gives 9% improvement on Kraken if without the change in bug #70418, and 1% if based on the code with bug #70418 change. Performance is neutral in V8 and SunSpider. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::fillDouble): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): 2011-10-19 Gavin Barraclough <barraclough@apple.com> Poisoning of strict caller,arguments inappropriately poisoning "in" https://bugs.webkit.org/show_bug.cgi?id=63398 Reviewed by Oliver Hunt. This fixes the problem by correctly implementing the spec - the error should actually be being thrown from a standard JS getter/setter. This implements spec correct behaviour for strict mode JS functions & bound functions, I'll follow up with a patch to do the same for arguments. * runtime/JSBoundFunction.cpp: (JSC::JSBoundFunction::finishCreation): - Add the poisoned caller/arguments properties. * runtime/JSBoundFunction.h: * runtime/JSFunction.cpp: (JSC::JSFunction::finishCreation): (JSC::JSFunction::getOwnPropertySlot): (JSC::JSFunction::getOwnPropertyDescriptor): (JSC::JSFunction::put): - If the caller/arguments are accessed on a strict mode function, lazily add the ThrowTypeError getter. * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::createThrowTypeError): (JSC::JSGlobalObject::visitChildren): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::throwTypeErrorGetterSetter): - Add a ThrowTypeError type, per ES5 13.2.3. * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncThrowTypeError): * runtime/JSGlobalObjectFunctions.h: - Implementation of ThrowTypeError. * runtime/JSObject.cpp: (JSC::JSObject::initializeGetterSetterProperty): * runtime/JSObject.h: - This function adds a new property (must not exist already) that is an initialized getter/setter. 2011-10-19 Yuqiang Xian <yuqiang.xian@intel.com> DFG JIT 32_64 - improve double boxing/unboxing https://bugs.webkit.org/show_bug.cgi?id=70418 Reviewed by Gavin Barraclough. Double boxing/unboxing in DFG JIT 32_64 is currently implemented inefficiently, which tries to exchange data through memory. On X86 some SSE instructions can help us on such operations with better performance. This improves 32-bit DFG performance by 29% on Kraken, 7% on SunSpider, and 2% on V8, tested on Linux X86 (Core i7 Nehalem). * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::lshiftPacked): (JSC::MacroAssemblerX86Common::rshiftPacked): (JSC::MacroAssemblerX86Common::orPacked): (JSC::MacroAssemblerX86Common::moveInt32ToPacked): (JSC::MacroAssemblerX86Common::movePackedToInt32): * assembler/X86Assembler.h: (JSC::X86Assembler::movd_rr): (JSC::X86Assembler::psllq_i8r): (JSC::X86Assembler::psrlq_i8r): (JSC::X86Assembler::por_rr): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::boxDouble): (JSC::DFG::JITCodeGenerator::unboxDouble): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToNumber): (JSC::DFG::JITCodeGenerator::nonSpeculativeUInt32ToNumber): (JSC::DFG::JITCodeGenerator::nonSpeculativeKnownConstantArithOp): (JSC::DFG::JITCodeGenerator::nonSpeculativeBasicArithOp): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::boxDouble): (JSC::DFG::JITCompiler::unboxDouble): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::convertToDouble): (JSC::DFG::SpeculativeJIT::compile): 2011-10-19 Gyuyoung Kim <gyuyoung.kim@samsung.com> [EFL] Fix DSO linkage of wtf_efl. Unreviewed build fix. Need to add -ldl to jsc_efl (requested by dladdr). * wtf/CMakeListsEfl.txt: 2011-10-19 Geoffrey Garen <ggaren@apple.com> Removed StringImplBase, fusing it into StringImpl https://bugs.webkit.org/show_bug.cgi?id=70443 Reviewed by Gavin Barraclough. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.order: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/CMakeLists.txt: * wtf/text/StringImpl.h: (WTF::StringImpl::StringImpl): (WTF::StringImpl::ref): (WTF::StringImpl::length): * wtf/text/StringImplBase.h: Removed. * wtf/wtf.pri: Removed! 2011-10-19 Mark Hahnenberg <mhahnenberg@apple.com> Add getConstructData to the MethodTable https://bugs.webkit.org/show_bug.cgi?id=70163 Reviewed by Geoffrey Garen. Adding getConstructData to the MethodTable in order to be able to remove all calls to getConstructDataVirtual soon. Part of the process of de-virtualizing JSCell. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/ClassInfo.h: 2011-10-18 Oliver Hunt <oliver@apple.com> Support CanvasPixelArray in the DFG https://bugs.webkit.org/show_bug.cgi?id=70384 Reviewed by Filip Pizlo. Add support for the old CanvasPixelArray optimisations to the DFG. This removes the regression seen in the DFG when using a CPA. * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::store8): (JSC::MacroAssemblerX86Common::truncateDoubleToInt32): * assembler/X86Assembler.h: (JSC::X86Assembler::movb_rm): (JSC::X86Assembler::X86InstructionFormatter::oneByteOp8): * bytecode/PredictedType.cpp: (JSC::predictionToString): (JSC::predictionFromClassInfo): * bytecode/PredictedType.h: (JSC::isByteArrayPrediction): * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::initialize): (JSC::DFG::AbstractState::execute): * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateByteArray): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::compileClampDoubleToByte): (JSC::DFG::SpeculativeJIT::compilePutByValForByteArray): (JSC::DFG::SpeculativeJIT::compileGetByValOnByteArray): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/JSByteArray.h: (JSC::JSByteArray::offsetOfStorage): * wtf/ByteArray.cpp: * wtf/ByteArray.h: (WTF::ByteArray::offsetOfSize): (WTF::ByteArray::offsetOfData): 2011-10-18 Geoffrey Garen <ggaren@apple.com> Some rope cleanup following r97827 https://bugs.webkit.org/show_bug.cgi?id=70398 Reviewed by Oliver Hunt. 9% speedup on date-format-xparb, neutral overall. - Removed RopeImpl*. - Removed JSString::m_fiberCount, since this can be deduced from other data. - Renamed a jsString() variant to jsStringFromArguments for clarity. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.order: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: Removed RopeImpl*. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileGetCharCodeAt): (JSC::DFG::SpeculativeJIT::compileGetByValOnString): * jit/JITInlineMethods.h: (JSC::JIT::emitLoadCharacterString): * jit/JITPropertyAccess.cpp: (JSC::JIT::stringGetByValStubGenerator): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::stringGetByValStubGenerator): * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::loadJSStringArgument): * jit/ThunkGenerators.cpp: (JSC::stringCharLoad): Use a NULL m_value to signal rope-iness, instead of testing m_fiberCount, since m_fiberCount is gone now. * runtime/JSString.cpp: (JSC::JSString::RopeBuilder::expand): (JSC::JSString::visitChildren): (JSC::JSString::resolveRope): (JSC::JSString::resolveRopeSlowCase): (JSC::JSString::outOfMemory): Use a NULL fiber to indicate "last fiber in the vector" instead of testing m_fiberCount, since m_fiberCount is gone now. * runtime/JSString.h: (JSC::RopeBuilder::JSString): (JSC::RopeBuilder::finishCreation): (JSC::RopeBuilder::offsetOfLength): (JSC::RopeBuilder::isRope): (JSC::RopeBuilder::string): Removed m_fiberCount. Renamed jsString => jsStringFromArguments for clarity. * runtime/Operations.h: (JSC::jsStringFromArguments): Renamed. * runtime/RopeImpl.cpp: Removed. * runtime/RopeImpl.h: Removed. * runtime/SmallStrings.cpp: (JSC::SmallStrings::createEmptyString): Switched to StringImpl::empty, which is slightly faster. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncConcat): Updated for rename. * wtf/text/StringImplBase.h: (WTF::StringImplBase::StringImplBase): Removed the concept of an invalid StringImpl, since this was only used by RopeImpl, which is now gone. 2011-10-19 Rafael Antognolli <antognolli@profusion.mobi> [EFL] Fix DSO linkage of jsc_efl. https://bugs.webkit.org/show_bug.cgi?id=70412 Unreviewed build fix. Need to add -ldl to jsc_efl (requested by dladdr). * shell/CMakeListsEfl.txt: 2011-10-18 Geoffrey Garen <ggaren@apple.com> Rolled out last Windows build fix because it was wrong. 2011-10-18 Geoffrey Garen <ggaren@apple.com> Rolled out last Windows build fix because it was wrong. 2011-10-18 Geoffrey Garen <ggaren@apple.com> Try to fix part of the Windows build. Export! 2011-10-18 Geoffrey Garen <ggaren@apple.com> Switched ropes from malloc memory to GC memory https://bugs.webkit.org/show_bug.cgi?id=70364 Reviewed by Gavin Barraclough. ~1% SunSpider speedup. Neutral elsewhere. Removes one cause for strings having C++ destructors. * heap/MarkStack.cpp: (JSC::visitChildren): Call the JSString visitChildren function now, since it's no longer a no-op. * runtime/JSString.cpp: (JSC::JSString::~JSString): Moved this destructor out of line because it's called virtually, so there's no value to inlining. (JSC::JSString::RopeBuilder::expand): Switched RopeBuilder to be a thin initializing wrapper around JSString. JSString now represents ropes directly, rather than relying on an underlying malloc object. (JSC::JSString::visitChildren): Visit our rope fibers, since they're GC objects now. (JSC::JSString::resolveRope): (JSC::JSString::resolveRopeSlowCase): (JSC::JSString::outOfMemory): Updated for operating on JSStrings instead of malloc objects. (JSC::JSString::replaceCharacter): Removed optimizations for substringing ropes and replacing subsections of ropes. We want to reimplement versions of these optimizations in the future, but this patch already has good performance without them. * runtime/JSString.h: (JSC::RopeBuilder::JSString): (JSC::RopeBuilder::finishCreation): (JSC::RopeBuilder::createNull): (JSC::RopeBuilder::create): (JSC::RopeBuilder::createHasOtherOwner): (JSC::jsSingleCharacterString): (JSC::jsSingleCharacterSubstring): (JSC::jsNontrivialString): (JSC::jsString): (JSC::jsSubstring): (JSC::jsOwnedString): Lots of mechanical changes here. The two important things are: (1) The fibers in JSString::m_fibers are JSStrings now, not malloc objects; (2) I simplified the JSString constructor interface to only accept PassRefPtr<StringImpl>, instead of variations on that like UString, reducing refcount churn. * runtime/JSValue.h: * runtime/JSValue.cpp: (JSC::JSValue::toPrimitiveString): Updated this function to return a JSString instead of a UString, since that's what clients want now. * runtime/Operations.cpp: (JSC::jsAddSlowCase): * runtime/Operations.h: (JSC::jsString): * runtime/SmallStrings.cpp: (JSC::SmallStrings::createEmptyString): Updated for interface changes above. * runtime/StringConstructor.cpp: (JSC::constructWithStringConstructor): * runtime/StringObject.h: (JSC::StringObject::create): Don't create a new JSString if we already have a JSString. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncConcat): Updated for interface changes above. 2011-10-18 Gavin Barraclough <barraclough@apple.com> Errrk, fix partial commit of r97825! * runtime/DatePrototype.cpp: (JSC::dateProtoFuncToISOString): 2011-10-18 Gavin Barraclough <barraclough@apple.com> Date.prototype.toISOString fails to throw exception https://bugs.webkit.org/show_bug.cgi?id=70394 Reviewed by Sam Weinig. * runtime/DatePrototype.cpp: (JSC::dateProtoFuncToISOString): - Should throw a range error if the internal value is not finite. 2011-10-18 Mark Hahnenberg <mhahnenberg@apple.com> Rename static put to putByIndex https://bugs.webkit.org/show_bug.cgi?id=70281 Reviewed by Geoffrey Garen. Renaming versions of deleteProperty that use an unsigned as the property name to "deletePropertyByIndex" in preparation for adding them to the MethodTable, which requires unique names for each method. * dfg/DFGOperations.cpp: (JSC::DFG::putByVal): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Arguments.cpp: (JSC::Arguments::putVirtual): (JSC::Arguments::putByIndex): * runtime/Arguments.h: * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncMap): * runtime/JSArray.cpp: (JSC::JSArray::put): (JSC::JSArray::putVirtual): (JSC::JSArray::putByIndex): * runtime/JSArray.h: * runtime/JSByteArray.cpp: (JSC::JSByteArray::putVirtual): (JSC::JSByteArray::putByIndex): * runtime/JSByteArray.h: * runtime/JSCell.cpp: (JSC::JSCell::putVirtual): (JSC::JSCell::putByIndex): * runtime/JSCell.h: * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::putVirtual): (JSC::JSNotAnObject::putByIndex): * runtime/JSNotAnObject.h: * runtime/JSObject.cpp: (JSC::JSObject::putVirtual): (JSC::JSObject::putByIndex): * runtime/JSObject.h: * runtime/RegExpConstructor.cpp: (JSC::RegExpMatchesArray::fillArrayInstance): * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::putVirtual): (JSC::RegExpMatchesArray::putByIndex): 2011-10-18 Gavin Barraclough <barraclough@apple.com> Array.prototype methods missing exception checks https://bugs.webkit.org/show_bug.cgi?id=70360 Reviewed by Geoff Garen. Missing exception checks after calls to the static getProperty helper, these may result in the wrong exception being thrown (or an ASSERT being hit, as is currently the case running test-262). No performance impact. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncConcat): (JSC::arrayProtoFuncReverse): (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSlice): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): (JSC::arrayProtoFuncReduce): (JSC::arrayProtoFuncReduceRight): (JSC::arrayProtoFuncIndexOf): (JSC::arrayProtoFuncLastIndexOf): 2011-10-18 Adam Barth <abarth@webkit.org> Always enable ENABLE(XPATH) https://bugs.webkit.org/show_bug.cgi?id=70217 Reviewed by Eric Seidel. * Configurations/FeatureDefines.xcconfig: 2011-10-18 Gavin Barraclough <barraclough@apple.com> Indexed arguments on the Arguments object should be enumerable. https://bugs.webkit.org/show_bug.cgi?id=70302 Reviewed by Sam Weinig. See ECMA-262 5.1 chapter 10.6 step 11b. This is visible through a number of means, including Object.keys, Object.getOwnPropertyDescriptor, and operator in. * runtime/Arguments.cpp: (JSC::Arguments::getOwnPropertyDescriptor): - The 'enumerable' property should be true for indexed arguments. (JSC::Arguments::getOwnPropertyNames): - Don't guard the adding of indexed properties with 'IncludeDontEnumProperties'. 2011-10-18 Gustavo Noronha Silva <gns@gnome.org> Fix distcheck. * GNUmakefile.list.am: fix a typo and add a missing header to the list. 2011-10-18 Balazs Kelemen <kbalazs@webkit.org> ParallelJobs: maximum number of threads should be determined dynamically https://bugs.webkit.org/show_bug.cgi?id=68540 Reviewed by Zoltan Herczeg. Add logic to determine the number of cores and use this as the maximum number of threads. The implementation currently covers Linux, Darwin, Windows, AIX, Solaris, OpenBSD and NetBSD. The patch was tested on Linux, Mac and Windows which was enough to cover all code path. It should work on the rest accoring to the documentation of those OS's. The hard coded constant is still used on uncovered OS's which should be fixed in the future. * wtf/ParallelJobs.h: Removed the default value of the requestedJobNumber argument because clients should always fill it and the 0 default value was incorrect anyway. (WTF::ParallelJobs::ParallelJobs): * wtf/ParallelJobsGeneric.cpp: (WTF::ParallelEnvironment::determineMaxNumberOfParallelThreads): * wtf/ParallelJobsGeneric.h: (WTF::ParallelEnvironment::ParallelEnvironment): 2011-10-17 Gavin Barraclough <barraclough@apple.com> Reverted r997709, this caused test failures. * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/JSObject.cpp: (JSC::JSObject::hasProperty): (JSC::JSObject::hasOwnProperty): 2011-10-17 Ryosuke Niwa <rniwa@webkit.org> Rename deregister* to unregister* https://bugs.webkit.org/show_bug.cgi?id=70272 Reviewed by Darin Adler. Renamed deregisterWeakMap to unregisterWeakMap. * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::unregisterWeakMap): 2011-10-17 Gavin Barraclough <barraclough@apple.com> Poisoning of strict caller/arguments inappropriately poisoning "in" https://bugs.webkit.org/show_bug.cgi?id=63398 Reviewed by Sam Weinig. The problem here is that the has[Own]Property methods get the slot rather than the descriptor, and getting the slot may cause the property to be eagerly accessed. * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): - We don't expect hasProperty to ever throw. If it does, it won't get caught (since it is after the exception check), so ASSERT to guard against this. * runtime/JSObject.cpp: (JSC::JSObject::hasProperty): (JSC::JSObject::hasOwnProperty): - These methods should not check for the presence of the descriptor; never get the value. 2011-10-17 Gavin Barraclough <barraclough@apple.com> Exception ordering in String.prototype.replace https://bugs.webkit.org/show_bug.cgi?id=70290 If pattern is not a regexp, it should be converted toString before the replacement value has it's toString conversion called. Reviewed by Oliver Hunt. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncReplace): 2011-10-17 Filip Pizlo <fpizlo@apple.com> DFG bytecode parser should understand inline stacks https://bugs.webkit.org/show_bug.cgi?id=70278 Reviewed by Oliver Hunt. The DFG bytecode parser is now capable of parsing multiple code blocks at once. This remains turned off since not all inlining functionality is implemented. This required making a few changes elsewhere in the system. The bytecode parser now may do some of the same things that the bytecode generator does, like allocating constants and identifiers. Basic block linking relies on bytecode indices, which are only meaningful within the context of one basic block. This is fine, so long as linking is done eagerly whenever switching from one code block to another. * bytecode/CodeOrigin.h: (JSC::CodeOrigin::CodeOrigin): * bytecompiler/BytecodeGenerator.h: * dfg/DFGBasicBlock.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::get): (JSC::DFG::ByteCodeParser::set): (JSC::DFG::ByteCodeParser::getThis): (JSC::DFG::ByteCodeParser::setThis): (JSC::DFG::ByteCodeParser::currentCodeOrigin): (JSC::DFG::ByteCodeParser::getPrediction): (JSC::DFG::ByteCodeParser::makeSafe): (JSC::DFG::ByteCodeParser::makeDivSafe): (JSC::DFG::ByteCodeParser::InlineStackEntry::executable): (JSC::DFG::ByteCodeParser::InlineStackEntry::~InlineStackEntry): (JSC::DFG::ByteCodeParser::InlineStackEntry::remapOperand): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::linkBlock): (JSC::DFG::ByteCodeParser::linkBlocks): (JSC::DFG::ByteCodeParser::setupPredecessors): (JSC::DFG::ByteCodeParser::buildOperandMapsIfNecessary): (JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry): (JSC::DFG::ByteCodeParser::parseCodeBlock): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGraph.h: (JSC::DFG::GetBytecodeBeginForBlock::GetBytecodeBeginForBlock): (JSC::DFG::GetBytecodeBeginForBlock::operator()): (JSC::DFG::Graph::blockIndexForBytecodeOffset): * dfg/DFGNode.h: * runtime/Identifier.h: (JSC::IdentifierMapIndexHashTraits::emptyValue): * runtime/JSValue.h: * wtf/StdLibExtras.h: (WTF::binarySearchWithFunctor): 2011-10-17 Gavin Barraclough <barraclough@apple.com> Incorrect behavior from String match/search & undefined pattern https://bugs.webkit.org/show_bug.cgi?id=70286 Reviewed by Sam weinig. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncMatch): - In case of undefined, pattern is "". (JSC::stringProtoFuncSearch): - In case of undefined, pattern is "". 2011-10-17 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=70207 After deleting __defineSetter__, it is absent but appears in name list Reviewed by Darin Adler. * runtime/JSObject.cpp: (JSC::JSObject::getOwnPropertyNames): - This should check whether static functions have been reified. 2011-10-17 Geoffrey Garen <ggaren@apple.com> Mac build fix. * JavaScriptCore.exp: Export! 2011-10-17 Geoffrey Garen <ggaren@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Export! 2011-10-17 Geoffrey Garen <ggaren@apple.com> Windows build fix. * heap/HandleStack.cpp: Added a missing #include. 2011-10-17 Geoffrey Garen <ggaren@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Removed no longer existant symbol. * heap/MarkStack.cpp: (JSC::MarkStackArray::shrinkAllocation): Cast to the right type. 2011-10-17 Geoffrey Garen <ggaren@apple.com> Simplified GC marking logic https://bugs.webkit.org/show_bug.cgi?id=70258 Reviewed by Filip Pizlo. No perf. change. This is a first step toward GC allocating string backing stores, starting with ropes. It also enables future simplifications and optimizations. - Replaced some complex mark stack logic with a simple linear stack of JSCell pointers. - Replaced logic for short-circuiting marking based on JSType and/or Structure flags with special cases for object, array, and string. - Fiddled with inlining for better codegen. * JavaScriptCore.exp: * heap/HandleStack.cpp: Build! * heap/Heap.cpp: (JSC::Heap::Heap): Provide more vptrs to SlotVisitor, for use in marking. * heap/HeapRootVisitor.h: Removed unused functions that no longer build. * heap/MarkStack.cpp: (JSC::MarkStackArray::MarkStackArray): (JSC::MarkStackArray::~MarkStackArray): (JSC::MarkStackArray::expand): (JSC::MarkStackArray::shrinkAllocation): (JSC::MarkStack::reset): (JSC::visitChildren): (JSC::SlotVisitor::drain): * heap/MarkStack.h: (JSC::MarkStack::MarkStack): (JSC::MarkStack::~MarkStack): (JSC::MarkStackArray::append): (JSC::MarkStackArray::removeLast): (JSC::MarkStackArray::isEmpty): (JSC::MarkStack::append): (JSC::MarkStack::appendUnbarrieredPointer): (JSC::MarkStack::internalAppend): Replaced complex mark set logic with simple linear stack. * heap/SlotVisitor.h: (JSC::SlotVisitor::SlotVisitor): Updated for above changes. * runtime/JSArray.cpp: (JSC::JSArray::visitChildren): * runtime/JSArray.h: * runtime/JSObject.cpp: (JSC::JSObject::visitChildren): * runtime/JSObject.h: Don't inline visitChildren; it's too big. * runtime/Structure.h: (JSC::MarkStack::internalAppend): Nixed the short-circuit for CompoundType because it prevented strings from owning GC pointers. * runtime/WriteBarrier.h: (JSC::MarkStack::appendValues): No need to validate; internalAppend will do that for us. 2011-10-17 Adam Roben <aroben@apple.com> Windows build fix after r97536, part 3 * runtime/JSAPIValueWrapper.h: * runtime/JSObject.h: Use JS_EXPORTDATA to export the s_info members. 2011-10-17 Adam Roben <aroben@apple.com> Interpreter build fix after r97564 * runtime/Executable.cpp: (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): Moved declaration of globalData variable into ENABLE(JIT) blocks, since it is only used there. 2011-10-17 Adam Roben <aroben@apple.com> Windows build fix after r97536, part 2 * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Added back JSC::setUpStaticFunctionSlot with its new mangled name. SOrted the rest of the file while I was at it. 2011-10-17 Adam Roben <aroben@apple.com> Windows build fix after r97536 * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Removed export of JSC::setUpStaticFunctionSlot, which no longer exists. Also removed incorrect exports of s_info members, which need to be exported via JS_EXPORTDATA instead. 2011-10-17 Patrick Gansterer <paroga@webkit.org> Interpreter build fix after r97436, r97506, r97532 and r97537. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): 2011-10-16 Adam Barth <abarth@webkit.org> Always disable ENABLE(ON_FIRST_TEXTAREA_FOCUS_SELECT_ALL) and delete associated code https://bugs.webkit.org/show_bug.cgi?id=70216 Reviewed by Eric Seidel. * wtf/Platform.h: 2011-10-16 Noel Gordon <noel.gordon@gmail.com> [chromium] Remove PageAllocatorSymbian.h, OSAllocatorSymbian.cpp, gtk/ThreadingGtk.cpp from gyp project files https://bugs.webkit.org/show_bug.cgi?id=70205 Reviewed by James Robinson. wtf/PageAllocatorSymbian.h and wtf/OSAllocatorSymbian.cpp were removed in r97557. wtf/gtk/ThreadingGtk.cpp was removed in r97269. * JavaScriptCore.gypi: 2011-10-16 Adam Barth <abarth@webkit.org> Always enable ENABLE(DOM_STORAGE) https://bugs.webkit.org/show_bug.cgi?id=70189 Reviewed by Eric Seidel. * Configurations/FeatureDefines.xcconfig: 2011-10-15 Dan Horák <dan@danny.cz> The s390 and s390x architectures both use 64-bit double type that conforms to the IEEE-754 standard. https://bugs.webkit.org/show_bug.cgi?id=69940 Reviewed by Gavin Barraclough. * wtf/dtoa/utils.h: 2011-10-14 Filip Pizlo <fpizlo@apple.com> FunctionExecutable should expose the ability to create unattached FunctionCodeBlocks https://bugs.webkit.org/show_bug.cgi?id=70157 Reviewed by Geoff Garen. Added FunctionExecutable::produceCodeBlockFor() and rewired compileForCallInternal() and compileForConstructInternal() to use this method. This required more cleanly exposing some of CodeBlock's tiering functionality and moving the CompilationKind enum to Executable.h, as this was the easiest way to make it available to the declarations/definitions of CodeBlock, FunctionExecutable, and BytecodeGenerator. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::copyDataFrom): (JSC::CodeBlock::copyDataFromAlternative): * bytecode/CodeBlock.h: (JSC::CodeBlock::setAlternative): * bytecompiler/BytecodeGenerator.h: * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::produceCodeBlockFor): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): * runtime/Executable.h: (JSC::FunctionExecutable::codeBlockFor): 2011-10-15 Laszlo Gombos <laszlo.1.gombos@nokia.com> [Qt] [Symbian] Remove support for the Symbian platform for the QtWebKit port https://bugs.webkit.org/show_bug.cgi?id=69920 Reviewed by Kenneth Rohde Christiansen. * JavaScriptCore.pri: * JavaScriptCore.pro: * heap/MarkStack.h: (JSC::::shrinkAllocation): * jit/ExecutableAllocator.cpp: * jit/ExecutableAllocator.h: (JSC::ExecutableAllocator::cacheFlush): * jit/JITStubs.cpp: * jsc.pro: * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncToString): * runtime/DatePrototype.cpp: (JSC::formatLocaleDate): * runtime/StringPrototype.cpp: (JSC::stringProtoFuncLastIndexOf): * runtime/TimeoutChecker.cpp: (JSC::getCPUTime): * wtf/Assertions.cpp: * wtf/Assertions.h: * wtf/Atomics.h: * wtf/MathExtras.h: * wtf/OSAllocator.h: (WTF::OSAllocator::decommitAndRelease): * wtf/OSAllocatorSymbian.cpp: Removed. * wtf/OSRandomSource.cpp: (WTF::cryptographicallyRandomValuesFromOS): * wtf/PageAllocation.h: * wtf/PageAllocatorSymbian.h: Removed. * wtf/PageBlock.cpp: * wtf/Platform.h: * wtf/StackBounds.cpp: * wtf/wtf.pri: 2011-10-15 Yuqiang Xian <yuqiang.xian@intel.com> Trivial fix for a missing change in r97512 https://bugs.webkit.org/show_bug.cgi?id=70166 Reviewed by Gavin Barraclough. * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::link): 2011-10-14 Mark Hahnenberg <mhahnenberg@apple.com> Rename getOwnPropertySlot to getOwnPropertySlotVirtual https://bugs.webkit.org/show_bug.cgi?id=69810 Reviewed by Geoffrey Garen. Renamed the virtual version of getOwnPropertySlot to getOwnPropertySlotVirtual in preparation for when we add the static getOwnPropertySlot to the MethodTable in ClassInfo. Also added a few static getOwnPropertySlot functions where they had been overlooked before (especially in CodeGeneratorJS.pm). * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::getOwnPropertySlotVirtual): (JSC::::getOwnPropertySlot): (JSC::::getOwnPropertyDescriptor): (JSC::::staticFunctionGetter): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::getOwnPropertySlotVirtual): (JSC::DebuggerActivation::getOwnPropertySlot): * debugger/DebuggerActivation.h: * runtime/Arguments.cpp: (JSC::Arguments::getOwnPropertySlotVirtual): (JSC::Arguments::getOwnPropertySlot): * runtime/Arguments.h: * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::getOwnPropertySlotVirtual): (JSC::ArrayConstructor::getOwnPropertySlot): * runtime/ArrayConstructor.h: * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::getOwnPropertySlotVirtual): * runtime/ArrayPrototype.h: * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::getOwnPropertySlotVirtual): * runtime/BooleanPrototype.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::getOwnPropertySlotVirtual): * runtime/DateConstructor.h: * runtime/DatePrototype.cpp: (JSC::DatePrototype::getOwnPropertySlotVirtual): * runtime/DatePrototype.h: * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::getOwnPropertySlotVirtual): * runtime/ErrorPrototype.h: * runtime/JSActivation.cpp: (JSC::JSActivation::getOwnPropertySlotVirtual): * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::getOwnPropertySlotVirtual): (JSC::JSArray::getOwnPropertySlot): * runtime/JSArray.h: * runtime/JSBoundFunction.cpp: (JSC::JSBoundFunction::getOwnPropertySlotVirtual): * runtime/JSBoundFunction.h: * runtime/JSByteArray.cpp: (JSC::JSByteArray::getOwnPropertySlotVirtual): * runtime/JSByteArray.h: * runtime/JSCell.cpp: (JSC::JSCell::getOwnPropertySlotVirtual): * runtime/JSCell.h: * runtime/JSFunction.cpp: (JSC::JSFunction::getOwnPropertySlotVirtual): (JSC::JSFunction::getOwnPropertyDescriptor): (JSC::JSFunction::getOwnPropertyNames): (JSC::JSFunction::put): * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::getOwnPropertySlotVirtual): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::hasOwnPropertyForWrite): * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::getOwnPropertySlotVirtual): * runtime/JSNotAnObject.h: * runtime/JSONObject.cpp: (JSC::Stringifier::Holder::appendNextProperty): (JSC::JSONObject::getOwnPropertySlotVirtual): (JSC::Walker::walk): * runtime/JSONObject.h: * runtime/JSObject.cpp: (JSC::JSObject::getOwnPropertySlotVirtual): (JSC::JSObject::getOwnPropertySlot): (JSC::JSObject::hasOwnProperty): * runtime/JSObject.h: (JSC::JSObject::getOwnPropertySlotVirtual): (JSC::JSCell::fastGetOwnPropertySlot): (JSC::JSObject::getPropertySlot): (JSC::JSValue::get): * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::getOwnPropertySlotVirtual): * runtime/JSStaticScopeObject.h: * runtime/JSString.cpp: (JSC::JSString::getOwnPropertySlotVirtual): (JSC::JSString::getOwnPropertySlot): * runtime/JSString.h: * runtime/Lookup.h: (JSC::getStaticPropertySlot): (JSC::getStaticFunctionSlot): (JSC::getStaticValueSlot): * runtime/MathObject.cpp: (JSC::MathObject::getOwnPropertySlotVirtual): * runtime/MathObject.h: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::getOwnPropertySlotVirtual): * runtime/NumberConstructor.h: * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::getOwnPropertySlotVirtual): * runtime/NumberPrototype.h: * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::getOwnPropertySlotVirtual): * runtime/ObjectConstructor.h: * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::getOwnPropertySlotVirtual): * runtime/ObjectPrototype.h: * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::getOwnPropertySlotVirtual): * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::getOwnPropertySlotVirtual): * runtime/RegExpObject.cpp: (JSC::RegExpObject::getOwnPropertySlotVirtual): * runtime/RegExpObject.h: * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::getOwnPropertySlotVirtual): * runtime/RegExpPrototype.h: * runtime/StringConstructor.cpp: (JSC::StringConstructor::getOwnPropertySlotVirtual): * runtime/StringConstructor.h: * runtime/StringObject.cpp: (JSC::StringObject::getOwnPropertySlotVirtual): * runtime/StringObject.h: * runtime/StringPrototype.cpp: (JSC::StringPrototype::getOwnPropertySlotVirtual): * runtime/StringPrototype.h: 2011-10-14 Gavin Barraclough <baraclough@apple.com> Most built-in properties are not deletable https://bugs.webkit.org/show_bug.cgi?id=61014 Reviewed by Filip Pizlo. Our static hash tables don't allow for deleting properties. This is the cause of a bunch of expected failures in LayoutTests/sputnik. This fixes the problem by reifying all static functions immediately prior to the first deletion. Reification is tracked by a flag on the structure, so properties will no longer 'bounce-back' on later access. Theoretically there could probably also be an issue with custom accessor properties, but we probably do not really require any of these to be Configurable anyway. I'll follow up with a separate patch to address this. * runtime/ClassInfo.h: (JSC::ClassInfo::hasStaticProperties): - detects static property tables. * runtime/JSObject.cpp: (JSC::JSObject::deleteProperty): - call reifyStaticFunctions before deletion. (JSC::JSObject::reifyStaticFunctions): - If the class has static functions, set them up now. * runtime/JSObject.h: (JSC::JSObject::staticFunctionsReified): - returns true if static functions have been reified, and as such should no longer be added. * runtime/Lookup.cpp: (JSC::setUpStaticFunctionSlot): - If static functions have been reified do not add. * runtime/Lookup.h: (JSC::HashTable::ConstIterator::ConstIterator): (JSC::HashTable::ConstIterator::operator->): (JSC::HashTable::ConstIterator::operator*): (JSC::HashTable::ConstIterator::operator!=): (JSC::HashTable::ConstIterator::operator++): (JSC::HashTable::ConstIterator::skipInvalidKeys): (JSC::HashTable::begin): (JSC::HashTable::end): (JSC::getStaticPropertySlot): (JSC::getStaticPropertyDescriptor): (JSC::getStaticFunctionSlot): (JSC::getStaticFunctionDescriptor): - setUpStaticFunctionSlot may not add, returns a bool. (JSC::lookupPut): - remove redundant branch. * runtime/Structure.cpp: (JSC::Structure::Structure): - initialize new flag in constructors. * runtime/Structure.h: (JSC::Structure::staticFunctionsReified): (JSC::Structure::setStaticFunctionsReified): - added flag 2011-10-14 Mark Hahnenberg <mhahnenberg@apple.com> Rename virtual put to putVirtual https://bugs.webkit.org/show_bug.cgi?id=69851 Reviewed by Darin Adler. Renamed virtual versions of put to putVirtual in prepration for adding the static put to the MethodTable in ClassInfo since the compiler gets mad if the virtual and static versions have the same name. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::putVirtual): * API/JSObjectRef.cpp: (JSObjectSetProperty): (JSObjectSetPropertyAtIndex): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::putVirtual): (JSC::DebuggerActivation::put): * debugger/DebuggerActivation.h: * dfg/DFGOperations.cpp: (JSC::DFG::putByVal): * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jsc.cpp: (GlobalObject::finishCreation): * runtime/Arguments.cpp: (JSC::Arguments::putVirtual): * runtime/Arguments.h: * runtime/ArrayPrototype.cpp: (JSC::putProperty): (JSC::arrayProtoFuncConcat): (JSC::arrayProtoFuncPush): (JSC::arrayProtoFuncReverse): (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSlice): (JSC::arrayProtoFuncSort): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): (JSC::arrayProtoFuncFilter): (JSC::arrayProtoFuncMap): * runtime/JSActivation.cpp: (JSC::JSActivation::putVirtual): * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::putVirtual): (JSC::JSArray::putSlowCase): (JSC::JSArray::push): (JSC::JSArray::shiftCount): (JSC::JSArray::unshiftCount): * runtime/JSArray.h: * runtime/JSByteArray.cpp: (JSC::JSByteArray::putVirtual): * runtime/JSByteArray.h: * runtime/JSCell.cpp: (JSC::JSCell::putVirtual): (JSC::JSCell::put): * runtime/JSCell.h: * runtime/JSFunction.cpp: (JSC::JSFunction::putVirtual): * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::putVirtual): (JSC::JSGlobalObject::putWithAttributes): * runtime/JSGlobalObject.h: * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::putVirtual): * runtime/JSNotAnObject.h: * runtime/JSONObject.cpp: (JSC::Walker::walk): * runtime/JSObject.cpp: (JSC::JSObject::putVirtual): (JSC::JSObject::put): (JSC::JSObject::defineOwnProperty): * runtime/JSObject.h: (JSC::JSValue::put): * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::putVirtual): * runtime/JSStaticScopeObject.h: * runtime/Lookup.h: (JSC::lookupPut): * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::putVirtual): * runtime/ObjectPrototype.h: * runtime/RegExpConstructor.cpp: (JSC::RegExpMatchesArray::fillArrayInstance): (JSC::RegExpConstructor::putVirtual): * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::putVirtual): * runtime/RegExpObject.cpp: (JSC::RegExpObject::putVirtual): * runtime/RegExpObject.h: * runtime/StringObject.cpp: (JSC::StringObject::putVirtual): * runtime/StringObject.h: * runtime/StringPrototype.cpp: (JSC::stringProtoFuncSplit): 2011-10-13 Filip Pizlo <fpizlo@apple.com> Reflective Arguments retrieval should be hardened for the possibility of inlining https://bugs.webkit.org/show_bug.cgi?id=70068 Reviewed by Oliver Hunt. CodeBlock can now track, as part of its RareData, the virtual inline stack at callsites. CallFrame walking can now rematerialize "inline" CallFrames by combining the meta-data in CodeBlock with the information already in the JS stack. Arguments can now safely retrieve the arguments from inline CallFrames. The DFG already had the notion of a "CodeOrigin" in preparation for inlining. This notion will now be saved into the CodeBlock, if the DFG had done inlining. So, CodeOrigin has been moved to bytecode/ and has been changed to behave more like a struct since that is how it's meant to be used. * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.h: (JSC::CodeBlock::inlineCallFrames): (JSC::CodeBlock::codeOrigins): (JSC::CodeBlock::hasCodeOrigins): (JSC::CodeBlock::codeOriginForReturn): * bytecode/CodeOrigin.h: Added. (JSC::CodeOrigin::CodeOrigin): (JSC::CodeOrigin::isSet): (JSC::getCallReturnOffsetForCodeOrigin): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGNode.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * interpreter/CallFrame.cpp: (JSC::CallFrame::isInlineCallFrame): (JSC::CallFrame::trueCallerFrame): * interpreter/CallFrame.h: (JSC::ExecState::inlineCallFrame): (JSC::ExecState::setInlineCallFrame): (JSC::ExecState::isInlineCallFrame): (JSC::ExecState::trueCallerFrame): * interpreter/Interpreter.cpp: (JSC::Interpreter::findFunctionCallFrame): * interpreter/Register.h: (JSC::Register::operator=): (JSC::Register::inlineCallFrame): * runtime/Arguments.h: (JSC::Arguments::getArgumentsData): (JSC::Arguments::finishCreationButDontCopyRegisters): (JSC::Arguments::finishCreation): (JSC::Arguments::finishCreationAndCopyRegisters): * runtime/Executable.h: (JSC::FunctionExecutable::parameterCount): 2011-10-14 Mark Hahnenberg <mhahnenberg@apple.com> Rename virtual deleteProperty to deletePropertyVirtual https://bugs.webkit.org/show_bug.cgi?id=69884 Reviewed by Darin Adler. Renamed virtual versions of deleteProperty to deletePropertyVirtual in prepration for adding the static deleteProperty to the MethodTable in ClassInfo since the compiler gets mad if the virtual and static versions have the same name. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::deletePropertyVirtual): (JSC::::deleteProperty): * API/JSObjectRef.cpp: (JSObjectDeleteProperty): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::deletePropertyVirtual): (JSC::DebuggerActivation::deleteProperty): * debugger/DebuggerActivation.h: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Arguments.cpp: (JSC::Arguments::deletePropertyVirtual): * runtime/Arguments.h: * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncPop): (JSC::arrayProtoFuncReverse): (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): * runtime/JSActivation.cpp: (JSC::JSActivation::deletePropertyVirtual): * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::deletePropertyVirtual): (JSC::JSArray::deleteProperty): * runtime/JSArray.h: * runtime/JSCell.cpp: (JSC::JSCell::deletePropertyVirtual): (JSC::JSCell::deleteProperty): * runtime/JSCell.h: * runtime/JSFunction.cpp: (JSC::JSFunction::deletePropertyVirtual): * runtime/JSFunction.h: * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::deletePropertyVirtual): * runtime/JSNotAnObject.h: * runtime/JSONObject.cpp: (JSC::Walker::walk): * runtime/JSObject.cpp: (JSC::JSObject::deletePropertyVirtual): (JSC::JSObject::deleteProperty): (JSC::JSObject::defineOwnProperty): * runtime/JSObject.h: * runtime/JSVariableObject.cpp: (JSC::JSVariableObject::deletePropertyVirtual): * runtime/JSVariableObject.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::deletePropertyVirtual): * runtime/StrictEvalActivation.cpp: (JSC::StrictEvalActivation::deletePropertyVirtual): * runtime/StrictEvalActivation.h: * runtime/StringObject.cpp: (JSC::StringObject::deletePropertyVirtual): * runtime/StringObject.h: 2011-10-14 Peter Beverloo <peter@chromium.org> [Chromium] Inherit settings from Chromium's envsetup.sh, address a NDK todo https://bugs.webkit.org/show_bug.cgi?id=70028 Reviewed by Adam Barth. * JavaScriptCore.gyp/JavaScriptCore.gyp: 2011-10-14 Yuqiang Xian <yuqiang.xian@intel.com> DFG JIT 32_64 - Performance fix for ResolveGlobal https://bugs.webkit.org/show_bug.cgi?id=70096 Reviewed by Gavin Barraclough. Structure check of global object should be a pointer comparison instead of a tag and payload pair comparison. This fix improves SunSpider by 7% on Linux 32, with bitops-bitwise-and improved by 4.75X. Also two trivial fixes for successful 32-bit build are included. * dfg/DFGSpeculativeJIT.cpp: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-13 Filip Pizlo <fpizlo@apple.com> Speculation failures in ValueToInt32 are causing a 2x slow-down in Kraken/stanford-crypto-pbkdf2 https://bugs.webkit.org/show_bug.cgi?id=70089 Reviewed by Gavin Barraclough. If we can't truncate to Int32 using machine code, then don't fail speculation. Just call JSC::toInt32. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileValueToInt32): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-13 Mark Hahnenberg <mhahnenberg@apple.com> Rename virtual getConstructData to getConstructDataVirtual https://bugs.webkit.org/show_bug.cgi?id=69872 Reviewed by Geoffrey Garen. Renamed virtual getConstructData functions to getConstructDataVirtual to avoid conflicts when we add static getConstructData to the MethodTable. * API/JSCallbackConstructor.cpp: (JSC::JSCallbackConstructor::getConstructDataVirtual): * API/JSCallbackConstructor.h: * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::getConstructDataVirtual): * API/JSObjectRef.cpp: (JSObjectIsConstructor): (JSObjectCallAsConstructor): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * dfg/DFGOperations.cpp: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::getConstructDataVirtual): * runtime/ArrayConstructor.h: * runtime/BooleanConstructor.cpp: (JSC::BooleanConstructor::getConstructDataVirtual): * runtime/BooleanConstructor.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::getConstructDataVirtual): * runtime/DateConstructor.h: * runtime/Error.h: (JSC::StrictModeTypeErrorFunction::getConstructDataVirtual): * runtime/ErrorConstructor.cpp: (JSC::ErrorConstructor::getConstructDataVirtual): * runtime/ErrorConstructor.h: * runtime/FunctionConstructor.cpp: (JSC::FunctionConstructor::getConstructDataVirtual): * runtime/FunctionConstructor.h: * runtime/JSCell.cpp: (JSC::JSCell::getConstructDataVirtual): * runtime/JSCell.h: (JSC::getConstructData): * runtime/JSFunction.cpp: (JSC::JSFunction::getConstructDataVirtual): * runtime/JSFunction.h: * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::getConstructDataVirtual): * runtime/NativeErrorConstructor.h: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::getConstructDataVirtual): * runtime/NumberConstructor.h: * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::getConstructDataVirtual): * runtime/ObjectConstructor.h: * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::getConstructDataVirtual): * runtime/RegExpConstructor.h: * runtime/StringConstructor.cpp: (JSC::StringConstructor::getConstructDataVirtual): * runtime/StringConstructor.h: 2011-10-13 Filip Pizlo <fpizlo@apple.com> Rubber stamped Stephanie Lewis. DFG_ENABLE() macro was always returning false. * dfg/DFGNode.h: 2011-10-13 Gavin Barraclough <baraclough@apple.com> Speculative build fix for !DFG builds. * jit/JIT.cpp: (JSC::JIT::privateCompile): 2011-10-13 Oliver Hunt <oliver@apple.com> Fix performance of ValueToInt32 node when predicting double https://bugs.webkit.org/show_bug.cgi?id=70063 Reviewed by Filip Pizlo. Currently we fail to inline double to int conversion when performing a ValueToInt32 operation on a value we predict to be a double. * dfg/DFGAbstractState.cpp: (JSC::DFG::AbstractState::execute): Apply correct filter for the double prediction path * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): Support double parameters even when value has been spilled. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileValueToInt32): Moved old valueToInt32 code to this function, and added path for double prediction * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): Made the two implementations of ValueToInt32 call a single shared compileValueToInt32 function. 2011-10-13 Chris Marrin <cmarrin@apple.com> Sync requestAnimationFrame callback to CVDisplayLink on Mac https://bugs.webkit.org/show_bug.cgi?id=68911 Reviewed by Simon Fraser. Add REQUEST_ANIMATION_FRAME_DISPLAY_MONITOR for implementations that use the DisplayRefreshMonitor logic. * wtf/Platform.h: 2011-10-13 Gavin Barraclough <baraclough@apple.com> DFG JIT should not be using ENABLE macro to enable features https://bugs.webkit.org/show_bug.cgi?id=70060 Reviewed by Oliver Hunt. The ENABLE macro is only intended to be used to detect features that are configured in Platform.h. Using its to detect settings defined in other headers is an error. The problem is that the ENABLE macro checks if the value is defined, so will silently return false if you fail to include the header defining the switch. This is not a problem if (1) the settings are defined in the same header that defines the macro that tests them, or (2) the header is included everywhere. In the case of ENABLE settings defined in Platform.h, both are true! To make this clear, add an explicit DFG_ENABLE macro. * bytecode/CodeBlock.cpp: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getPrediction): (JSC::DFG::ByteCodeParser::makeSafe): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGJITCodeGenerator.cpp: * dfg/DFGJITCodeGenerator.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::noticeOSREntry): * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::link): * dfg/DFGNode.h: * dfg/DFGOSREntry.cpp: (JSC::DFG::prepareOSREntry): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::fixpoint): (JSC::DFG::Propagator::propagateArithNodeFlags): (JSC::DFG::Propagator::propagateArithNodeFlagsForward): (JSC::DFG::Propagator::propagateArithNodeFlagsBackward): (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::propagatePredictionsForward): (JSC::DFG::Propagator::propagatePredictionsBackward): (JSC::DFG::Propagator::propagatePredictions): (JSC::DFG::Propagator::toDouble): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::fixup): (JSC::DFG::Propagator::startIndexForChildren): (JSC::DFG::Propagator::endIndexForPureCSE): (JSC::DFG::Propagator::setReplacement): (JSC::DFG::Propagator::eliminate): (JSC::DFG::Propagator::performNodeCSE): (JSC::DFG::Propagator::localCSE): (JSC::DFG::Propagator::allocateVirtualRegisters): (JSC::DFG::Propagator::performBlockCFA): (JSC::DFG::Propagator::performForwardCFA): (JSC::DFG::Propagator::globalCFA): * dfg/DFGScoreBoard.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compile): * jit/JIT.cpp: (JSC::JIT::privateCompile): 2011-10-13 Gavin Barraclough <baraclough@apple.com> terminateSpeculativeExecution for fillSpeculateDouble with DataFormatCell Rubber stamped by Filip Pizlo This is breaking fast/canvas/canvas-composite-alpha.html on 32_64 DFG JIT. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): 2011-10-13 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualized JSCell::toNumber https://bugs.webkit.org/show_bug.cgi?id=69858 Reviewed by Sam Weinig. Removed JSCallbackObject::toNumber because its no longer necessary since JSObject::toNumber now suffices since we implicitly add valueOf to an object's prototype whenever a convertToType callback is provided. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: De-virtualized JSCell::toNumber, JSObject::toNumber, and JSString::toNumber. * runtime/JSCell.cpp: (JSC::JSCell::toNumber): * runtime/JSCell.h: * runtime/JSObject.h: * runtime/JSString.h: Removed JSNotAnObject::toNumber because its result doesn't matter and it implements defaultValue, therefore JSObject::toNumber can cover its case. * runtime/JSNotAnObject.cpp: * runtime/JSNotAnObject.h: 2011-10-13 Xianzhu Wang <wangxianzhu@chromium.org> Use realloc() to expand/shrink StringBuilder buffer https://bugs.webkit.org/show_bug.cgi?id=69913 Reviewed by Darin Adler. * wtf/text/StringBuilder.cpp: (WTF::StringBuilder::reserveCapacity): (WTF::StringBuilder::reallocateBuffer): (WTF::StringBuilder::appendUninitialized): (WTF::StringBuilder::shrinkToFit): * wtf/text/StringBuilder.h: * wtf/text/StringImpl.cpp: (WTF::StringImpl::reallocate): Added to allow StringBuilder to reallocate the buffer. * wtf/text/StringImpl.h: 2011-10-12 Filip Pizlo <fpizlo@apple.com> If an Arguments object is being used to copy the arguments, then make this explicit https://bugs.webkit.org/show_bug.cgi?id=69995 Reviewed by Sam Weinig. * interpreter/Interpreter.cpp: (JSC::Interpreter::retrieveArguments): * runtime/Arguments.h: (JSC::Arguments::createAndCopyRegisters): (JSC::Arguments::finishCreationButDontCopyRegisters): (JSC::Arguments::finishCreation): (JSC::Arguments::finishCreationAndCopyRegisters): 2011-10-12 Filip Pizlo <fpizlo@apple.com> DFG CFA does not filter structures aggressively enough. https://bugs.webkit.org/show_bug.cgi?id=69989 Reviewed by Oliver Hunt. * dfg/DFGAbstractValue.h: (JSC::DFG::AbstractValue::clear): (JSC::DFG::AbstractValue::makeTop): (JSC::DFG::AbstractValue::clobberStructures): (JSC::DFG::AbstractValue::set): (JSC::DFG::AbstractValue::merge): (JSC::DFG::AbstractValue::filter): (JSC::DFG::AbstractValue::checkConsistency): 2011-10-12 Adam Barth <abarth@webkit.org> Remove ENABLE(XHTMLMP) and associated code https://bugs.webkit.org/show_bug.cgi?id=69729 Reviewed by David Levin. * Configurations/FeatureDefines.xcconfig: 2011-10-12 Gavin Barraclough <baraclough@apple.com> MacroAssemblerX86 8-bit register ops unsafe on CPU(X86) https://bugs.webkit.org/show_bug.cgi?id=69978 Reviewed by Filip Pizlo. Certain ops are unsafe if the register passed is esp..edi (will instead test/set the ). compare32/test8/test32 Call setCC, which sets an 8-bit register - we can fix this by adding a couple of xchg instructions. branchTest8 with a register argument is also affected. In all cases this is currently used this is testing a value that is correct to 32 or more bits, so we can simply switch these to branchTest32 & remove the corresponding branchTest8 (this is desirable anyway, since the 32-bit form is cheaper to implement on platforms that don't have an 8-bit compare instruction). This fixes the remaining fast/js failures with the DFG JIT 32_64. * assembler/MacroAssemblerARMv7.h - removed branchTest8. * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::compare32): (JSC::MacroAssemblerX86Common::test8): (JSC::MacroAssemblerX86Common::test32): (JSC::MacroAssemblerX86Common::set32): - added set32 helper that is 'h' register safe. - removed branchTest8. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeStrictEq): - switch uses of branchTest8 to branchTest32. * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeStrictEq): - switch uses of branchTest8 to branchTest32. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitBranch): - switch uses of branchTest8 to branchTest32. * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::emitBranch): - switch uses of branchTest8 to branchTest32. 2011-10-12 Gavin Barraclough <baraclough@apple.com> Errrk, revert accidental commit! * wtf/Platform.h: 2011-10-12 Gavin Barraclough <baraclough@apple.com> Unreviewed, re-land changes from #69890, #69903. These were reverted due to bug #69897, but #69903 fixed this problem. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentFillGPR): 2011-10-12 Filip Pizlo <fpizlo@apple.com> ValueProfile::computeUpdatedPrediction doesn't merge statistics correctly https://bugs.webkit.org/show_bug.cgi?id=69906 Reviewed by Gavin Barraclough. It turns out that the simplest fix is to switch computeUpdatedPredictions() to using predictionFromValue() combined with mergePrediction(). Doing so allowed me to kill off weakBuckets and visitWeakReferences(). Hence this not only fixes a performance bug but kills off a lot of code that I never liked to begin with. This appears to be a 1% win on V8. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): * bytecode/CodeBlock.h: * bytecode/PredictedType.cpp: (JSC::predictionFromValue): * bytecode/ValueProfile.cpp: (JSC::ValueProfile::computeStatistics): (JSC::ValueProfile::computeUpdatedPrediction): * bytecode/ValueProfile.h: (JSC::ValueProfile::classInfo): (JSC::ValueProfile::numberOfSamples): (JSC::ValueProfile::isLive): (JSC::ValueProfile::dump): 2011-10-12 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSCell::toString https://bugs.webkit.org/show_bug.cgi?id=69677 Reviewed by Sam Weinig. Removed toString from JSCallbackObject, since it is no longer necessary since we now implicitly add toString and valueOf functions to object prototypes when a convertToType callback is provided, which is now the standard way to override toString and valueOf in the JSC C API. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Removed toString from InterruptedExecutionError and TerminatedExecutionError and replaced it with defaultValue, which JSObject::toString calls. We'll probably have to de-virtualize defaultValue eventually, but we'll cross that bridge when we come to it. * runtime/ExceptionHelpers.cpp: (JSC::InterruptedExecutionError::defaultValue): (JSC::TerminatedExecutionError::defaultValue): * runtime/ExceptionHelpers.h: Removed toString from JSNotAnObject, since its return value doesn't actually matter and JSObject::toString can cover it. * runtime/JSNotAnObject.cpp: * runtime/JSNotAnObject.h: De-virtualized JSCell::toString, JSObject::toString and JSString::toString. Added handling of all cases for JSCell to JSCell::toString. * runtime/JSObject.h: * runtime/JSString.h: * runtime/JSCell.cpp: (JSC::JSCell::toString): * runtime/JSCell.h: 2011-10-12 Oliver Hunt <oliver@apple.com> Global stringStructure caches its prototype chain, abandoning a web page https://bugs.webkit.org/show_bug.cgi?id=69952 Reviewed by Filip Pizlo. When visiting a structure, we don't keep the prototype chain alive if we're not the structure for an object type. * runtime/Structure.cpp: (JSC::Structure::visitChildren): 2011-10-12 Yuqiang Xian <yuqiang.xian@intel.com> DFG JIT 32_64 - Fix ArrayPop https://bugs.webkit.org/show_bug.cgi?id=69918 Reviewed by Filip Pizlo. The storageLengthGPR is polluted by EmptyValueTag and later used to index the array, which results in abnormal behaviors in execution. This fix makes 32_64 DFG pass v8-deltablue and kraken crypto-sha256-iterative on Linux ia32. * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::store32): * assembler/X86Assembler.h: (JSC::X86Assembler::movl_i32m): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-12 Gustavo Noronha Silva <gustavo.noronha@collabora.co.uk> Fix build with GLib 2.31 https://bugs.webkit.org/show_bug.cgi?id=69840 Reviewed by Martin Robinson. * GNUmakefile.list.am: removed ThreadingGtk.cpp. * wtf/ThreadingPrimitives.h: remove GTK+-specific definitions. * wtf/gobject/GOwnPtr.cpp: remove GCond and GMutex specializations. * wtf/gobject/GOwnPtr.h: ditto. * wtf/gobject/GTypedefs.h: remove GCond and GMutex forward declarations. * wtf/gtk/ThreadingGtk.cpp: Removed. 2011-10-12 Filip Pizlo <fpizlo@apple.com> Layout tests crashing in DFG JIT code https://bugs.webkit.org/show_bug.cgi?id=69897 Reviewed by Gavin Barraclough. Abstract value filtration didn't take into account cases where a structure set filter, combined with predicted type knowledge, could lead to a stronger filter for the structure abstract value. This bug would have been benign in release builds; it would have just meant that the analysis was less precise and some optimization opportunities would be missed. I have an ASSERT that is meant to catch such cases, and it was triggering sporadically in one of the LayoutTests. * dfg/DFGAbstractValue.h: (JSC::DFG::AbstractValue::filter): 2011-10-11 Gavin Barraclough <baraclough@apple.com> Unreviewed, temporarily reverted r97216 due to bug #69897. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentFillGPR): 2011-10-11 Yuqiang Xian <yuqiang.xian@intel.com> DFG 32_64 - fix silentFillGPR https://bugs.webkit.org/show_bug.cgi?id=69903 Reviewed by Filip Pizlo. Fix a small bug in silentFillGPR, and add the newly introduced DFG file to CMakeListsEfl. * CMakeListsEfl.txt: * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentFillGPR): 2011-10-08 Filip Pizlo <fpizlo@apple.com> DFG does not have flow-sensitive intraprocedural control flow analysis https://bugs.webkit.org/show_bug.cgi?id=69690 Reviewed by Gavin Barraclough. Implemented a control flow analysis (CFA). It currently propagates type proofs only. For example, if all predecessors to a basic block have checks that variable X is a JSFinalObject with structure 0xabcdef, then this basic block will now know this fact and will know that it does not have to emit either JSFinalObject checks or any structure checks since the structure is precisely known. The CFA takes heap side-effects into account (though somewhat conservatively), so that if the object pointed to by variable X could have possibly undergone a structure transition then this is reflected: the analysis may simply say that X's structure is unknown. This also propagates a wealth of other type information which is currently not being used. For example, we now know when a variable can only hold doubles. Even if a variable may hold other types at different points in its live range, we can still prove exactly when it will only be double. There's a bunch of stuff that the CFA could do that it still does not do, like precise handling of PutStructure (i.e. structure transitions), precise handling of CheckFunction and CheckMethod, etc. So this is very much intended to be a starting point rather than an end unto itself. This is a 1% win on V8 (mostly due to a 3% win on richards and deltablue) and a 1% win on Kraken (mostly due to a 6% win on imaging-desaturate). Neutral on SunSpider. * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/ActionablePrediction.h: Removed. * bytecode/PredictedType.cpp: (JSC::predictionToString): * bytecode/PredictedType.h: * dfg/DFGAbstractState.cpp: Added. (JSC::DFG::AbstractState::AbstractState): (JSC::DFG::AbstractState::~AbstractState): (JSC::DFG::AbstractState::beginBasicBlock): (JSC::DFG::AbstractState::initialize): (JSC::DFG::AbstractState::endBasicBlock): (JSC::DFG::AbstractState::reset): (JSC::DFG::AbstractState::execute): (JSC::DFG::AbstractState::clobberStructures): (JSC::DFG::AbstractState::mergeStateAtTail): (JSC::DFG::AbstractState::merge): (JSC::DFG::AbstractState::mergeToSuccessors): (JSC::DFG::AbstractState::mergeVariableBetweenBlocks): (JSC::DFG::AbstractState::dump): * dfg/DFGAbstractState.h: Added. (JSC::DFG::AbstractState::forNode): (JSC::DFG::AbstractState::isValid): * dfg/DFGAbstractValue.h: Added. (JSC::DFG::StructureAbstractValue::StructureAbstractValue): (JSC::DFG::StructureAbstractValue::clear): (JSC::DFG::StructureAbstractValue::makeTop): (JSC::DFG::StructureAbstractValue::top): (JSC::DFG::StructureAbstractValue::add): (JSC::DFG::StructureAbstractValue::addAll): (JSC::DFG::StructureAbstractValue::contains): (JSC::DFG::StructureAbstractValue::isSubsetOf): (JSC::DFG::StructureAbstractValue::doesNotContainAnyOtherThan): (JSC::DFG::StructureAbstractValue::isSupersetOf): (JSC::DFG::StructureAbstractValue::filter): (JSC::DFG::StructureAbstractValue::isClear): (JSC::DFG::StructureAbstractValue::isTop): (JSC::DFG::StructureAbstractValue::size): (JSC::DFG::StructureAbstractValue::at): (JSC::DFG::StructureAbstractValue::operator[]): (JSC::DFG::StructureAbstractValue::last): (JSC::DFG::StructureAbstractValue::predictionFromStructures): (JSC::DFG::StructureAbstractValue::operator==): (JSC::DFG::StructureAbstractValue::dump): (JSC::DFG::AbstractValue::AbstractValue): (JSC::DFG::AbstractValue::clear): (JSC::DFG::AbstractValue::isClear): (JSC::DFG::AbstractValue::makeTop): (JSC::DFG::AbstractValue::clobberStructures): (JSC::DFG::AbstractValue::isTop): (JSC::DFG::AbstractValue::top): (JSC::DFG::AbstractValue::set): (JSC::DFG::AbstractValue::operator==): (JSC::DFG::AbstractValue::merge): (JSC::DFG::AbstractValue::filter): (JSC::DFG::AbstractValue::validate): (JSC::DFG::AbstractValue::dump): * dfg/DFGBasicBlock.h: Added. (JSC::DFG::BasicBlock::BasicBlock): (JSC::DFG::BasicBlock::getBytecodeBegin): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getLocal): (JSC::DFG::ByteCodeParser::setLocal): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::setArgument): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::processPhiStack): (JSC::DFG::ByteCodeParser::setupPredecessors): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: * dfg/DFGJITCodeGenerator.h: (JSC::DFG::block): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranchNull): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeStrictEq): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranchNull): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeStrictEq): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::noticeOSREntry): * dfg/DFGNode.h: (JSC::DFG::NodeIndexTraits::defaultValue): (JSC::DFG::Node::variableAccessData): (JSC::DFG::Node::takenBytecodeOffsetDuringParsing): (JSC::DFG::Node::notTakenBytecodeOffsetDuringParsing): (JSC::DFG::Node::setTakenBlockIndex): (JSC::DFG::Node::setNotTakenBlockIndex): (JSC::DFG::Node::takenBlockIndex): (JSC::DFG::Node::notTakenBlockIndex): * dfg/DFGOSREntry.cpp: (JSC::DFG::prepareOSREntry): * dfg/DFGOSREntry.h: * dfg/DFGOperands.h: Added. (JSC::DFG::operandIsArgument): (JSC::DFG::OperandValueTraits::defaultValue): (JSC::DFG::Operands::Operands): (JSC::DFG::Operands::numberOfArguments): (JSC::DFG::Operands::numberOfLocals): (JSC::DFG::Operands::argument): (JSC::DFG::Operands::local): (JSC::DFG::Operands::setLocal): (JSC::DFG::Operands::setArgumentFirstTime): (JSC::DFG::Operands::setLocalFirstTime): (JSC::DFG::Operands::operand): (JSC::DFG::Operands::setOperand): (JSC::DFG::Operands::clear): (JSC::DFG::dumpOperands): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::fixpoint): (JSC::DFG::Propagator::propagateArithNodeFlags): (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::propagatePredictions): (JSC::DFG::Propagator::performBlockCFA): (JSC::DFG::Propagator::performForwardCFA): (JSC::DFG::Propagator::globalCFA): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleDoubleBranch): (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality): (JSC::DFG::SpeculativeJIT::compilePeepHoleIntegerBranch): (JSC::DFG::SpeculativeJIT::compilePeepHoleBranch): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::compileGetCharCodeAt): (JSC::DFG::SpeculativeJIT::compileGetByValOnString): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGStructureSet.h: (JSC::DFG::StructureSet::clear): (JSC::DFG::StructureSet::predictionFromStructures): (JSC::DFG::StructureSet::operator==): (JSC::DFG::StructureSet::dump): * dfg/DFGVariableAccessData.h: Added. 2011-10-11 Gavin Barraclough <baraclough@apple.com> DFG JIT 32_64 - Fix silentFillGPR for non-integer constants. https://bugs.webkit.org/show_bug.cgi?id=69890 Reviewed by Oliver Hunt. Cell constants are currently hitting the valueOfInt32Constant case, there is no constant handling for JSValues. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentFillGPR): 2011-10-11 Ryosuke Niwa <rniwa@webkit.org> GTK build fix attempt after r97197. * wtf/BitVector.h: 2011-10-11 Oliver Hunt <oliver@apple.com> Remove unintentional logging. * heap/Heap.cpp: 2011-10-11 Oliver Hunt <oliver@apple.com> Tidy up card walking logic https://bugs.webkit.org/show_bug.cgi?id=69883 Reviewed by Gavin Barraclough. Special case common cell sizes when walking a block's cards. * heap/CardSet.h: (JSC::::testAndClear): * heap/Heap.cpp: (JSC::GCTimer::GCCounter::GCCounter): (JSC::GCTimer::GCCounter::count): (JSC::GCTimer::GCCounter::~GCCounter): (JSC::Heap::markRoots): * heap/MarkStack.cpp: (JSC::MarkStack::reset): * heap/MarkStack.h: (JSC::MarkStack::visitCount): (JSC::MarkStack::MarkStack): (JSC::MarkStack::append): * heap/MarkedBlock.h: (JSC::MarkedBlock::gatherDirtyCellsWithSize): (JSC::MarkedBlock::gatherDirtyCells): * runtime/Structure.h: (JSC::MarkStack::internalAppend): 2011-10-11 Filip Pizlo <fpizlo@apple.com> DFG virtual register allocator should be more aggressive in reusing temporary slots https://bugs.webkit.org/show_bug.cgi?id=69868 Reviewed by Oliver Hunt. 1.2% win on V8, neutral elsewhere. The win is probably because it increases precision of GC conservative scans. This required making the DFG::ScoreBoard operate over a bitvector of preserved variables, rather than just a preserved variable threshold. To do this, I improved the WTF::BitVector class to make it more user-friendly. It still retains all previous functionality. Also made changes to PackedIntVector to accomodate those changes. Finally, this adds more debugging to the virtual register allocator and to the OSR exit code, as this was necessary to track down bugs in an earlier version of this patch. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::getLocal): * dfg/DFGGraph.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::allocateVirtualRegisters): * dfg/DFGScoreBoard.h: (JSC::DFG::ScoreBoard::ScoreBoard): (JSC::DFG::ScoreBoard::~ScoreBoard): (JSC::DFG::ScoreBoard::allocate): (JSC::DFG::ScoreBoard::use): (JSC::DFG::ScoreBoard::highWatermark): (JSC::DFG::ScoreBoard::dump): (JSC::DFG::ScoreBoard::max): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::ValueRecovery::dump): * wtf/BitVector.cpp: (WTF::BitVector::setSlow): (WTF::BitVector::resizeOutOfLine): (WTF::BitVector::dump): * wtf/BitVector.h: (WTF::BitVector::BitVector): (WTF::BitVector::operator=): (WTF::BitVector::quickGet): (WTF::BitVector::quickSet): (WTF::BitVector::quickClear): (WTF::BitVector::get): (WTF::BitVector::set): (WTF::BitVector::clear): * wtf/PackedIntVector.h: (WTF::PackedIntVector::get): (WTF::PackedIntVector::set): 2011-10-11 Gavin Barraclough <baraclough@apple.com> DFG JIT 32_64 - Switch to cdecl calling convention. https://bugs.webkit.org/show_bug.cgi?id=69863 Reviewed by Oliver Hunt. This makes it easier to keep the stack correctly aligned, which is required on OS X. * assembler/MacroAssemblerCodeRef.h: (JSC::FunctionPtr::FunctionPtr): - Provide default FunctionPtr constructors for CDECL functions on STDCALL platforms. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): - Switch calls to poke arguments rather than pushing them. (JSC::DFG::resetCallArguments): (JSC::DFG::addCallArgument): (JSC::DFG::addCallArgumentBoxed): - Helper functions to stack up call arguments on X86. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::emitCall): - Don't push, poke! * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::compileBody): - Don't push, poke! * dfg/DFGOperations.cpp: - Switch ReturnAddress wrappers to push return address last, update asm trampolines. * dfg/DFGOperations.h: - switch DFG_OPERATION to assert CDECL on STDCALL platforms. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::fmodWithCDecl): (JSC::DFG::SpeculativeJIT::compile): - On STDCALL platforms wrap fmod, since DFG_OPERATION wrappers are CDECL. 2011-10-11 Gavin Barraclough <baraclough@apple.com> Switch RegisterSizedBoolean/dfgConvertJSValueToInt32 return type to size_t https://bugs.webkit.org/show_bug.cgi?id=69821 Reviewed by Filip Pizlo. Operations returning types Z (int32_t) and B (RegisterSizedBoolean - implemented as an intptr_t) are indistinguishable on 32-bit Linux, preventing the DFG JIT from building. dfgConvertJSValueToInt32 would be better returning a value known to be register sized, for JSVALUE64 (we currently zero-extend in JIT code, potentially introducing an unnecessary move), so by switching all associated operations to return a size_t we can fix the type problem on Linux & make it a small tweak that removes an unnecessary instruction. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeCompare): - comparisons now return a size_t. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): - Removed Z_DFGOperation_EJ form. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): - comparisons now return a size_t. * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): - comparisons now return a size_t. * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: - Change return types for comparison operations & dfgConvertJSValueToInt32 to size_t, Both need to return values zero extended to fill a register. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleBranch): - comparisons now return a size_t. * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compare): - comparisons now return a size_t. * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compare): - comparisons now return a size_t. 2011-10-11 Tor Arne Vestbø <tor.arne.vestbo@nokia.com> [Qt] Remove all references to QTDIR_build and standalone_package Qt is now modularized, which means we no longer import WebKit into the Qt source tree. Instead we use git submodules, and building QtWebKit as "part of Qt" is really building QtWebKit as from trunk. To decrease the number of buildsystem configurations we also remove the standalone_package code-path used when we were providing tarballs with the derived sources pre-generated. Reviewed by Simon Hausmann. * DerivedSources.pro: * JavaScriptCore.pri: * JavaScriptCore.pro: 2011-10-11 Yuqiang Xian <yuqiang.xian@intel.com> Add missing copyright notice in DFG JIT files https://bugs.webkit.org/show_bug.cgi?id=69809 Reviewed by Gavin Barraclough. * dfg/DFGJITCodeGenerator32_64.cpp: * dfg/DFGJITCompiler32_64.cpp: * dfg/DFGJITCompilerInlineMethods.h: * dfg/DFGSpeculativeJIT32_64.cpp: 2011-10-10 Filip Pizlo <fpizlo@apple.com> DFG JSVALUE64 spill/fill code should not box integers and doubles https://bugs.webkit.org/show_bug.cgi?id=69782 Reviewed by Oliver Hunt. Added the notion of DataFormatInteger and DataFormatDouble to the spillFormat. This required changing all of the places that spill registers (both silently and not) and filling registers (both silently and on demand). It also required changing OSR exit to recognize that a spilled value (DisplacedInRegisterFile) may have the wrong format for the old JIT (unboxed int or double). This is a slight win on Kraken (0.25%) and neutral elsewhere. * dfg/DFGGenerationInfo.h: (JSC::DFG::GenerationInfo::spill): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentFillFPR): (JSC::DFG::JITCodeGenerator::spill): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::fillInteger): (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::ValueRecovery::displacedInRegisterFile): (JSC::DFG::ValueRecovery::virtualRegister): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): 2011-10-10 Gavin Barraclough <baraclough@apple.com> DFG JIT switch dfgConvert methods to use callOperation https://bugs.webkit.org/show_bug.cgi?id=69806 Reviewed by Filip Pizlo. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToNumber): (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToNumber): (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): * dfg/DFGOperations.h: 2011-10-10 Gavin Barraclough <baraclough@apple.com> Remove some unused methods from the DFG JIT. Rubber stamped by Oliver Hunt Thee methods were only used by the non-speculative JIT, and can be removed. * dfg/DFGJITCodeGenerator.h: * dfg/DFGJITCodeGenerator32_64.cpp: * dfg/DFGJITCodeGenerator64.cpp: - removed: nonSpeculativeAdd nonSpeculativeArithSub nonSpeculativeArithMod nonSpeculativeCheckHasInstance nonSpeculativeInstanceOf * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: - removed: operationArithMod operationInstanceOf operationThrowHasInstanceError 2011-10-10 Gavin Barraclough <baraclough@apple.com> Switch most calls in DFGJITCodeGenerator to use callOperation. https://bugs.webkit.org/show_bug.cgi?id=69802 Reviewed by Oliver Hunt. Compares, add, mod are the easy cases. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeKnownConstantArithOp): (JSC::DFG::JITCodeGenerator::nonSpeculativeBasicArithOp): (JSC::DFG::JITCodeGenerator::nonSpeculativeArithMod): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeStrictEq): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeStrictEq): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeKnownConstantArithOp): (JSC::DFG::JITCodeGenerator::nonSpeculativeBasicArithOp): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeStrictEq): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeStrictEq): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: 2011-10-10 Gavin Barraclough <baraclough@apple.com> DFG: Switch GetById / PutById to use callOperation https://bugs.webkit.org/show_bug.cgi?id=69795 Reviewed by Oliver Hunt. Also make the take base as a cell, so 32_64 doesn't have to set up the cell tag. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGRepatch.cpp: (JSC::DFG::appropriatePutByIdFunction): 2011-10-10 Filip Pizlo <fpizlo@apple.com> REGRESSIoN (r95399): Web process hangs when opening documents on Google Docs https://bugs.webkit.org/show_bug.cgi?id=69412 Reviewed by Oliver Hunt. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JIT.h: 2011-10-10 Mark Hahnenberg <mhahnenberg@apple.com> Remove getCallDataVirtual methods https://bugs.webkit.org/show_bug.cgi?id=69186 Reviewed by Geoffrey Garen. Removed all getCallDataVirtual methods and replaced their call sites with an explicit lookup in the MethodTable. * API/JSCallbackFunction.cpp: * API/JSCallbackFunction.h: * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: * API/JSObjectRef.cpp: (JSObjectIsFunction): (JSObjectCallAsFunction): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ArrayConstructor.cpp: * runtime/ArrayConstructor.h: * runtime/BooleanConstructor.cpp: * runtime/BooleanConstructor.h: * runtime/DateConstructor.cpp: * runtime/DateConstructor.h: Moved StrictModeTypeErrorFunction to Error.h in order to be able to include the class definition in JSGlobalObject.cpp. * runtime/Error.cpp: (JSC::createTypeErrorFunction): * runtime/Error.h: (JSC::StrictModeTypeErrorFunction::StrictModeTypeErrorFunction): (JSC::StrictModeTypeErrorFunction::create): (JSC::StrictModeTypeErrorFunction::constructThrowTypeError): (JSC::StrictModeTypeErrorFunction::getConstructData): (JSC::StrictModeTypeErrorFunction::callThrowTypeError): (JSC::StrictModeTypeErrorFunction::getCallData): (JSC::StrictModeTypeErrorFunction::createStructure): * runtime/ErrorConstructor.cpp: * runtime/ErrorConstructor.h: * runtime/FunctionConstructor.cpp: * runtime/FunctionConstructor.h: * runtime/FunctionPrototype.cpp: * runtime/FunctionPrototype.h: To allow subclasses of InternalFunction (e.g. QtRuntimeMethod) to not have to declare their own ClassInfo if they don't override getCallData, provided an implementation that calls ASSERT_NOT_REACHED if called, providing roughly the same functionality as of the pure virtual method InternalFunction used to have. Also made this new implementation protected rather than private for the same reason. Also added an ASSERT in InternalFunction::finishCreation to make sure that whatever object is being created provides their own implementation of getCallData. This just makes execution fail earlier in a place where the source of the error is easy to trace. These ASSERTs are better than putting a null in the MethodTable because they appear much more intentional to anybody who fails to provide their own implementation or who tries to explicitly call InternalFunction::getCallData. * runtime/InternalFunction.cpp: (JSC::InternalFunction::finishCreation): (JSC::InternalFunction::getCallData): * runtime/InternalFunction.h: * runtime/JSCell.cpp: * runtime/JSCell.h: * runtime/JSFunction.cpp: * runtime/JSFunction.h: Added a global structure to JSGlobalObject for StrictModeTypeErrorFunction to enable it to be reused rather than creating a new Structure every time we instantiate it. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): (JSC::JSGlobalObject::visitChildren): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::strictModeTypeErrorFunctionStructure): * runtime/JSONObject.cpp: (JSC::Stringifier::Stringifier): (JSC::Stringifier::toJSON): (JSC::Stringifier::appendStringifiedValue): * runtime/JSObject.cpp: (JSC::JSObject::put): * runtime/JSObject.h: (JSC::getCallData): * runtime/NativeErrorConstructor.cpp: * runtime/NativeErrorConstructor.h: * runtime/NumberConstructor.cpp: * runtime/NumberConstructor.h: * runtime/ObjectConstructor.cpp: * runtime/ObjectConstructor.h: * runtime/Operations.cpp: (JSC::jsTypeStringForValue): (JSC::jsIsObjectType): (JSC::jsIsFunctionType): * runtime/PropertySlot.cpp: (JSC::PropertySlot::functionGetter): * runtime/RegExpConstructor.cpp: * runtime/RegExpConstructor.h: * runtime/StringConstructor.cpp: * runtime/StringConstructor.h: * runtime/Structure.h: 2011-10-10 Gavin Barraclough <barraclough@apple.com> Switch last calls from DFGSpeculativeJIT to use callOperation. https://bugs.webkit.org/show_bug.cgi?id=69780 Reviewed by Oliver Hunt. Also, rename type in operations for booleans from Z to B, since Z is the mathematical symbol for integers. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeCompare): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleBranch): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compile): * wtf/Platform.h: 2011-10-10 Yuqiang Xian <yuqiang.xian@intel.com> JSVALUE32_64 DFG JIT - bug fix for V8 benchmark cases "crypto" and "raytrace" https://bugs.webkit.org/show_bug.cgi?id=69748 Reviewed by Filip Pizlo. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::cachedGetMethod): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compileObjectEquality): 2011-10-10 Adam Roben <aroben@apple.com> Build fix * wtf/MainThread.h: Pull in Platform.h since this file uses PLATFORM() macros. 2011-10-10 Yuqiang Xian <yuqiang.xian@intel.com> JSVALUE32_64 DFG JIT - Bug fix for BranchNull https://bugs.webkit.org/show_bug.cgi?id=69743 Reviewed by Darin Adler. This fixes the error in access-binary-trees. All SunSpider cases passed. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranchNull): 2011-10-07 Gavin Barraclough <barraclough@apple.com> DFG JIT: callOperation should return the Call. https://bugs.webkit.org/show_bug.cgi?id=69682 Reviewed by Oliver Hunt. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): (JSC::DFG::appendCallWithExceptionCheckSetResult): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::appendCall): * wtf/Platform.h: 2011-10-10 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r97045. http://trac.webkit.org/changeset/97045 https://bugs.webkit.org/show_bug.cgi?id=69746 makes apple bots very crashy :( (Requested by kling on #webkit). * config.h: 2011-10-10 Andreas Kling <kling@webkit.org> Shrink BorderValue. https://bugs.webkit.org/show_bug.cgi?id=69521 Reviewed by Antti Koivisto. * config.h: Touch to force full rebuild. 2011-10-09 Yuqiang Xian <yuqiang.xian@intel.com> Improve Null or Undefined test in 32_64 DFG https://bugs.webkit.org/show_bug.cgi?id=69734 Reviewed by Darin Adler. Currently Null or Undefined value test in 32_64 DFG will check Null and Undefined tag separately and introduce one more branch. It can be improved in the way how the baseline JIT is doing - by relying on the fact that "UndefinedTag + 1 == NullTag and NullTag & 1". * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompareNull): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): 2011-10-09 Yuqiang Xian <yuqiang.xian@intel.com> JSVALUE32_64 DFG JIT - Bug fix for ConvertThis https://bugs.webkit.org/show_bug.cgi?id=69721 Reviewed by Darin Adler. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-09 Yuqiang Xian <yuqiang.xian@intel.com> Remove unused callOperation code of DFG JIT on X86 https://bugs.webkit.org/show_bug.cgi?id=69722 Reviewed by Filip Pizlo. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): 2011-10-09 Yuqiang Xian <yuqiang.xian@intel.com> JSVALUE32_64 DFG JIT - fillJSValue with a pair of GPRs should not set the registerFormat to be DataFormatJSDouble https://bugs.webkit.org/show_bug.cgi?id=69720 Reviewed by Filip Pizlo. In JSVALUE32_64 DFG, DataFormatJSDouble is assumed to be represented by a FPR and will be used for further optimizations, though we currently don't fully utilize it. For now when filling a JS value which was spilled as a JSDouble with a pair of GPRs, we'll set the registerFormat to DataFormatJS to avoid compilation errors. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::fillJSValue): 2011-10-09 Filip Pizlo <fpizlo@apple.com> DFG should not always speculate that a ByVal access has an integer index https://bugs.webkit.org/show_bug.cgi?id=69716 Reviewed by Oliver Hunt. 1% win on SunSpider, neutral elsewhere. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): * dfg/DFGNode.h: * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::byValHasIntBase): (JSC::DFG::Propagator::clobbersWorld): (JSC::DFG::Propagator::getMethodLoadElimination): (JSC::DFG::Propagator::checkStructureLoadElimination): (JSC::DFG::Propagator::getByOffsetLoadElimination): (JSC::DFG::Propagator::getPropertyStorageLoadElimination): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-09 Yuqiang Xian <yuqiang.xian@intel.com> Fix value profiling in 32_64 JIT https://bugs.webkit.org/show_bug.cgi?id=69717 Reviewed by Filip Pizlo. Current value profiling for 32_64 JIT is broken and cannot record correct predicated types, which results in many speculation failures in the 32_64 DFG JIT, fallbacks to baseline JIT, and re-optimizations again and again. With this fix 32_64 DFG JIT can demonstrate real performance gains. * bytecode/ValueProfile.cpp: (JSC::ValueProfile::computeStatistics): * bytecode/ValueProfile.h: (JSC::ValueProfile::classInfo): (JSC::ValueProfile::numberOfSamples): (JSC::ValueProfile::isLive): (JSC::ValueProfile::numberOfInt32s): (JSC::ValueProfile::numberOfDoubles): (JSC::ValueProfile::numberOfBooleans): (JSC::ValueProfile::dump): Empty value check should be performed on decoded JSValue, as for 32_64 empty value is not identical to encoded 0. * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JITInlineMethods.h: (JSC::JIT::emitValueProfilingSite): * jit/JITStubCall.h: (JSC::JITStubCall::callWithValueProfiling): Record the right profiling result for 32_64. 2011-10-09 Yuqiang Xian <yuqiang.xian@intel.com> Remove 32 bit restrictions in DFG JIT https://bugs.webkit.org/show_bug.cgi?id=69711 Reviewed by Filip Pizlo. op_call/op_construct support was disabled for 32 bit DFG JIT because there was regression in javascriptcore tests. Now the bugs are fixed and there should be no regression. This makes 32 bit DFG have the same capability as 64 bit DFG, and improves the coverage. * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): 2011-10-08 Mark Hahnenberg <mhahnenberg@apple.com> Add static version of JSCell::getConstructData https://bugs.webkit.org/show_bug.cgi?id=69673 Reviewed by Geoffrey Garen. Added static version of getConstructData to all classes that override it and changed the virtual versions to call the static versions. This is the first step in de-virtualizing JSCell::getConstructData. * API/JSCallbackConstructor.cpp: (JSC::JSCallbackConstructor::getConstructData): * API/JSCallbackConstructor.h: * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::getConstructData): * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::getConstructData): * runtime/ArrayConstructor.h: * runtime/BooleanConstructor.cpp: (JSC::BooleanConstructor::getConstructData): * runtime/BooleanConstructor.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::getConstructData): * runtime/DateConstructor.h: * runtime/ErrorConstructor.cpp: (JSC::ErrorConstructor::getConstructData): * runtime/ErrorConstructor.h: * runtime/FunctionConstructor.cpp: (JSC::FunctionConstructor::getConstructData): * runtime/FunctionConstructor.h: * runtime/JSCell.cpp: (JSC::JSCell::getConstructData): * runtime/JSCell.h: * runtime/JSFunction.cpp: (JSC::JSFunction::getConstructData): * runtime/JSFunction.h: * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::getConstructData): * runtime/NativeErrorConstructor.h: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::getConstructData): * runtime/NumberConstructor.h: * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::getConstructData): * runtime/ObjectConstructor.h: * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::getConstructData): * runtime/RegExpConstructor.h: * runtime/StringConstructor.cpp: (JSC::StringConstructor::getConstructData): * runtime/StringConstructor.h: 2011-10-08 Mark Hahnenberg <mhahnenberg@apple.com> Add static version of JSCell::getOwnPropertySlot https://bugs.webkit.org/show_bug.cgi?id=69593 Reviewed by Geoffrey Garen. Added static version of getOwnPropertySlot to every class that overrides JSCell::getOwnPropertySlot. The virtual versions now call the static versions. This is the first step in de-virtualizing JSCell::getOwnPropertySlot. * JavaScriptCore.exp: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::getOwnPropertySlot): * debugger/DebuggerActivation.h: * runtime/Arguments.cpp: (JSC::Arguments::getOwnPropertySlot): * runtime/Arguments.h: * runtime/ArrayConstructor.h: * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::getOwnPropertySlot): * runtime/ArrayPrototype.h: * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::getOwnPropertySlot): * runtime/BooleanPrototype.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::getOwnPropertySlot): * runtime/DateConstructor.h: * runtime/DatePrototype.cpp: (JSC::DatePrototype::getOwnPropertySlot): * runtime/DatePrototype.h: * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::getOwnPropertySlot): * runtime/ErrorPrototype.h: * runtime/JSActivation.cpp: (JSC::JSActivation::getOwnPropertySlot): * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::getOwnPropertySlot): * runtime/JSArray.h: * runtime/JSBoundFunction.cpp: (JSC::JSBoundFunction::getOwnPropertySlot): * runtime/JSBoundFunction.h: * runtime/JSByteArray.cpp: (JSC::JSByteArray::getOwnPropertySlot): * runtime/JSByteArray.h: * runtime/JSCell.cpp: (JSC::JSCell::getOwnPropertySlot): * runtime/JSCell.h: * runtime/JSFunction.cpp: (JSC::JSFunction::getOwnPropertySlot): * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::getOwnPropertySlot): * runtime/JSGlobalObject.h: * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::getOwnPropertySlot): * runtime/JSNotAnObject.h: * runtime/JSONObject.cpp: (JSC::JSONObject::getOwnPropertySlot): * runtime/JSONObject.h: * runtime/JSObject.cpp: (JSC::JSObject::getOwnPropertySlot): * runtime/JSObject.h: (JSC::JSObject::getOwnPropertySlot): * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::getOwnPropertySlot): * runtime/JSStaticScopeObject.h: * runtime/JSString.cpp: (JSC::JSString::getOwnPropertySlot): * runtime/JSString.h: * runtime/MathObject.cpp: (JSC::MathObject::getOwnPropertySlot): * runtime/MathObject.h: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::getOwnPropertySlot): * runtime/NumberConstructor.h: * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::getOwnPropertySlot): * runtime/NumberPrototype.h: * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::getOwnPropertySlot): * runtime/ObjectConstructor.h: * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::getOwnPropertySlot): * runtime/ObjectPrototype.h: * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::getOwnPropertySlot): * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::getOwnPropertySlot): * runtime/RegExpObject.cpp: (JSC::RegExpObject::getOwnPropertySlot): * runtime/RegExpObject.h: * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::getOwnPropertySlot): * runtime/RegExpPrototype.h: * runtime/StringConstructor.cpp: (JSC::StringConstructor::getOwnPropertySlot): * runtime/StringConstructor.h: * runtime/StringObject.cpp: (JSC::StringObject::getOwnPropertySlot): * runtime/StringObject.h: * runtime/StringPrototype.cpp: (JSC::StringPrototype::getOwnPropertySlot): * runtime/StringPrototype.h: 2011-10-08 Yuqiang Xian <yuqiang.xian@intel.com> JSVALUE32_64 DFG JIT - GetLocal should produce a cell result for Array predictions https://bugs.webkit.org/show_bug.cgi?id=69699 Reviewed by Filip Pizlo. It should match SetLocal where only payload is stored for array predictions. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-08 Yuqiang Xian <yuqiang.xian@intel.com> JSVALUE32_64 DFG JIT - Bug fixes for Branch and LogicalNot https://bugs.webkit.org/show_bug.cgi?id=69702 Reviewed by Filip Pizlo. There are some errors in generating code for Branch and LogicalNot, when the operand is predicted as ObjectOrOther. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): 2011-10-08 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r96996. http://trac.webkit.org/changeset/96996 https://bugs.webkit.org/show_bug.cgi?id=69697 It broke all tests on the Qt bot (Requested by Ossy_night on #webkit). * API/JSCallbackFunction.cpp: (JSC::JSCallbackFunction::getCallDataVirtual): * API/JSCallbackFunction.h: * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::getCallDataVirtual): * API/JSObjectRef.cpp: (JSObjectIsFunction): (JSObjectCallAsFunction): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::getCallDataVirtual): * runtime/ArrayConstructor.h: * runtime/BooleanConstructor.cpp: (JSC::BooleanConstructor::getCallDataVirtual): * runtime/BooleanConstructor.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::getCallDataVirtual): * runtime/DateConstructor.h: * runtime/Error.cpp: (JSC::StrictModeTypeErrorFunction::StrictModeTypeErrorFunction): (JSC::StrictModeTypeErrorFunction::create): (JSC::StrictModeTypeErrorFunction::constructThrowTypeError): (JSC::StrictModeTypeErrorFunction::getConstructData): (JSC::StrictModeTypeErrorFunction::callThrowTypeError): (JSC::StrictModeTypeErrorFunction::getCallDataVirtual): (JSC::StrictModeTypeErrorFunction::getCallData): (JSC::StrictModeTypeErrorFunction::createStructure): (JSC::createTypeErrorFunction): * runtime/Error.h: * runtime/ErrorConstructor.cpp: (JSC::ErrorConstructor::getCallDataVirtual): * runtime/ErrorConstructor.h: * runtime/FunctionConstructor.cpp: (JSC::FunctionConstructor::getCallDataVirtual): * runtime/FunctionConstructor.h: * runtime/FunctionPrototype.cpp: (JSC::FunctionPrototype::getCallDataVirtual): * runtime/FunctionPrototype.h: * runtime/InternalFunction.cpp: (JSC::InternalFunction::finishCreation): * runtime/InternalFunction.h: * runtime/JSCell.cpp: (JSC::JSCell::getCallDataVirtual): * runtime/JSCell.h: (JSC::getCallData): * runtime/JSFunction.cpp: (JSC::JSFunction::getCallDataVirtual): * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): (JSC::JSGlobalObject::visitChildren): * runtime/JSGlobalObject.h: * runtime/JSONObject.cpp: (JSC::Stringifier::Stringifier): (JSC::Stringifier::toJSON): (JSC::Stringifier::appendStringifiedValue): * runtime/JSObject.cpp: (JSC::JSObject::put): * runtime/JSObject.h: * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::getCallDataVirtual): * runtime/NativeErrorConstructor.h: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::getCallDataVirtual): * runtime/NumberConstructor.h: * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::getCallDataVirtual): * runtime/ObjectConstructor.h: * runtime/Operations.cpp: (JSC::jsTypeStringForValue): (JSC::jsIsObjectType): (JSC::jsIsFunctionType): * runtime/PropertySlot.cpp: (JSC::PropertySlot::functionGetter): * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::getCallDataVirtual): * runtime/RegExpConstructor.h: * runtime/StringConstructor.cpp: (JSC::StringConstructor::getCallDataVirtual): * runtime/StringConstructor.h: * runtime/Structure.h: 2011-10-08 Yuqiang Xian <yuqiang.xian@intel.com> DFG JIT - only Array predictions can result in unboxed cells in register file https://bugs.webkit.org/show_bug.cgi?id=69695 Reviewed by Filip Pizlo. In current DFG JIT, only array predictions can result in unboxed cells in register file, not for the other cell predictions. * dfg/DFGSpeculativeJIT.h: (JSC::DFG::ValueSource::forPrediction): 2011-10-07 Yuqiang Xian <yuqiang.xian@intel.com> bug fixes for ArrayPush and ArrayPop in 32_64 DFG JIT https://bugs.webkit.org/show_bug.cgi?id=69696 Reviewed by Filip Pizlo. On 32-bit, we should use TimesEight (8) instead of ScalePtr (4) to compute the address of a JS array element. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-07 Mark Hahnenberg <mhahnenberg@apple.com> Add static version of JSCell::deleteProperty https://bugs.webkit.org/show_bug.cgi?id=69659 Reviewed by Geoffrey Garen. Added static version of both versions of put to all classes that override them and changed the virtual versions to call the static versions. This is the first step in de-virtualizing JSCell::deleteProperty. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::deleteProperty): * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::deleteProperty): * debugger/DebuggerActivation.h: * runtime/Arguments.cpp: (JSC::Arguments::deleteProperty): * runtime/Arguments.h: * runtime/JSActivation.cpp: (JSC::JSActivation::deleteProperty): * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::deleteProperty): * runtime/JSArray.h: * runtime/JSCell.cpp: (JSC::JSCell::deleteProperty): * runtime/JSCell.h: * runtime/JSFunction.cpp: (JSC::JSFunction::deleteProperty): * runtime/JSFunction.h: * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::deleteProperty): * runtime/JSNotAnObject.h: * runtime/JSObject.cpp: (JSC::JSObject::deleteProperty): * runtime/JSObject.h: * runtime/JSVariableObject.cpp: (JSC::JSVariableObject::deleteProperty): * runtime/JSVariableObject.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::deleteProperty): * runtime/StrictEvalActivation.cpp: (JSC::StrictEvalActivation::deleteProperty): * runtime/StrictEvalActivation.h: * runtime/StringObject.cpp: (JSC::StringObject::deleteProperty): * runtime/StringObject.h: 2011-10-07 Mark Hahnenberg <mhahnenberg@apple.com> Remove getCallDataVirtual methods https://bugs.webkit.org/show_bug.cgi?id=69186 Reviewed by Geoffrey Garen. Removed all getCallDataVirtual methods and replaced their call sites with an explicit lookup in the MethodTable. * API/JSCallbackFunction.cpp: * API/JSCallbackFunction.h: * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: * API/JSObjectRef.cpp: (JSObjectIsFunction): (JSObjectCallAsFunction): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ArrayConstructor.cpp: * runtime/ArrayConstructor.h: * runtime/BooleanConstructor.cpp: * runtime/BooleanConstructor.h: * runtime/DateConstructor.cpp: * runtime/DateConstructor.h: * runtime/Error.cpp: (JSC::createTypeErrorFunction): Moved StrictModeTypeErrorFunction to Error.h in order to be able to include the class definition in JSGlobalObject.cpp. * runtime/Error.h: (JSC::StrictModeTypeErrorFunction::StrictModeTypeErrorFunction): (JSC::StrictModeTypeErrorFunction::create): (JSC::StrictModeTypeErrorFunction::constructThrowTypeError): (JSC::StrictModeTypeErrorFunction::getConstructData): (JSC::StrictModeTypeErrorFunction::callThrowTypeError): (JSC::StrictModeTypeErrorFunction::getCallData): (JSC::StrictModeTypeErrorFunction::createStructure): * runtime/ErrorConstructor.cpp: * runtime/ErrorConstructor.h: * runtime/FunctionConstructor.cpp: * runtime/FunctionConstructor.h: * runtime/FunctionPrototype.cpp: * runtime/FunctionPrototype.h: To allow subclasses of InternalFunction (e.g. QtRuntimeMethod) to not have to declare their own ClassInfo if they don't override getCallData, provided an implementation that calls ASSERT_NOT_REACHED if called, providing roughly the same functionality as of the pure virtual method InternalFunction used to have. Also made this new implementation protected rather than private for the same reason. Also added an ASSERT in InternalFunction::finishCreation to make sure that whatever object is being created provides their own implementation of getCallData. This just makes execution fail earlier in a place where the source of the error is easy to trace. These ASSERTs are better than putting a null in the MethodTable because they appear much more intentional to anybody who fails to provide their own implementation or who tries to explicitly call InternalFunction::getCallData. * runtime/InternalFunction.cpp: (JSC::InternalFunction::finishCreation): (JSC::InternalFunction::getCallData): * runtime/InternalFunction.h: * runtime/JSCell.cpp: * runtime/JSCell.h: * runtime/JSFunction.cpp: * runtime/JSFunction.h: Added a global structure to JSGlobalObject for StrictModeTypeErrorFunction to enable it to be reused rather than creating a new Structure every time we instantiate it. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): (JSC::JSGlobalObject::visitChildren): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::strictModeTypeErrorFunctionStructure): * runtime/JSONObject.cpp: (JSC::Stringifier::Stringifier): (JSC::Stringifier::toJSON): (JSC::Stringifier::appendStringifiedValue): * runtime/JSObject.cpp: (JSC::JSObject::put): * runtime/JSObject.h: (JSC::getCallData): * runtime/NativeErrorConstructor.cpp: * runtime/NativeErrorConstructor.h: * runtime/NumberConstructor.cpp: * runtime/NumberConstructor.h: * runtime/ObjectConstructor.cpp: * runtime/ObjectConstructor.h: * runtime/Operations.cpp: (JSC::jsTypeStringForValue): (JSC::jsIsObjectType): (JSC::jsIsFunctionType): * runtime/PropertySlot.cpp: (JSC::PropertySlot::functionGetter): * runtime/RegExpConstructor.cpp: * runtime/RegExpConstructor.h: * runtime/StringConstructor.cpp: * runtime/StringConstructor.h: * runtime/Structure.h: 2011-10-07 Oliver Hunt <oliver@apple.com> Add missing break statement. Reviewed by Gavin Barraclough. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): 2011-10-07 Oliver Hunt <oliver@apple.com> Support some string intrinsics in the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=69678 Reviewed by Gavin Barraclough. Add support for charAt and charCodeAt intrinsics in the DFG. * create_hash_table: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleIntrinsic): * dfg/DFGIntrinsic.h: * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileGetCharCodeAt): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-07 Mark Hahnenberg <mhahnenberg@apple.com> Add static version of JSCell::put https://bugs.webkit.org/show_bug.cgi?id=69382 Reviewed by Geoffrey Garen. Added static version of both versions of put to all classes that override them and changed the virtual versions to call the static versions. * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::put): * JavaScriptCore.exp: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::put): * debugger/DebuggerActivation.h: * runtime/Arguments.cpp: (JSC::Arguments::put): * runtime/Arguments.h: * runtime/JSActivation.cpp: (JSC::JSActivation::put): * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::put): * runtime/JSArray.h: * runtime/JSByteArray.cpp: (JSC::JSByteArray::put): * runtime/JSByteArray.h: * runtime/JSCell.cpp: (JSC::JSCell::put): * runtime/JSCell.h: * runtime/JSFunction.cpp: (JSC::JSFunction::put): * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::put): * runtime/JSGlobalObject.h: * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::put): * runtime/JSNotAnObject.h: * runtime/JSObject.cpp: (JSC::JSObject::put): * runtime/JSObject.h: * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::put): * runtime/JSStaticScopeObject.h: * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::put): * runtime/ObjectPrototype.h: * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::put): * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::put): * runtime/RegExpObject.cpp: (JSC::RegExpObject::put): * runtime/RegExpObject.h: * runtime/StringObject.cpp: (JSC::StringObject::put): * runtime/StringObject.h: 2011-10-07 Gavin Barraclough <barraclough@apple.com> Refactor DFG to make for use of callOperation https://bugs.webkit.org/show_bug.cgi?id=69672 Reviewed by Oliver Hunt. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): - Added new callOperation calls, don't ASSERT flushed (use helpers for unexpected calls, too). * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: - Switch operationNewObject/operationCreateThis to return Cells, - Added C_DFGOperation_E/C_DFGOperation_EC/J_DFGOperation_EA/J_DFGOperation_EJA call types. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): - Replace code plating calls to operations to with calls to callOperation. * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): - Replace code plating calls to operations to with calls to callOperation. 2011-10-07 Oliver Hunt <oliver@apple.com> Support string indexing in the DFG https://bugs.webkit.org/show_bug.cgi?id=69671 Reviewed by Gavin Barraclough. Emit code to support inline indexing of strings * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compileGetByValOnString): Shared code to perform string indexing. * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): Use compileGetByValOnString if we predict that the base object is a string in GetByVal. * runtime/JSString.h: (JSC::JSString::offsetOfFiberCount): (JSC::JSString::offsetOfValue): 2011-10-07 Filip Pizlo <fpizlo@apple.com> DFG ConvertThis speculation logic is wrong https://bugs.webkit.org/show_bug.cgi?id=69663 Reviewed by Oliver Hunt. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::fixupNode): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-07 Oliver Hunt <oliver@apple.com> Verify that our call speculation is valid. Reviewed by Filip Pizlo. Before specialising an intrinsic we need to verify that we our speculation is correct. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): 2011-10-07 Brent Fulgham <bfulgham@webkit.org> [WinCairo] Unreviewed build correction for the build bot. * JavaScriptCore.vcproj/JavaScriptCore.sln: Add the missing Release_Cairo_CFLite and Debug_Cairo_CFLite targets so that build-jsc can find the target it needs to run the JSC tests. 2011-10-07 Oliver Hunt <oliver@apple.com> Fix 32-bit build. * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCall): 2011-10-07 Oliver Hunt <oliver@apple.com> Support direct calls to intrinsic functions https://bugs.webkit.org/show_bug.cgi?id=69646 Reviewed by Gavin Barraclough. Add support for optimising non-method_check calls to intrinsic functions (eg. when Math.abs, etc are cached in local variables). * bytecode/CodeBlock.h: (JSC::getCallLinkInfoBytecodeIndex): Support searching CallLinkInfos by bytecode index * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): Add support for linked calls in addition to method_check when searching for intrinsics * dfg/DFGNode.h: (JSC::DFG::Node::hasFunctionCheckData): (JSC::DFG::Node::function): Add ability to store a JSFunction* in a node - this is safe as the function will be marked by the codeblock we're compiling * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::checkFunctionElimination): (JSC::DFG::Propagator::performNodeCSE): Add support for new CheckFunction node, and implement CSE pass. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): Rather trivial implementation of CheckFunction * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JIT.h: * jit/JITCall.cpp: (JSC::JIT::compileOpCall): * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCall): Need to propagate bytecode index for calls now. 2011-10-07 Dominic Cooney <dominicc@chromium.org> [JSC] Disable ThreadRestrictionVerifier for JIT ExecutableMemoryHandles https://bugs.webkit.org/show_bug.cgi?id=69599 Reviewed by Sam Weinig. DFG JIT manipulates MetaAllocatorHandles across threads, eg in allocating JITCode buffers on a background thread to execute a proxy autoconfiguration PAC file but garbage collecting it in response to allocation on the main thread. Disabling ThreadRestrictionVerification until there is a verification scheme that understands this handoff. * wtf/MetaAllocator.cpp: (WTF::MetaAllocator::allocate): 2011-10-06 Filip Pizlo <fpizlo@apple.com> DFG should not always speculate that ConvertThis is operating on an object https://bugs.webkit.org/show_bug.cgi?id=69570 Reviewed by Oliver Hunt. Mostly neutral, but with a slight regression in Kraken since it increases coverage in DFG and thus reveals some performance pathologies (which I prefer to think of as performance opportunities, in a good way). * bytecode/PredictedType.cpp: (JSC::predictionToString): * bytecode/PredictedType.h: (JSC::isOtherPrediction): (JSC::mergePredictions): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-06 Mark Hahnenberg <mhahnenberg@apple.com> Windows build fix Unreviewed build fix. Weird runtime failures on Windows due to linking issues caused by the ClassInfo struct in JSByteArray not being declared with JS_EXPORTDATA. * runtime/JSByteArray.h: 2011-10-06 Filip Pizlo <fpizlo@apple.com> Structure does not reset m_previous when pinning the property map https://bugs.webkit.org/show_bug.cgi?id=69583 Reviewed by Gavin Barraclough. This is an 0.6% performance improvement in V8, and 0.2% overall. * runtime/Structure.cpp: (JSC::Structure::changePrototypeTransition): (JSC::Structure::despecifyFunctionTransition): (JSC::Structure::getterSetterTransition): (JSC::Structure::toDictionaryTransition): (JSC::Structure::preventExtensionsTransition): (JSC::Structure::addPropertyWithoutTransition): (JSC::Structure::removePropertyWithoutTransition): (JSC::Structure::pin): * runtime/Structure.h: 2011-10-06 Anders Carlsson <andersca@apple.com> When building with clang, enable -Wglobal-constructors and -Wexit-time-destructors https://bugs.webkit.org/show_bug.cgi?id=69586 Reviewed by Darin Adler. * Configurations/Base.xcconfig: Add -Wglobal-constructors and -Wexit-time-destructors when building with clang. * JavaScriptCore.xcodeproj/project.pbxproj: When building with clang, we don't need to run the check-for-global-initializers and check-for-exit-time-destructors anymore. * jsc.cpp: (runInteractive): Move interpreterName into runInteractive. * wtf/StdLibExtras.h: When building with clang, disable the -Wglobal-constructors and -Wexit-time-destructors warnings around the variable declaration. 2011-10-06 Anders Carlsson <andersca@apple.com> Add DEFINE_DEBUG_ONLY_GLOBAL for globals that should be defined in debug builds https://bugs.webkit.org/show_bug.cgi?id=69584 Reviewed by Darin Adler. Add DEFINE_DEBUG_ONLY_GLOBAL macro. * wtf/StdLibExtras.h: 2011-10-06 Oliver Hunt <oliver@apple.com> Write barrier shouldn't allocate temporaries inside control flow https://bugs.webkit.org/show_bug.cgi?id=69582 Reviewed by Gavin Barraclough. Reorder the code to avoid spill-related badness. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::writeBarrier): 2011-10-06 Filip Pizlo <fpizlo@apple.com> DFG::shouldSpeculate methods are too complicated https://bugs.webkit.org/show_bug.cgi?id=69560 Reviewed by Geoffrey Garen. Moved shouldSpeculate methods to DFG::Node, and cleaned them up to just use node predictions. By itself this would have meant that SpeculativeJIT code would have had to say things like m_jit.graph()[nodeIndex].shouldSpeculateXYZ(). So this adds an at(NodeIndex) method to JITCodeGenerator. I replaced all uses of the m_jit.graph()[nodeIndex] idiom with at(nodeIndex). This is an 0.4% progression overall that shows up in all benchmarks, for reasons unknown. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::at): (JSC::DFG::JITCodeGenerator::canReuse): (JSC::DFG::JITCodeGenerator::isFilled): (JSC::DFG::JITCodeGenerator::isFilledDouble): (JSC::DFG::JITCodeGenerator::use): (JSC::DFG::JITCodeGenerator::silentSpillFPR): (JSC::DFG::JITCodeGenerator::silentFillGPR): (JSC::DFG::JITCodeGenerator::silentFillFPR): (JSC::DFG::detectPeepHoleBranch): (JSC::DFG::integerResult): (JSC::DFG::noResult): (JSC::DFG::cellResult): (JSC::DFG::jsValueResult): (JSC::DFG::storageResult): (JSC::DFG::doubleResult): (JSC::DFG::initConstantInfo): (JSC::DFG::appendCallWithExceptionCheck): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::fillInteger): (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToNumber): (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): (JSC::DFG::JITCodeGenerator::nonSpeculativeUInt32ToNumber): (JSC::DFG::JITCodeGenerator::nonSpeculativeKnownConstantArithOp): (JSC::DFG::JITCodeGenerator::nonSpeculativeBasicArithOp): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranchNull): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeStrictEq): (JSC::DFG::JITCodeGenerator::emitCall): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::fillInteger): (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranchNull): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeStrictEq): (JSC::DFG::JITCodeGenerator::emitCall): * dfg/DFGNode.h: (JSC::DFG::Node::shouldSpeculateInteger): (JSC::DFG::Node::shouldSpeculateDouble): (JSC::DFG::Node::shouldSpeculateNumber): (JSC::DFG::Node::shouldNotSpeculateInteger): (JSC::DFG::Node::shouldSpeculateFinalObject): (JSC::DFG::Node::shouldSpeculateFinalObjectOrOther): (JSC::DFG::Node::shouldSpeculateArray): (JSC::DFG::Node::shouldSpeculateArrayOrOther): (JSC::DFG::Node::shouldSpeculateObject): (JSC::DFG::Node::shouldSpeculateCell): (JSC::DFG::Node::canSpeculateInteger): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleDoubleBranch): (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality): (JSC::DFG::SpeculativeJIT::compilePeepHoleIntegerBranch): (JSC::DFG::SpeculativeJIT::compilePeepHoleBranch): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isInteger): (JSC::DFG::SpeculativeJIT::isKnownArray): (JSC::DFG::SpeculativeJIT::isKnownString): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::convertToDouble): (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitBranch): (JSC::DFG::SpeculativeJIT::compile): 2011-10-06 Gavin Peters <gavinp@chromium.org> REGRESSION (r96595): First frame in assertion backtraces is no longer labeled "1" https://bugs.webkit.org/show_bug.cgi?id=69556 Reviewed by Adam Roben. * wtf/Assertions.cpp: 2011-10-06 Filip Pizlo <fpizlo@apple.com> DFG implementation of UInt32ToNumber is missing a break statement https://bugs.webkit.org/show_bug.cgi?id=69552 Reviewed by Oliver Hunt. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-06 Gavin Barraclough <barraclough@apple.com> Unreviewed build fix for DFG JIT 32_64 release builds. * dfg/DFGJITCompiler.cpp: * dfg/DFGJITCompiler.h: * dfg/DFGJITCompiler32_64.cpp: - Remove three unused methods. 2011-10-06 Gavin Barraclough <barraclough@apple.com> DFG JIT 32_64 should check type of values being filled by fillSpeculateInt https://bugs.webkit.org/show_bug.cgi?id=69549 Reviewed by Oliver Hunt. This breaks sunspider/3d-cube. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): - Speculation check on the tag. 2011-10-06 Mark Hahnenberg <mhahnenberg@apple.com> Snow Leopard build fix Unreviewed build fix * JavaScriptCore.exp: 2011-10-05 Gavin Barraclough <barraclough@apple.com> Add explicit JSGlobalThis type. https://bugs.webkit.org/show_bug.cgi?id=69478 Reviewed by Darin Adler. JSC supports a split global object, as used by WebCore for the Window. As a stage of making this visible to JSC, make it so that if the global this value is not the global object itself, it must be a subclass of JSGlobalThis. * API/JSCallbackObjectFunctions.h: (JSC::::finishCreation): - Don't pass the thisValue to JSGlobalObject::finishCreation. * JavaScriptCore.xcodeproj/project.pbxproj: - Added JSGlobalThis.h * jsc.cpp: (GlobalObject::finishCreation): - Don't pass the thisValue to JSGlobalObject::finishCreation. * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::create): (JSC::JSGlobalObject::finishCreation): - finishCreation takes a JSGlobalThis, or thisValue is implicit. * runtime/JSGlobalThis.h: Added. (JSC::JSGlobalThis::create): (JSC::JSGlobalThis::JSGlobalThis): (JSC::JSGlobalThis::finishCreation): - Thin wrapper on JSNonFinalObject to allow type checking. * testRegExp.cpp: (GlobalObject::finishCreation): - Don't pass the thisValue to JSGlobalObject::finishCreation. 2011-10-06 Mark Hahnenberg <mhahnenberg@apple.com> JSC objects need to know their own cell size at runtime. https://bugs.webkit.org/show_bug.cgi?id=69390 Reviewed by Geoffrey Garen. Added the cellSize field to ClassInfo and the static calculation of size of each class to the CREATE_METHOD_TABLE macro, which will be renamed in a followup patch to make its name match its broader use. Also added a few ClassInfo structs so that each object that is allocated has its correct size. * JavaScriptCore.exp: * runtime/ClassInfo.h: Changed JSByteArray s_defaultInfo to s_info so that the template will get the correct ClassInfo struct from it when it's allocated. * runtime/JSByteArray.cpp: * runtime/JSByteArray.h: * runtime/JSCell.h: (JSC::allocateCell): * runtime/JSNotAnObject.cpp: * runtime/JSNotAnObject.h: * runtime/JSObject.cpp: * runtime/JSObject.h: (JSC::JSCell::cellSize): * runtime/JSStaticScopeObject.cpp: * runtime/JSStaticScopeObject.h: * runtime/StrictEvalActivation.cpp: * runtime/StrictEvalActivation.h: 2011-10-06 Gavin Peters <gavinp@chromium.org> export new stack dumping method https://bugs.webkit.org/show_bug.cgi?id=69018 The original landing of bug 69018 didn't export WTFGetBacktrace, so that when bug 69453 landed, the first use of this function, many builds broke. So here we add the exports, so that the function is usable. Reviewed by Adam Roben. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-10-06 Csaba Osztrogonác <ossy@webkit.org> REGRESSION(r96347): Build is broken with MSVC compiler if !PLATFORM(WINDOWS) https://bugs.webkit.org/show_bug.cgi?id=69413 Reviewed by Darin Adler. * assembler/MacroAssemblerCodeRef.h: Define STDCALL for MSVC in a proper way. 2011-10-05 Filip Pizlo <fpizlo@apple.com> SpeculativeJIT::isKnownString() is wrong https://bugs.webkit.org/show_bug.cgi?id=69501 Reviewed by Oliver Hunt. Removed the wrong case (GetLocal predicted String) and added a case that works (StrCat). * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isKnownString): 2011-10-05 Ryosuke Niwa <rniwa@webkit.org> Windows build fix attempt after r96760. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-10-05 Chris Rogers <crogers@google.com> Define a log2f() function for Windows in wtf/MathExtras.h https://bugs.webkit.org/show_bug.cgi?id=69491 Reviewed by Darin Adler. * wtf/MathExtras.h: (log2f): 2011-10-05 Jer Noble <jer.noble@apple.com> Enable WEB_AUDIO by default in the WebKit/mac port. https://bugs.webkit.org/show_bug.cgi?id=68587 Reviewed by Simon Fraser. * Configurations/FeatureDefines.xcconfig: * wtf/Platform.h: 2011-10-05 Filip Pizlo <fpizlo@apple.com> Assertion hit in JSC::DFG::SpeculativeJIT::compile on SL bots https://bugs.webkit.org/show_bug.cgi?id=69346 Reviewed by Oliver Hunt. Removed the assertion, since it was completely wrong for op_post_inc. Short of having specialized PostInc nodes in the DFG, there is no robust way of asserting what this assertion was trying to assert while also supporting op_post_inc. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-05 Geoffrey Garen <ggaren@apple.com> Added a simpler mechanism for registering one-off finalizers https://bugs.webkit.org/show_bug.cgi?id=69466 Reviewed by Oliver Hunt. * heap/Heap.cpp: (JSC::Heap::addFinalizer): (JSC::Heap::FinalizerOwner::finalize): * heap/Heap.h: New function for adding an arbitrary finalizer for an arbitrary cell without declaring any special classes or Handles yourself. * JavaScriptCore.exp: Fix build. * runtime/Executable.cpp: (JSC::ExecutableBase::clearCode): (JSC::ExecutableBase::clearCodeVirtual): (JSC::EvalExecutable::clearCodeVirtual): (JSC::ProgramExecutable::clearCodeVirtual): (JSC::FunctionExecutable::discardCode): (JSC::FunctionExecutable::clearCodeVirtual): * runtime/Executable.h: (JSC::ExecutableBase::finishCreation): Use the new mechanism for eager finalization of executables. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::clearRareData): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::createRareDataIfNeeded): (JSC::JSGlobalObject::registerWeakMap): Use the new mechanism for eager finalization of weak maps. 2011-10-05 Adam Roben <aroben@apple.com> Ensure RetainPtr::hashTableDeletedValue returns a pointer, not a pointer to a pointer RetainPtr's behavior of allowing the template parameter to be either a pointer type or a pointed-to type confused us when we implemented hashTableDeletedValue. Fixes <http://webkit.org/b/69414> <rdar://problem/10236833> Using RetainPtr as the key type in HashMap/HashSet fails to compile Reviewed by John Sullivan. * wtf/RetainPtr.h: (WTF::RetainPtr::hashTableDeletedValue): Changed to use the PtrType typedef rather than T*, since T might itself be a pointer. (WTF::PtrHash<RetainPtr<P> >): Updated this to use PtrType everywhere, even though T* didn't seem to be causing a problem. 2011-10-05 Oliver Hunt <oliver@apple.com> Remove last vestiges of anonymous storage. Reviewed by Gavin Barraclough. One anonymous storage function escaped my prior purge of this feature, this patch removes it. * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::finishCreation): * runtime/JSObject.h: 2011-10-04 Filip Pizlo <fpizlo@apple.com> DFG should be capable of a broader range of speculations on branch and not https://bugs.webkit.org/show_bug.cgi?id=69322 Reviewed by Oliver Hunt. * bytecode/PredictedType.h: (JSC::isFinalObjectOrOtherPrediction): (JSC::isArrayOrOtherPrediction): * dfg/DFGJITCodeGenerator.cpp: * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::JITCodeGenerator): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::shouldSpeculateFinalObjectOrOther): (JSC::DFG::SpeculativeJIT::shouldSpeculateArrayOrOther): (JSC::DFG::SpeculativeJIT::SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitBranch): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compileObjectOrOtherLogicalNot): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::emitObjectOrOtherBranch): (JSC::DFG::SpeculativeJIT::emitBranch): 2011-10-05 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r96733. http://trac.webkit.org/changeset/96733 https://bugs.webkit.org/show_bug.cgi?id=69454 Broke GCC for some reason (Requested by andersca on #webkit). * wtf/ListHashSet.h: (WTF::ListHashSetReverseIterator::ListHashSetReverseIterator): (WTF::ListHashSetReverseIterator::get): (WTF::ListHashSetReverseIterator::operator*): (WTF::ListHashSetReverseIterator::operator->): (WTF::ListHashSetReverseIterator::operator++): (WTF::ListHashSetReverseIterator::operator--): (WTF::ListHashSetReverseIterator::operator==): (WTF::ListHashSetReverseIterator::operator!=): (WTF::ListHashSetReverseIterator::operator const_reverse_iterator): (WTF::ListHashSetReverseIterator::node): (WTF::ListHashSetConstReverseIterator::ListHashSetConstReverseIterator): (WTF::ListHashSetConstReverseIterator::get): (WTF::ListHashSetConstReverseIterator::operator*): (WTF::ListHashSetConstReverseIterator::operator->): (WTF::ListHashSetConstReverseIterator::operator++): (WTF::ListHashSetConstReverseIterator::operator--): (WTF::ListHashSetConstReverseIterator::operator==): (WTF::ListHashSetConstReverseIterator::operator!=): (WTF::ListHashSetConstReverseIterator::node): (WTF::::rbegin): (WTF::::rend): (WTF::::makeReverseIterator): (WTF::::makeConstReverseIterator): 2011-10-04 Oliver Hunt <oliver@apple.com> Add rudimentary filtering to write barriers https://bugs.webkit.org/show_bug.cgi?id=69392 Reviewed by Filip Pizlo. Add approximate filtering for write barriers based on the target's mark bit. Also add some macros to support dumping GC phase timings. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::markCellCard): * heap/Heap.cpp: (JSC::GCTimer::GCTimerScope::GCTimerScope): (JSC::GCTimer::GCTimerScope::~GCTimerScope): (JSC::Heap::markRoots): (JSC::Heap::collect): Add phase timing information. * heap/MarkedBlock.h: (JSC::MarkedBlock::offsetOfMarks): (JSC::MarkedBlock::gatherDirtyCells): * jit/JITPropertyAccess.cpp: (JSC::JIT::emitWriteBarrier): 2011-10-05 Anders Carlsson <andersca@apple.com> Use std::reverse_iterator for ListHashSet reverse iterators https://bugs.webkit.org/show_bug.cgi?id=69446 Reviewed by Darin Adler. * wtf/ListHashSet.h: Use the std::reverse_iterator iterator adaptor for the ListHashSet reverse iterators and get rid of the ListHashSetReverseIterator and ListHashSetConstReverseIterator classes. 2011-10-04 Gavin Barraclough <barraclough@apple.com> Make Object.prototype getter/setter methods match ES5 behaviour https://bugs.webkit.org/show_bug.cgi?id=69393 Reviewed by Sam Weinig. The rest of Object.prototype no longer substitute Null/Undefined with the global object, this is old ES3 behaviour. Remove it here too. * runtime/ObjectPrototype.cpp: (JSC::objectProtoFuncDefineGetter): (JSC::objectProtoFuncDefineSetter): (JSC::objectProtoFuncLookupGetter): (JSC::objectProtoFuncLookupSetter): 2011-10-05 Patrick Gansterer <paroga@webkit.org> Get rid of posixThread in MachineStackMarker::Thread https://bugs.webkit.org/show_bug.cgi?id=54836 Reviewed by Oliver Hunt. * heap/MachineStackMarker.cpp: (JSC::MachineThreads::Thread::Thread): (JSC::getCurrentPlatformThread): (JSC::equalThread): (JSC::MachineThreads::addCurrentThread): (JSC::MachineThreads::removeCurrentThread): (JSC::MachineThreads::gatherConservativeRoots): 2011-10-04 Geoffrey Garen <ggaren@apple.com> Removed JSValue::toJSNumber https://bugs.webkit.org/show_bug.cgi?id=69399 No perf. change. toJSNumber() used to provide an implicit fast path for immediate numbers, but those fast paths are all explicit now, so it's just cruft. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/JSValue.h: * runtime/JSValueInlineMethods.h: 2011-10-05 Gavin Peters <gavinp@chromium.org> REGRESSION (r96595): WTFReportBacktrace listed as the top frame in all assertion backtraces https://bugs.webkit.org/show_bug.cgi?id=69424 Skip an extra frame in WTFReportBacktrace. As well, I now don't count skipped frames in maxFrames, so I've updated maxFrames to 31, as with one skipped frame the previous value was effectively 31 reported frames. Reviewed by Adam Roben. * wtf/Assertions.cpp: * wtf/Assertions.h: 2011-10-05 Patrick Gansterer <paroga@webkit.org> Unreviewed WinCE build fix for r96595. * wtf/Assertions.cpp: RtlCaptureStackBackTrace() isn't available on WinCE. 2011-10-04 Kent Tamura <tkent@chromium.org> Introduce feature flags for incomplete input types https://bugs.webkit.org/show_bug.cgi?id=68971 Reviewed by Hajime Morita. * Configurations/FeatureDefines.xcconfig: Add ENABLE_INPUT_TYPE_* flags. They are enabled only for iOS. 2011-10-04 Geoffrey Garen <ggaren@apple.com> Build fix. * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): Use an explicit cast when shortening. 2011-10-04 Mark Hahnenberg <mhahnenberg@apple.com> Add static ClassInfo structs to classes that override JSCell::getCallData https://bugs.webkit.org/show_bug.cgi?id=69311 Reviewed by Darin Adler. Added ClassInfo structs to each class that defined its own getCallData function but did not already have its own ClassInfo struct. This is a necessary addition for when we switch over to looking up getCallData from the MethodTable in ClassInfo rather than doing the virtual call (which we are removing). These new ClassInfo structs are public because we often use these structs in other areas of the code to uniquely identify JSC classes and to enforce runtime invariants based on those class identities using ASSERTs. Also added new createStructure methods to those classes that didn't have them so that the new ClassInfo structs would be used when creating the Structures in these classes. * runtime/BooleanConstructor.cpp: * runtime/BooleanConstructor.h: (JSC::BooleanConstructor::createStructure): getCallData was not marked as static in StrictModeTypeErrorFunction. * runtime/Error.cpp: (JSC::StrictModeTypeErrorFunction::getCallDataVirtual): (JSC::StrictModeTypeErrorFunction::getCallData): (JSC::StrictModeTypeErrorFunction::createStructure): * runtime/ErrorConstructor.cpp: * runtime/ErrorConstructor.h: (JSC::ErrorConstructor::createStructure): * runtime/FunctionConstructor.cpp: * runtime/FunctionConstructor.h: (JSC::FunctionConstructor::createStructure): * runtime/FunctionPrototype.cpp: * runtime/FunctionPrototype.h: 2011-10-03 Geoffrey Garen <ggaren@apple.com> Some JSValue cleanup https://bugs.webkit.org/show_bug.cgi?id=69320 Reviewed by Darin Adler. No measurable performance change. Removed some JSValue::get* functions. get* used to be an optimization when every value operation was a virtual function call: get* would combine two virtual calls into one. Now, with non-virtual, inlined functions, get* isn't faster, and may be slightly slower. Merged getBoolean(bool&) and getBoolean() into asBoolean(). Merged uncheckedGetNumber(), getJSNumber() and getNumber() into asNumber(). * runtime/JSValue.h: * runtime/JSValueInlineMethods.h: (JSC::JSValue::asNumber): (JSC::JSValue::asBoolean): As promised! * runtime/NumberPrototype.cpp: (JSC::toThisNumber): (JSC::numberProtoFuncToExponential): (JSC::numberProtoFuncToFixed): (JSC::numberProtoFuncToPrecision): (JSC::numberProtoFuncToString): (JSC::numberProtoFuncToLocaleString): (JSC::numberProtoFuncValueOf): Removed a bunch of uses of getJSNumber() by switching to toThisNumber(). * API/JSCallbackObjectFunctions.h: (JSC::::toNumber): * dfg/DFGGraph.h: (JSC::DFG::Graph::valueOfNumberConstant): (JSC::DFG::Graph::valueOfBooleanConstant): * dfg/DFGOperations.cpp: (JSC::DFG::putByVal): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/DateInstance.h: (JSC::DateInstance::internalNumber): * runtime/FunctionPrototype.cpp: (JSC::functionProtoFuncBind): * runtime/JSArray.cpp: (JSC::compareNumbersForQSort): Replaced getNumber() => isNumber() / asNumber(). getBoolean() => isBoolean() / asBoolean(), uncheckedGetNumber() => asNumber(). * runtime/JSCell.cpp: * runtime/JSCell.h: Nixed getJSNumber(). * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncParseInt): * runtime/JSONObject.cpp: (JSC::gap): (JSC::Stringifier::Stringifier): (JSC::Stringifier::appendStringifiedValue): * runtime/NumberObject.cpp: * runtime/NumberObject.h: (JSC::NumberObject::createStructure): * runtime/Operations.h: (JSC::JSValue::equalSlowCaseInline): (JSC::JSValue::strictEqual): (JSC::jsLess): (JSC::jsLessEq): (JSC::jsAdd): Replaced getNumber() => isNumber() / asNumber(). getBoolean() => isBoolean() / asBoolean(), uncheckedGetNumber() => asNumber(). 2011-10-04 Scott Graham <scottmg@chromium.org> Add GAMEPAD feature flag https://bugs.webkit.org/show_bug.cgi?id=66859 Reviewed by Darin Fisher. * Configurations/FeatureDefines.xcconfig: 2011-10-03 Filip Pizlo <fpizlo@apple.com> JITCodeGenerator should no longer have code that tries too hard to be both speculative and non-speculative https://bugs.webkit.org/show_bug.cgi?id=69321 Reviewed by Gavin Barraclough. Removed m_isSpeculative and speculationCheck() from JITCodeGenerator. This required moving emitBranch() to SpeculativeJIT, since it was the main user of that field and method. Other than trvial clean-ups in emitBranch(), the code is unchanged (and still has some disparity between 64 and 32_64, and still lacks some obvious optimizations). * dfg/DFGJITCodeGenerator.cpp: * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::JITCodeGenerator): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): * dfg/DFGJITCodeGenerator64.cpp: (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitBranch): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::emitBranch): 2011-10-04 David Hyatt <hyatt@apple.com> https://bugs.webkit.org/show_bug.cgi?id=69372 [CSS3 Regions] Make sure overflow:visible lets content spill out of regions. Add support for reverse iteration to ListHashSet to support being able to walk them backwards easily. Reviewed by Anders Carlsson. * wtf/ListHashSet.h: (WTF::ListHashSetReverseIterator::ListHashSetReverseIterator): (WTF::ListHashSetReverseIterator::get): (WTF::ListHashSetReverseIterator::operator*): (WTF::ListHashSetReverseIterator::operator->): (WTF::ListHashSetReverseIterator::operator++): (WTF::ListHashSetReverseIterator::operator--): (WTF::ListHashSetReverseIterator::operator==): (WTF::ListHashSetReverseIterator::operator!=): (WTF::ListHashSetReverseIterator::operator const_reverse_iterator): (WTF::ListHashSetReverseIterator::node): (WTF::ListHashSetConstReverseIterator::ListHashSetConstReverseIterator): (WTF::ListHashSetConstReverseIterator::get): (WTF::ListHashSetConstReverseIterator::operator*): (WTF::ListHashSetConstReverseIterator::operator->): (WTF::ListHashSetConstReverseIterator::operator++): (WTF::ListHashSetConstReverseIterator::operator--): (WTF::ListHashSetConstReverseIterator::operator==): (WTF::ListHashSetConstReverseIterator::operator!=): (WTF::ListHashSetConstReverseIterator::node): (WTF::::rbegin): (WTF::::rend): (WTF::::makeReverseIterator): (WTF::::makeConstReverseIterator): (WTF::::makeConstIterator): 2011-10-04 Gavin Peters <gavinp@chromium.org> fix gtk breakage caused by changeset 96595 https://bugs.webkit.org/show_bug.cgi?id=69371 ews did not catch build breakage in the gtk WebKitPluginProcess target; this patch removes the pretty printer on gtk, which should fix the build on that platform. Reviewed by NOBODY, this is a build fix. * wtf/Assertions.cpp: 2011-10-04 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r96630. http://trac.webkit.org/changeset/96630 https://bugs.webkit.org/show_bug.cgi?id=69368 Caused assertion failures in validateCell (Requested by mhahnenberg on #webkit). * runtime/BooleanConstructor.cpp: * runtime/BooleanConstructor.h: * runtime/Error.cpp: (JSC::StrictModeTypeErrorFunction::getCallDataVirtual): (JSC::StrictModeTypeErrorFunction::getCallData): * runtime/ErrorConstructor.cpp: * runtime/ErrorConstructor.h: * runtime/FunctionConstructor.cpp: * runtime/FunctionConstructor.h: * runtime/FunctionPrototype.cpp: * runtime/FunctionPrototype.h: 2011-10-04 Mark Hahnenberg <mhahnenberg@apple.com> Add static ClassInfo structs to classes that override JSCell::getCallData https://bugs.webkit.org/show_bug.cgi?id=69311 Reviewed by Darin Adler. Added ClassInfo structs to each class that defined its own getCallData function but did not already have its own ClassInfo struct. This is a necessary addition for when we switch over to looking up getCallData from the MethodTable in ClassInfo rather than doing the virtual call (which we are removing). These new ClassInfo structs are public because we often use these structs in other areas of the code to uniquely identify JSC classes and to enforce runtime invariants based on those class identities using ASSERTs. * runtime/BooleanConstructor.cpp: * runtime/BooleanConstructor.h: getCallData was not marked as static is StrictModeTypeErrorFunction. * runtime/Error.cpp: (JSC::StrictModeTypeErrorFunction::getCallDataVirtual): (JSC::StrictModeTypeErrorFunction::getCallData): * runtime/ErrorConstructor.cpp: * runtime/ErrorConstructor.h: * runtime/FunctionConstructor.cpp: * runtime/FunctionConstructor.h: * runtime/FunctionPrototype.cpp: * runtime/FunctionPrototype.h: 2011-10-04 Ryosuke Niwa <rniwa@webkit.org> Leopard build fix after r96613. * wtf/Platform.h: 2011-10-04 Mark Hahnenberg <mhahnenberg@apple.com> Implicitly add toString and valueOf to prototype when convertToType callback is provided https://bugs.webkit.org/show_bug.cgi?id=69156 Reviewed by Geoffrey Garen. Added callbacks for toString and valueOf which are implicitly added to a client object's prototype if they provide a convertToType callback when declaring their class through the JSC API. * API/JSCallbackFunction.cpp: (JSC::JSCallbackFunction::toStringCallback): (JSC::JSCallbackFunction::valueOfCallback): * API/JSCallbackFunction.h: * API/JSClassRef.cpp: (OpaqueJSClass::prototype): * API/tests/testapi.js: 2011-10-03 Jon Lee <jonlee@apple.com> Extend DOM WheelEvent to differentiate between physical and logical scroll directions https://bugs.webkit.org/show_bug.cgi?id=68959 <rdar://problem/10036688> Reviewed by Sam Weinig. * wtf/Platform.h: Added HAVE_INVERTED_WHEEL_EVENTS for Lion and later. 2011-10-04 Csaba Osztrogonác <ossy@webkit.org> MinGW warning fix after r96286. Avoid redefining STDCALL, because STDCALL is also defined in mingw32/include/windef.h: #define __stdcall __attribute__((stdcall)) #define STDCALL __stdcall Reviewed by Tor Arne Vestbø. * assembler/MacroAssemblerCodeRef.h: 2011-10-04 Gavin Peters <gavinp@chromium.org> add more stack dumping methods https://bugs.webkit.org/show_bug.cgi?id=69018 In addition to WTFReportBacktrace, this adds the cross-platform WTFGetBacktrace, which lets WebKit programmatically retrieve the current stack. This is useful if you need to add more reporting to field crash report uploads, if you're tracking down an irreproducable bug, for instance. Reviewed by Darin Adler. * wtf/Assertions.cpp: * wtf/Assertions.h: 2011-10-03 Filip Pizlo <fpizlo@apple.com> DFG should inline Array.push and Array.pop https://bugs.webkit.org/show_bug.cgi?id=69314 Reviewed by Geoff Garen. Fix 32-bit. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-03 Filip Pizlo <fpizlo@apple.com> DFG should inline Array.push and Array.pop https://bugs.webkit.org/show_bug.cgi?id=69314 Reviewed by Oliver Hunt. 1% speed-up in V8 due to 6% speed-up in V8-deltablue. * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::storePtr): * create_hash_table: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleIntrinsic): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGIntrinsic.h: * dfg/DFGNode.h: (JSC::DFG::Node::hasHeapPrediction): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::getByValLoadElimination): (JSC::DFG::Propagator::getMethodLoadElimination): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-03 Filip Pizlo <fpizlo@apple.com> JSC ASSERT Opening the Web Inspector https://bugs.webkit.org/show_bug.cgi?id=69293 Reviewed by Oliver Hunt. If a polymorphic access structure list has a duplicated structure, then don't crash. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): 2011-10-03 Gavin Barraclough <barraclough@apple.com> On X86, switch bucketCount into a register, timeoutCheck into memory https://bugs.webkit.org/show_bug.cgi?id=69299 Reviewed by Geoff Garen. We don't have sufficient registers to keep both in registers, and DFG JIT will trample esi; it doesn't matter if the bucketCount gets stomped on (in fact it may add to randomness!), but it if the timeoutCheck gets trashed we may make calls out to the timout_check stub function too frequently (regressing performance). This patch has no perf impact on sunspider. * JavaScriptCore.xcodeproj/project.pbxproj: * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::branchAdd32): (JSC::MacroAssemblerX86::branchSub32): - Added branchSub32 with AbsoluteAddress. * jit/JIT.cpp: (JSC::JIT::emitTimeoutCheck): - Keep timeout count in memory on X86. * jit/JITInlineMethods.h: (JSC::JIT::emitValueProfilingSite): - remove X86 specific code, switch bucket count back into a register. * jit/JITStubs.cpp: - Stop initializing esi (it is no longer the timeoutCheck!) * jit/JSInterfaceJIT.h: - change definition of esi to be the bucketCountRegister. * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: - Add timeoutCount as a property to global data (the counter should be per-thread). 2011-10-03 Filip Pizlo <fpizlo@apple.com> DFG backends don't have access to per-node predictions from the propagator https://bugs.webkit.org/show_bug.cgi?id=69291 Reviewed by Oliver Hunt. Nodes now have two notion of predictions: the heap prediction, which is what came directly from value profiling, and the propagator's predictions, which arise out of abstract interpretation. Every node has a propagator prediction, but not every node has a heap prediction; and there is no guarantee that a node that has both will keep them consistent as the propagator may have additional information available to it. This is performance neutral. * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::getPrediction): * dfg/DFGNode.h: (JSC::DFG::Node::Node): (JSC::DFG::Node::hasHeapPrediction): (JSC::DFG::Node::getHeapPrediction): (JSC::DFG::Node::predictHeap): (JSC::DFG::Node::prediction): (JSC::DFG::Node::predict): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::Propagator): (JSC::DFG::Propagator::setPrediction): (JSC::DFG::Propagator::mergePrediction): (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::isPredictedNumerical): (JSC::DFG::Propagator::logicalNotIsPure): (JSC::DFG::Propagator::setReplacement): 2011-10-03 Jer Noble <jer.noble@apple.com> Unreviewed, rolling out r96526. http://trac.webkit.org/changeset/96526 https://bugs.webkit.org/show_bug.cgi?id=68587 WEB_AUDIO has numerous 64->32 bit casting warnings, causing build breakages where -Wall is enabled. * Configurations/FeatureDefines.xcconfig: * wtf/Platform.h: 2011-10-03 Gavin Barraclough <barraclough@apple.com> Unreviewed build fix for DFG JIT 32_64. * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-10-02 Filip Pizlo <fpizlo@apple.com> DFG should speculate more aggressively on obvious cases on polymorphic get_by_id https://bugs.webkit.org/show_bug.cgi?id=69235 Reviewed by Oliver Hunt. This implements trivial polymorphic get_by_id. It also fixes problems in the CSE for CheckStructure in the put_by_id transition case. Doing this required knowing whether a polymorphic get_by_id stub was doing a direct access rather than a call of some kind. Slight speed-up on Kraken and SunSpider. 0.5% speed-up in the scaled mean of all benchmarks. * GNUmakefile.list.am: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/Instruction.h: (JSC::PolymorphicAccessStructureList::PolymorphicStubInfo::set): (JSC::PolymorphicAccessStructureList::PolymorphicAccessStructureList): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::cellConstant): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::Graph::addStructureSet): (JSC::DFG::Graph::addStructureTransitionData): * dfg/DFGNode.h: (JSC::DFG::StructureTransitionData::StructureTransitionData): (JSC::DFG::Node::hasStructureTransitionData): (JSC::DFG::Node::structureTransitionData): (JSC::DFG::Node::hasStructureSet): (JSC::DFG::Node::structureSet): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::checkStructureLoadElimination): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGRepatch.cpp: (JSC::DFG::tryBuildGetByIDList): (JSC::DFG::tryBuildGetByIDProtoList): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGStructureSet.h: Added. (JSC::DFG::StructureSet::StructureSet): (JSC::DFG::StructureSet::add): (JSC::DFG::StructureSet::addAll): (JSC::DFG::StructureSet::remove): (JSC::DFG::StructureSet::contains): (JSC::DFG::StructureSet::isSubsetOf): (JSC::DFG::StructureSet::isSupersetOf): (JSC::DFG::StructureSet::size): (JSC::DFG::StructureSet::at): (JSC::DFG::StructureSet::operator[]): (JSC::DFG::StructureSet::last): * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): (JSC::getPolymorphicAccessStructureListSlot): 2011-10-03 Jer Noble <jer.noble@apple.com> Enable WEB_AUDIO by default in the WebKit/mac port. https://bugs.webkit.org/show_bug.cgi?id=68587 Reviewed by Simon Fraser. * Configurations/FeatureDefines.xcconfig: * wtf/Platform.h: 2011-10-03 Carlos Garcia Campos <cgarcia@igalia.com> [GTK] Fix make distcheck build https://bugs.webkit.org/show_bug.cgi?id=69243 Reviewed by Martin Robinson. * GNUmakefile.list.am: 2011-10-03 Pierre Rossi <pierre.rossi@gmail.com> [Qt] Build fix: Qt::escape is deprecated in Qt5 https://bugs.webkit.org/show_bug.cgi?id=69162 Use QString::toHtmlEscaped in the Qt5 case. Reviewed by Andreas Kling. * JavaScriptCore.pri: * wtf/qt/UtilsQt.h: Added. (escapeHtml): * wtf/wtf.pri: 2011-10-03 Balazs Kelemen <kbalazs@webkit.org> libdispatch based ParallelJobs is not enough parallel https://bugs.webkit.org/show_bug.cgi?id=66378 Reviewed by Zoltan Herczeg. Use the appropriate libdispatch API for our use case. Throw away the hard coded limit of parallel threads and use dispatch_apply with the default priority normal queue istead of using our own custom serial queue (which was a misuse of the API). Enabling PARALLEL_JOBS is now a 60% win (2.63x as fast) on the methanol benchmark (https://gitorious.org/methanol) with an SVG centric test set while the old implementation was almost identical (less than 5% win). * wtf/ParallelJobsLibdispatch.h: (WTF::ParallelEnvironment::ParallelEnvironment): (WTF::ParallelEnvironment::execute): 2011-10-02 Zoltan Herczeg <zherczeg@webkit.org> [Qt]REGRESSION(r95912): It made sputnik tests flakey https://bugs.webkit.org/show_bug.cgi?id=68990 Reviewed by Geoffrey Garen. Changing signed char to int in r96354 solved the problem. However transitionCount still returns with a signed char and should be changed to int. * runtime/Structure.h: (JSC::Structure::transitionCount): 2011-10-02 Filip Pizlo <fpizlo@apple.com> DFG misses some obvious opportunities for common subexpression elimination https://bugs.webkit.org/show_bug.cgi?id=69233 Reviewed by Oliver Hunt. 0.7% speed-up on SunSpider. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::getByValLoadElimination): (JSC::DFG::Propagator::getMethodLoadElimination): (JSC::DFG::Propagator::checkStructureLoadElimination): (JSC::DFG::Propagator::getByOffsetLoadElimination): (JSC::DFG::Propagator::getPropertyStorageLoadElimination): (JSC::DFG::Propagator::performNodeCSE): 2011-10-02 Gavin Barraclough <barraclough@apple.com> Bug 67455 - Different regular expression result Reviewed by Darin Adler. Fix a regression introduced in r72140. A return was added to the backtracking loop for backtrackParentheses with QuantifierNonGreedy, so it always returns after one iteration. This is incorrect. The additional return should only trigger to force an early return if an error has occured. * yarr/YarrInterpreter.cpp: (JSC::Yarr::Interpreter::matchParentheses): - Simplify some nested if else logic. (JSC::Yarr::Interpreter::backtrackParentheses): - Simplify some nested if else logic. - Only return early from backtrackParentheses on success/error, not on failure. 2011-10-01 Geoffrey Garen <ggaren@apple.com> Removed redundant helper functions for allocating Strong handles https://bugs.webkit.org/show_bug.cgi?id=69218 Reviewed by Sam Weinig. * heap/Heap.h: (JSC::Heap::handleHeap): * runtime/JSGlobalData.h: Removed these helper functions, since they just created indirection. * heap/StrongInlines.h: Added. Broke out a header for inline functions to resolve circular dependencies created by inlining. I'm told this is the future for JavaScriptCore. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: Go forth and build. * API/JSCallbackObjectFunctions.h: (JSC::::init): * runtime/WeakGCMap.h: (JSC::WeakGCMap::add): (JSC::WeakGCMap::set): * runtime/StructureTransitionTable.h: (JSC::StructureTransitionTable::setSingleTransition): * heap/Local.h: (JSC::::Local): * heap/Strong.h: (JSC::::Strong): (JSC::::set): * heap/Weak.h: (JSC::Weak::Weak): (JSC::Weak::set): Allocate handles directly instead of going through a chain of forwarding functions. * bytecompiler/BytecodeGenerator.cpp: * runtime/JSGlobalData.cpp: * runtime/LiteralParser.cpp: * runtime/RegExpCache.cpp: Updated for header changes. 2011-09-30 Filip Pizlo <fpizlo@apple.com> All of JSC's heuristics should be in one place for easier tuning https://bugs.webkit.org/show_bug.cgi?id=69201 Reviewed by Oliver Hunt. This makes it possible to change tiered compilation heuristics in one place (Heuristics.cpp) without recompiling the whole project. It also makes it possible to enable setting heuristics using environment variables. This is off by default. When turned on, it makes tuning the system much easier. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::shouldOptimizeNow): * bytecode/CodeBlock.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * jit/JIT.cpp: (JSC::JIT::emitOptimizationCheck): * runtime/Heuristics.cpp: Added. (JSC::Heuristics::parse): (JSC::Heuristics::setHeuristic): (JSC::Heuristics::initializeHeuristics): * runtime/Heuristics.h: Added. * runtime/InitializeThreading.cpp: (JSC::initializeThreadingOnce): 2011-10-01 Oliver Hunt <oliver@apple.com> Support string length in the DFG https://bugs.webkit.org/show_bug.cgi?id=69215 Reviewed by Geoff Garen. Adds a GetStringLength node to the DFG so that we can support string.length inline. * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isKnownString): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/JSString.h: (JSC::JSString::offsetOfLength): 2011-10-01 Yuqiang Xian <yuqiang.xian@intel.com> JSVALUE32_64 DFG JIT - unboxed integers and cells in register file must be reboxed before exiting from DFG JIT https://bugs.webkit.org/show_bug.cgi?id=69205 Reviewed by Gavin Barraclough. If there are unboxed integers and cells in register file (e.g. by SetLocal), they must be reboxed before exiting from the speculative DFG JIT execution. This patch also adds a new ValueSourceKind (CellInRegisterFile) and a new ValueRecoveryTechnique (AlreadyInRegisterFileAsCell). * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::ValueSource::dump): (JSC::DFG::ValueRecovery::dump): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::ValueSource::forPrediction): (JSC::DFG::ValueRecovery::alreadyInRegisterFileAsUnboxedCell): 2011-10-01 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r96421. http://trac.webkit.org/changeset/96421 https://bugs.webkit.org/show_bug.cgi?id=69206 It broke Qt-WK2 build (Requested by ossy on #webkit). * JavaScriptCore.pri: * wtf/qt/UtilsQt.h: Removed. * wtf/wtf.pri: 2011-09-30 Daniel Bates <dbates@webkit.org> Attempt to fix the Apple Windows and WinCairo Debug builds after <http://trac.webkit.org/changeset/96446> (https://bugs.webkit.org/show_bug.cgi?id=69203). * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Remove the symbol ?toStrictThisObject@JSObject@JSC@@UBE?AVJSValue@2@PAVExecState@2@@Z since the corresponding function, JSValue::toStrictThisObject(), was removed. 2011-09-30 Yuqiang Xian <yuqiang.xian@intel.com> DFG operation results are not set correctly in JSVALUE32_64 DFG JIT https://bugs.webkit.org/show_bug.cgi?id=69126 Reviewed by Gavin Barraclough. The setupResults routine has the bug of reversing the source and destination. Also some other trivial (but stupid) bugs need to be fixed in JSVALUE32_64 DFG JIT. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::setupTwoStubArgs): (JSC::DFG::setupResults): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): 2011-09-30 Gavin Barraclough <barraclough@apple.com> Remove toStrictThisObject, toThisString, toThisJSString https://bugs.webkit.org/show_bug.cgi?id=69203 Rubber stamped by Sam Weinig These are no longer used. * JavaScriptCore.exp: * runtime/JSActivation.cpp: * runtime/JSActivation.h: * runtime/JSObject.cpp: * runtime/JSObject.h: * runtime/JSStaticScopeObject.cpp: * runtime/JSStaticScopeObject.h: * runtime/JSValue.h: * runtime/StrictEvalActivation.cpp: * runtime/StrictEvalActivation.h: 2011-09-30 Filip Pizlo <fpizlo@apple.com> DFG does not speculate aggressively enough on put_by_id https://bugs.webkit.org/show_bug.cgi?id=69114 Reviewed by Oliver Hunt. This adds new nodes along with optimizations for those nodes: GetPropertyStorage: CheckStructure used to do both the structure check and retrieve the storage pointer. Now CheckStructure just checks the structure, and GetPropertyStorage retrieves the storage pointer. PutStructure: Changes the structure, and has the expected store to load optimization with CheckStructure. PutByOffset: Directly sets the value. Has store to load optimization with GetByOffset. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::cellConstant): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::writeBarrier): * dfg/DFGJITCodeGenerator.h: * dfg/DFGNode.h: (JSC::DFG::Node::hasStructure): (JSC::DFG::Node::hasStorageAccessData): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::impureCSE): (JSC::DFG::Propagator::checkStructureLoadElimination): (JSC::DFG::Propagator::getByOffsetLoadElimination): (JSC::DFG::Propagator::getPropertyStorageLoadElimination): (JSC::DFG::Propagator::eliminate): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-30 Gavin Barraclough <barraclough@apple.com> StringRecursionChecker should not work in terms of EncodedJSValue https://bugs.webkit.org/show_bug.cgi?id=69188 Reviewed by Oliver Hunt. 0 is not the empty value on 32_64. Code that casts literals to EncodedJSValues may be unsafe if we change our internal representation. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncToString): (JSC::arrayProtoFuncToLocaleString): (JSC::arrayProtoFuncJoin): * runtime/ErrorPrototype.cpp: (JSC::errorProtoFuncToString): * runtime/RegExpPrototype.cpp: (JSC::regExpProtoFuncToString): * runtime/StringRecursionChecker.cpp: (JSC::StringRecursionChecker::throwStackOverflowError): (JSC::StringRecursionChecker::emptyString): * runtime/StringRecursionChecker.h: (JSC::StringRecursionChecker::performCheck): (JSC::StringRecursionChecker::earlyReturnValue): 2011-09-30 Gavin Barraclough <barraclough@apple.com> DFG JIT, Branch on integer can always be a 32-bit compare. https://bugs.webkit.org/show_bug.cgi?id=69174 Reviewed by Sam Weinig. if (shouldSpeculateInteger(node.child1()) && !isStrictInt32(node.child1())), the JSVALUE64 JIT will currently compare all 64bits in the register, but in these cases the DataFormat is always a JS boxed integer. In these cases we can just compare the low 32bits anyway - no need to check the tag. This allows the code to be unified with the JSVALUE32_64 JIT. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-30 Oliver Hunt <oliver@apple.com> Need a sensible GGC policy Reviewed by Geoff Garen. This replaces the existing random collection policy with a deterministic policy based on nursery size. * heap/AllocationSpace.cpp: (JSC::AllocationSpace::allocateSlowCase): * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::markRoots): (JSC::Heap::collect): * heap/Heap.h: * heap/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::resetAllocator): * heap/MarkedSpace.h: (JSC::MarkedSpace::nurseryWaterMark): (JSC::MarkedSpace::allocate): 2011-09-30 Filip Pizlo <fpizlo@apple.com> DFG 32-bit support for op_call and op_construct causes run-javascriptcore-tests to fail https://bugs.webkit.org/show_bug.cgi?id=69171 Reviewed by Gavin Barraclough. This fixes one obvious bug that was causing test failures (no support for dummy slow case for op_add in 32_64), and disables op_call and op_construct by default. * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * jit/JITArithmetic32_64.cpp: (JSC::JIT::emit_op_add): (JSC::JIT::emitSlow_op_add): 2011-09-30 Geoffrey Garen <ggaren@apple.com> Crash due to out of bounds read/write in MarkedSpace https://bugs.webkit.org/show_bug.cgi?id=69148 This was a case of being surprised by a poorly aritulcated cell size limit, plus an incorrect ASSERT guarding the cell size limit. Reviewed by Oliver Hunt. * heap/MarkedSpace.h: (JSC::MarkedSpace::sizeClassFor): Changed heap size ranges to be inclusive, since it makes the ranges easier to understand. Bumped up the max cell size to support the use case in this bug. Since the atomSize is much bigger than it used to be, there isn't much accounting cost to handling more size classes. Switched to FixedArray, to help catch SizeClass indexing bugs in the future. * heap/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::resetAllocator): (JSC::MarkedSpace::canonicalizeCellLivenessData): Updated for size ranges being inclusive. 2011-09-30 Pierre Rossi <pierre.rossi@gmail.com> [Qt] Build fix: Qt::escape is deprecated in Qt5 https://bugs.webkit.org/show_bug.cgi?id=69162 Use QString::toHtmlEscaped in the Qt5 case. Reviewed by Andreas Kling. * JavaScriptCore.pri: * wtf/qt/UtilsQt.h: Added. (escapeHtml): * wtf/wtf.pri: 2011-09-30 Yuqiang Xian <yuqiang.xian@intel.com> Fix bug in getHostCallReturnValue of DFG JIT on X86 https://bugs.webkit.org/show_bug.cgi?id=69133 Reviewed by Gavin Barraclough. We need to insert the additional argument in the stack slot before return address instead of simply pushing it afterwards. Also getHostCallReturnValue* should be attributed as stdcall to make the stack cleaned up by the callee. * dfg/DFGOperations.cpp: 2011-09-30 Pierre Rossi <pierre.rossi@gmail.com> [Qt] wtf header files are unknown to Qt Creator https://bugs.webkit.org/show_bug.cgi?id=69158 Adding the HEADERS variable in wtf.pri so that the header files can be accessed easily. Reviewed by Andreas Kling. * wtf/wtf.pri: 2011-09-30 Gavin Barraclough <barraclough@apple.com> Merge some more of DFGSpeculativeJIT 32_64/64 https://bugs.webkit.org/show_bug.cgi?id=69164 Reviewed by Oliver Hunt. * dfg/DFGJITCodeGenerator.h: * dfg/DFGJITCodeGenerator32_64.cpp: * dfg/DFGJITCodeGenerator64.cpp: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleBranch): * dfg/DFGSpeculativeJIT.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compileValueAdd): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compileValueAdd): (JSC::DFG::SpeculativeJIT::compileLogicalNot): (JSC::DFG::SpeculativeJIT::compile): 2011-09-30 Mark Hahnenberg <mhahnenberg@apple.com> Add getCallData to MethodTable in ClassInfo https://bugs.webkit.org/show_bug.cgi?id=69024 Reviewed by Sam Weinig. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Added the getCallData to the MethodTable in the ClassInfo struct. * runtime/ClassInfo.h: 2011-09-29 Yuqiang Xian <yuqiang.xian@intel.com> Add op_call/op_constructor support to JSVALUE32_64 DFG JIT https://bugs.webkit.org/show_bug.cgi?id=69120 Reviewed by Gavin Barraclough. Improve the coverage of JSVALUE32_64 DFG JIT. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::tagOfCallData): (JSC::DFG::payloadOfCallData): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::emitCall): 2011-09-29 Yuqiang Xian <yuqiang.xian@intel.com> DFG JIT - register not unlocked after usage in ArithDiv https://bugs.webkit.org/show_bug.cgi?id=69122 Reviewed by Geoffrey Garen. Some allocated register is not unlocked after the usage in ArithDiv. Also there's a typo in "ENBALE_DFG_CONSISTENTCY_CHECK". * dfg/DFGNode.h: * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-29 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSCell::toObject https://bugs.webkit.org/show_bug.cgi?id=68937 Reviewed by Darin Adler. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: De-virtualized JSCell::toObject and changed its implementation to manually check the cases for JSString and JSObject rather than leaving it up to the virtual method call. * runtime/JSCell.cpp: (JSC::JSCell::toObject): * runtime/JSCell.h: Removed JSNotAnObject::toObject because the case for JSObject works for it. Also removed JSObject::toObject because it was essentially the identity function, which is not necessary since toObject is no longer virtual. * runtime/JSNotAnObject.cpp: * runtime/JSNotAnObject.h: * runtime/JSObject.cpp: * runtime/JSObject.h: De-virtualized JSObject::toObject and JSString::toObject. * runtime/JSString.h: 2011-09-29 Gavin Barraclough <barraclough@apple.com> Start refactoring DFGSpeculativeJIT https://bugs.webkit.org/show_bug.cgi?id=69112 Reviewed by Oliver Hunt. Again, move JSVALUE64 code into a DFJSpeculativeJIT64.cpp * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::ValueSource::dump): (JSC::DFG::ValueRecovery::dump): (JSC::DFG::OSRExit::OSRExit): (JSC::DFG::OSRExit::dump): (JSC::DFG::SpeculativeJIT::compilePeepHoleDoubleBranch): (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality): (JSC::DFG::SpeculativeJIT::compilePeepHoleIntegerBranch): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::emitAllocateJSFinalObject): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compare): * dfg/DFGSpeculativeJIT64.cpp: Copied from Source/JavaScriptCore/dfg/DFGSpeculativeJIT.cpp. (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::compile): 2011-09-29 Gavin Barraclough <barraclough@apple.com> Refactor out trivially duplicated code in DFGJITCodeGenerator. https://bugs.webkit.org/show_bug.cgi?id=69109 Reviewed by Oliver Hunt. Some code is trivially redundant between DFGJITCodeGenerator.cpp & DFGJITCodeGenerator32_64.cpp Basically move a JSVALUE64 specific code into a new DFGJITCodeGenerator64.cpp, leave common code in DFGJITCodeGenerator.cpp, and remove copies from DFGJITCodeGenerator32_64.cpp. For some function differences are trivial & make more sense to ifdef individually, and some Operand methods make more sense left in DFGJITCodeGenerator.cpp alongside similar constructors. * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::isKnownNotNumber): (JSC::DFG::JITCodeGenerator::isKnownBoolean): (JSC::DFG::JITCodeGenerator::writeBarrier): (JSC::DFG::JITCodeGenerator::dump): (JSC::DFG::JITCodeGenerator::checkConsistency): (JSC::DFG::GPRTemporary::GPRTemporary): (JSC::DFG::FPRTemporary::FPRTemporary): * dfg/DFGJITCodeGenerator32_64.cpp: * dfg/DFGJITCodeGenerator64.cpp: Copied from Source/JavaScriptCore/dfg/DFGJITCodeGenerator.cpp. * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::branchIfNotCell): * dfg/DFGJITCompilerInlineMethods.h: 2011-09-28 Filip Pizlo <fpizlo@apple.com> DFG JIT should infer which uses of a variable are not aliased https://bugs.webkit.org/show_bug.cgi?id=68593 Reviewed by Oliver Hunt. This separates how a variable is stored (i.e. its virtual register) from how it's predicted. Each variable now takes a VariableAccessData as its operand, instead of the virtual register. The VariableAccessData stores the operand and the prediction. If multiple uses of a variable are aliased, their VariableAccessDatas are unified. This also adds tracking of which argument values are used. It correctly observes that an argument value is not used, if the argument is assigned to inside the function before being used. This also adds tracking of which variables are live at the head of a basic block, and separates that from a variable being live at the tail. Finally, this communicates to both OSR entry and OSR exit code how a variable is predicted at a particular point in the code, rather than just communicating how it was predicted in the entire code block (since with this patch there is no longer the notion of a variable having just one prediction for a code block). * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/ActionablePrediction.h: Added. (JSC::actionablePredictionFromPredictedType): (JSC::valueObeysPrediction): (JSC::actionablePredictionToString): (JSC::ActionablePredictions::ActionablePredictions): (JSC::ActionablePredictions::setArgument): (JSC::ActionablePredictions::argument): (JSC::ActionablePredictions::setVariable): (JSC::ActionablePredictions::variable): (JSC::ActionablePredictions::argumentUpperBound): (JSC::ActionablePredictions::variableUpperBound): (JSC::ActionablePredictions::pack): (JSC::ActionablePredictions::packVector): * bytecode/CodeBlock.h: * bytecode/PredictionTracker.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::newVariableAccessData): (JSC::DFG::ByteCodeParser::getLocal): (JSC::DFG::ByteCodeParser::setLocal): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::setArgument): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::processPhiStack): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::nameOfVariableAccessData): (JSC::DFG::Graph::dump): (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGGraph.h: (JSC::DFG::operandIsArgument): (JSC::DFG::VariableRecord::setFirstTime): (JSC::DFG::BasicBlock::BasicBlock): (JSC::DFG::Graph::predict): (JSC::DFG::Graph::getPrediction): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::noticeOSREntry): * dfg/DFGNode.h: (JSC::DFG::Node::hasVariableAccessData): (JSC::DFG::Node::hasLocal): (JSC::DFG::Node::variableAccessData): (JSC::DFG::Node::local): * dfg/DFGOSREntry.cpp: (JSC::DFG::prepareOSREntry): * dfg/DFGOSREntry.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::ValueSource::dump): (JSC::DFG::OSRExit::OSRExit): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::ValueSource::ValueSource): (JSC::DFG::ValueSource::forPrediction): (JSC::DFG::ValueSource::isSet): (JSC::DFG::ValueSource::kind): (JSC::DFG::ValueSource::nodeIndex): (JSC::DFG::ValueSource::nodeIndexFromKind): (JSC::DFG::ValueSource::kindFromNodeIndex): (JSC::DFG::SpeculativeJIT::isKnownArray): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): (JSC::DFG::SpeculativeJIT::SpeculativeJIT): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::OSRExit::OSRExit): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * wtf/PackedIntVector.h: Added. (WTF::PackedIntVector::PackedIntVector): (WTF::PackedIntVector::operator=): (WTF::PackedIntVector::size): (WTF::PackedIntVector::ensureSize): (WTF::PackedIntVector::resize): (WTF::PackedIntVector::clearAll): (WTF::PackedIntVector::get): (WTF::PackedIntVector::set): (WTF::PackedIntVector::mask): * wtf/Platform.h: * wtf/UnionFind.h: Added. (WTF::UnionFind::UnionFind): (WTF::UnionFind::find): (WTF::UnionFind::unify): 2011-09-29 Oliver Hunt <oliver@apple.com> Build fix. * heap/AllocationSpace.h: 2011-09-29 Oliver Hunt <oliver@apple.com> Add logic to collect dirty objects as roots https://bugs.webkit.org/show_bug.cgi?id=69100 Reviewed by Geoff Garen. This gives us the ability to walk all the MarkedBlocks in an AllocationSpace and collect the dirty objects, and then use them as GC roots. I also rearranged the order of these instructions because it makes them smaller on some platforms with some card sizes. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::markCellCard): * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::markCellCard): * heap/AllocationSpace.cpp: Tidy up the write barrier logic a bit. (JSC::MarkedBlock::gatherDirtyObjects): (JSC::TakeIfDirty::returnValue): (JSC::TakeIfDirty::TakeIfDirty): (JSC::TakeIfDirty::operator()): (JSC::AllocationSpace::gatherDirtyObjects): * heap/AllocationSpace.h: * heap/CardSet.h: (JSC::::isCardMarked): (JSC::::clearCard): * heap/Heap.cpp: (JSC::Heap::markRoots): * heap/Heap.h: (JSC::Heap::writeBarrier): * heap/MarkStack.cpp: (JSC::SlotVisitor::visitChildren): * heap/MarkedBlock.h: (JSC::MarkedBlock::setDirtyObject): (JSC::MarkedBlock::addressOfCardFor): * heap/SlotVisitor.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::emitWriteBarrier): Tidy the write barrier a bit. 2011-09-29 Gavin Barraclough <barraclough@apple.com> Unreviewed windows build fix. * assembler/MacroAssemblerCodeRef.h: * dfg/DFGOperations.h: 2011-09-29 Filip Pizlo <fpizlo@apple.com> Structure transitions involving many (> 64) properties sometimes cause structure corruption https://bugs.webkit.org/show_bug.cgi?id=69102 Reviewed by Darin Adler. Made m_offset an int instead of a signed char. Changed the code to ensure that transitions don't lead to the dictionary kind being forgotten. * runtime/Structure.cpp: (JSC::Structure::Structure): * runtime/Structure.h: 2011-09-29 Yuqiang Xian <yuqiang.xian@intel.com> DFG operation calls should be stdcall in Linux JSVALUE32_64 DFG JIT https://bugs.webkit.org/show_bug.cgi?id=69058 Reviewed by Gavin Barraclough. Also Fixed the stdcall FunctionPtr constructors to make them compiled correctly on Linux * assembler/MacroAssemblerCodeRef.h: (JSC::FunctionPtr::FunctionPtr): 2011-09-29 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSCell::visitChildrenVirtual and remove all other visitChildrenVirtual methods https://bugs.webkit.org/show_bug.cgi?id=68839 Reviewed by Geoffrey Garen. Removed the remaining visitChildrenVirtual methods. This patch completes the process of de-virtualizing visitChildren. * API/JSCallbackObject.h: * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: * debugger/DebuggerActivation.h: * runtime/Arguments.cpp: * runtime/Arguments.h: * runtime/Executable.cpp: * runtime/Executable.h: * runtime/GetterSetter.cpp: * runtime/GetterSetter.h: * runtime/JSActivation.cpp: * runtime/JSActivation.h: * runtime/JSArray.cpp: * runtime/JSArray.h: * runtime/JSFunction.cpp: * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: * runtime/JSGlobalObject.h: * runtime/JSObject.cpp: * runtime/JSPropertyNameIterator.cpp: * runtime/JSPropertyNameIterator.h: * runtime/JSStaticScopeObject.cpp: * runtime/JSStaticScopeObject.h: * runtime/JSValue.h: * runtime/NativeErrorConstructor.cpp: * runtime/NativeErrorConstructor.h: * runtime/RegExpObject.cpp: * runtime/RegExpObject.h: * runtime/Structure.cpp: * runtime/Structure.h: * runtime/StructureChain.cpp: * runtime/StructureChain.h: Inlined the method table access and call to the visitChildren function (the only call sites to visitChildren are here). * heap/MarkStack.cpp: (JSC::SlotVisitor::visitChildren): Changed the field name for the visitChildren function pointer to visitChildren (from visitChildrenFunctionPtr) to make call sites less verbose. * runtime/ClassInfo.h: Discovered JSBoundFunction doesn't have its own ClassInfo (it used JSFunction's ClassInfo) but overrides visitChildren, so it needs to have its own ClassInfo. * runtime/JSBoundFunction.cpp: * runtime/JSBoundFunction.h: Had to move className up to make sure that the virtual destructor in JSObject wasn't the first non-inline virtual method in JSObject (as per the comment in the file). Also moved JSCell::visitChildrenVirtual into JSObject.h in order for it be inline-able to mitigate the cost of an extra method call. Also added a convenience accessor function methodTable() to JSCell to return the MethodTable to make call sites more concise. Implementation is inline in JSObject.h. * runtime/JSObject.h: (JSC::JSCell::methodTable): * runtime/JSCell.h: Added an out of line virtual destructor to JSWrapperObject and ScopeChainNode to appease the vtable gods. It refused to compile if there were no virtual methods in both of these classes due to the presence of a weak vtable pointer. * runtime/JSWrapperObject.cpp: (JSC::JSWrapperObject::~JSWrapperObject): * runtime/JSWrapperObject.h: * runtime/ScopeChain.cpp: (JSC::ScopeChainNode::~ScopeChainNode): * runtime/ScopeChain.h: 2011-09-29 Yuqiang Xian <yuqiang.xian@intel.com> Bug fixes for CreateThis, NewObject and GetByOffset in JSVALUE32_64 DFG JIT https://bugs.webkit.org/show_bug.cgi?id=69075 Reviewed by Gavin Barraclough. * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-29 Yuqiang Xian <yuqiang.xian@intel.com> JSVALUE32_64 DFG JIT failed to be built on 32-bit Linux due to incorrect overloaded OpInfo constructor https://bugs.webkit.org/show_bug.cgi?id=69054 Reviewed by Gavin Barraclough. size_t is equal to uint32_t on most 32-bit platforms, except for Mac OS. * dfg/DFGNode.h: 2011-09-28 Filip Pizlo <fpizlo@apple.com> DFG checkArgumentTypes fails to check boolean predictions https://bugs.webkit.org/show_bug.cgi?id=69059 Reviewed by Gavin Barraclough. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::checkArgumentTypes): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::checkArgumentTypes): 2011-09-28 Gavin Barraclough <barraclough@apple.com> Build fix pt 2 for r96286. * assembler/MacroAssemblerCodeRef.h: 2011-09-28 Ryosuke Niwa <rniwa@webkit.org> Build fix attempt for r96286. * assembler/MacroAssemblerCodeRef.h: 2011-09-28 Gavin Barraclough <barraclough@apple.com> DFG JIT Operations on 32_64 should use stdcall calling convention. https://bugs.webkit.org/show_bug.cgi?id=69046 Reviewed by Sam Weinig. All calls out are expecting stdcall conventions, but the default on OS X are cdecl. Leave D_DFGOperation_DD calls as the one exception, since we want to be able to link directly to std library functions like fmod - leave these calls obeying the default platform calling convention. * assembler/MacroAssemblerCodeRef.h: (JSC::FunctionPtr::FunctionPtr): - Add implicit constructors for std calls. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): - Make this work non-Mac platforms. * dfg/DFGOperations.cpp: (JSC::DFG::operationPutByValInternal): * dfg/DFGOperations.h: - Mark all operations as stdcalls. 2011-09-28 Filip Pizlo <fpizlo@apple.com> DFG JIT falls back on numerical comparisons when it does not recognize a prediction https://bugs.webkit.org/show_bug.cgi?id=68977 Reviewed by Geoffrey Garen. This fixes both the way comparison implementations are selected. It also fixes a bug where comparisons other than equality (like < or >) on objects are compiled as if the comparison was equality. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compare): 2011-09-28 Gavin Barraclough <barraclough@apple.com> Implement callOperation(D_DFGOperation_DD) for DFG JIT 32_64 https://bugs.webkit.org/show_bug.cgi?id=69026 Reviewed by Sam Weinig. * assembler/X86Assembler.h: (JSC::X86Assembler::fstpl): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): 2011-09-28 Gavin Barraclough <barraclough@apple.com> Merge bug#68580, bug#68932 for DFG JIT with JSVALUE32_64 https://bugs.webkit.org/show_bug.cgi?id=69017 Reviewed by Oliver Hunt. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::emitAllocateJSFinalObject): (JSC::DFG::SpeculativeJIT::compile): 2011-09-28 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64679 Fix bugs in Array.prototype this handling. Reviewed by Oliver Hunt. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncJoin): (JSC::arrayProtoFuncConcat): (JSC::arrayProtoFuncPop): (JSC::arrayProtoFuncPush): (JSC::arrayProtoFuncReverse): (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSlice): (JSC::arrayProtoFuncSort): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): (JSC::arrayProtoFuncFilter): (JSC::arrayProtoFuncMap): (JSC::arrayProtoFuncEvery): (JSC::arrayProtoFuncForEach): (JSC::arrayProtoFuncSome): (JSC::arrayProtoFuncReduce): (JSC::arrayProtoFuncReduceRight): (JSC::arrayProtoFuncIndexOf): (JSC::arrayProtoFuncLastIndexOf): - These methods should throw if this value is undefined. 2011-09-27 Yuqiang Xian <yuqiang.xian@intel.com> Value profiling in baseline JIT for JSVALUE32_64 https://bugs.webkit.org/show_bug.cgi?id=68750 Reviewed by Geoff Garen. * jit/JITArithmetic32_64.cpp: (JSC::JIT::emit_op_mul): (JSC::JIT::emit_op_div): * jit/JITCall32_64.cpp: (JSC::JIT::emit_op_call_put_result): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_resolve): (JSC::JIT::emit_op_resolve_base): (JSC::JIT::emit_op_resolve_skip): (JSC::JIT::emit_op_resolve_global): (JSC::JIT::emitSlow_op_resolve_global): (JSC::JIT::emit_op_resolve_with_base): (JSC::JIT::emit_op_resolve_with_this): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_method_check): (JSC::JIT::emit_op_get_by_val): (JSC::JIT::emitSlow_op_get_by_val): (JSC::JIT::emit_op_get_by_id): (JSC::JIT::emitSlow_op_get_by_id): (JSC::JIT::emit_op_get_scoped_var): (JSC::JIT::emit_op_get_global_var): * jit/JITStubCall.h: (JSC::JITStubCall::callWithValueProfiling): 2011-09-28 Yuqiang Xian <yuqiang.xian@intel.com> Wrong integer checks in JSVALUE32_64 DFG JIT https://bugs.webkit.org/show_bug.cgi?id=68985 Reviewed by Geoffrey Garen. * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::fillDouble): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): 2011-09-28 Adam Barth <abarth@webkit.org> Remove empty directories. * wtf/brew: Removed. * wtf/unicode/brew: Removed. 2011-09-27 Filip Pizlo <fpizlo@apple.com> DFG JIT cannot compile op_new_object, op_new_array, op_new_array_buffer, or op_new_regexp https://bugs.webkit.org/show_bug.cgi?id=68580 Reviewed by Oliver Hunt. This implements all four opcodes, but has op_new_regexp turns off by default because it unveils some bad speculation logic when compiling string-validate-input. With op_new_regexp turned off, this is a 5% win on Kraken and a 0.7% speed-up on V8. Neutral on SunSpider. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::callOperation): * dfg/DFGNode.h: (JSC::DFG::Node::hasConstantBuffer): (JSC::DFG::Node::startConstant): (JSC::DFG::Node::numConstants): (JSC::DFG::Node::hasRegexpIndex): (JSC::DFG::Node::regexpIndex): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::emitAllocateJSFinalObject): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isKnownArray): 2011-09-27 Filip Pizlo <fpizlo@apple.com> DFG JIT should speculate more aggressively on reads of array.length https://bugs.webkit.org/show_bug.cgi?id=68932 Reviewed by Oliver Hunt. This is a 2% speed-up on Kraken, neutral elsewhere. * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-27 Gavin Barraclough <barraclough@apple.com> DFG JIT - merge changes between 95905 - 96175 https://bugs.webkit.org/show_bug.cgi?id=68963 Reviewed by Sam Weinig. Merge missing changes from bug#68677, bug#68784, bug#68785. * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): (JSC::DFG::JITCompiler::compileEntry): (JSC::DFG::JITCompiler::compileBody): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-27 Gavin Barraclough <barraclough@apple.com> Get JSVALUE32_64 DFG JIT building on OS X. https://bugs.webkit.org/show_bug.cgi?id=68961 Reviewed by Geoff Garen. * Merge bug #68763 (DFG JIT should not eagerly initialize integer tags in the register file). * Forward-declare functions in DFGOperations.cpp * UNUSED_PARAM for unused arguments * NO_RETURN for unimplemented function that ASSERT_NOT_REACHED * Fix argument types handled by OpInfo constructor. * Use SYMBOL_STRING instead of STRINGIZE for asm symbols. * Add files to Xcode project. 2011-09-27 Yuqiang Xian <yuqiang.xian@intel.com> Bug fixes for GetById, PutById, and GetByOffset in JSVALUE32_64 DFG JIT https://bugs.webkit.org/show_bug.cgi?id=68755 Reviewed by Gavin Barraclough. We need to load/store and repatch both tag and payload of a property for GetById/PutById. Also reorder the loads of tag and payload for GetByOffset as the result tag GPR could reuse the storage GPR. * bytecode/StructureStubInfo.h: * dfg/DFGJITCodeGenerator32_64.cpp: (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::addPropertyAccess): (JSC::DFG::JITCompiler::PropertyAccessRecord::PropertyAccessRecord): * dfg/DFGJITCompiler32_64.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGRepatch.cpp: (JSC::DFG::dfgRepatchByIdSelfAccess): * dfg/DFGSpeculativeJIT32_64.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-24 Gavin Barraclough <barraclough@apple.com> Macro assembler branch8 & 16 methods vary in treatment of upper bits https://bugs.webkit.org/show_bug.cgi?id=68301 Reviewed by Sam Weinig. Fix for branch16 - remove it! No performance impact. * assembler/MacroAssembler.h: * assembler/MacroAssemblerARM.h: * assembler/MacroAssemblerARMv7.h: * assembler/MacroAssemblerMIPS.h: * assembler/MacroAssemblerSH4.h: * assembler/MacroAssemblerX86Common.h: * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::jumpIfCharNotEquals): (JSC::Yarr::YarrGenerator::generatePatternCharacterOnce): (JSC::Yarr::YarrGenerator::generatePatternCharacterFixed): (JSC::Yarr::YarrGenerator::generatePatternCharacterGreedy): (JSC::Yarr::YarrGenerator::backtrackPatternCharacterNonGreedy): 2011-09-27 Mark Hahnenberg <mhahnenberg@apple.com> Add static version of JSCell::getCallData https://bugs.webkit.org/show_bug.cgi?id=68741 Reviewed by Darin Adler. In this patch we just extract the bodies of the virtual getCallData methods throughout the JSCell inheritance hierarchy out into static methods, which are now called from the virtual methods. This is an intermediate step in trying to move the virtual-ness of getCallData into our own method table stored in ClassInfo. We need to convert the methods to static methods because static methods can be represented as function pointers rather than pointers to member functions, and function pointers are smaller and faster to call than pointers to member functions. * API/JSCallbackFunction.cpp: (JSC::JSCallbackFunction::getCallDataVirtual): (JSC::JSCallbackFunction::getCallData): * API/JSCallbackFunction.h: * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::getCallDataVirtual): (JSC::::getCallData): * API/JSObjectRef.cpp: (JSObjectIsFunction): (JSObjectCallAsFunction): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::getCallDataVirtual): (JSC::ArrayConstructor::getCallData): * runtime/ArrayConstructor.h: * runtime/BooleanConstructor.cpp: (JSC::BooleanConstructor::getCallDataVirtual): (JSC::BooleanConstructor::getCallData): * runtime/BooleanConstructor.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::getCallDataVirtual): (JSC::DateConstructor::getCallData): * runtime/DateConstructor.h: * runtime/Error.cpp: (JSC::StrictModeTypeErrorFunction::getCallDataVirtual): (JSC::StrictModeTypeErrorFunction::getCallData): * runtime/ErrorConstructor.cpp: (JSC::ErrorConstructor::getCallDataVirtual): (JSC::ErrorConstructor::getCallData): * runtime/ErrorConstructor.h: * runtime/FunctionConstructor.cpp: (JSC::FunctionConstructor::getCallDataVirtual): (JSC::FunctionConstructor::getCallData): * runtime/FunctionConstructor.h: * runtime/FunctionPrototype.cpp: (JSC::FunctionPrototype::getCallDataVirtual): (JSC::FunctionPrototype::getCallData): * runtime/FunctionPrototype.h: * runtime/InternalFunction.h: * runtime/JSCell.cpp: (JSC::JSCell::getCallDataVirtual): (JSC::JSCell::getCallData): * runtime/JSCell.h: (JSC::getCallData): * runtime/JSFunction.cpp: (JSC::JSFunction::getCallDataVirtual): (JSC::JSFunction::getCallData): * runtime/JSFunction.h: * runtime/JSONObject.cpp: (JSC::Stringifier::Stringifier): (JSC::Stringifier::toJSON): (JSC::Stringifier::appendStringifiedValue): * runtime/JSObject.cpp: (JSC::JSObject::put): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::getCallDataVirtual): (JSC::NativeErrorConstructor::getCallData): * runtime/NativeErrorConstructor.h: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::getCallDataVirtual): (JSC::NumberConstructor::getCallData): * runtime/NumberConstructor.h: * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::getCallDataVirtual): (JSC::ObjectConstructor::getCallData): * runtime/ObjectConstructor.h: * runtime/Operations.cpp: (JSC::jsTypeStringForValue): (JSC::jsIsObjectType): (JSC::jsIsFunctionType): * runtime/PropertySlot.cpp: (JSC::PropertySlot::functionGetter): * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::getCallDataVirtual): (JSC::RegExpConstructor::getCallData): * runtime/RegExpConstructor.h: * runtime/StringConstructor.cpp: (JSC::StringConstructor::getCallDataVirtual): (JSC::StringConstructor::getCallData): * runtime/StringConstructor.h: 2011-09-27 Tim Horton <timothy_horton@apple.com> Rapidly refreshing a feMorphology[erode] with r=0 can sometimes cause display corruption https://bugs.webkit.org/show_bug.cgi?id=68816 <rdar://problem/10186468> Reviewed by Simon Fraser. Add ByteArray::clear, which zeros the memory in the ByteArray. * wtf/ByteArray.h: (WTF::ByteArray::clear): Added. 2011-09-27 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r96131. http://trac.webkit.org/changeset/96131 https://bugs.webkit.org/show_bug.cgi?id=68927 It made 18+ tests crash on all platform (Requested by Ossy_night on #webkit). * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::Interpreter::throwException): * interpreter/Interpreter.h: * jsc.cpp: (GlobalObject::finishCreation): * parser/Parser.h: (JSC::Parser::parse): * runtime/CommonIdentifiers.h: * runtime/Error.cpp: (JSC::addErrorInfo): * runtime/Error.h: 2011-09-27 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSCell::getPrimitiveNumber https://bugs.webkit.org/show_bug.cgi?id=68851 Reviewed by Darin Adler. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Changed JSCell::getPrimitiveNumber to manually handle the dispatch for JSCells (JSObject and JSString in this case). * runtime/JSCell.cpp: (JSC::JSCell::getPrimitiveNumber): * runtime/JSCell.h: Removed JSNotAnObject::getPrimitiveNumber since its return value doesn't matter and it already implements defaultValue, so JSObject::getPrimitiveNumber can cover the case for JSNotAnObject. * runtime/JSNotAnObject.cpp: * runtime/JSNotAnObject.h: De-virtualized JSObject::getPrimitiveNumber and JSString::getPrimitiveNumber and changed them to be const. Also made JSString::getPrimitiveNumber public because it needs to be called from JSCell::getPrimitiveNumber and also since it's no longer virtual, we want people who have a more specific pointer (JSString* instead of JSCell*) to not have to pay the cost of a virtual method call. * runtime/JSObject.cpp: (JSC::JSObject::getPrimitiveNumber): * runtime/JSObject.h: * runtime/JSString.cpp: (JSC::JSString::getPrimitiveNumber): * runtime/JSString.h: 2011-09-27 Juan Carlos Montemayor Elosua <j.mont@me.com> Implement Error.stack https://bugs.webkit.org/show_bug.cgi?id=66994 Reviewed by Oliver Hunt. This patch utilizes topCallFrame to create a stack trace when an error is thrown. Users will also be able to use the stack() command in jsc to get arrays with stack trace information. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::getCallerLine): (JSC::getSourceURLFromCallFrame): (JSC::getStackFrameCodeType): (JSC::Interpreter::getStackTrace): (JSC::Interpreter::throwException): * interpreter/Interpreter.h: (JSC::StackFrame::toString): * jsc.cpp: (GlobalObject::finishCreation): (functionJSCStack): * parser/Parser.h: (JSC::Parser::parse): * runtime/CommonIdentifiers.h: * runtime/Error.cpp: (JSC::addErrorInfo): * runtime/Error.h: 2011-09-27 Carlos Garcia Campos <cgarcia@igalia.com> [GTK] Reorganize header files https://bugs.webkit.org/show_bug.cgi?id=65616 Reviewed by Martin Robinson. Install header files under $libwebkitgtkincludedir/JavaScriptCore. * GNUmakefile.am: Use $libwebkitgtkincludedir. * javascriptcoregtk.pc.in: Use webkitgtk-<api-version> as include dir. 2011-09-26 Geoffrey Garen <ggaren@apple.com> REGRESSION (r95912): Conservative marking doesn't filter out pointers to MarkedBlock metadata https://bugs.webkit.org/show_bug.cgi?id=68860 Reviewed by Oliver Hunt. Bencher says no performance change, maybe a 7% speedup on kraken-imaging-darkroom. * heap/MarkedBlock.h: (JSC::MarkedBlock::isAtomAligned): Renamed atomMask to atomAlignment mask because the mask doesn't produce the actual atom number. (JSC::MarkedBlock::isLiveCell): Testing just for alignment isn't good enough; we also need to test that a pointer is beyond the metadata section of a MarkedBlock, to avoid treating random metadata as a JSCell. 2011-09-26 Mark Hahnenberg <mhahnenberg@apple.com> Make JSCell::toBoolean non-virtual https://bugs.webkit.org/show_bug.cgi?id=67727 Reviewed by Geoffrey Garen. JSCell::toBoolean now manually performs the toBoolean check for objects and strings (where before it was simply virtual and would crash if its implementation was called). Its descendants in JSObject and JSString have also been made non-virtual. JSCell now explicitly covers all cases of toBoolean, so having a virtual implementation of JSCell::toBoolean is no longer necessary. This is part of a larger process of un-virtualizing JSCell. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSCell.cpp: * runtime/JSCell.h: * runtime/JSNotAnObject.cpp: * runtime/JSNotAnObject.h: * runtime/JSObject.h: * runtime/JSString.h: (JSC::JSCell::toBoolean): (JSC::JSValue::toBoolean): 2011-09-26 Chris Marrin <cmarrin@apple.com> Enable requestAnimationFrame on Windows https://bugs.webkit.org/show_bug.cgi?id=68397 Reviewed by Simon Fraser. Enabled REQUEST_ANIMATION_FRAME_TIMER for Windows * wtf/Platform.h: 2011-09-26 Noel Gordon <noel.gordon@gmail.com> [Chromium] Remove DFGAliasTracker.h references from gyp project files https://bugs.webkit.org/show_bug.cgi?id=68787 Reviewed by Geoffrey Garen. DFG/DFGAliasTracker.h was removed in r95389. Cleanup (remove) references to that file from the gyp project files. * JavaScriptCore.gypi: 2011-09-26 Zoltan Herczeg <zherczeg@webkit.org> [Qt]REGRESSION(r95865): It made 4 tests crash https://bugs.webkit.org/show_bug.cgi?id=68780 Reviewed by Oliver Hunt. emitJumpSlowCaseIfNotJSCell(...) cannot be moved away since the next load depends on it. * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_put_by_val): 2011-09-25 Mark Hahnenberg <mhahnenberg@apple.com> Add custom vtable struct to ClassInfo struct https://bugs.webkit.org/show_bug.cgi?id=68567 Reviewed by Oliver Hunt. Declared/defined the MethodTable struct and added it to the ClassInfo struct. Also defined the CREATE_METHOD_TABLE macro to generate these method tables succinctly where they need to be defined. Also added to it the first function to use this macro, visitChildren. This is part of the process of getting rid of all C++ virtual methods in JSCell. Eventually all virtual functions in JSCell that can't easily be converted to non-virtual functions will be put into this custom vtable structure. * runtime/ClassInfo.h: Added the CREATE_METHOD_TABLE macro call as the last argument to each of the ClassInfo structs declared in these classes. This saves us from having to visit each s_info definition in the future when we add more methods to the MethodTable. * API/JSCallbackConstructor.cpp: * API/JSCallbackFunction.cpp: * API/JSCallbackObject.cpp: * JavaScriptCore.exp: * runtime/Arguments.cpp: * runtime/ArrayConstructor.cpp: * runtime/ArrayPrototype.cpp: * runtime/BooleanObject.cpp: * runtime/BooleanPrototype.cpp: * runtime/DateConstructor.cpp: * runtime/DateInstance.cpp: * runtime/DatePrototype.cpp: * runtime/ErrorInstance.cpp: * runtime/ErrorPrototype.cpp: * runtime/ExceptionHelpers.cpp: * runtime/Executable.cpp: * runtime/GetterSetter.cpp: * runtime/InternalFunction.cpp: * runtime/JSAPIValueWrapper.cpp: * runtime/JSActivation.cpp: * runtime/JSArray.cpp: * runtime/JSByteArray.cpp: * runtime/JSFunction.cpp: * runtime/JSGlobalObject.cpp: * runtime/JSONObject.cpp: * runtime/JSObject.cpp: * runtime/JSPropertyNameIterator.cpp: * runtime/JSString.cpp: * runtime/MathObject.cpp: * runtime/NativeErrorConstructor.cpp: * runtime/NumberConstructor.cpp: * runtime/NumberObject.cpp: * runtime/NumberPrototype.cpp: * runtime/ObjectConstructor.cpp: * runtime/ObjectPrototype.cpp: * runtime/RegExp.cpp: * runtime/RegExpConstructor.cpp: * runtime/RegExpObject.cpp: * runtime/RegExpPrototype.cpp: * runtime/ScopeChain.cpp: * runtime/StringConstructor.cpp: * runtime/StringObject.cpp: * runtime/StringPrototype.cpp: * runtime/Structure.cpp: * runtime/StructureChain.cpp: Had to make visitChildren and visitChildrenVirtual protected instead of private because some of the subclasses of JSWrapperObject need access to JSWrapperObject's visitChildren function pointer in their vtable since they don't provide their own implementation. Same for RegExpObject. * runtime/JSWrapperObject.h: * runtime/RegExpObject.h: 2011-09-25 Adam Barth <abarth@webkit.org> Finish removing PLATFORM(BREWMP) by removing associated code https://bugs.webkit.org/show_bug.cgi?id=68779 Reviewed by Sam Weinig. * JavaScriptCore.gyp/JavaScriptCore.gyp: * JavaScriptCore.gypi: * gyp/JavaScriptCore.gyp: * wscript: * wtf/FastMalloc.cpp: (WTF::fastMallocSize): * wtf/Vector.h: * wtf/brew: Removed. * wtf/brew/MainThreadBrew.cpp: Removed. * wtf/brew/OwnPtrBrew.cpp: Removed. * wtf/brew/RefPtrBrew.h: Removed. * wtf/brew/ShellBrew.h: Removed. * wtf/brew/StringBrew.cpp: Removed. * wtf/brew/SystemMallocBrew.h: Removed. * wtf/unicode/brew: Removed. * wtf/unicode/brew/UnicodeBrew.cpp: Removed. * wtf/unicode/brew/UnicodeBrew.h: Removed. 2011-09-25 Filip Pizlo <fpizlo@apple.com> DFG JIT does not count speculation successes correctly https://bugs.webkit.org/show_bug.cgi?id=68785 Reviewed by Geoffrey Garen. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileEntry): (JSC::DFG::JITCompiler::compileBody): * dfg/DFGOperations.cpp: 2011-09-25 Filip Pizlo <fpizlo@apple.com> DFG support for op_resolve_global is not enabled https://bugs.webkit.org/show_bug.cgi?id=68786 Reviewed by Geoffrey Garen. * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): 2011-09-25 Filip Pizlo <fpizlo@apple.com> DFG static prediction code is no longer needed and should be removed https://bugs.webkit.org/show_bug.cgi?id=68784 Reviewed by Oliver Hunt. This gets rid of static prediction code, and ensures that we do not try to compile code where dynamic predictions are not available. This is accomplished by immediately performing an OSR exit wherever a value is retrieved for which no predictions exist. This also adds value profiling for this on functions used for calls. The heuristics for deciding when to optimize code are also tweaked, since it is now profitable to optimize sooner. This may need to be tweaked further, but this patch only makes minimal changes. This results in a 16% speed-up on Kraken/ai-astar, leading to a 3% overall win on Kraken. It's neutral elsewhere. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::shouldOptimizeNow): (JSC::CodeBlock::dumpValueProfiles): * bytecode/CodeBlock.h: * bytecode/PredictedType.cpp: (JSC::predictionToString): * bytecode/PredictedType.h: (JSC::isCellPrediction): (JSC::isObjectPrediction): (JSC::isFinalObjectPrediction): (JSC::isStringPrediction): (JSC::isArrayPrediction): (JSC::isInt32Prediction): (JSC::isDoublePrediction): (JSC::isNumberPrediction): (JSC::isBooleanPrediction): (JSC::mergePredictions): * bytecode/PredictionTracker.h: (JSC::PredictionTracker::predictArgument): (JSC::PredictionTracker::predict): (JSC::PredictionTracker::predictGlobalVar): * bytecode/ValueProfile.cpp: (JSC::ValueProfile::computeUpdatedPrediction): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::set): (JSC::DFG::ByteCodeParser::addCall): (JSC::DFG::ByteCodeParser::getPrediction): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGGraph.h: (JSC::DFG::Graph::predict): (JSC::DFG::Graph::predictGlobalVar): (JSC::DFG::Graph::getMethodCheckPrediction): (JSC::DFG::Graph::getJSConstantPrediction): (JSC::DFG::Graph::getPrediction): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::writeBarrier): (JSC::DFG::JITCodeGenerator::emitBranch): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::getPrediction): * dfg/DFGNode.h: (JSC::DFG::Node::valueOfJSConstantNode): (JSC::DFG::Node::isInt32Constant): (JSC::DFG::Node::isDoubleConstant): (JSC::DFG::Node::isNumberConstant): (JSC::DFG::Node::isBooleanConstant): (JSC::DFG::Node::predict): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::Propagator): (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::isPredictedNumerical): (JSC::DFG::Propagator::logicalNotIsPure): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::shouldSpeculateInteger): (JSC::DFG::SpeculativeJIT::shouldSpeculateDouble): (JSC::DFG::SpeculativeJIT::shouldSpeculateNumber): (JSC::DFG::SpeculativeJIT::shouldNotSpeculateInteger): (JSC::DFG::SpeculativeJIT::shouldSpeculateFinalObject): (JSC::DFG::SpeculativeJIT::shouldSpeculateArray): (JSC::DFG::SpeculativeJIT::shouldSpeculateObject): (JSC::DFG::SpeculativeJIT::shouldSpeculateCell): * jit/JIT.cpp: (JSC::JIT::privateCompile): 2011-09-25 Filip Pizlo <fpizlo@apple.com> DFG JIT Construct opcode takes a this argument even though it's not passed https://bugs.webkit.org/show_bug.cgi?id=68782 Reviewed by Oliver Hunt. This is performance-neutral, mostly. It's a slight speed-up on v8-splay. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::addCall): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::emitCall): 2011-09-25 Filip Pizlo <fpizlo@apple.com> DFG tracking of the value in cachedResultRegister does not handle op_mov correctly https://bugs.webkit.org/show_bug.cgi?id=68781 Reviewed by Oliver Hunt. This takes the simplest approach: it makes the old JIT dumber rather than making the DFG JIT smarter. This is performance-neutral. * jit/JIT.h: (JSC::JIT::canBeOptimized): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_mov): 2011-09-25 Adam Barth <abarth@webkit.org> Remove PLATFORM(HAIKU) and associated code https://bugs.webkit.org/show_bug.cgi?id=68774 Reviewed by Sam Weinig. * JavaScriptCore.gyp/JavaScriptCore.gyp: * JavaScriptCore.gypi: * gyp/JavaScriptCore.gyp: * heap/MachineStackMarker.cpp: * wtf/PageAllocation.h: * wtf/Platform.h: * wtf/StackBounds.cpp: * wtf/haiku: Removed. * wtf/haiku/MainThreadHaiku.cpp: Removed. * wtf/haiku/StringHaiku.cpp: Removed. * wtf/text/WTFString.h: 2011-09-24 Adam Barth <abarth@webkit.org> Always enable ENABLE(OFFLINE_WEB_APPLICATIONS) https://bugs.webkit.org/show_bug.cgi?id=68767 Reviewed by Eric Seidel. * Configurations/FeatureDefines.xcconfig: 2011-09-24 Filip Pizlo <fpizlo@apple.com> JIT implementation of put_by_val increments m_length instead of setting it to index+1 https://bugs.webkit.org/show_bug.cgi?id=68766 Reviewed by Geoffrey Garen. * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_put_by_val): 2011-09-24 Geoffrey Garen <ggaren@apple.com> More build fixage. * heap/ConservativeRoots.cpp: Our system of #includes, it is chaos. 2011-09-24 Filip Pizlo <fpizlo@apple.com> The DFG should not attempt to guess types in the absence of value profiles https://bugs.webkit.org/show_bug.cgi?id=68677 Reviewed by Oliver Hunt. This adds the ForceOSRExit node, which is ignored by the propagator and virtual register allocator (and hence ensuring that liveness analysis works correctly), but forces terminateSpeculativeExecution() in the back-end. This appears to be a slight speed-up on benchmark averages, with ~5% swings on individual benchmarks, in both directions. But it's never a regression on any average, and appears to be a ~1% progression in the SunSpider average. This also adds a bit better debugging support in the old JIT and in DFG, as this was necessary to debug the much more frequent OSR transitions that occur with this change. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::addCall): (JSC::DFG::ByteCodeParser::getStrongPrediction): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompileSlowCases): (JSC::JIT::privateCompile): * jit/JIT.h: 2011-09-24 Geoffrey Garen <ggaren@apple.com> Some Windows build fixage. * heap/MarkedBlock.cpp: (JSC::MarkedBlock::sweep): * heap/MarkedBlock.h: (JSC::MarkedBlock::isLive): Show the compiler that all control paths return a value. There, there, compiler. Everything's going to be OK. * runtime/JSCell.h: (JSC::JSCell::setVPtr): Oops! Unrename this function. 2011-09-24 Geoffrey Garen <ggaren@apple.com> Allocate new objects unmarked https://bugs.webkit.org/show_bug.cgi?id=68764 Reviewed by Oliver Hunt. This is a pre-requisite to using the mark bit to determine object age. ~2% v8 speedup, mostly due to a 12% v8-splay speedup. * heap/MarkedBlock.h: (JSC::MarkedBlock::isLive): (JSC::MarkedBlock::isLiveCell): These two functions are the reason for this patch. They can now determine object liveness without relying on newly allocated objects having their mark bits set. Each MarkedBlock now has a state variable that tells us how to determine whether its cells are live. (This new state variable supercedes the old one about destructor state. The rest of this patch is just refactoring to support the invariants of this new state variable without introducing a performance regression.) (JSC::MarkedBlock::didConsumeFreeList): New function for updating interal state when a block becomes fully allocated. (JSC::MarkedBlock::clearMarks): Folded a state change to 'Marked' into this function because, logically, clearing all mark bits is the first step in saying "mark bits now exactly reflect object liveness". (JSC::MarkedBlock::markCountIsZero): Renamed from isEmpty() to clarify that this function only tells you about the mark bits, so it's only meaningful if you've put the mark bits into a meaningful state before calling it. (JSC::MarkedBlock::forEachCell): Changed to use isLive() helper function instead of testing mark bits, since mark bits are not always the right way to find out if an object is live anymore. (New objects are live, but not marked.) * heap/MarkedBlock.cpp: (JSC::MarkedBlock::recycle): (JSC::MarkedBlock::MarkedBlock): Folded all initialization -- even initialization when recycling an old block -- into the MarkedBlock constructor, for simplicity. (JSC::MarkedBlock::callDestructor): Inlined for speed. Always check for a zapped cell before running a destructor, and always zap after running a destructor. This does not seem to be expensive, and the alternative just creates a too-confusing matrix of possible cell states ((zombie undestructed cell + zombie destructed cell + zapped destructed cell) * 5! permutations for progressing through block states = "Oh my!"). (JSC::MarkedBlock::specializedSweep): (JSC::MarkedBlock::sweep): Maintained and expanded a pre-existing optimization to use template specialization to constant fold lots of branches and elide certain operations entirely during a sweep. Merged four or five functions that were logically about sweeping into this one function pair, so there's only one way to do things now, it's automatically correct, and it's always fast. (JSC::MarkedBlock::zapFreeList): Renamed this function to be more explicit about exactly what it does, and to honor the new block state system. * heap/AllocationSpace.cpp: (JSC::AllocationSpace::allocateBlock): Updated for rename. (JSC::AllocationSpace::freeBlocks): Updated for changed interface. (JSC::TakeIfUnmarked::TakeIfUnmarked): (JSC::TakeIfUnmarked::operator()): (JSC::TakeIfUnmarked::returnValue): Just like isEmpty() above, renamed to clarify that this functor only tests the mark bits, so it's only valid if you've put the mark bits into a meaningful state before calling it. (JSC::AllocationSpace::shrink): Updated for rename. * heap/AllocationSpace.h: (JSC::AllocationSpace::canonicalizeCellLivenessData): Renamed to be a little more specific about what we're making canonical. (JSC::AllocationSpace::forEachCell): Updated for rename. (JSC::AllocationSpace::forEachBlock): No need to canonicalize cell liveness data before iterating blocks -- clients that want iterated blocks to have valid cell lieveness data should make this call for themselves. (And not all clients want it.) * heap/ConservativeRoots.cpp: (JSC::ConservativeRoots::genericAddPointer): Updated for rename. Removed obsolete comment. * heap/Heap.cpp: (JSC::CountFunctor::ClearMarks::operator()): Removed call to notify...() because clearMarks() now does that implicitly. (JSC::Heap::destroy): Make sure to canonicalize before tear-down, since tear-down tests cell liveness when running destructors. (JSC::Heap::markRoots): (JSC::Heap::collect): Moved weak reference harvesting out of markRoots() and into collect, since it strictly depends on root marking, and does not contribute to root marking. (JSC::Heap::canonicalizeCellLivenessData): Renamed to be a little more specific about what we're making canonical. * heap/Heap.h: (JSC::Heap::forEachProtectedCell): No need to canonicalize cell liveness data before iterating protected cells, since we know they're all live, and don't need to test for it. * heap/Local.h: (JSC::::set): Can't make the same ASSERT we used to because we just don't have the mark bits for it anymore. Perhaps we can bring this ASSERT back in a weaker form in the future. * heap/MarkedSpace.cpp: (JSC::MarkedSpace::addBlock): (JSC::MarkedSpace::removeBlock): Updated for interface change. (JSC::MarkedSpace::canonicalizeCellLivenessData): Renamed to be a little more specific about what we're making canonical. * heap/MarkedSpace.h: (JSC::MarkedSpace::allocate): (JSC::MarkedSpace::SizeClass::SizeClass): (JSC::MarkedSpace::SizeClass::resetAllocator): (JSC::MarkedSpace::SizeClass::zapFreeList): Simplified this allocator functionality a bit. We now track only one block -- "currentBlock" -- and rely on its internal state to know whether it has more cells to allocate. * heap/Weak.h: (JSC::Weak::set): Can't make the same ASSERT we used to because we just don't have the mark bits for it anymore. Perhaps we can bring this ASSERT back in a weaker form in the future. * runtime/JSCell.h: (JSC::JSCell::vptr): (JSC::JSCell::zap): (JSC::JSCell::isZapped): (JSC::isZapped): Made zapping a property of JSCell, for a little abstraction. In the future, exactly how a JSCell zaps itself will change, as the internal representation of JSCell changes. 2011-09-24 Filip Pizlo <fpizlo@apple.com> DFG JIT should not eagerly initialize integer tags in the register file https://bugs.webkit.org/show_bug.cgi?id=68763 Reviewed by Oliver Hunt. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::ValueRecovery::dump): (JSC::DFG::OSRExit::OSRExit): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::ValueRecovery::alreadyInRegisterFileAsUnboxedInt32): (JSC::DFG::OSRExit::operandForArgument): (JSC::DFG::OSRExit::operandForIndex): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): 2011-09-23 Yuqiang Xian <yuqiang.xian@intel.com> Add JSVALUE32_64 support to DFG JIT https://bugs.webkit.org/show_bug.cgi?id=67460 Reviewed by Gavin Barraclough. This is the initial attempt to add JSVALUE32_64 support to DFG JIT. It's tested on IA32 Linux EFL port currently. It still cannot run all the test cases and benchmarks so should be turned off now. The major work includes: 1) dealing with JSVALUE32_64 data format in DFG JIT; 2) bindings between 64-bit JS Value and 32-bit registers; 3) handling of function calls. Currently for DFG operation function calls we follow the X86 cdecl calling convention on Linux, and the implementation is in a naive way by pushing the arguments into stack one by one. The known issues include: 1) some code duplicates unnecessarily, especially in Speculative JIT code generation, where most of the operations on SpeculataInteger / SpeculateDouble should be identical to the JSVALUE64 code. Refactoring is needed in the future; 2) lack of op_call and op_construct support, comparing to current JSVALUE64 DFG; 3) currently integer speculations assume to be StrictInt32; 4) lack of JSBoolean speculations; 5) boxing and unboxing doubles could be improved; 6) DFG X86 register description is different with the baseline JIT, the timeoutCheckRegister is used for general purpose usage; 7) calls to runtime functions with primitive double parameters (e.g. fmod) don't work. Support needs to be added to the assembler to implement the mechanism of passing double parameters for X86 cdecl convention. And there should be many other hidden bugs which should be exposed and resolved in later debugging process. * CMakeListsEfl.txt: * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::loadDouble): (JSC::MacroAssemblerX86::storeDouble): * assembler/X86Assembler.h: (JSC::X86Assembler::movsd_rm): * bytecode/StructureStubInfo.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGFPRInfo.h: (JSC::DFG::FPRInfo::debugName): * dfg/DFGGPRInfo.h: (JSC::DFG::GPRInfo::toRegister): (JSC::DFG::GPRInfo::toIndex): (JSC::DFG::GPRInfo::debugName): * dfg/DFGGenerationInfo.h: (JSC::DFG::needDataFormatConversion): (JSC::DFG::GenerationInfo::initJSValue): (JSC::DFG::GenerationInfo::initDouble): (JSC::DFG::GenerationInfo::gpr): (JSC::DFG::GenerationInfo::tagGPR): (JSC::DFG::GenerationInfo::payloadGPR): (JSC::DFG::GenerationInfo::fpr): (JSC::DFG::GenerationInfo::fillJSValue): (JSC::DFG::GenerationInfo::fillCell): (JSC::DFG::GenerationInfo::fillDouble): * dfg/DFGJITCodeGenerator.cpp: * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::allocate): (JSC::DFG::JITCodeGenerator::use): (JSC::DFG::JITCodeGenerator::registersMatched): (JSC::DFG::JITCodeGenerator::silentSpillGPR): (JSC::DFG::JITCodeGenerator::silentFillGPR): (JSC::DFG::JITCodeGenerator::silentFillFPR): (JSC::DFG::JITCodeGenerator::silentSpillAllRegisters): (JSC::DFG::JITCodeGenerator::silentFillAllRegisters): (JSC::DFG::JITCodeGenerator::boxDouble): (JSC::DFG::JITCodeGenerator::unboxDouble): (JSC::DFG::JITCodeGenerator::spill): (JSC::DFG::addressOfDoubleConstant): (JSC::DFG::integerResult): (JSC::DFG::jsValueResult): (JSC::DFG::setupResults): (JSC::DFG::callOperation): (JSC::JSValueOperand::JSValueOperand): (JSC::JSValueOperand::~JSValueOperand): (JSC::JSValueOperand::isDouble): (JSC::JSValueOperand::fill): (JSC::JSValueOperand::tagGPR): (JSC::JSValueOperand::payloadGPR): (JSC::JSValueOperand::fpr): (JSC::GPRTemporary::~GPRTemporary): (JSC::GPRTemporary::gpr): (JSC::GPRResult2::GPRResult2): * dfg/DFGJITCodeGenerator32_64.cpp: Added. (JSC::DFG::JITCodeGenerator::clearGenerationInfo): (JSC::DFG::JITCodeGenerator::fillInteger): (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::fillStorage): (JSC::DFG::JITCodeGenerator::useChildren): (JSC::DFG::JITCodeGenerator::isStrictInt32): (JSC::DFG::JITCodeGenerator::isKnownInteger): (JSC::DFG::JITCodeGenerator::isKnownNumeric): (JSC::DFG::JITCodeGenerator::isKnownCell): (JSC::DFG::JITCodeGenerator::isKnownNotInteger): (JSC::DFG::JITCodeGenerator::isKnownNotNumber): (JSC::DFG::JITCodeGenerator::isKnownBoolean): (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToNumber): (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): (JSC::DFG::JITCodeGenerator::nonSpeculativeUInt32ToNumber): (JSC::DFG::JITCodeGenerator::nonSpeculativeKnownConstantArithOp): (JSC::DFG::JITCodeGenerator::nonSpeculativeBasicArithOp): (JSC::DFG::JITCodeGenerator::nonSpeculativeArithMod): (JSC::DFG::JITCodeGenerator::nonSpeculativeCheckHasInstance): (JSC::DFG::JITCodeGenerator::nonSpeculativeInstanceOf): (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::writeBarrier): (JSC::DFG::JITCodeGenerator::cachedPutById): (JSC::DFG::JITCodeGenerator::cachedGetMethod): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompareNull): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranchNull): (JSC::DFG::JITCodeGenerator::nonSpeculativeCompareNull): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): (JSC::DFG::JITCodeGenerator::nonSpeculativeCompare): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeStrictEq): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeStrictEq): (JSC::DFG::JITCodeGenerator::nonSpeculativeStrictEq): (JSC::DFG::JITCodeGenerator::emitBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativeLogicalNot): (JSC::DFG::JITCodeGenerator::emitCall): (JSC::DFG::JITCodeGenerator::speculationCheck): (JSC::DFG::dataFormatString): (JSC::DFG::JITCodeGenerator::dump): (JSC::DFG::JITCodeGenerator::checkConsistency): (JSC::DFG::GPRTemporary::GPRTemporary): (JSC::DFG::FPRTemporary::FPRTemporary): * dfg/DFGJITCompiler.cpp: * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::tagForGlobalVar): (JSC::DFG::JITCompiler::payloadForGlobalVar): (JSC::DFG::JITCompiler::appendCallWithExceptionCheck): (JSC::DFG::JITCompiler::addressOfDoubleConstant): (JSC::DFG::JITCompiler::boxDouble): (JSC::DFG::JITCompiler::unboxDouble): (JSC::DFG::JITCompiler::addPropertyAccess): (JSC::DFG::JITCompiler::PropertyAccessRecord::PropertyAccessRecord): * dfg/DFGJITCompiler32_64.cpp: Added. (JSC::DFG::JITCompiler::fillNumericToDouble): (JSC::DFG::JITCompiler::fillInt32ToInteger): (JSC::DFG::JITCompiler::fillToJS): (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): (JSC::DFG::JITCompiler::linkOSRExits): (JSC::DFG::JITCompiler::compileEntry): (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::link): (JSC::DFG::JITCompiler::compile): (JSC::DFG::JITCompiler::compileFunction): (JSC::DFG::JITCompiler::jitAssertIsInt32): (JSC::DFG::JITCompiler::jitAssertIsJSInt32): (JSC::DFG::JITCompiler::jitAssertIsJSNumber): (JSC::DFG::JITCompiler::jitAssertIsJSDouble): (JSC::DFG::JITCompiler::jitAssertIsCell): (JSC::DFG::JITCompiler::emitCount): (JSC::DFG::JITCompiler::setSamplingFlag): (JSC::DFG::JITCompiler::clearSamplingFlag): * dfg/DFGJITCompilerInlineMethods.h: Added. (JSC::DFG::JITCompiler::emitLoadTag): (JSC::DFG::JITCompiler::emitLoadPayload): (JSC::DFG::JITCompiler::emitLoad): (JSC::DFG::JITCompiler::emitLoad2): (JSC::DFG::JITCompiler::emitLoadDouble): (JSC::DFG::JITCompiler::emitLoadInt32ToDouble): (JSC::DFG::JITCompiler::emitStore): (JSC::DFG::JITCompiler::emitStoreInt32): (JSC::DFG::JITCompiler::emitStoreCell): (JSC::DFG::JITCompiler::emitStoreBool): (JSC::DFG::JITCompiler::emitStoreDouble): * dfg/DFGNode.h: * dfg/DFGOperations.cpp: * dfg/DFGRepatch.cpp: (JSC::DFG::generateProtoChainAccessStub): (JSC::DFG::tryCacheGetByID): (JSC::DFG::tryBuildGetByIDList): (JSC::DFG::tryCachePutByID): * dfg/DFGSpeculativeJIT.cpp: * dfg/DFGSpeculativeJIT.h: (JSC::DFG::ValueRecovery::inGPR): (JSC::DFG::ValueRecovery::inPair): (JSC::DFG::ValueRecovery::tagGPR): (JSC::DFG::ValueRecovery::payloadGPR): * dfg/DFGSpeculativeJIT32_64.cpp: Added. (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::ValueSource::dump): (JSC::DFG::ValueRecovery::dump): (JSC::DFG::OSRExit::OSRExit): (JSC::DFG::OSRExit::dump): (JSC::DFG::SpeculativeJIT::fillSpeculateInt): (JSC::DFG::SpeculativeJIT::fillSpeculateIntStrict): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compilePeepHoleIntegerBranch): (JSC::DFG::SpeculativeJIT::convertToDouble): (JSC::DFG::SpeculativeJIT::compilePeepHoleDoubleBranch): (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality): (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::compileMovHint): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::initializeVariableTypes): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * runtime/JSValue.h: 2011-09-23 Filip Pizlo <fpizlo@apple.com> wtf/BitVector.h has a variety of bugs which manifest when the vector grows beyond 63 bits https://bugs.webkit.org/show_bug.cgi?id=68746 Reviewed by Oliver Hunt. Out-of-lined slow path code in BitVector so that not every user of CodeBlock ends up having to compile it. Fixed a variety of index computation and size computation bugs. I have not seen these issues manifest themselves, but they are blocking a patch that uses BitVector more aggressively. * GNUmakefile.list.am: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/BitVector.cpp: Added. (BitVector::BitVector): (BitVector::operator=): (BitVector::resize): (BitVector::clearAll): (BitVector::OutOfLineBits::create): (BitVector::OutOfLineBits::destroy): (BitVector::resizeOutOfLine): * wtf/BitVector.h: (WTF::BitVector::ensureSize): (WTF::BitVector::get): (WTF::BitVector::set): (WTF::BitVector::clear): (WTF::BitVector::byteCount): (WTF::BitVector::OutOfLineBits::numWords): (WTF::BitVector::OutOfLineBits::bits): (WTF::BitVector::outOfLineBits): * wtf/CMakeLists.txt: * wtf/wtf.pri: 2011-09-23 Adam Klein <adamk@chromium.org> Add ENABLE_MUTATION_OBSERVERS feature flag https://bugs.webkit.org/show_bug.cgi?id=68732 Reviewed by Ojan Vafai. This flag will guard an implementation of the "Mutation Observers" proposed in http://lists.w3.org/Archives/Public/public-webapps/2011JulSep/1622.html * Configurations/FeatureDefines.xcconfig: 2011-09-23 Mark Hahnenberg <mhahnenberg@apple.com> De-virtualize JSCell::getJSNumber https://bugs.webkit.org/show_bug.cgi?id=68651 Reviewed by Oliver Hunt. Added a new JSType to check whether or not something is a NumberObject (which includes NumberPrototype) in TypeInfo::isNumberObject because there's not currently a better way to determine whether something is indeed a NumberObject. Also de-virtualized JSCell::getJSNumber, having it check the TypeInfo for whether the object is a NumberObject or not. This patch is part of the larger process of de-virtualizing JSCell. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSCell.cpp: (JSC::JSCell::getJSNumber): * runtime/JSCell.h: (JSC::JSValue::getJSNumber): * runtime/JSType.h: * runtime/JSTypeInfo.h: (JSC::TypeInfo::isNumberObject): * runtime/JSValue.h: * runtime/NumberObject.cpp: (JSC::NumberObject::getJSNumber): * runtime/NumberObject.h: (JSC::NumberObject::createStructure): * runtime/NumberPrototype.h: (JSC::NumberPrototype::createStructure): 2011-09-23 Filip Pizlo <fpizlo@apple.com> Resolve opcodes should have value profiling. https://bugs.webkit.org/show_bug.cgi?id=68723 Reviewed by Oliver Hunt. This adds value profiling to all forms of op_resolve in the old JIT, and patches that information into the DFG along with performing the appropriate type propagation. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.h: (JSC::DFG::Graph::predict): * dfg/DFGNode.h: (JSC::DFG::Node::hasIdentifier): (JSC::DFG::Node::resolveGlobalDataIndex): (JSC::DFG::Node::hasPrediction): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_resolve): (JSC::JIT::emit_op_resolve_base): (JSC::JIT::emit_op_resolve_skip): (JSC::JIT::emit_op_resolve_global): (JSC::JIT::emitSlow_op_resolve_global): (JSC::JIT::emit_op_resolve_with_base): (JSC::JIT::emit_op_resolve_with_this): (JSC::JIT::emitSlow_op_resolve_global_dynamic): * jit/JITStubCall.h: (JSC::JITStubCall::callWithValueProfiling): 2011-09-23 Oliver Hunt <oliver@apple.com> Fix windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-09-23 Gavin Barraclough <barraclough@apple.com> Strict mode does not work in non-trivial nested functions. https://bugs.webkit.org/show_bug.cgi?id=68740 Reviewed by Oliver Hunt. Function-info caching does not preserve all state that it should. * parser/JSParser.cpp: (JSC::JSParser::Scope::saveFunctionInfo): (JSC::JSParser::Scope::restoreFunctionInfo): (JSC::JSParser::parseFunctionInfo): * parser/SourceProviderCacheItem.h: 2011-09-23 Filip Pizlo <fpizlo@apple.com> ValueToDouble handling in prediction propagation should be ASSERT_NOT_REACHED https://bugs.webkit.org/show_bug.cgi?id=68724 Reviewed by Oliver Hunt. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): 2011-09-23 Oliver Hunt <oliver@apple.com> Build fix. * JavaScriptCore.xcodeproj/project.pbxproj: 2011-09-23 Filip Pizlo <fpizlo@apple.com> DFG implementation of PutScopedVar corrupts register allocation https://bugs.webkit.org/show_bug.cgi?id=68735 Reviewed by Oliver Hunt. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-23 Oliver Hunt <oliver@apple.com> Make write barriers actually do something when enabled https://bugs.webkit.org/show_bug.cgi?id=68717 Reviewed by Geoffrey Garen. Add a basic card marking style write barrier to JSC (currently turned off). This requires two scratch registers in the JIT so there was some register re-arranging to satisfy that requirement. Happily this produced a minor perf bump in sunspider (~0.5%). Turning the barriers on causes an overall regression of around 1.5% * JavaScriptCore.exp: * JavaScriptCore.xcodeproj/project.pbxproj: * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::store8): * assembler/X86Assembler.h: (JSC::X86Assembler::movb_i8m): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::isKnownNotCell): (JSC::DFG::JITCodeGenerator::writeBarrier): (JSC::DFG::JITCodeGenerator::markCellCard): (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCodeGenerator.h: * dfg/DFGRepatch.cpp: (JSC::DFG::tryCachePutByID): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * heap/CardSet.h: Added. (JSC::CardSet::CardSet): (JSC::::cardForAtom): (JSC::::cardMarkedForAtom): (JSC::::markCardForAtom): * heap/Heap.cpp: * heap/Heap.h: (JSC::Heap::addressOfCardFor): (JSC::Heap::writeBarrierFastCase): * heap/MarkedBlock.h: (JSC::MarkedBlock::setDirtyObject): (JSC::MarkedBlock::addressOfCardFor): (JSC::MarkedBlock::offsetOfCards): * jit/JIT.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_put_by_val): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::emit_op_put_scoped_var): (JSC::JIT::emit_op_put_global_var): (JSC::JIT::emitWriteBarrier): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_put_by_val): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::emitSlow_op_put_by_id): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::emit_op_put_scoped_var): (JSC::JIT::emit_op_put_global_var): 2011-09-23 Thouraya ANDOLSI <thouraya.andolsi@st.com> https://bugs.webkit.org/show_bug.cgi?id=68077 SH4 assemblers doesn't refer to executable memory handle. Reviewed by Gavin Barraclough. * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::branch8): * assembler/SH4Assembler.h: (JSC::SH4Assembler::executableCopy): 2011-09-23 Oliver Hunt <oliver@apple.com> PutScopedVar nodes should report that it has a var number https://bugs.webkit.org/show_bug.cgi?id=68721 Reviewed by Anders Carlsson. Another assertion fix. * dfg/DFGNode.h: (JSC::DFG::Node::hasVarNumber): 2011-09-23 Oliver Hunt <oliver@apple.com> Add a bunch of unhandled node types to the propagator https://bugs.webkit.org/show_bug.cgi?id=68716 Reviewed by Darin Adler. Remove the ASSERT_NOT_REACHED() default for debug builds in the prediction propagator, this way unhandled nodes will just cause compile time failures rather than failing at some point in the future. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): 2011-09-23 Mark Hahnenberg <mhahnenberg@apple.com> Add static version of JSCell::visitChildren https://bugs.webkit.org/show_bug.cgi?id=68404 Reviewed by Darin Adler. In this patch we just extract the bodies of the virtual visitChildren methods throughout the JSCell inheritance hierarchy out into static methods, which are now called from the virtual methods. This is an intermediate step in trying to move the virtual-ness of visitChildren into our own custom vtable stored in ClassInfo. We need to convert the methods to static methods in order to be able to more easily store and refer to them in our custom vtable since normal member methods store some implicit information in their types, making it impossible to store them generically in ClassInfo. * API/JSCallbackObject.h: (JSC::JSCallbackObject::visitChildrenVirtual): (JSC::JSCallbackObject::visitChildren): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::visitChildrenVirtual): (JSC::DebuggerActivation::visitChildren): * debugger/DebuggerActivation.h: * heap/MarkStack.cpp: (JSC::SlotVisitor::visitChildren): (JSC::SlotVisitor::drain): * runtime/Arguments.cpp: (JSC::Arguments::visitChildrenVirtual): (JSC::Arguments::visitChildren): * runtime/Arguments.h: * runtime/Executable.cpp: (JSC::EvalExecutable::visitChildrenVirtual): (JSC::EvalExecutable::visitChildren): (JSC::ProgramExecutable::visitChildrenVirtual): (JSC::ProgramExecutable::visitChildren): (JSC::FunctionExecutable::visitChildrenVirtual): (JSC::FunctionExecutable::visitChildren): * runtime/Executable.h: * runtime/GetterSetter.cpp: (JSC::GetterSetter::visitChildrenVirtual): (JSC::GetterSetter::visitChildren): * runtime/GetterSetter.h: * runtime/JSActivation.cpp: (JSC::JSActivation::visitChildrenVirtual): (JSC::JSActivation::visitChildren): * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::visitChildrenVirtual): (JSC::JSArray::visitChildren): * runtime/JSArray.h: * runtime/JSBoundFunction.cpp: (JSC::JSBoundFunction::visitChildrenVirtual): (JSC::JSBoundFunction::visitChildren): * runtime/JSBoundFunction.h: * runtime/JSCell.h: (JSC::JSCell::visitChildrenVirtual): (JSC::JSCell::visitChildren): * runtime/JSFunction.cpp: (JSC::JSFunction::visitChildrenVirtual): (JSC::JSFunction::visitChildren): * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::visitChildrenVirtual): (JSC::JSGlobalObject::visitChildren): * runtime/JSGlobalObject.h: * runtime/JSObject.cpp: (JSC::JSObject::visitChildrenVirtual): (JSC::JSObject::visitChildren): * runtime/JSObject.h: (JSC::JSObject::visitChildrenDirect): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::visitChildrenVirtual): (JSC::JSPropertyNameIterator::visitChildren): * runtime/JSPropertyNameIterator.h: * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::visitChildrenVirtual): (JSC::JSStaticScopeObject::visitChildren): * runtime/JSStaticScopeObject.h: * runtime/JSWrapperObject.cpp: (JSC::JSWrapperObject::visitChildrenVirtual): (JSC::JSWrapperObject::visitChildren): * runtime/JSWrapperObject.h: * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::visitChildrenVirtual): (JSC::NativeErrorConstructor::visitChildren): * runtime/NativeErrorConstructor.h: * runtime/RegExpObject.cpp: (JSC::RegExpObject::visitChildrenVirtual): (JSC::RegExpObject::visitChildren): * runtime/RegExpObject.h: * runtime/ScopeChain.cpp: (JSC::ScopeChainNode::visitChildrenVirtual): (JSC::ScopeChainNode::visitChildren): * runtime/ScopeChain.h: * runtime/Structure.cpp: (JSC::Structure::visitChildrenVirtual): (JSC::Structure::visitChildren): * runtime/Structure.h: * runtime/StructureChain.cpp: (JSC::StructureChain::visitChildrenVirtual): (JSC::StructureChain::visitChildren): * runtime/StructureChain.h: 2011-09-23 Oliver Hunt <oliver@apple.com> Node propagation doesn't handle PutScopedVar https://bugs.webkit.org/show_bug.cgi?id=68713 Reviewed by Sam Weinig. This was causing assertion failures. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): 2011-09-23 Anders Carlsson <andersca@apple.com> Make sure to define OVERRIDE and FINAL for older builds of clang. * wtf/Compiler.h: 2011-09-23 Gavin Barraclough <barraclough@apple.com> Implement op_resolve_global in the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=68704 Reviewed by Oliver Hunt. This is performance neutral, but increases coverage. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGNode.h: (JSC::DFG::Node::hasIdentifier): (JSC::DFG::Node::resolveInfoIndex): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-23 Mark Rowe <mrowe@apple.com> Define BUILDING_ON_LION / TARGETING_LION when appropriate in Platform.h. * wtf/Platform.h: 2011-09-22 Anders Carlsson <andersca@apple.com> We should add support for OVERRIDE and FINAL annotations https://bugs.webkit.org/show_bug.cgi?id=68654 Reviewed by David Hyatt. Add OVERRIDE and FINAL macros for compilers that support them. * wtf/Compiler.h: 2011-09-22 Filip Pizlo <fpizlo@apple.com> GetScopedVar should have value profiling https://bugs.webkit.org/show_bug.cgi?id=68676 Reviewed by Oliver Hunt. Added GetScopedVar value profiling and predictin propagation. Added GetScopeChain to CSE. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.h: (JSC::DFG::Graph::predict): * dfg/DFGNode.h: (JSC::DFG::Node::hasPrediction): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::getScopeChainLoadElimination): (JSC::DFG::Propagator::performNodeCSE): * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_get_scoped_var): 2011-09-22 Filip Pizlo <fpizlo@apple.com> PPC build fix, part 3. * runtime/Executable.cpp: (JSC::FunctionExecutable::compileForConstructInternal): 2011-09-22 Filip Pizlo <fpizlo@apple.com> Another PPC build fix. * runtime/Executable.cpp: * runtime/Executable.h: 2011-09-22 Dean Jackson <dino@apple.com> Add ENABLE_CSS_FILTERS https://bugs.webkit.org/show_bug.cgi?id=68652 Reviewed by Simon Fraser. * Configurations/FeatureDefines.xcconfig: 2011-09-22 Gavin Barraclough <barraclough@apple.com> Incorrect this value passed to callbacks. https://bugs.webkit.org/show_bug.cgi?id=68668 Reviewed by Oliver Hunt. From Array/String prototype function. Should be undefined, but global object is passed instead (this is visible for strict callbacks). * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncSort): (JSC::arrayProtoFuncFilter): (JSC::arrayProtoFuncMap): (JSC::arrayProtoFuncEvery): (JSC::arrayProtoFuncForEach): (JSC::arrayProtoFuncSome): * runtime/JSArray.cpp: (JSC::AVLTreeAbstractorForArrayCompare::compare_key_key): (JSC::JSArray::sort): * runtime/StringPrototype.cpp: (JSC::stringProtoFuncReplace): 2011-09-22 Gavin Barraclough <barraclough@apple.com> Function.prototype.bind.length shoudl be 1. Rubber stamped by Olier Hunt. * runtime/FunctionPrototype.cpp: (JSC::FunctionPrototype::addFunctionProperties): 2011-09-22 Filip Pizlo <fpizlo@apple.com> PPC build fix. * bytecode/CodeBlock.h: 2011-09-22 Gavin Barraclough <barraclough@apple.com> Windows build fix pt. 2 * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-09-22 Gavin Barraclough <barraclough@apple.com> Windows build fix pt. 1 * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-09-21 Filip Pizlo <fpizlo@apple.com> DFG JIT does not support to_primitive or strcat https://bugs.webkit.org/show_bug.cgi?id=68582 Reviewed by Darin Adler. This adds functional support for to_primitive and strcat. It focuses on minimizing the amount of code emitted on to_primitive (if we know that it is a primitive or can speculate cheaply, then we omit the slow path) and on keeping the implementation of strcat simple while leveraging whatever optimizations we have already. In particular, unlike the Call and Construct nodes which require extending the size of the DFG's callee registers, StrCat takes advantage of the fact that no JS code can run while StrCat is in progress and uses a scratch buffer, rather than the register file, to store the list of values to concatenate. This was done mainly to keep the code simple, but there are probably other benefits to keeping call frame sizes down. Essentially, this patch ensures that the presence of an op_strcat does not mess up any other optimizations we might do while ensuring that if you do execute it, it'll work about as well as you'd expect. When combined with the previous patch for integer division, this is a 14% speed-up on Kraken. Without it, it would have been a 2% loss. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::TrustedImmPtr::TrustedImmPtr): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::callOperation): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): * dfg/DFGNode.h: * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::~JSGlobalData): * runtime/JSGlobalData.h: (JSC::JSGlobalData::scratchBufferForSize): 2011-09-22 Filip Pizlo <fpizlo@apple.com> DFG JIT should support integer division https://bugs.webkit.org/show_bug.cgi?id=68597 Reviewed by Darin Adler. This adds support for ArithDiv speculating integer, and speculating that the result is integer (i.e. remainder = 0). This is a 4% win on Kraken and a 1% loss on V8. * bytecode/CodeBlock.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::makeDivSafe): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGNode.h: (JSC::DFG::Node::hasArithNodeFlags): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateArithNodeFlags): (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::fixupNode): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JITArithmetic.cpp: (JSC::JIT::emit_op_div): 2011-09-22 Oliver Hunt <oliver@apple.com> Implement put_scoped_var in the DFG jit https://bugs.webkit.org/show_bug.cgi?id=68653 Reviewed by Gavin Barraclough. Naive implementation of put_scoped_var. Same story as the get_scoped_var implementation, although I've hoisted scope object acquisition into a separate dfg node. Ideally in the future we would reuse the resolved scope chain object, but for now we don't. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGNode.h: (JSC::DFG::Node::hasScopeChainDepth): (JSC::DFG::Node::scopeChainDepth): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-22 Gavin Barraclough <barraclough@apple.com> Implement Function.prototype.bind https://bugs.webkit.org/show_bug.cgi?id=26382 Reviewed by Sam Weinig. This patch provides a basic functional implementation for Function.bind. It should (hopefully!) be fully functionally correct, and the bound functions can be called to quickly (since they are a subclass of JSFunction, not InternalFunction), but we'll probably want to follow up with some optimization work to keep bound calls in JIT code. * JavaScriptCore.JSVALUE32_64only.exp: * JavaScriptCore.JSVALUE64only.exp: * JavaScriptCore.exp: * JavaScriptCore.xcodeproj/project.pbxproj: * jit/JITStubs.cpp: (JSC::JITThunks::hostFunctionStub): * jit/JITStubs.h: * jsc.cpp: (GlobalObject::addFunction): * runtime/CommonIdentifiers.h: * runtime/ConstructData.h: * runtime/Executable.h: (JSC::NativeExecutable::NativeExecutable): * runtime/FunctionPrototype.cpp: (JSC::FunctionPrototype::addFunctionProperties): (JSC::functionProtoFuncBind): * runtime/FunctionPrototype.h: * runtime/JSBoundFunction.cpp: Added. (JSC::boundFunctionCall): (JSC::boundFunctionConstruct): (JSC::JSBoundFunction::create): (JSC::JSBoundFunction::hasInstance): (JSC::JSBoundFunction::getOwnPropertySlot): (JSC::JSBoundFunction::getOwnPropertyDescriptor): (JSC::JSBoundFunction::JSBoundFunction): (JSC::JSBoundFunction::finishCreation): * runtime/JSBoundFunction.h: Added. (JSC::JSBoundFunction::targetFunction): (JSC::JSBoundFunction::boundThis): (JSC::JSBoundFunction::boundArgs): (JSC::JSBoundFunction::createStructure): * runtime/JSFunction.cpp: (JSC::JSFunction::create): (JSC::JSFunction::finishCreation): (JSC::createDescriptorForThrowingProperty): (JSC::JSFunction::getOwnPropertySlot): * runtime/JSFunction.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::getHostFunction): * runtime/JSGlobalData.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): (JSC::JSGlobalObject::visitChildren): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::boundFunctionStructure): * runtime/Lookup.cpp: (JSC::setUpStaticFunctionSlot): 2011-09-22 Oliver Hunt <oliver@apple.com> Implement get_scoped_var in the DFG https://bugs.webkit.org/show_bug.cgi?id=68640 Reviewed by Gavin Barraclough. Naive implementation of get_scoped_var in the DFG. Essentially this is the bare minimum required to get correct behaviour, so there's no load/store coalescing or type profiling involved, even though these would be wins. No impact on SunSpider or V8. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGNode.h: (JSC::DFG::Node::hasVarNumber): (JSC::DFG::Node::hasScopeChainDepth): (JSC::DFG::Node::scopeChainDepth): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-22 Adam Roben <aroben@apple.com> Remove FindSafari from all our .sln files It isn't used anymore, so there's no point in building it. Part of <http://webkit.org/b/68628> Remove FindSafari Reviewed by Steve Falkenburg. * JavaScriptCore.vcproj/JavaScriptCore.sln: 2011-09-22 Filip Pizlo <fpizlo@apple.com> 32-bit call code clobbers the function cell tag https://bugs.webkit.org/show_bug.cgi?id=68606 Reviewed by Csaba Osztrogonác. This is a minimalistic fix: it simply emits code to restore the cell tag on the slow path, if we know that we failed due to emitCallIfNotType. * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCallVarargsSlowCase): (JSC::JIT::compileOpCallSlowCase): 2011-09-21 Gavin Barraclough <barraclough@apple.com> Add missing addPtr->add32 mapping for X86. Rubber stamped by Sam Weinig. * assembler/MacroAssembler.h: (JSC::MacroAssembler::addPtr): 2011-09-21 Gavin Barraclough <barraclough@apple.com> Add missing addDouble for AbsoluteAddress to X86 Rubber stamped by Geoff Garen. * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::addDouble): * assembler/X86Assembler.h: (JSC::X86Assembler::addsd_mr): (JSC::X86Assembler::cvtsi2sd_rr): (JSC::X86Assembler::cvtsi2sd_mr): 2011-09-21 Gavin Barraclough <barraclough@apple.com> Build fix following fix for bug #68586. * jit/JIT.cpp: * jit/JITInlineMethods.h: 2011-09-21 Filip Pizlo <fpizlo@apple.com> DFG JIT should be able to compile op_throw https://bugs.webkit.org/show_bug.cgi?id=68571 Reviewed by Geoffrey Garen. This compiles op_throw in the simplest way possible: it's an OSR point back to the old JIT. This is a good step towards increasing coverage, particularly on Kraken, but it's neutral because the same functions that do throw also use some other unsupported opcodes. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-21 Filip Pizlo <fpizlo@apple.com> DFG should support continuous optimization https://bugs.webkit.org/show_bug.cgi?id=68329 Reviewed by Geoffrey Garen. This adds the ability to reoptimize a code block if speculation failures happen frequently. 6% speed-up on Kraken, 1% slow-down on V8, neutral on SunSpider. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::CodeBlock): (JSC::ProgramCodeBlock::jettison): (JSC::EvalCodeBlock::jettison): (JSC::FunctionCodeBlock::jettison): (JSC::CodeBlock::shouldOptimizeNow): (JSC::CodeBlock::dumpValueProfiles): * bytecode/CodeBlock.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getStrongPrediction): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): (JSC::DFG::JITCompiler::compileEntry): (JSC::DFG::JITCompiler::compileBody): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::noticeOSREntry): * dfg/DFGOSREntry.cpp: (JSC::DFG::prepareOSREntry): * dfg/DFGOSREntry.h: (JSC::DFG::getOSREntryDataBytecodeIndex): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * heap/ConservativeRoots.cpp: (JSC::ConservativeRoots::ConservativeRoots): (JSC::ConservativeRoots::~ConservativeRoots): (JSC::DummyMarkHook::mark): (JSC::ConservativeRoots::genericAddPointer): (JSC::ConservativeRoots::genericAddSpan): (JSC::ConservativeRoots::add): * heap/ConservativeRoots.h: * heap/Heap.cpp: (JSC::Heap::addJettisonCodeBlock): (JSC::Heap::markRoots): * heap/Heap.h: * heap/JettisonedCodeBlocks.cpp: Added. (JSC::JettisonedCodeBlocks::JettisonedCodeBlocks): (JSC::JettisonedCodeBlocks::~JettisonedCodeBlocks): (JSC::JettisonedCodeBlocks::addCodeBlock): (JSC::JettisonedCodeBlocks::clearMarks): (JSC::JettisonedCodeBlocks::deleteUnmarkedCodeBlocks): (JSC::JettisonedCodeBlocks::traceCodeBlocks): * heap/JettisonedCodeBlocks.h: Added. (JSC::JettisonedCodeBlocks::mark): * interpreter/RegisterFile.cpp: (JSC::RegisterFile::gatherConservativeRoots): * interpreter/RegisterFile.h: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Executable.cpp: (JSC::jettisonCodeBlock): (JSC::EvalExecutable::jettisonOptimizedCode): (JSC::ProgramExecutable::jettisonOptimizedCode): (JSC::FunctionExecutable::jettisonOptimizedCodeForCall): (JSC::FunctionExecutable::jettisonOptimizedCodeForConstruct): * runtime/Executable.h: (JSC::FunctionExecutable::jettisonOptimizedCodeFor): * wtf/BitVector.h: Added. (WTF::BitVector::BitVector): (WTF::BitVector::~BitVector): (WTF::BitVector::operator=): (WTF::BitVector::size): (WTF::BitVector::ensureSize): (WTF::BitVector::resize): (WTF::BitVector::clearAll): (WTF::BitVector::get): (WTF::BitVector::set): (WTF::BitVector::clear): (WTF::BitVector::bitsInPointer): (WTF::BitVector::maxInlineBits): (WTF::BitVector::byteCount): (WTF::BitVector::makeInlineBits): (WTF::BitVector::OutOfLineBits::numBits): (WTF::BitVector::OutOfLineBits::numWords): (WTF::BitVector::OutOfLineBits::bits): (WTF::BitVector::OutOfLineBits::create): (WTF::BitVector::OutOfLineBits::destroy): (WTF::BitVector::OutOfLineBits::OutOfLineBits): (WTF::BitVector::isInline): (WTF::BitVector::outOfLineBits): (WTF::BitVector::resizeOutOfLine): (WTF::BitVector::bits): 2011-09-21 Gavin Barraclough <barraclough@apple.com> Add X86 GPRInfo for DFG JIT. https://bugs.webkit.org/show_bug.cgi?id=68586 Reviewed by Geoff Garen. * dfg/DFGGPRInfo.h: (JSC::DFG::GPRInfo::toRegister): (JSC::DFG::GPRInfo::toIndex): (JSC::DFG::GPRInfo::debugName): 2011-09-21 Gavin Barraclough <barraclough@apple.com> Should support value profiling on CPU(X86) https://bugs.webkit.org/show_bug.cgi?id=68575 Reviewed by Sam Weinig. Fix verbose profiling in ToT (SlowCaseProfile had been partially renamed to RareCaseProfile), add in-memory bucket counter for CPU(X86), move JIT::m_canBeOptimized out of the DFG_JIT ifdef. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::resetRareCaseProfiles): (JSC::CodeBlock::dumpValueProfiles): * bytecode/CodeBlock.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::makeSafe): * jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases): (JSC::JIT::privateCompile): * jit/JIT.h: * jit/JITInlineMethods.h: (JSC::JIT::emitValueProfilingSite): 2011-09-21 Filip Pizlo <fpizlo@apple.com> DFG does not support compiling functions as constructors https://bugs.webkit.org/show_bug.cgi?id=68500 Reviewed by Oliver Hunt. This adds support for compiling constructors to the DFG. It's a 1% speed-up on V8, mostly due to a 6% speed-up on early-boyer. It's also a 13% win on access-binary-trees, but it's neutral in the SunSpider and Kraken averages. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCapabilities.h: (JSC::DFG::mightCompileFunctionForConstruct): (JSC::DFG::canCompileOpcode): * dfg/DFGNode.h: * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/Executable.cpp: (JSC::FunctionExecutable::compileOptimizedForConstruct): (JSC::FunctionExecutable::compileForConstructInternal): * runtime/Executable.h: (JSC::FunctionExecutable::compileForConstruct): (JSC::FunctionExecutable::compileFor): (JSC::FunctionExecutable::compileOptimizedFor): 2011-09-21 Gavin Barraclough <barraclough@apple.com> Replace jsFunctionVPtr compares with a type check on the Structure. https://bugs.webkit.org/show_bug.cgi?id=68557 Reviewed by Oliver Hunt. This will permit calls to still optimize to subclasses of JSFunction that have the correct type (but a different C++ vptr). This patch stops passing the globalData into numerous functions. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.h: (JSC::DFG::Graph::isFunctionConstant): (JSC::DFG::Graph::valueOfFunctionConstant): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::isFunctionConstant): (JSC::DFG::JITCompiler::valueOfFunctionConstant): * dfg/DFGOperations.cpp: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JIT.h: * jit/JITCall.cpp: (JSC::JIT::compileOpCallVarargs): (JSC::JIT::compileOpCallSlowCase): * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCallVarargs): (JSC::JIT::compileOpCallSlowCase): * jit/JITInlineMethods.h: (JSC::JIT::emitJumpIfNotType): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Executable.h: (JSC::isHostFunction): * runtime/JSFunction.h: (JSC::JSFunction::createStructure): * runtime/JSObject.cpp: (JSC::JSObject::put): (JSC::JSObject::putWithAttributes): * runtime/JSObject.h: (JSC::getJSFunction): (JSC::JSObject::putDirect): (JSC::JSObject::putDirectWithoutTransition): * runtime/JSType.h: 2011-09-21 Geoffrey Garen <ggaren@apple.com> Removed WTFTHREADDATA_MULTITHREADED, making it always true https://bugs.webkit.org/show_bug.cgi?id=68549 Reviewed by Darin Adler. Another part of making threads exist in WebKit. * wtf/WTFThreadData.cpp: * wtf/WTFThreadData.h: (WTF::wtfThreadData): 2011-09-21 Dan Bernstein <mitz@apple.com> JavaScriptCore Part of: Prevent the WebKit frameworks from defining inappropriately-named Objective-C classes https://bugs.webkit.org/show_bug.cgi?id=68451 Reviewed by Darin Adler. * JavaScriptCore.xcodeproj/project.pbxproj: Added a script build phase that invokes check-for-inappropriate-objc-class-names, allowing only class names prefixed with "JS". 2011-09-20 Gavin Barraclough <barraclough@apple.com> MacroAssembler fixes. https://bugs.webkit.org/show_bug.cgi?id=68494 Reviewed by Sam Weinig. Add X86-64's 3 operand or32 to other MacroAssembler, fix load32's [const] void* mismatch * assembler/MacroAssembler.h: (JSC::MacroAssembler::orPtr): (JSC::MacroAssembler::loadPtr): * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::or32): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::or32): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::or32): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::or32): (JSC::MacroAssemblerSH4::load32): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::load32): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::load32): 2011-09-20 Geoffrey Garen <ggaren@apple.com> Some Heap cleanup. Reviewed by Beth Dakin. * heap/MarkedBlock.cpp: (JSC::MarkedBlock::blessNewBlock): Removed blessNewBlockForSlowPath() because it was unused; renamed blessNewBlockForFastPath() to blessNewBlock() since there is only one now. * heap/MarkedBlock.h: Removed ownerSet-related stuff since it was unused. Updated mark bit overhead calculation. Deployed atomsPerBlock in one place where we were recalculating it. * heap/MarkedSpace.cpp: (JSC::MarkedSpace::addBlock): Updated for rename. 2011-09-20 Filip Pizlo <fpizlo@apple.com> DFG JIT always speculates integer on modulo https://bugs.webkit.org/show_bug.cgi?id=68485 Reviewed by Oliver Hunt. Added support for double modulo, which is a call to fmod(). Also added support for recording the old JIT's statistics on op_mod and propagating them along the graph. Finally, fixed a goof in the ArithNodeFlags propagation logic that was made obvious when I started testing ArithMod. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::makeSafe): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGNode.h: (JSC::DFG::Node::hasArithNodeFlags): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateArithNodeFlags): (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::fixupNode): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-20 ChangSeok Oh <shivamidow@gmail.com> [GTK] requestAnimationFrame support for gtk port https://bugs.webkit.org/show_bug.cgi?id=66280 Reviewed by Martin Robinson. Let GTK port use REQUEST_ANIMATION_FRAME_TIMER. * wtf/Platform.h: 2011-09-20 Filip Pizlo <fpizlo@apple.com> DFG JIT performs too many negative zero checks, and too many overflow checks https://bugs.webkit.org/show_bug.cgi?id=68430 Reviewed by Oliver Hunt. This adds comprehensive support for deciding how to perform an arithmetic operations based on a combination of overflow profiling, negative zero profiling, value profiling, and a static analysis of how the results of these operations get used. This is a 72% speed-up on stanford-crypto-sha256-iterative, and a 2.5% speed-up on the Kraken average, a 1.4% speed-up on the V8 geomean, and neutral on SunSpider. It's also an 8.5% speed-up on V8-crypto, because apparenty everything we do speeds up crypto. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::toInt32): (JSC::DFG::ByteCodeParser::toNumber): (JSC::DFG::ByteCodeParser::isSmallInt32Constant): (JSC::DFG::ByteCodeParser::valueOfInt32Constant): (JSC::DFG::ByteCodeParser::weaklyPredictInt32): (JSC::DFG::ByteCodeParser::makeSafe): (JSC::DFG::ByteCodeParser::handleMinMax): (JSC::DFG::ByteCodeParser::handleIntrinsic): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::processPhiStack): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeBasicArithOp): * dfg/DFGNode.h: (JSC::DFG::nodeUsedAsNumber): (JSC::DFG::nodeCanTruncateInteger): (JSC::DFG::nodeCanIgnoreNegativeZero): (JSC::DFG::nodeCanSpeculateInteger): (JSC::DFG::arithNodeFlagsAsString): (JSC::DFG::Node::Node): (JSC::DFG::Node::hasArithNodeFlags): (JSC::DFG::Node::rawArithNodeFlags): (JSC::DFG::Node::arithNodeFlags): (JSC::DFG::Node::arithNodeFlagsForCompare): (JSC::DFG::Node::setArithNodeFlag): (JSC::DFG::Node::mergeArithNodeFlags): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::fixpoint): (JSC::DFG::Propagator::isNotNegZero): (JSC::DFG::Propagator::isNotZero): (JSC::DFG::Propagator::propagateArithNodeFlags): (JSC::DFG::Propagator::propagateArithNodeFlagsForward): (JSC::DFG::Propagator::propagateArithNodeFlagsBackward): (JSC::DFG::Propagator::propagateNodePredictions): (JSC::DFG::Propagator::propagatePredictionsForward): (JSC::DFG::Propagator::propagatePredictionsBackward): (JSC::DFG::Propagator::toDouble): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::fixup): (JSC::DFG::Propagator::startIndexForChildren): (JSC::DFG::Propagator::endIndexForPureCSE): (JSC::DFG::Propagator::pureCSE): (JSC::DFG::Propagator::clobbersWorld): (JSC::DFG::Propagator::setReplacement): (JSC::DFG::Propagator::performNodeCSE): (JSC::DFG::Propagator::localCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): 2011-09-19 Oliver Hunt <oliver@apple.com> Refactor Heap allocation logic into separate AllocationSpace class https://bugs.webkit.org/show_bug.cgi?id=68409 Reviewed by Gavin Barraclough. This patch hoists direct manipulation of the MarkedSpace and related data out of Heap and into a separate class. This will allow us to have multiple allocation spaces in future, so easing the way towards having GC'd backing stores for objects. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * debugger/Debugger.cpp: (JSC::Debugger::recompileAllJSFunctions): * heap/AllocationSpace.cpp: Added. (JSC::AllocationSpace::tryAllocate): (JSC::AllocationSpace::allocateSlowCase): (JSC::AllocationSpace::allocateBlock): (JSC::AllocationSpace::freeBlocks): (JSC::TakeIfEmpty::TakeIfEmpty): (JSC::TakeIfEmpty::operator()): (JSC::TakeIfEmpty::returnValue): (JSC::AllocationSpace::shrink): * heap/AllocationSpace.h: Added. (JSC::AllocationSpace::AllocationSpace): (JSC::AllocationSpace::blocks): (JSC::AllocationSpace::sizeClassFor): (JSC::AllocationSpace::setHighWaterMark): (JSC::AllocationSpace::highWaterMark): (JSC::AllocationSpace::canonicalizeBlocks): (JSC::AllocationSpace::resetAllocator): (JSC::AllocationSpace::forEachCell): (JSC::AllocationSpace::forEachBlock): (JSC::AllocationSpace::allocate): * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::reportExtraMemoryCostSlowCase): (JSC::Heap::getConservativeRegisterRoots): (JSC::Heap::markRoots): (JSC::Heap::clearMarks): (JSC::Heap::sweep): (JSC::Heap::objectCount): (JSC::Heap::size): (JSC::Heap::capacity): (JSC::Heap::globalObjectCount): (JSC::Heap::objectTypeCounts): (JSC::Heap::collect): (JSC::Heap::canonicalizeBlocks): (JSC::Heap::resetAllocator): (JSC::Heap::freeBlocks): (JSC::Heap::shrink): * heap/Heap.h: (JSC::Heap::objectSpace): (JSC::Heap::sizeClassForObject): (JSC::Heap::allocate): * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateBasicJSObject): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::recompileAllJSFunctions): (JSC::JSGlobalData::releaseExecutableMemory): 2011-09-19 Geoffrey Garen <ggaren@apple.com> Removed BREWMP* platform #ifdefs https://bugs.webkit.org/show_bug.cgi?id=68425 BREWMP* has no maintainer, and this is dead code. Reviewed by Darin Adler. * heap/MarkStack.h: (JSC::::shrinkAllocation): * jit/ExecutableAllocator.h: (JSC::ExecutableAllocator::cacheFlush): * runtime/TimeoutChecker.cpp: (JSC::getCPUTime): * wtf/Assertions.cpp: * wtf/Assertions.h: * wtf/CurrentTime.cpp: * wtf/DateMath.cpp: (WTF::calculateUTCOffset): * wtf/FastMalloc.cpp: (WTF::fastMalloc): (WTF::fastCalloc): (WTF::fastMallocSize): * wtf/FastMalloc.h: * wtf/MainThread.cpp: * wtf/MathExtras.h: * wtf/OwnPtrCommon.h: * wtf/Platform.h: * wtf/RandomNumber.cpp: (WTF::randomNumber): * wtf/RandomNumberSeed.h: (WTF::initializeRandomNumberGenerator): * wtf/text/WTFString.h: * wtf/unicode/Unicode.h: 2011-09-20 Adam Roben <aroben@apple.com> Windows build fix after r95523 * wtf/CheckedArithmetic.h: Added stdint.h so we can have int64_t defined. 2011-09-18 Filip Pizlo <fpizlo@apple.com> DFG JIT does not speculate aggressively enough on GetById https://bugs.webkit.org/show_bug.cgi?id=68320 Reviewed by Oliver Hunt. This adds the ability to access properties directly, by offset. This optimization kicks in when at the time of DFG compilation, it appears that the given get_by_id is self-cached by the old JIT. Two new opcodes get introduced: CheckStructure and GetByOffset. CheckStructure performs a speculation check on the object's structure, and returns the storage pointer. GetByOffset performs a direct read of the field from the storage pointer. Both CheckStructure and GetByOffset can be CSE'd, so that we can eliminate redundant structure checks, and redundant reads of the same field. This is a 4% speed-up on V8, a 2% slow-down on Kraken, and neutral on SunSpider. * bytecode/PredictedType.cpp: (JSC::predictionFromClassInfo): (JSC::predictionFromStructure): (JSC::predictionFromCell): * bytecode/PredictedType.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGenerationInfo.h: (JSC::DFG::dataFormatToString): (JSC::DFG::needDataFormatConversion): (JSC::DFG::GenerationInfo::initStorage): (JSC::DFG::GenerationInfo::spill): (JSC::DFG::GenerationInfo::fillStorage): * dfg/DFGGraph.h: (JSC::DFG::Graph::predict): (JSC::DFG::Graph::getPrediction): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillInteger): (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::fillStorage): (JSC::DFG::GPRTemporary::GPRTemporary): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentSpillGPR): (JSC::DFG::JITCodeGenerator::silentFillGPR): (JSC::DFG::JITCodeGenerator::spill): (JSC::DFG::JITCodeGenerator::storageResult): (JSC::DFG::StorageOperand::StorageOperand): (JSC::DFG::StorageOperand::~StorageOperand): (JSC::DFG::StorageOperand::index): (JSC::DFG::StorageOperand::gpr): (JSC::DFG::StorageOperand::use): * dfg/DFGNode.h: (JSC::DFG::OpInfo::OpInfo): (JSC::DFG::Node::Node): (JSC::DFG::Node::hasPrediction): (JSC::DFG::Node::hasStructure): (JSC::DFG::Node::structure): (JSC::DFG::Node::hasStorageAccessData): (JSC::DFG::Node::storageAccessDataIndex): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNode): (JSC::DFG::Propagator::globalVarLoadElimination): (JSC::DFG::Propagator::getMethodLoadElimination): (JSC::DFG::Propagator::checkStructureLoadElimination): (JSC::DFG::Propagator::getByOffsetLoadElimination): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compile): * wtf/StdLibExtras.h: (WTF::safeCast): 2011-09-19 Mark Hahnenberg <mhahnenberg@apple.com> Remove toPrimitive from JSCell https://bugs.webkit.org/show_bug.cgi?id=67875 Reviewed by Darin Adler. Part of the refactoring process to un-virtualize JSCell. We move all of the implicit functionality provided by the virtual toPrimitive method in JSCell to be explicit in JSValue::toPrimitive and JSCell:toPrimitive while also de-virtualizing JSCell::toPrimitive. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSCell.cpp: (JSC::JSCell::toPrimitive): * runtime/JSCell.h: We replace JSNotAnObject::toPrimitive with defaultValue, which it overrides from JSObject. This pushes the virtual method further down, enabling us to get rid of the virtual call in JSCell. Eventually we'll probably have to deal with this again, but we'll cross that bridge when we come to it. * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::defaultValue): * runtime/JSNotAnObject.h: * runtime/JSObject.h: * runtime/JSString.h: 2011-09-19 Geoffrey Garen <ggaren@apple.com> Removed ENABLE_LAZY_BLOCK_FREEING and related #ifdefs https://bugs.webkit.org/show_bug.cgi?id=68424 As discussed on webkit-dev. All ports build with threads enabled in JSC now. This may break WinCE and other ports that have not built and tested with this configuration. I've filed bugs for port maintainers. It's time for WebKit to move forward. Reviewed by Mark Rowe. * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::~Heap): (JSC::Heap::destroy): (JSC::Heap::blockFreeingThreadMain): (JSC::Heap::allocateBlock): (JSC::Heap::freeBlocks): (JSC::Heap::releaseFreeBlocks): * heap/Heap.h: * wtf/Platform.h: 2011-09-19 Geoffrey Garen <ggaren@apple.com> Removed ENABLE_WTF_MULTIPLE_THREADS and related #ifdefs https://bugs.webkit.org/show_bug.cgi?id=68423 As discussed on webkit-dev. All ports build with threads enabled in WTF now. This may break WinCE and other ports that have not built and tested with this configuration. I've filed bugs for port maintainers. It's time for WebKit to move forward. Reviewed by Mark Rowe. * wtf/CryptographicallyRandomNumber.cpp: (WTF::ARC4Stream::ARC4RandomNumberGenerator::randomNumber): (WTF::ARC4Stream::ARC4RandomNumberGenerator::randomValues): * wtf/FastMalloc.cpp: * wtf/Platform.h: * wtf/RandomNumber.cpp: (WTF::randomNumber): * wtf/RefCountedLeakCounter.cpp: (WTF::RefCountedLeakCounter::increment): (WTF::RefCountedLeakCounter::decrement): * wtf/ThreadingPthreads.cpp: (WTF::initializeThreading): * wtf/ThreadingWin.cpp: (WTF::initializeThreading): * wtf/dtoa.cpp: (WTF::pow5mult): * wtf/gtk/ThreadingGtk.cpp: (WTF::initializeThreading): * wtf/qt/ThreadingQt.cpp: (WTF::initializeThreading): 2011-09-19 Geoffrey Garen <ggaren@apple.com> Removed ENABLE_JSC_MULTIPLE_THREADS and related #ifdefs. https://bugs.webkit.org/show_bug.cgi?id=68422 As discussed on webkit-dev. All ports build with threads enabled in JSC now. This may break WinCE and other ports that have not built and tested with this configuration. I've filed bugs for port maintainers. It's time for WebKit to move forward. Reviewed by Sam Weinig. * API/APIShims.h: (JSC::APIEntryShimWithoutLock::APIEntryShimWithoutLock): * API/JSContextRef.cpp: * heap/MachineStackMarker.cpp: (JSC::MachineThreads::MachineThreads): (JSC::MachineThreads::~MachineThreads): (JSC::MachineThreads::gatherConservativeRoots): * heap/MachineStackMarker.h: * runtime/InitializeThreading.cpp: (JSC::initializeThreadingOnce): (JSC::initializeThreading): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::sharedInstance): * runtime/JSGlobalData.h: (JSC::JSGlobalData::makeUsableFromMultipleThreads): * runtime/JSLock.cpp: * runtime/Structure.cpp: * wtf/Platform.h: 2011-09-19 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r95493 and r95496. http://trac.webkit.org/changeset/95493 http://trac.webkit.org/changeset/95496 https://bugs.webkit.org/show_bug.cgi?id=68418 Broke Windows build (Requested by rniwa on #webkit). * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * debugger/Debugger.cpp: (JSC::Debugger::recompileAllJSFunctions): * heap/AllocationSpace.cpp: Removed. * heap/AllocationSpace.h: Removed. * heap/Heap.cpp: (JSC::CountFunctor::TakeIfEmpty::TakeIfEmpty): (JSC::CountFunctor::TakeIfEmpty::operator()): (JSC::CountFunctor::TakeIfEmpty::returnValue): (JSC::Heap::Heap): (JSC::Heap::reportExtraMemoryCostSlowCase): (JSC::Heap::tryAllocate): (JSC::Heap::allocateSlowCase): (JSC::Heap::getConservativeRegisterRoots): (JSC::Heap::markRoots): (JSC::Heap::clearMarks): (JSC::Heap::sweep): (JSC::Heap::objectCount): (JSC::Heap::size): (JSC::Heap::capacity): (JSC::Heap::globalObjectCount): (JSC::Heap::objectTypeCounts): (JSC::Heap::collect): (JSC::Heap::canonicalizeBlocks): (JSC::Heap::resetAllocator): (JSC::Heap::allocateBlock): (JSC::Heap::freeBlocks): (JSC::Heap::shrink): * heap/Heap.h: (JSC::Heap::markedSpace): (JSC::Heap::forEachCell): (JSC::Heap::forEachBlock): (JSC::Heap::sizeClassFor): (JSC::Heap::allocate): * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateBasicJSObject): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::recompileAllJSFunctions): (JSC::JSGlobalData::releaseExecutableMemory): 2011-09-19 Gavin Barraclough <barraclough@apple.com> Errrk, missed stylebot comments in last commit. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncSplit): 2011-09-19 Gavin Barraclough <barraclough@apple.com> String#split is buggy https://bugs.webkit.org/show_bug.cgi?id=68348 Reviewed by Sam Weinig. * runtime/StringPrototype.cpp: (JSC::jsStringWithReuse): - added helper function to reuse original JSString value. (JSC::stringProtoFuncSplit): - Rewritten from the spec. * tests/mozilla/ecma/String/15.5.4.8-2.js: (getTestCases): - This test is not ES5 compliant. 2011-09-19 Geoffrey Garen <ggaren@apple.com> Removed lots of friend declarations from JSCell, so we can more effectively make use of private and protected. Reviewed by Sam Weinig. * runtime/JSCell.h: Removed MSVCBugWorkaround because it was a lot of confusion for not much safety. (JSC::JSCell::operator new): Made this public because it is used by a few clients, and not really dangerous. * runtime/JSObject.cpp: (JSC::JSObject::put): (JSC::JSObject::deleteProperty): (JSC::JSObject::defineGetter): (JSC::JSObject::defineSetter): (JSC::JSObject::getPropertySpecificValue): (JSC::JSObject::getOwnPropertyNames): (JSC::JSObject::seal): (JSC::JSObject::freeze): (JSC::JSObject::preventExtensions): (JSC::JSObject::removeDirect): (JSC::JSObject::createInheritorID): (JSC::JSObject::allocatePropertyStorage): (JSC::JSObject::getOwnPropertyDescriptor): * runtime/JSObject.h: (JSC::JSObject::getDirect): (JSC::JSObject::getDirectLocation): (JSC::JSObject::hasCustomProperties): (JSC::JSObject::hasGetterSetterProperties): (JSC::JSObject::isSealed): (JSC::JSObject::isFrozen): (JSC::JSObject::isExtensible): (JSC::JSObject::flattenDictionaryObject): (JSC::JSObject::finishCreation): (JSC::JSObject::prototype): (JSC::JSObject::setPrototype): (JSC::JSObject::inlineGetOwnPropertySlot): (JSC::JSCell::fastGetOwnProperty): (JSC::JSObject::putDirectInternal): (JSC::JSObject::putDirectWithoutTransition): (JSC::JSObject::transitionTo): (JSC::JSObject::visitChildrenDirect): Changed all use of m_structure to structure() / setStructure(), so we don't have to be a friend of JSCell. * runtime/Structure.h: (JSC::JSCell::setStructure): Added, to avoid direct access by JSObject to JSCell::m_structure. 2011-09-19 Adam Barth <abarth@webkit.org> Always enable ENABLE(EVENTSOURCE) https://bugs.webkit.org/show_bug.cgi?id=68414 Reviewed by Eric Seidel. * Configurations/FeatureDefines.xcconfig: 2011-09-19 Eli Fidler <efidler@rim.com> Enable JSC_MULTIPLE_THREADS for OS(QNX). https://bugs.webkit.org/show_bug.cgi?id=68047 Reviewed by Daniel Bates. SA_RESTART was required for SIGUSR2-based debugging, but is not present on QNX. This debugging doesn't seem critical to JSC_MULTIPLE_THREADS, so allow it to proceed. * heap/MachineStackMarker.cpp: (JSC::MachineThreads::Thread::Thread): (JSC::getPlatformThreadRegisters): (JSC::otherThreadStackPointer): (JSC::freePlatformThreadRegisters): * wtf/Platform.h: enable PTHREADS for OS(QNX) 2011-09-19 Oliver Hunt <oliver@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-09-19 Oliver Hunt <oliver@apple.com> Refactor Heap allocation logic into separate AllocationSpace class https://bugs.webkit.org/show_bug.cgi?id=68409 Reviewed by Gavin Barraclough. This patch hoists direct manipulation of the MarkedSpace and related data out of Heap and into a separate class. This will allow us to have multiple allocation spaces in future, so easing the way towards having GC'd backing stores for objects. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * debugger/Debugger.cpp: (JSC::Debugger::recompileAllJSFunctions): * heap/AllocationSpace.cpp: Added. (JSC::AllocationSpace::tryAllocate): (JSC::AllocationSpace::allocateSlowCase): (JSC::AllocationSpace::allocateBlock): (JSC::AllocationSpace::freeBlocks): (JSC::TakeIfEmpty::TakeIfEmpty): (JSC::TakeIfEmpty::operator()): (JSC::TakeIfEmpty::returnValue): (JSC::AllocationSpace::shrink): * heap/AllocationSpace.h: Added. (JSC::AllocationSpace::AllocationSpace): (JSC::AllocationSpace::blocks): (JSC::AllocationSpace::sizeClassFor): (JSC::AllocationSpace::setHighWaterMark): (JSC::AllocationSpace::highWaterMark): (JSC::AllocationSpace::canonicalizeBlocks): (JSC::AllocationSpace::resetAllocator): (JSC::AllocationSpace::forEachCell): (JSC::AllocationSpace::forEachBlock): (JSC::AllocationSpace::allocate): * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::reportExtraMemoryCostSlowCase): (JSC::Heap::getConservativeRegisterRoots): (JSC::Heap::markRoots): (JSC::Heap::clearMarks): (JSC::Heap::sweep): (JSC::Heap::objectCount): (JSC::Heap::size): (JSC::Heap::capacity): (JSC::Heap::globalObjectCount): (JSC::Heap::objectTypeCounts): (JSC::Heap::collect): (JSC::Heap::canonicalizeBlocks): (JSC::Heap::resetAllocator): (JSC::Heap::freeBlocks): (JSC::Heap::shrink): * heap/Heap.h: (JSC::Heap::objectSpace): (JSC::Heap::sizeClassForObject): (JSC::Heap::allocate): * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateBasicJSObject): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::recompileAllJSFunctions): (JSC::JSGlobalData::releaseExecutableMemory): 2011-09-19 Adam Roben <aroben@apple.com> Windows build fix after r95310 * JavaScriptCore.vcproj/testRegExp/testRegExpCommon.vsprops: Added include\private\JavaScriptCore to the include path so DFGIntrinsic.h can be found. 2011-09-19 Filip Pizlo <fpizlo@apple.com> DFG speculation failures should act as additional value profiles https://bugs.webkit.org/show_bug.cgi?id=68335 Reviewed by Oliver Hunt. This adds slow-case counters to the old JIT. It also ensures that negative zero in multiply is handled carefully. The old JIT previously took slow path if the result of a multiply was zero, which, without any changes, would cause the DFG to think that every such multiply produced a double result. This also fixes a bug in the old JIT's handling of decrements. It would take the slow path if the result was zero, but not if it underflowed. By itself, this would be a 1% slow-down on V8 and Kraken. But then I wrote optimizations in the DFG that take advantage of this new information. It's no longer the case that every multiply needs to do a check for negative zero; it only happens if the negative zero is ignored. This results in a 12% speed-up on v8-crypto, for a 1.4% geomean speed-up in V8. It's mostly neutral on Kraken. I can see an 0.5% slow-down and it appears to be significant. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::resetRareCaseProfiles): (JSC::CodeBlock::dumpValueProfiles): * bytecode/CodeBlock.h: * bytecode/ValueProfile.h: (JSC::RareCaseProfile::RareCaseProfile): (JSC::getRareCaseProfileBytecodeOffset): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::toInt32): (JSC::DFG::ByteCodeParser::makeSafe): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::GPRTemporary::GPRTemporary): * dfg/DFGJITCodeGenerator.h: * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNode): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::clobbersWorld): (JSC::DFG::Propagator::performNodeCSE): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases): * jit/JIT.h: (JSC::JIT::linkDummySlowCase): * jit/JITArithmetic.cpp: (JSC::JIT::emit_op_post_dec): (JSC::JIT::emit_op_pre_dec): (JSC::JIT::compileBinaryArithOp): (JSC::JIT::emit_op_add): (JSC::JIT::emitSlow_op_add): * jit/JITInlineMethods.h: (JSC::JIT::addSlowCase): 2011-09-19 Adam Roben <aroben@apple.com> Windows build fix after r94575 * JavaScriptCore.vcproj/JavaScriptCore.sln: Relinearized project dependencies. testRegExp now builds just before FindSafari. 2011-09-19 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r95466. http://trac.webkit.org/changeset/95466 https://bugs.webkit.org/show_bug.cgi?id=68389 Incorrect version of the patch. (Requested by mhahnenberg on #webkit). * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSCell.cpp: (JSC::JSCell::toPrimitive): * runtime/JSCell.h: (JSC::JSCell::JSValue::toPrimitive): * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::toPrimitive): * runtime/JSNotAnObject.h: * runtime/JSObject.h: * runtime/JSString.h: 2011-09-19 Mark Hahnenberg <mhahnenberg@apple.com> Remove toPrimitive from JSCell https://bugs.webkit.org/show_bug.cgi?id=67875 Reviewed by Geoffrey Garen. Part of the refactoring process to un-virtualize JSCell. We move all of the implicit functionality provided by the virtual toPrimitive method in JSCell to be explicit in JSValue::toPrimitive and JSCell:toPrimitive while also de-virtualizing JSCell::toPrimitive. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSCell.cpp: (JSC::JSCell::toPrimitive): * runtime/JSCell.h: We replace JSNotAnObject::toPrimitive with defaultValue, which it overrides from JSObject. This pushes the virtual method further down, enabling us to get rid of the virtual call in JSCell. Eventually we'll probably have to deal with this again, but we'll cross that bridge when we come to it. * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::defaultValue): * runtime/JSNotAnObject.h: * runtime/JSObject.h: * runtime/JSString.h: (JSC::JSValue::toPrimitive): 2011-09-19 Oliver Hunt <oliver@apple.com> Build fix. * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::compileGetDirectOffset): 2011-09-19 Oliver Hunt <oliver@apple.com> Rename NewSpace.{h,cpp} to MarkedSpace.{h,cpp} https://bugs.webkit.org/show_bug.cgi?id=68376 Reviewed by Gavin Barraclough. Renamed the the MarkedSpace files to match new name, and updated the relevant references. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * heap/Heap.h: * heap/MarkedSpace.cpp: Renamed from Source/JavaScriptCore/heap/NewSpace.cpp. (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::addBlock): (JSC::MarkedSpace::removeBlock): (JSC::MarkedSpace::resetAllocator): (JSC::MarkedSpace::canonicalizeBlocks): * heap/MarkedSpace.h: Renamed from Source/JavaScriptCore/heap/NewSpace.h. (JSC::MarkedSpace::waterMark): (JSC::MarkedSpace::highWaterMark): (JSC::MarkedSpace::setHighWaterMark): (JSC::MarkedSpace::sizeClassFor): (JSC::MarkedSpace::allocate): (JSC::MarkedSpace::forEachBlock): (JSC::MarkedSpace::SizeClass::SizeClass): (JSC::MarkedSpace::SizeClass::resetAllocator): (JSC::MarkedSpace::SizeClass::canonicalizeBlock): * runtime/JSCell.h: 2011-09-19 Oliver Hunt <oliver@apple.com> Rename NewSpace to MarkedSpace https://bugs.webkit.org/show_bug.cgi?id=68375 Reviewed by Gavin Barraclough. Rename NewSpace to a more accurate name, and update all uses. This patch doesn't rename the files themselves as that will just make the patch appear bigger than it is. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * heap/Heap.cpp: (JSC::CountFunctor::TakeIfEmpty::TakeIfEmpty): (JSC::CountFunctor::TakeIfEmpty::operator()): (JSC::Heap::Heap): (JSC::Heap::reportExtraMemoryCostSlowCase): (JSC::Heap::tryAllocate): (JSC::Heap::allocateSlowCase): (JSC::Heap::collect): (JSC::Heap::canonicalizeBlocks): (JSC::Heap::resetAllocator): (JSC::Heap::isValidAllocation): (JSC::Heap::shrink): * heap/Heap.h: (JSC::Heap::markedSpace): (JSC::Heap::sizeClassFor): (JSC::Heap::allocate): * heap/NewSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::addBlock): (JSC::MarkedSpace::removeBlock): (JSC::MarkedSpace::resetAllocator): (JSC::MarkedSpace::canonicalizeBlocks): * heap/NewSpace.h: (JSC::MarkedSpace::waterMark): (JSC::MarkedSpace::highWaterMark): (JSC::MarkedSpace::setHighWaterMark): (JSC::MarkedSpace::sizeClassFor): (JSC::MarkedSpace::allocate): (JSC::MarkedSpace::forEachBlock): (JSC::MarkedSpace::SizeClass::SizeClass): (JSC::MarkedSpace::SizeClass::resetAllocator): (JSC::MarkedSpace::SizeClass::canonicalizeBlock): * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateBasicJSObject): 2011-09-19 Peter Rybin <peter.rybin@gmail.com> TextPosition refactoring: Merge ZeroBasedNumber and OneBasedNumber classes https://bugs.webkit.org/show_bug.cgi?id=63541 Reviewed by Adam Barth. * parser/SourceProvider.h: (JSC::SourceProvider::startPosition): * wtf/text/TextPosition.h: (WTF::OrdinalNumber::fromZeroBasedInt): (WTF::OrdinalNumber::fromOneBasedInt): (WTF::OrdinalNumber::OrdinalNumber): (WTF::OrdinalNumber::zeroBasedInt): (WTF::OrdinalNumber::oneBasedInt): (WTF::OrdinalNumber::operator==): (WTF::OrdinalNumber::operator!=): (WTF::OrdinalNumber::first): (WTF::OrdinalNumber::beforeFirst): (WTF::TextPosition::TextPosition): (WTF::TextPosition::minimumPosition): (WTF::TextPosition::belowRangePosition): 2011-09-19 Dan Bernstein <mitz@apple.com> JavaScriptCore part of [mac] WebKit contains Objective-C classes that are not prefixed with its standard prefixes https://bugs.webkit.org/show_bug.cgi?id=68323 Reviewed by Sam Weinig. Renamed WTFMainThreadCaller to JSWTFMainThreadCaller. * wtf/mac/MainThreadMac.mm: (WTF::initializeMainThreadPlatform): (WTF::initializeMainThreadToProcessMainThreadPlatform): 2011-09-19 Oliver Hunt <oliver@apple.com> Remove direct property slot pointers from the instruction stream https://bugs.webkit.org/show_bug.cgi?id=68373 Reviewed by Gavin Barraclough. Use an indirect load to access prototype properties rather than directly storing the property address in the instruction stream. This should allow further optimisations in future, and also provides a 0.5% win to sunspider. * dfg/DFGRepatch.cpp: (JSC::DFG::generateProtoChainAccessStub): * jit/JITPropertyAccess.cpp: (JSC::JIT::compileGetDirectOffset): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::compileGetDirectOffset): * runtime/JSObject.h: (JSC::JSObject::addressOfPropertyStorage): 2011-09-19 Oliver Hunt <oliver@apple.com> Remove bump allocator https://bugs.webkit.org/show_bug.cgi?id=68370 Reviewed by Sam Weinig. Can't do anything with this allocator currently, and it's increasing the complexity of the GC code. Slight progression on SunSpider, slight regression (undoing the original progression) in V8. * heap/Heap.cpp: (JSC::Heap::collect): * heap/Heap.h: * heap/NewSpace.cpp: (JSC::NewSpace::NewSpace): * heap/NewSpace.h: (JSC::NewSpace::allocate): * runtime/JSObject.cpp: (JSC::JSObject::allocatePropertyStorage): * runtime/JSObject.h: (JSC::JSObject::~JSObject): (JSC::JSObject::visitChildrenDirect): * runtime/StorageBarrier.h: (JSC::StorageBarrier::set): 2011-09-19 Carlos Garcia Campos <cgarcia@igalia.com> [GTK] Fix distcheck build https://bugs.webkit.org/show_bug.cgi?id=68346 Reviewed by Philippe Normand. * GNUmakefile.list.am: 2011-09-19 Carlos Garcia Campos <cgarcia@igalia.com> [GTK] Fix distcheck build https://bugs.webkit.org/show_bug.cgi?id=68241 Reviewed by Martin Robinson. * GNUmakefile.list.am: 2011-09-18 Dan Bernstein <mitz@apple.com> Removed ProfilerServer. Reviewed by Mark Rowe. * JavaScriptCore.gypi: * JavaScriptCore.xcodeproj/project.pbxproj: * profiler/ProfilerServer.h: Removed. * profiler/ProfilerServer.mm: Removed. * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * wscript: 2011-09-17 Filip Pizlo <fpizlo@apple.com> DFG JIT should inline Math.min, Math.max, and Math.sqrt https://bugs.webkit.org/show_bug.cgi?id=68318 Reviewed by Gavin Barraclough. Adds Math.min, Math.max, and Math.sqrt intrinsics. Adds support for a function to have an intrinsic but not a thunk generator. This is a 7% speed-up on access-nbody, and neutral elsewhere, mainly because we're still not DFG compiling the bulk of the hot code in Kraken audio benchmarks. * create_hash_table: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::handleMinMax): (JSC::DFG::ByteCodeParser::handleIntrinsic): * dfg/DFGIntrinsic.h: * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNode): (JSC::DFG::Propagator::fixupNode): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JITStubs.cpp: (JSC::JITThunks::hostFunctionStub): * runtime/Lookup.cpp: (JSC::setUpStaticFunctionSlot): 2011-09-18 Nico Weber <thakis@chromium.org> Remove two files from JavaScriptCore.gypi that were removed in r95240 https://bugs.webkit.org/show_bug.cgi?id=68327 Unreviewed, build warning fix. * JavaScriptCore.gypi: 2011-09-17 Oliver Hunt <oliver@apple.com> Remove special case handling of inline storage from the JIT https://bugs.webkit.org/show_bug.cgi?id=68319 Reviewed by Gavin Barraclough. Simplify logic used for reading and writing to property storage by removing the special cases for inline storage. This has no perf impact. * dfg/DFGRepatch.cpp: (JSC::DFG::generateProtoChainAccessStub): (JSC::DFG::tryBuildGetByIDList): * jit/JIT.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::compilePutDirectOffset): (JSC::JIT::compileGetDirectOffset): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompileGetByIdSelfList): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::compilePutDirectOffset): (JSC::JIT::compileGetDirectOffset): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompileGetByIdSelfList): 2011-09-17 Filip Pizlo <fpizlo@apple.com> DFG JIT does not have full block-local CSE https://bugs.webkit.org/show_bug.cgi?id=68316 Reviewed by Oliver Hunt. This adds block-local CSE to the DFG. CSE runs in the propagator just after type propagation. It is part of the propagator itself because it needs to use the propagator's internal data structures to determine which operations may have side effects. Because it changes the live-ranges of nodes, the virtual register allocator had to be moved into the propagator so that it runs after CSE. To ensure that the back-end knows to keep the inputs to any eliminated node alive for OSR, a new node type, Phantom, was introduced. It is a no-op but prolonges the live-range of its inputs. This is an 80% speed-up on imaging-gaussian-blur, and a 10% speed-up on Kraken. * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGAliasTracker.h: Removed. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::MethodCheckData::operator==): (JSC::DFG::MethodCheckData::operator!=): * dfg/DFGNode.h: (JSC::DFG::Node::hasVirtualRegister): (JSC::DFG::Node::setRefCount): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::Propagator): (JSC::DFG::Propagator::fixpoint): (JSC::DFG::Propagator::propagateNode): (JSC::DFG::Propagator::canonicalize): (JSC::DFG::Propagator::computeStartIndex): (JSC::DFG::Propagator::startIndex): (JSC::DFG::Propagator::pureCSE): (JSC::DFG::Propagator::globalVarLoadElimination): (JSC::DFG::Propagator::getByValLoadElimination): (JSC::DFG::Propagator::getMethodLoadElimination): (JSC::DFG::Propagator::performSubstitution): (JSC::DFG::Propagator::setReplacement): (JSC::DFG::Propagator::performNodeCSE): (JSC::DFG::Propagator::performBlockCSE): (JSC::DFG::Propagator::localCSE): (JSC::DFG::Propagator::allocateVirtualRegisters): (JSC::DFG::propagate): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-16 Filip Pizlo <fpizlo@apple.com> method_check should repatch itself if it finds that the new structure(s) are the result of transitions from the old structure(s) https://bugs.webkit.org/show_bug.cgi?id=68294 Reviewed by Gavin Barraclough. Previously a patched method_check would slow-path to get_by_id. Now it slow-paths to method_check_update, which attempts to correct the method_check due to structure transitions before bailing to get_by_id. This is a 1-2% speed-up on some benchmarks and is not a slow-down anywhere, leading to a 0.6% speed-up on the Kraken geomean. * jit/JITPropertyAccess.cpp: (JSC::JIT::patchMethodCallProto): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jit/JITStubs.h: * runtime/Structure.h: (JSC::Structure::transitivelyTransitionedFrom): 2011-09-16 Ryosuke Niwa <rniwa@webkit.org> Touch Platform.h in the hope to fix SnowLeopard Intel Release (WebKit2 Tests). * wtf/Platform.h: 2011-09-16 Sam Weinig <sam@webkit.org> Rename APIValueWrapper type to APIValueWrapperType for consistency https://bugs.webkit.org/show_bug.cgi?id=68306 Reviewed by Anders Carlsson. * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): Update name. * runtime/JSType.h: Update name and un-indent. * runtime/Structure.h: (JSC::JSCell::isAPIValueWrapper): Update name. 2011-09-16 Sam Weinig <sam@webkit.org> Remove unused isStrictModeFunction function https://bugs.webkit.org/show_bug.cgi?id=68305 Reviewed by Anders Carlsson. * runtime/JSObject.h: (JSC::JSObject::isStrictModeFunction): 2011-09-16 Sam Weinig <sam@webkit.org> Cleanup JSTypeInfo a bit https://bugs.webkit.org/show_bug.cgi?id=68289 Reviewed by Anders Carlsson. * dfg/DFGOperations.cpp: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): Replace direct access to flags() with predicate. * runtime/JSObject.h: (JSC::JSFinalObject::createStructure): Pass FinalObjectType instead of using special IsJSFinalObject. * runtime/JSTypeInfo.h: (JSC::TypeInfo::TypeInfo): Add additional assert that you should no object should OverridesHasInstance but not have ImplementsHasInstance set. (JSC::TypeInfo::isFinalObject): Added. (JSC::TypeInfo::masqueradesAsUndefined): (JSC::TypeInfo::implementsHasInstance): (JSC::TypeInfo::isEnvironmentRecord): (JSC::TypeInfo::overridesHasInstance): (JSC::TypeInfo::implementsDefaultHasInstance): (JSC::TypeInfo::overridesGetOwnPropertySlot): (JSC::TypeInfo::overridesVisitChildren): (JSC::TypeInfo::overridesGetPropertyNames): (JSC::TypeInfo::prohibitsPropertyCaching): (JSC::TypeInfo::isSetOnFlags1): (JSC::TypeInfo::isSetOnFlags2): Replace direct bit twiddling with helper functions. * runtime/Structure.cpp: (JSC::Structure::Structure): Use new isFinalObject() predicate. 2011-09-16 Gavin Barraclough <barraclough@apple.com> Unsigned bit shift fails under certain conditions in 32 bit builds https://bugs.webkit.org/show_bug.cgi?id=68166 Reviewed by Geoff Garen. The major bug here is that the slow case (which handles shifts of doubles) doesn't check for negative results from an unsigned shift (which should be unsigned, and as such can't be represented by a signed integer immediate). The implementation is also flawed for shifts by negative shift amounts (treats as shift by zero). * jit/JITArithmetic32_64.cpp: (JSC::JIT::emitRightShift): (JSC::JIT::emitRightShiftSlowCase): 2011-09-16 Geoffrey Garen <ggaren@apple.com> Removed undetectable style.filter. Reviewed by Sam Weinig. This feature was added in http://trac.webkit.org/changeset/15557 to support housingmaps.com. But housingmaps.com no longer needs this hack, we don't know of other websites that need it, and we don't know of any other browsers that have implemented this feature. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * runtime/JSTypeInfo.h: * runtime/StringObjectThatMasqueradesAsUndefined.h: Removed. 2011-09-15 Sam Weinig <sam@webkit.org> Prepare JSTypes for more Object subtypes https://bugs.webkit.org/show_bug.cgi?id=68200 Reviewed by Gavin Barraclough. * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::branchIfNotObject): * jit/JITInlineMethods.h: (JSC::JIT::emitJumpIfNotObject): * runtime/JSGlobalObject.h: (JSC::Structure::prototypeForLookup): * runtime/JSObject.h: (JSC::JSObject::finishCreation): * runtime/JSType.h: * runtime/JSTypeInfo.h: (JSC::TypeInfo::type): (JSC::TypeInfo::isObject): (JSC::TypeInfo::isFinal): (JSC::TypeInfo::prohibitsPropertyCaching): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::finishCreation): * runtime/Operations.cpp: (JSC::jsIsObjectType): * runtime/Structure.cpp: (JSC::Structure::addPropertyTransitionToExistingStructure): (JSC::Structure::addPropertyTransition): * runtime/Structure.h: (JSC::Structure::isObject): (JSC::JSCell::isObject): 2011-09-16 Geoffrey Garen <ggaren@apple.com> Rolled back in r95201 with test failure fixed. I missed two cases of jumpSlowToHot in rshift -- these cases need to be sure to initialize regT1 to the int tag, since it will otherwise hold the top 32 bits of a double. * jit/JIT.h: * jit/JITArithmetic32_64.cpp: (JSC::JIT::emit_op_lshift): (JSC::JIT::emitRightShift): (JSC::JIT::emitRightShiftSlowCase): (JSC::JIT::emit_op_bitand): (JSC::JIT::emit_op_bitor): (JSC::JIT::emit_op_bitxor): (JSC::JIT::emit_op_bitnot): (JSC::JIT::emit_op_post_inc): (JSC::JIT::emit_op_post_dec): (JSC::JIT::emit_op_pre_inc): (JSC::JIT::emit_op_pre_dec): * jit/JITInlineMethods.h: (JSC::JIT::emitStoreAndMapInt32): 2011-09-16 Filip Pizlo <fpizlo@apple.com> Unreviewed Windows build fix after 95318. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-09-16 Adam Roben <aroben@apple.com> Windows build fix after r95310 * JavaScriptCore.vcproj/jsc/jscCommon.vsprops: Added include\private\JavaScriptCore to the include path so DFGIntrinsic.h can be found. 2011-09-16 Gavin Barraclough <barraclough@apple.com> Rationalize JSObject::putDirect* methods https://bugs.webkit.org/show_bug.cgi?id=68274 Reviewed by Sam Weinig. Delete the *Function variants. These are overall inefficient, in the way they get the name back from the function rather than just passing it in. * JavaScriptCore.exp: * jsc.cpp: (GlobalObject::finishCreation): (GlobalObject::addFunction): * runtime/FunctionPrototype.cpp: (JSC::FunctionPrototype::addFunctionProperties): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): * runtime/JSObject.cpp: (JSC::JSObject::put): (JSC::JSObject::putWithAttributes): (JSC::JSObject::defineGetter): (JSC::JSObject::defineSetter): * runtime/JSObject.h: (JSC::JSObject::putDirect): (JSC::JSObject::putDirectWithoutTransition): * runtime/Lookup.cpp: (JSC::setUpStaticFunctionSlot): * runtime/Lookup.h: (JSC::lookupPut): 2011-09-16 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix for Windows. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: 2011-09-16 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix for non-DFG builds. * runtime/Executable.h: (JSC::NativeExecutable::finishCreation): 2011-09-16 Filip Pizlo <fpizlo@apple.com> DFG JIT should inline Math.abs https://bugs.webkit.org/show_bug.cgi?id=68227 Reviewed by Oliver Hunt. This adds the ability to track intrinsic functions throughout the host function infrastructure, so that the DFG can easily query whether or not a call's target is intrinsic, and if so, which intrinsic it is. On top of this, it adds Math.abs intrinsics to DFG. Call(Math.abs) is transformed into ValueToNumber<-ArithAbs nodes. These nodes then get optimized using the usual tricks. Also had to make a completely unrelated change to DateInstanceCache.h in order to fix a preexisting alphabetical sorting problem in JSGlobalData.h This results in a big win in imaging-gaussian-blur: 61% faster than before. The net win on Kraken is around 13%. * JavaScriptCore.xcodeproj/project.pbxproj: * create_hash_table: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.h: (JSC::DFG::Graph::isFunctionConstant): (JSC::DFG::Graph::valueOfFunctionConstant): * dfg/DFGIntrinsic.h: Added. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::isFunctionConstant): (JSC::DFG::JITCodeGenerator::valueOfFunctionConstant): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::isFunctionConstant): (JSC::DFG::JITCompiler::valueOfFunctionConstant): * dfg/DFGNode.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNode): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JITStubs.cpp: (JSC::JITThunks::hostFunctionStub): * jit/JITStubs.h: * runtime/DateInstanceCache.h: * runtime/Executable.cpp: (JSC::ExecutableBase::intrinsic): (JSC::NativeExecutable::intrinsic): * runtime/Executable.h: (JSC::NativeExecutable::create): (JSC::NativeExecutable::finishCreation): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::getHostFunction): * runtime/JSGlobalData.h: * runtime/Lookup.cpp: (JSC::HashTable::createTable): (JSC::setUpStaticFunctionSlot): * runtime/Lookup.h: (JSC::HashEntry::initialize): (JSC::HashEntry::intrinsic): 2011-09-16 Filip Pizlo <fpizlo@apple.com> REGRESSION: Reproducible crash below SlotVisitor::harvestWeakReferences using Domino's online ordering https://bugs.webkit.org/show_bug.cgi?id=68220 Reviewed by Oliver Hunt. Weak handle processing can result in new objects being marked, which results in new WeakReferencesHarvesters being added. But weak reference harvesters are only processed before weak handle processing, so there's the risk that a weak reference harvester will persist until the next collection, by which time it may have been deleted. * heap/Heap.cpp: (JSC::Heap::markRoots): 2011-09-16 Csaba Osztrogonác <ossy@webkit.org> REGRESSION(r95201): It made two tests fail https://bugs.webkit.org/show_bug.cgi?id=68230 Unreviewed rolling out r95201. * jit/JIT.h: * jit/JITArithmetic32_64.cpp: (JSC::JIT::emit_op_lshift): (JSC::JIT::emitRightShift): (JSC::JIT::emit_op_bitand): (JSC::JIT::emit_op_bitor): (JSC::JIT::emit_op_bitxor): (JSC::JIT::emit_op_bitnot): (JSC::JIT::emit_op_post_inc): (JSC::JIT::emit_op_post_dec): (JSC::JIT::emit_op_pre_inc): (JSC::JIT::emit_op_pre_dec): * jit/JITInlineMethods.h: 2011-09-15 Filip Pizlo <fpizlo@apple.com> DFG JIT does not optimize method_check https://bugs.webkit.org/show_bug.cgi?id=68215 Reviewed by Oliver Hunt. MethodCallLinkInfo and StructureStubInfo are now searchable by bytecodeIndex, so that DFG::ByteCodeParser can use that information to determine how to optimize GetMethod. A new node op has been added to DFG: CheckMethod. This is a variant of GetMethod that has been optimized for the case that GetMethod always takes the fast path. CheckMethod results in only a very small amount of code (two loads and two branches in the worst case, one load and one branch in the best case). CheckMethod behaves as if it were a constant. Introduced the notion that a DFG node that is not JSConstant behaves as a constant. CheckMethod uses this functionality. This is a 3% speed-up on Kraken, and a small speed-up on V8. Appears to be neutral on SunSpider. * bytecode/CodeBlock.h: (JSC::getStructureStubInfoBytecodeIndex): (JSC::getMethodCallLinkInfoBytecodeIndex): * bytecode/PredictedType.cpp: (JSC::predictionFromCell): (JSC::predictionFromValue): * bytecode/PredictedType.h: * bytecode/StructureStubInfo.h: * dfg/DFGAliasTracker.h: (JSC::DFG::AliasTracker::recordGetMethod): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::Graph::getMethodCheckPrediction): (JSC::DFG::Graph::getPrediction): (JSC::DFG::Graph::isConstant): (JSC::DFG::Graph::isJSConstant): (JSC::DFG::Graph::valueOfJSConstant): (JSC::DFG::Graph::valueOfInt32Constant): (JSC::DFG::Graph::valueOfNumberConstant): (JSC::DFG::Graph::valueOfBooleanConstant): (JSC::DFG::Graph::valueOfJSConstantNode): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillInteger): (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::isKnownNotInteger): (JSC::DFG::JITCodeGenerator::isKnownNotNumber): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentSpillFPR): (JSC::DFG::JITCodeGenerator::silentFillGPR): (JSC::DFG::JITCodeGenerator::silentFillFPR): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::fillNumericToDouble): (JSC::DFG::JITCompiler::fillInt32ToInteger): (JSC::DFG::JITCompiler::fillToJS): * dfg/DFGNode.h: (JSC::DFG::Node::hasConstant): (JSC::DFG::Node::hasIdentifier): (JSC::DFG::Node::hasMethodCheckData): (JSC::DFG::Node::methodCheckDataIndex): (JSC::DFG::Node::valueOfJSConstant): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNode): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compile): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JIT.h: (JSC::PropertyStubCompilationInfo::PropertyStubCompilationInfo): (JSC::MethodCallCompilationInfo::MethodCallCompilationInfo): * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_method_check): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::emit_op_put_by_id): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_method_check): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::emit_op_put_by_id): * runtime/JSCell.h: (JSC::JSCell::JSCell::structureAddress): 2011-09-15 Adam Barth <abarth@webkit.org> Rename ENABLE(DATABASE) to ENABLE(SQL_DATABASE) https://bugs.webkit.org/show_bug.cgi?id=68205 Reviewed by Eric Seidel. * Configurations/FeatureDefines.xcconfig: * wtf/Platform.h: 2011-09-15 Mark Hahnenberg <mhahnenberg@apple.com> Unzip initialization lists and constructors in JSCell hierarchy (7/7) https://bugs.webkit.org/show_bug.cgi?id=68122 Reviewed by Geoffrey Garen. Completed the seventh and final level of the refactoring to add finishCreation() methods to all classes within the JSCell hierarchy with non-trivial constructor bodies. JSCallbackObject was missed in previous patches due to the fact that it's non-obvious (at least to my script) that it is in the JSCell hierarchy, so this is just a bit of retroactive cleanup. * API/JSCallbackObject.h: (JSC::JSCallbackObject::create): * API/JSCallbackObjectFunctions.h: (JSC::::JSCallbackObject): 2011-09-15 Filip Pizlo <fpizlo@apple.com> The DFG non-speculative JIT is no longer used and should be removed. https://bugs.webkit.org/show_bug.cgi?id=68177 Reviewed by Geoffrey Garen. This removes the non-speculative JIT and everything that relied on it, including the ability to turn on DFG but not tiered compilation the, ability to perform speculation failure into non-speculative JIT code, and the ability to statically terminate speculation. * GNUmakefile.list.am: * JavaScriptCore.pro: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.h: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitLoopHint): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::getStrongPrediction): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGGenerationInfo.h: * dfg/DFGGraph.cpp: (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGJITCodeGenerator.cpp: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::linkOSRExits): (JSC::DFG::JITCompiler::compileBody): * dfg/DFGJITCompiler.h: * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: Removed. * dfg/DFGNonSpeculativeJIT.h: Removed. * dfg/DFGOSREntry.cpp: (JSC::DFG::prepareOSREntry): * dfg/DFGPropagator.cpp: * dfg/DFGPropagator.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::osrExits): (JSC::DFG::SpeculativeJIT::speculationRecovery): (JSC::DFG::SpeculativeJIT::speculationCheck): (JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompile): * jit/JIT.h: * jit/JITCode.h: (JSC::JITCode::bottomTierJIT): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::~JSGlobalData): * runtime/JSGlobalData.h: * wtf/Platform.h: 2011-09-15 Eric Seidel <eric@webkit.org> Remove ENABLE(SVG_AS_IMAGE) since all major ports have it on by default https://bugs.webkit.org/show_bug.cgi?id=68182 Reviewed by Adam Barth. * Configurations/FeatureDefines.xcconfig: 2011-09-15 Filip Pizlo <fpizlo@apple.com> DFG speculative JIT sometimes asserts that a value is not a number even when it doesn't know anything about the number https://bugs.webkit.org/show_bug.cgi?id=68189 Reviewed by Oliver Hunt. * dfg/DFGGenerationInfo.h: (JSC::DFG::GenerationInfo::isUnknownJS): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::isKnownNotNumber): 2011-09-15 Filip Pizlo <fpizlo@apple.com> All of the functionality in the non-speculative JIT should be available to the speculative JIT via helper methods https://bugs.webkit.org/show_bug.cgi?id=68186 Reviewed by Oliver Hunt. Stole all of the goodness from NonSpeculativeJIT and placed it in JITCodeGenerator. Left all of the badness (i.e. subtle code duplication with SpeculativeJIT, etc). This is in preparation for removing the NonSpeculativeJIT entirely, but having its goodness available for reuse in the SpeculativeJIT if necessary. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToNumber): (JSC::DFG::JITCodeGenerator::nonSpeculativeValueToInt32): (JSC::DFG::JITCodeGenerator::nonSpeculativeUInt32ToNumber): (JSC::DFG::JITCodeGenerator::nonSpeculativeKnownConstantArithOp): (JSC::DFG::JITCodeGenerator::nonSpeculativeBasicArithOp): (JSC::DFG::JITCodeGenerator::nonSpeculativeArithMod): (JSC::DFG::JITCodeGenerator::nonSpeculativeCheckHasInstance): (JSC::DFG::JITCodeGenerator::nonSpeculativeInstanceOf): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::nonSpeculativeAdd): (JSC::DFG::JITCodeGenerator::nonSpeculativeArithSub): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGNonSpeculativeJIT.h: 2011-09-15 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r95167. http://trac.webkit.org/changeset/95167 https://bugs.webkit.org/show_bug.cgi?id=68191 Patch needs further work. (Requested by mhahnenberg on #webkit). * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSCell.cpp: (JSC::JSCell::toBoolean): * runtime/JSCell.h: (JSC::JSCell::JSValue::toBoolean): * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::toBoolean): * runtime/JSNotAnObject.h: * runtime/JSObject.h: * runtime/JSString.h: * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::toBoolean): 2011-09-15 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix for platforms that expect a linkable symbol for primitive static const's. * bytecode/CodeBlock.h: * jit/JIT.cpp: (JSC::JIT::emitOptimizationCheck): 2011-09-15 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix for assertion on existence of alternative CodeBlock. * dfg/DFGGraph.cpp: (JSC::DFG::Graph::predictArgumentTypes): 2011-09-14 Filip Pizlo <fpizlo@apple.com> Value profiles collect no information for global variables https://bugs.webkit.org/show_bug.cgi?id=68143 Reviewed by Geoffrey Garen. 17% speed-up on string-fasta. Neutral elsewhere. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getStrongPrediction): (JSC::DFG::ByteCodeParser::stronglyPredict): (JSC::DFG::ByteCodeParser::parseBlock): * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_get_global_var): 2011-09-15 Eric Seidel <eric@webkit.org> Remove ENABLE_SVG_ANIMATION as all major ports have it on by default https://bugs.webkit.org/show_bug.cgi?id=68022 Reviewed by Ryosuke Niwa. * Configurations/FeatureDefines.xcconfig: 2011-09-15 Gavin Barraclough <barraclough@apple.com> Ooops, revert accidentally commited unreviewed changes. * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_jfalse): (JSC::JIT::emit_op_jtrue): * jit/JSInterfaceJIT.h: * runtime/JSValue.h: 2011-09-15 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r95163. http://trac.webkit.org/changeset/95163 https://bugs.webkit.org/show_bug.cgi?id=68180 [Qt] The QT_GCC_X variables were removed in Qt5 by accident. (Requested by darktears on #webkit). * JavaScriptCore.pro: 2011-09-15 Gavin Barraclough <barraclough@apple.com> Windows build fix p1. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_jfalse): (JSC::JIT::emit_op_jtrue): * jit/JSInterfaceJIT.h: * runtime/JSValue.h: 2011-09-14 Filip Pizlo <fpizlo@apple.com> Tiered compilation should be enabled by default on platforms that support the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=68136 Reviewed by Sam Weinig. Neutral on SunSpider, 4% speed-up on V8, and 19% speed-up on Kraken. Large progressions on some benchmarks, including 3x on imaging-desaturate. * wtf/Platform.h: 2011-09-15 Gavin Barraclough <barraclough@apple.com> devirtualize preventExtensions https://bugs.webkit.org/show_bug.cgi?id=68176 Reviewed by Oliver Hunt. This is virtual due to problems in JSFunction putting the prototype property, but we can fix this problem a different way, just setting the checkReadOnly flag to false in the put. * runtime/JSFunction.cpp: (JSC::JSFunction::getOwnPropertySlot): * runtime/JSFunction.h: * runtime/JSObject.h: 2011-09-15 Geoffrey Garen <ggaren@apple.com> Value chaining for JSValue32_64 bitops. Reviewed by Sam Weinig. SunSpider says 2.3% faster, v8 ~1% faster (mostly due to crypto). * jit/JIT.h: * jit/JITInlineMethods.h: (JSC::JIT::emitStoreAndMapInt32): New int32 helper function for stores that can chain their results, which is the common case. * jit/JITArithmetic32_64.cpp: (JSC::JIT::emit_op_lshift): (JSC::JIT::emitRightShift): (JSC::JIT::emit_op_bitand): (JSC::JIT::emit_op_bitor): (JSC::JIT::emit_op_bitxor): (JSC::JIT::emit_op_bitnot): (JSC::JIT::emit_op_pre_inc): (JSC::JIT::emit_op_pre_dec): Deployed new function. (JSC::JIT::emit_op_post_inc): (JSC::JIT::emit_op_post_dec): Had to reorder these functions so they computed their result values last, to make them elligible for chaining. 2011-09-15 Adam Roben <aroben@apple.com> Clang build fix after r95172 * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::shouldSpeculateFinalObject): (JSC::DFG::SpeculativeJIT::shouldSpeculateArray): Added parentheses to make precendence clear. 2011-09-14 Filip Pizlo <fpizlo@apple.com> DFG does not speculate aggressively enough on comparisons https://bugs.webkit.org/show_bug.cgi?id=68138 Reviewed by Oliver Hunt. This is a 75% speed-up on Kraken/ai-astar. It's a 1% win on V8 and an 8.5% win on Kraken. Neutral on SunSpider. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleDoubleBranch): (JSC::DFG::SpeculativeJIT::compilePeepHoleObjectEquality): (JSC::DFG::SpeculativeJIT::compileObjectEquality): (JSC::DFG::SpeculativeJIT::compare): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::shouldSpeculateFinalObject): (JSC::DFG::SpeculativeJIT::shouldSpeculateArray): (JSC::DFG::SpeculativeJIT::shouldSpeculateObject): (JSC::DFG::SpeculativeJIT::shouldSpeculateCell): 2011-09-14 Filip Pizlo <fpizlo@apple.com> DFG JIT does not leverage integer speculations on branches https://bugs.webkit.org/show_bug.cgi?id=68140 Reviewed by Oliver Hunt. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::isStrictInt32): * dfg/DFGJITCodeGenerator.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-14 Gavin Barraclough <barraclough@apple.com> [n]stricteq code is bogus in JSValue32_64 JIT https://bugs.webkit.org/show_bug.cgi?id=68141 Reviewed by Sam Weinig. The code tries to check for both ints or cells, but this check also catches cases where values that are undefined, null, etc (probably was incorrectly assuming cell was the 2nd highest tag?). Also, there is no need not to handle int on the fast path. stricteq is just a case of comparing the payloads, if we: * handle cases of differing tags on a slow path * handle doubles a slow path * handle both-are-string on a slow path * jit/JITOpcodes32_64.cpp: (JSC::JIT::compileOpStrictEq): (JSC::JIT::emitSlow_op_stricteq): (JSC::JIT::emitSlow_op_nstricteq): 2011-09-14 Mark Hahnenberg <mhahnenberg@apple.com> Make JSCell::toBoolean non-virtual https://bugs.webkit.org/show_bug.cgi?id=67727 Reviewed by Sam Weinig. JSCell::toBoolean now manually performs the toBoolean check for objects and strings (where before it was simply virtual and would crash if its implementation was called). Its descendants in JSObject and JSString have also been made non-virtual. JSCell now explicitly covers all cases of toBoolean, so having a virtual implementation of JSCell::toBoolean is no longer necessary. This is part of a larger process of un-virtualizing JSCell. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSCell.cpp: * runtime/JSCell.h: * runtime/JSNotAnObject.cpp: * runtime/JSNotAnObject.h: * runtime/JSObject.h: * runtime/JSString.h: (JSC::JSCell::toBoolean): (JSC::JSValue::toBoolean): * runtime/StringObjectThatMasqueradesAsUndefined.h: 2011-09-14 Alexis Menard <alexis.menard@openbossa.org> [Qt] Replace QT_GCC_X as they don't exist in Qt5 anymore. https://bugs.webkit.org/show_bug.cgi?id=68114 Reviewed by Kenneth Rohde Christiansen. Use the new GCC_X variables defined in WebKit.pri to replace the usage of QT_GCC_X. * JavaScriptCore.pro: 2011-09-14 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r95145. http://trac.webkit.org/changeset/95145 https://bugs.webkit.org/show_bug.cgi?id=68139 The GTK+ build is working now, so revert this trial build fix. (Requested by mrobinson on #webkit). * GNUmakefile.list.am: 2011-09-14 Patrick Gansterer <paroga@webkit.org> Port MachineStackMarker to Windows ARM and MIPS https://bugs.webkit.org/show_bug.cgi?id=68068 Reviewed by Geoffrey Garen. Use the correct memeber of the CONTEXT struct for the stackpointer for CPU(ARM) and CPU(MIPS). Only query CONTEXT_INTEGER and CONTEXT_CONTROL, since CONTEXT_SEGMENTS isn't defined for CPU(ARM) and CPU(MIPS) and the stackpointer is defined in the CONTEXT_CONTROL section for CPU(ARM), CPU(X86) and CPU(X86_64) and in the CONTEXT_INTEGER section for CPU(MIPS). * heap/MachineStackMarker.cpp: (JSC::getPlatformThreadRegisters): (JSC::otherThreadStackPointer): 2011-09-12 Filip Pizlo <fpizlo@apple.com> DFG JIT always speculates that ValueAdd is a numeric addition https://bugs.webkit.org/show_bug.cgi?id=67956 Reviewed by Geoffrey Garen. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::isKnownNotNumber): * dfg/DFGJITCodeGenerator.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::knownConstantArithOp): (JSC::DFG::NonSpeculativeJIT::basicArithOp): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::shouldSpeculateNumber): 2011-09-14 Anders Carlsson <andersca@apple.com> Stop building BinarySemaphore to see if that's what's breaking the GTK+ build. * GNUmakefile.list.am: 2011-09-14 Anders Carlsson <andersca@apple.com> This is getting old. Yet another build fix attempt. * JavaScriptCore.vcproj/WTF/WTFCommon.vsprops: 2011-09-14 Anders Carlsson <andersca@apple.com> Yet another build fix attempt. * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: 2011-09-14 Anders Carlsson <andersca@apple.com> How I "love" Visual Studio... Try to fix build again. * JavaScriptCore.vcproj/WTF/WTFCommon.vsprops: 2011-09-14 Anders Carlsson <andersca@apple.com> Try to fix Windows build. * JavaScriptCore.vcproj/WTF/WTFCommon.vsprops: 2011-09-14 Anders Carlsson <andersca@apple.com> Add BinarySemaphore class from WebKit2 to WTF https://bugs.webkit.org/show_bug.cgi?id=68132 Reviewed by Sam Weinig. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/CMakeLists.txt: Update build systems. * wtf/threads: Added. * wtf/threads/BinarySemaphore.cpp: Copied from Source/WebKit2/Platform/CoreIPC/BinarySemaphore.cpp. * wtf/threads/BinarySemaphore.h: Copied from Source/WebKit2/Platform/CoreIPC/BinarySemaphore.h. * wtf/threads/win: Added. * wtf/threads/win/BinarySemaphoreWin.cpp: Copied from Source/WebKit2/Platform/CoreIPC/win/BinarySemaphoreWin.cpp. 2011-09-14 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix for Interpreter. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): 2011-09-14 Anders Carlsson <andersca@apple.com> Add wtf/threads and wtf/threads/win, so we can be sure that the EWS bots can correctly build the patch in https://bugs.webkit.org/show_bug.cgi?id=68132 Rubber-stamped by Sam Weinig. * wtf/threads: Added. * wtf/threads/win: Added. 2011-09-14 Filip Pizlo <fpizlo@apple.com> DFG JIT should not speculate integer if the value is always going to be used as a double anyway https://bugs.webkit.org/show_bug.cgi?id=68127 Reviewed by Oliver Hunt. Added a ValueToDouble node, which is a variant of ValueToNumber that hints that it will only be used as a double and never as an integer. Thus, it turns off integer speculation even if the value profiler told us that the value source is an int. The logic for converting a ValueToNumber into a ValueToDouble is found in Propagator. This appears to be a 22% speed-up in imaging-darkroom. * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::fixpoint): (JSC::DFG::Propagator::toDouble): (JSC::DFG::Propagator::fixupNode): (JSC::DFG::Propagator::fixup): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): 2011-09-14 Filip Pizlo <fpizlo@apple.com> Tiered compilation heuristics do not account for value profile fullness https://bugs.webkit.org/show_bug.cgi?id=68116 Reviewed by Oliver Hunt. Tiered compilation avoids invoking the DFG JIT if it finds that value profiles contain insufficient information. Instead, it produces a prediction from the current value profile, and then clears the value profile. This allows the value profile to heat up from scratch for some number of additional executions. The new profiles will then be merged with the previous prediction. Once the amount of information in predictions is enough according to heuristics in CodeBlock.cpp, DFG optimization is allowed to proceed. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::CodeBlock): (JSC::CodeBlock::~CodeBlock): (JSC::CodeBlock::visitAggregate): (JSC::CodeBlock::visitWeakReferences): (JSC::CodeBlock::shouldOptimizeNow): (JSC::CodeBlock::dumpValueProfiles): * bytecode/CodeBlock.h: * bytecode/PredictedType.cpp: (JSC::predictionToString): * bytecode/PredictedType.h: * bytecode/ValueProfile.cpp: Added. (JSC::ValueProfile::computeStatistics): (JSC::ValueProfile::computeUpdatedPrediction): * bytecode/ValueProfile.h: (JSC::ValueProfile::ValueProfile): (JSC::ValueProfile::classInfo): (JSC::ValueProfile::numberOfSamples): (JSC::ValueProfile::totalNumberOfSamples): (JSC::ValueProfile::isLive): (JSC::ValueProfile::numberOfInt32s): (JSC::ValueProfile::numberOfDoubles): (JSC::ValueProfile::numberOfBooleans): (JSC::ValueProfile::dump): (JSC::getValueProfileBytecodeOffset): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::stronglyPredict): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): * jit/JIT.cpp: (JSC::JIT::emitOptimizationCheck): * jit/JITInlineMethods.h: (JSC::JIT::emitValueProfilingSite): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): 2011-09-14 Filip Pizlo <fpizlo@apple.com> DFG should not speculate that the child of LogicalNot is a boolean if predictions tell us otherwise https://bugs.webkit.org/show_bug.cgi?id=68118 Reviewed by Geoffrey Garen. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativeLogicalNot): * dfg/DFGJITCodeGenerator.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-09-14 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix. Turn off tiered compilation. * wtf/Platform.h: 2011-09-13 Filip Pizlo <fpizlo@apple.com> Prediction tracking is not precise enough https://bugs.webkit.org/show_bug.cgi?id=67993 Reviewed by Oliver Hunt. Added a richer set of type predictions, including JSFinalObject, JSString, object that is not a JSFinalObject or JSArray (ObjectOther), some object but we don't or care know what kind (SomeObject), definitely an object, cell that is not an object or JSString, an value that is none of the above (so either Undefined or Null). Made the propagator and value profiler work with the new types. Performance is neutral, because the DFG JIT does not take advantage of this new knowledge yet. In the process of writing predictionToString() (which is now considerably more complex) I decided to finally add a BoundsCheckedPointer, which should come in handy in other places, like at least the OSR scratch buffer and the CompactJITCodeMap. It's great for cases where you want to do pointer arithmetic, you want to have assertions about the pointer not going out of bounds, but you don't want to write those assertions yourself. This also required refactoring inherits(), since the ValueProfiler may want to do the equivalent of inherits() but given two ClassInfo's. * GNUmakefile.list.am: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/PredictedType.cpp: Added. (JSC::predictionToString): (JSC::makePrediction): (JSC::predictionFromValue): * bytecode/PredictedType.h: (JSC::isCellPrediction): (JSC::isObjectPrediction): (JSC::isFinalObjectPrediction): (JSC::isStringPrediction): (JSC::mergePredictions): * bytecode/ValueProfile.h: (JSC::ValueProfile::numberOfObjects): (JSC::ValueProfile::numberOfFinalObjects): (JSC::ValueProfile::numberOfStrings): (JSC::ValueProfile::probabilityOfObject): (JSC::ValueProfile::probabilityOfFinalObject): (JSC::ValueProfile::probabilityOfString): (JSC::ValueProfile::dump): (JSC::ValueProfile::Statistics::Statistics): (JSC::ValueProfile::computeStatistics): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::stronglyPredict): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGNode.h: (JSC::DFG::Node::predict): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::propagateNode): * runtime/ClassInfo.h: (JSC::ClassInfo::isSubClassOf): * runtime/JSObject.h: (JSC::JSCell::inherits): * wtf/BoundsCheckedPointer.h: Added. (WTF::BoundsCheckedPointer::BoundsCheckedPointer): (WTF::BoundsCheckedPointer::operator=): (WTF::BoundsCheckedPointer::operator+=): (WTF::BoundsCheckedPointer::operator-=): (WTF::BoundsCheckedPointer::operator+): (WTF::BoundsCheckedPointer::operator-): (WTF::BoundsCheckedPointer::operator++): (WTF::BoundsCheckedPointer::operator--): (WTF::BoundsCheckedPointer::operator<): (WTF::BoundsCheckedPointer::operator<=): (WTF::BoundsCheckedPointer::operator>): (WTF::BoundsCheckedPointer::operator>=): (WTF::BoundsCheckedPointer::operator==): (WTF::BoundsCheckedPointer::operator!=): (WTF::BoundsCheckedPointer::operator!): (WTF::BoundsCheckedPointer::get): (WTF::BoundsCheckedPointer::operator*): (WTF::BoundsCheckedPointer::operator[]): (WTF::BoundsCheckedPointer::strcat): (WTF::BoundsCheckedPointer::validate): * wtf/CMakeLists.txt: 2011-09-14 Csaba Osztrogonác <ossy@webkit.org> [Qt] Win32 builds with threads turned off https://bugs.webkit.org/show_bug.cgi?id=67864 Reviewed by Geoffrey Garen. * JavaScriptCore.pri: Link pthread library on Windows platform. * wtf/Platform.h: Enable multiple threads. 2011-09-14 Mark Hahnenberg <mhahnenberg@apple.com> Unzip initialization lists and constructors in JSCell hierarchy (6/7) https://bugs.webkit.org/show_bug.cgi?id=67692 Reviewed by Geoffrey Garen. Completed the sixth level of the refactoring to add finishCreation() methods to all classes within the JSCell hierarchy with non-trivial constructor bodies. This primarily consists of pushing the calls to finishCreation() down into the constructors of the subclasses of the fifth level of the hierarchy as well as pulling the finishCreation() calls out into the class's corresponding create() method if it has one. Doing both simultaneously allows us to maintain the invariant that the finishCreation() method chain is called exactly once during the creation of an object, since calling it any other number of times (0, 2, or more) will cause an assertion failure. * API/JSCallbackFunction.cpp: (JSC::JSCallbackFunction::JSCallbackFunction): * API/JSCallbackFunction.h: (JSC::JSCallbackFunction::create): * jsc.cpp: (GlobalObject::create): (GlobalObject::GlobalObject): * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::ArrayConstructor): * runtime/ArrayConstructor.h: (JSC::ArrayConstructor::create): * runtime/BooleanConstructor.cpp: (JSC::BooleanConstructor::BooleanConstructor): * runtime/BooleanConstructor.h: (JSC::BooleanConstructor::create): * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::BooleanPrototype): * runtime/BooleanPrototype.h: (JSC::BooleanPrototype::create): * runtime/DateConstructor.cpp: (JSC::DateConstructor::DateConstructor): * runtime/DateConstructor.h: (JSC::DateConstructor::create): * runtime/DatePrototype.cpp: (JSC::DatePrototype::DatePrototype): * runtime/DatePrototype.h: (JSC::DatePrototype::create): * runtime/Error.cpp: (JSC::StrictModeTypeErrorFunction::StrictModeTypeErrorFunction): (JSC::StrictModeTypeErrorFunction::create): * runtime/ErrorConstructor.cpp: (JSC::ErrorConstructor::ErrorConstructor): * runtime/ErrorConstructor.h: (JSC::ErrorConstructor::create): * runtime/FunctionConstructor.cpp: (JSC::FunctionConstructor::FunctionConstructor): * runtime/FunctionConstructor.h: (JSC::FunctionConstructor::create): * runtime/FunctionPrototype.cpp: (JSC::FunctionPrototype::FunctionPrototype): * runtime/FunctionPrototype.h: (JSC::FunctionPrototype::create): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::NativeErrorConstructor): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::create): * runtime/NativeErrorPrototype.cpp: (JSC::NativeErrorPrototype::NativeErrorPrototype): (JSC::NativeErrorPrototype::finishCreation): * runtime/NativeErrorPrototype.h: (JSC::NativeErrorPrototype::create): * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::NumberConstructor): * runtime/NumberConstructor.h: (JSC::NumberConstructor::create): * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::NumberPrototype): * runtime/NumberPrototype.h: (JSC::NumberPrototype::create): * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::ObjectConstructor): * runtime/ObjectConstructor.h: (JSC::ObjectConstructor::create): * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::RegExpConstructor): * runtime/RegExpConstructor.h: (JSC::RegExpConstructor::create): * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::RegExpPrototype): * runtime/RegExpPrototype.h: (JSC::RegExpPrototype::create): * runtime/StringConstructor.cpp: (JSC::StringConstructor::StringConstructor): * runtime/StringConstructor.h: (JSC::StringConstructor::create): * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::create): (JSC::StringObjectThatMasqueradesAsUndefined::StringObjectThatMasqueradesAsUndefined): * runtime/StringPrototype.cpp: (JSC::StringPrototype::StringPrototype): * runtime/StringPrototype.h: (JSC::StringPrototype::create): 2011-09-13 Eric Seidel <eric@webkit.org> Remove ENABLE_SVG_USE as <use> is required by HTML5 https://bugs.webkit.org/show_bug.cgi?id=68019 Reviewed by Ryosuke Niwa. * Configurations/FeatureDefines.xcconfig: 2011-09-14 Iain Merrick <husky@google.com> HashTraits.h should include template specialization for WTF::String https://bugs.webkit.org/show_bug.cgi?id=67851 Ensure that the template specialization for HashTraits<String> is always picked up. (Previously it was possible to include HashSet and String but not the correct HashTraits, so you would get an inefficient template instantiation.) Reviewed by Darin Adler. * wtf/HashTraits.h: * wtf/text/StringHash.h: 2011-09-13 Filip Pizlo <fpizlo@apple.com> SpeculativeJIT::shouldSpeculateInteger(NodeIndex, NodeIndex) should return false if either node can be double https://bugs.webkit.org/show_bug.cgi?id=67985 Reviewed by Geoffrey Garen. This is a 17% speed-up on 3d-cube. This required allowing us to check if a constant is double but not integer, and making the shouldSpeculateInteger() check test for any hints of doubly-ness in its operands. This also required changing some terminology: previously "isDouble" often meant "isDouble or isInt32". Now "isDouble" means exactly what the name suggests, and "isNumber" means "isDouble or isInt32". * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::toNumber): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGenerationInfo.h: (JSC::DFG::isJSFormat): (JSC::DFG::isJSInteger): (JSC::DFG::isJSDouble): (JSC::DFG::isJSCell): (JSC::DFG::isJSBoolean): (JSC::DFG::GenerationInfo::isJSFormat): (JSC::DFG::GenerationInfo::isJSInteger): (JSC::DFG::GenerationInfo::isJSDouble): (JSC::DFG::GenerationInfo::isJSCell): (JSC::DFG::GenerationInfo::isJSBoolean): * dfg/DFGGraph.h: (JSC::DFG::Graph::isNumberConstant): (JSC::DFG::Graph::valueOfNumberConstant): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillInteger): (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::isKnownInteger): (JSC::DFG::JITCodeGenerator::isKnownNumeric): (JSC::DFG::JITCodeGenerator::isKnownCell): (JSC::DFG::JITCodeGenerator::isKnownNotInteger): (JSC::DFG::JITCodeGenerator::isKnownBoolean): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentFillFPR): (JSC::DFG::JITCodeGenerator::isNumberConstant): (JSC::DFG::JITCodeGenerator::valueOfNumberConstant): (JSC::DFG::JITCodeGenerator::initConstantInfo): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::fillNumericToDouble): (JSC::DFG::JITCompiler::fillToJS): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::isNumberConstant): (JSC::DFG::JITCompiler::valueOfNumberConstant): * dfg/DFGNode.h: (JSC::DFG::Node::isDoubleConstant): (JSC::DFG::Node::isNumberConstant): (JSC::DFG::Node::valueOfNumberConstant): (JSC::DFG::Node::hasNumberResult): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::knownConstantArithOp): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isInteger): (JSC::DFG::SpeculativeJIT::shouldSpeculateDouble): (JSC::DFG::SpeculativeJIT::shouldNotSpeculateInteger): (JSC::DFG::SpeculativeJIT::shouldSpeculateInteger): 2011-09-13 Anders Carlsson <andersca@apple.com> Disable C++ exceptions when building with clang https://bugs.webkit.org/show_bug.cgi?id=68031 <rdar://problem/9556880> Reviewed by Mark Rowe. * Configurations/Base.xcconfig: 2011-09-13 Eric Seidel <eric@webkit.org> Remove ENABLE_SVG_FOREIGN_OBJECT as it is a required part of HTML5 https://bugs.webkit.org/show_bug.cgi?id=68018 Reviewed by Ryosuke Niwa. * Configurations/FeatureDefines.xcconfig: 2011-09-13 Sam Weinig <sam@webkit.org> Object.getPrototypeOf should use JSValue::get() https://bugs.webkit.org/show_bug.cgi?id=67973 Reviewed by Darin Adler. * runtime/ObjectConstructor.cpp: (JSC::objectConstructorGetPrototypeOf): Pipe through JSValue::get() to allow overrides. 2011-09-12 Filip Pizlo <fpizlo@apple.com> JavaScriptCore does not have baseline->speculative OSR https://bugs.webkit.org/show_bug.cgi?id=67920 Reviewed by Oliver Hunt. This adds the ability to on-stack-replace (OSR) from code that is running hot in the old JIT to code compiled by the new JIT. This ensures that long-running loops benefit from DFG optimization. It also ensures that if code experiences a speculation failure in DFG code, it has an opportunity to reenter the DFG once every 1,000 loop iterations or so. This results in a 2.88x speed-up on Kraken/imaging-desaturate, and is a pure win on the main three benchmark suites (SunSpider, V8, Kraken), when tiered compilation is enabled. * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): (JSC::CodeBlock::CodeBlock): (JSC::ProgramCodeBlock::compileOptimized): (JSC::EvalCodeBlock::compileOptimized): (JSC::FunctionCodeBlock::compileOptimized): * bytecode/CodeBlock.h: * bytecode/Opcode.h: * bytecode/PredictedType.h: Added. (JSC::isCellPrediction): (JSC::isArrayPrediction): (JSC::isInt32Prediction): (JSC::isDoublePrediction): (JSC::isNumberPrediction): (JSC::isBooleanPrediction): (JSC::isStrongPrediction): (JSC::predictionToString): (JSC::mergePredictions): (JSC::mergePrediction): (JSC::makePrediction): * bytecode/PredictionTracker.h: Added. (JSC::operandIsArgument): (JSC::PredictionSlot::PredictionSlot): (JSC::PredictionTracker::PredictionTracker): (JSC::PredictionTracker::initializeSimilarTo): (JSC::PredictionTracker::copyLocalsFrom): (JSC::PredictionTracker::numberOfArguments): (JSC::PredictionTracker::numberOfVariables): (JSC::PredictionTracker::argumentOffsetForOperand): (JSC::PredictionTracker::predictArgument): (JSC::PredictionTracker::predict): (JSC::PredictionTracker::predictGlobalVar): (JSC::PredictionTracker::getArgumentPrediction): (JSC::PredictionTracker::getPrediction): (JSC::PredictionTracker::getGlobalVarPrediction): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitLoopHint): * bytecompiler/BytecodeGenerator.h: * bytecompiler/NodesCodegen.cpp: (JSC::DoWhileNode::emitBytecode): (JSC::WhileNode::emitBytecode): (JSC::ForNode::emitBytecode): (JSC::ForInNode::emitBytecode): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGCapabilities.h: (JSC::DFG::canCompileOpcode): * dfg/DFGDriver.cpp: (JSC::DFG::compile): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::BasicBlock::BasicBlock): (JSC::DFG::Graph::predict): (JSC::DFG::Graph::getPrediction): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): (JSC::DFG::JITCompiler::compileEntry): (JSC::DFG::JITCompiler::compileBody): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::noticeOSREntry): * dfg/DFGNode.h: * dfg/DFGOSREntry.cpp: Added. (JSC::DFG::predictionIsValid): (JSC::DFG::prepareOSREntry): * dfg/DFGOSREntry.h: Added. (JSC::DFG::prepareOSREntry): * dfg/DFGPredictionTracker.h: Removed. * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::mergeUse): (JSC::DFG::Propagator::mergePrediction): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/CompactJITCodeMap.h: (JSC::CompactJITCodeMap::numberOfEntries): (JSC::CompactJITCodeMap::decode): (JSC::CompactJITCodeMap::Decoder::Decoder): (JSC::CompactJITCodeMap::Decoder::numberOfEntriesRemaining): (JSC::CompactJITCodeMap::Decoder::read): * jit/JIT.cpp: (JSC::JIT::emitOptimizationCheck): (JSC::JIT::emitTimeoutCheck): (JSC::JIT::privateCompileMainPass): * jit/JIT.h: (JSC::JIT::emit_op_loop_hint): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): 2011-09-12 Sam Weinig <sam@webkit.org> Don't allow setting __proto__ to be a getter or setter https://bugs.webkit.org/show_bug.cgi?id=67982 Reviewed by Gavin Barraclough. * runtime/JSObject.cpp: (JSC::JSObject::defineGetter): (JSC::JSObject::defineSetter): Disallow setting a getter or setter on __proto__. 2011-09-12 James Robinson <jamesr@chromium.org> Unreviewed build fix for chromium. Guard access to UString::latin1() with USE(JSC) since it is defined in JavaScriptCore/runtime/UString.cpp, which is currently only compiled in by ports that use JavaScriptCore. This code is currently unreachable in builds so no change in functionality. * yarr/YarrInterpreter.cpp: (JSC::Yarr::Interpreter::CharAccess::CharAccess): 2011-09-09 Filip Pizlo <fpizlo@apple.com> JavaScriptCore does not have speculative->baseline OSR https://bugs.webkit.org/show_bug.cgi?id=67826 Reviewed by Oliver Hunt. This adds the ability to bail out of DFG speculative JIT execution by performing an on-stack replacement (OSR) that results in the control flow going to the equivalent code generated by the old JIT. This required a number of new features, as well as taking advantage of some features that happened to already be present: We already had a policy of storing the bytecode index for which a DFG node was generated inside the DFG::Node class. This was previously called exceptionInfo. It's now renamed to codeOrigin to reflect that it's used for more than just excpetions. OSR uses this to figure out which bytecode index to use to look up the machine code location in the code generated by the old JIT that we should be jumping to. CodeBlock now stores a mapping between bytecode indices and machine code offsets for code generated by the old JIT. This is implemented by CompactJITCodeMap, which tries to compress this data a bit. The OSR compiler decodes this and uses it to find the machine code locations it should be jumping to. We already had a mechanism that emitted SetLocal nodes in the DFG graph that told us the time at which the old JIT would have stored something into its register file, and the DFG::Node that corresponds to the value that it would have stored. These SetLocal's were mostly dead-code- eliminated, but our DCE leaves the nodes intact except for making them have 0 as the ref count. This allows the OSR compiler to construct a mapping between the state as it would have been seen by the old JIT and the state as the DFG JIT sees it. The OSR compiler uses this to generate code that reshapes the call frame so that it is like what the old JIT would expect. Finally, when DFG_OSR is enabled (the default for TIERED_COMPILATION) we no longer emit the non-speculative path. * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::currentCodeOrigin): (JSC::DFG::ByteCodeParser::addToGraph): * dfg/DFGGPRInfo.h: * dfg/DFGGenerationInfo.h: (JSC::DFG::GenerationInfo::alive): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::emitCall): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::appendCallWithExceptionCheck): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::exitSpeculativeWithOSR): (JSC::DFG::JITCompiler::linkOSRExits): (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: (JSC::DFG::CallRecord::CallRecord): (JSC::DFG::JITCompiler::notifyCall): (JSC::DFG::JITCompiler::appendCallWithExceptionCheck): (JSC::DFG::JITCompiler::appendCallWithFastExceptionCheck): (JSC::DFG::JITCompiler::addJSCall): (JSC::DFG::JITCompiler::JSCallRecord::JSCallRecord): * dfg/DFGNode.h: (JSC::DFG::CodeOrigin::CodeOrigin): (JSC::DFG::CodeOrigin::isSet): (JSC::DFG::CodeOrigin::bytecodeIndex): (JSC::DFG::Node::Node): (JSC::DFG::Node::child1Unchecked): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::ValueSource::dump): (JSC::DFG::ValueRecovery::dump): (JSC::DFG::OSRExit::OSRExit): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::compileMovHint): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::ValueSource::ValueSource): (JSC::DFG::ValueSource::isSet): (JSC::DFG::ValueSource::nodeIndex): (JSC::DFG::ValueRecovery::ValueRecovery): (JSC::DFG::ValueRecovery::alreadyInRegisterFile): (JSC::DFG::ValueRecovery::inGPR): (JSC::DFG::ValueRecovery::inFPR): (JSC::DFG::ValueRecovery::displacedInRegisterFile): (JSC::DFG::ValueRecovery::constant): (JSC::DFG::ValueRecovery::technique): (JSC::DFG::ValueRecovery::gpr): (JSC::DFG::ValueRecovery::fpr): (JSC::DFG::ValueRecovery::virtualRegister): (JSC::DFG::OSRExit::numberOfRecoveries): (JSC::DFG::OSRExit::valueRecovery): (JSC::DFG::OSRExit::isArgument): (JSC::DFG::OSRExit::argumentForIndex): (JSC::DFG::OSRExit::variableForIndex): (JSC::DFG::OSRExit::operandForIndex): (JSC::DFG::SpeculativeJIT::osrExits): (JSC::DFG::SpeculativeJIT::speculationCheck): (JSC::DFG::SpeculativeJIT::valueSourceForOperand): (JSC::DFG::SpeculativeJIT::setNodeIndexForOperand): (JSC::DFG::SpeculativeJIT::valueSourceReferenceForOperand): (JSC::DFG::SpeculativeJIT::computeValueRecoveryFor): (JSC::DFG::SpeculationCheckIndexIterator::SpeculationCheckIndexIterator): (JSC::DFG::SpeculativeJIT::SpeculativeJIT): * jit/CompactJITCodeMap.h: Added. (JSC::BytecodeAndMachineOffset::BytecodeAndMachineOffset): (JSC::BytecodeAndMachineOffset::getBytecodeIndex): (JSC::BytecodeAndMachineOffset::getMachineCodeOffset): (JSC::CompactJITCodeMap::~CompactJITCodeMap): (JSC::CompactJITCodeMap::decode): (JSC::CompactJITCodeMap::CompactJITCodeMap): (JSC::CompactJITCodeMap::at): (JSC::CompactJITCodeMap::decodeNumber): (JSC::CompactJITCodeMap::Encoder::Encoder): (JSC::CompactJITCodeMap::Encoder::~Encoder): (JSC::CompactJITCodeMap::Encoder::append): (JSC::CompactJITCodeMap::Encoder::finish): (JSC::CompactJITCodeMap::Encoder::appendByte): (JSC::CompactJITCodeMap::Encoder::encodeNumber): (JSC::CompactJITCodeMap::Encoder::ensureCapacityFor): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompile): * jit/JIT.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::~JSGlobalData): * runtime/JSGlobalData.h: (JSC::JSGlobalData::osrScratchBufferForSize): * runtime/JSValue.cpp: (JSC::JSValue::description): 2011-09-12 Geoffrey Garen <ggaren@apple.com> Re-enabled ENABLE(LAZY_BLOCK_FREEING). Reviewed by Stephanie Lewis. I accidentally disabled this in r94890, causing a big performance regression. * wtf/Platform.h: 2011-09-12 Michael Saboff <msaboff@apple.com> Broken Build for ARM - lshift32() needs TrustedImm32 arg https://bugs.webkit.org/show_bug.cgi?id=67965 Change lshift32(16, ARMRegisters::S1); to lshift32(TrustedImm32(16), ARMRegisters::S1); Reviewed by Anders Carlsson. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::branch16): 2011-09-12 Michael Saboff <msaboff@apple.com> Broken ARM build - missing semicolon in JavaScriptCore/assembler/MacroAssemblerARM.h https://bugs.webkit.org/show_bug.cgi?id=67961 Added missing semicolon. Reviewed by Ryosuke Niwa. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::branch16): 2011-09-12 Michael Saboff <msaboff@apple.com> Update RegExp and related classes to use 8 bit strings when available https://bugs.webkit.org/show_bug.cgi?id=67337 Modified both the Yarr interpreter and JIT to handle 8 bit subject strings. The code paths are triggered by the UString::is8bit() method which currently returns false. Implemented JIT changes for all current architectures. Tested X86_64 and ARM v7. This includes some code that will likely change as we complete the 8 bit string changes. This includes the way the raw buffer pointers are accessed as well as replacing the CharAccess class with a string interator returned from UString. Fixed build breakage in testRegExp.cpp due to globalObject construction changes. Reviewed by Gavin Barraclough. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * testRegExp.cpp: (GlobalObject::finishCreation): (GlobalObject::GlobalObject): * assembler/ARMAssembler.cpp: (JSC::ARMAssembler::baseIndexTransfer32): * assembler/ARMAssembler.h: * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::ubfx): (JSC::ARMv7Assembler::ARMInstructionFormatter::twoWordOp12Reg40Imm3Reg4Imm20Imm5): * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::load8): (JSC::MacroAssemblerARM::branch8): (JSC::MacroAssemblerARM::branch16): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::load8): (JSC::MacroAssemblerARMv7::branch16): (JSC::MacroAssemblerARMv7::branch8): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::load8): (JSC::MacroAssemblerMIPS::branch8): (JSC::MacroAssemblerMIPS::branch16): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::load8): (JSC::MacroAssemblerSH4::branch8): (JSC::MacroAssemblerSH4::branch16): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::load8): (JSC::MacroAssemblerX86Common::branch16): (JSC::MacroAssemblerX86Common::branch8): * assembler/SH4Assembler.h: (JSC::SH4Assembler::extub): (JSC::SH4Assembler::printInstr): * assembler/X86Assembler.h: (JSC::X86Assembler::cmpw_ir): (JSC::X86Assembler::movzbl_mr): * runtime/RegExp.cpp: (JSC::RegExp::compile): (JSC::RegExp::compileIfNecessary): (JSC::RegExp::match): (JSC::RegExp::matchCompareWithInterpreter): * runtime/RegExp.h: * runtime/UString.h: (JSC::UString::is8Bit): * yarr/Yarr.h: * yarr/YarrInterpreter.cpp: (JSC::Yarr::Interpreter::CharAccess::CharAccess): (JSC::Yarr::Interpreter::CharAccess::~CharAccess): (JSC::Yarr::Interpreter::CharAccess::operator[]): (JSC::Yarr::Interpreter::InputStream::InputStream): (JSC::Yarr::Interpreter::Interpreter): (JSC::Yarr::interpret): * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::jumpIfCharNotEquals): (JSC::Yarr::YarrGenerator::readCharacter): (JSC::Yarr::YarrGenerator::generatePatternCharacterOnce): (JSC::Yarr::YarrGenerator::generatePatternCharacterFixed): (JSC::Yarr::YarrGenerator::generatePatternCharacterGreedy): (JSC::Yarr::YarrGenerator::backtrackPatternCharacterNonGreedy): (JSC::Yarr::YarrGenerator::generateCharacterClassFixed): (JSC::Yarr::YarrGenerator::generateDotStarEnclosure): (JSC::Yarr::YarrGenerator::YarrGenerator): (JSC::Yarr::YarrGenerator::compile): (JSC::Yarr::jitCompile): (JSC::Yarr::execute): * yarr/YarrJIT.h: (JSC::Yarr::YarrCodeBlock::has8BitCode): (JSC::Yarr::YarrCodeBlock::has16BitCode): (JSC::Yarr::YarrCodeBlock::set8BitCode): (JSC::Yarr::YarrCodeBlock::set16BitCode): (JSC::Yarr::YarrCodeBlock::execute): * yarr/YarrParser.h: (JSC::Yarr::Parser::Parser): 2011-09-12 Andras Becsi <andras.becsi@nokia.com> [Qt] Build fails after r94920 with strict compiler https://bugs.webkit.org/show_bug.cgi?id=67928 Reviewed by Csaba Osztrogonác. * wtf/RedBlackTree.h: (WTF::RedBlackTree::insert): Remove dead variables updateStart and newSubTreeRoot. 2011-09-12 Patrick Gansterer <paroga@webkit.org> Unreviewed build fix after r94871. * runtime/InitializeThreading.cpp: (JSC::initializeThreadingOnce): * wtf/FastMalloc.cpp: * wtf/RefCountedLeakCounter.h: 2011-09-11 Filip Pizlo <fpizlo@apple.com> DFGNode.h has macros that indicate the enabling of a feature, but they do not use the ENABLE() idiom. https://bugs.webkit.org/show_bug.cgi?id=67907 Reviewed by Oliver Hunt. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::stronglyPredict): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGJITCodeGenerator.cpp: * dfg/DFGJITCodeGenerator.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::fillInt32ToInteger): (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::fixpoint): (JSC::DFG::Propagator::propagateNode): (JSC::DFG::Propagator::propagateForward): (JSC::DFG::Propagator::propagateBackward): (JSC::DFG::propagate): * dfg/DFGScoreBoard.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution): * jit/JIT.cpp: (JSC::JIT::privateCompile): 2011-09-11 Fumitoshi Ukai <ukai@chromium.org> Unreviewed build fix for chromium/mac & clang. Fix the macro redefinition error by r94927, because chromium set ENABLE_JSC_MULTIPLE_THREADS=0 in WebKit/chromium/features.gypi and it is not PLATFORM(QT). ../../JavaScriptCore/wtf/Platform.h:512:9: error: 'ENABLE_JSC_MULTIPLE_THREADS' macro redefined [-Werror] #define ENABLE_JSC_MULTIPLE_THREADS 1 <command line>:43:9: note: previous definition is here #define ENABLE_JSC_MULTIPLE_THREADS 0 1 error generated. * wtf/Platform.h: 2011-09-11 Sam Weinig <sam@webkit.org> Remove JSCell::isPropertyNameIterator(), it is unused https://bugs.webkit.org/show_bug.cgi?id=67911 Reviewed by Oliver Hunt. * runtime/JSCell.h: * runtime/JSPropertyNameIterator.h: 2011-09-11 Sam Weinig <sam@webkit.org> De-virtualize JSCell::isAPIValueWrapper https://bugs.webkit.org/show_bug.cgi?id=67909 Reviewed by Oliver Hunt. * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): Set the correct type on structure creation. * runtime/JSCell.h: Remove virtual keyword and default implementation. * runtime/JSType.h: Add type for APIValueWrapper. It must come after CompoundType since the APIValueWrapper has children in need of marking. * runtime/Structure.h: (JSC::JSCell::isAPIValueWrapper): Implement predicate using type info. 2011-09-10 Sam Weinig <sam@webkit.org> De-virtualize JSCell::isGetterSetter, type information is available for it https://bugs.webkit.org/show_bug.cgi?id=67902 Reviewed by Dan Bernstein. * runtime/GetterSetter.cpp: * runtime/GetterSetter.h: Remove override of isGetterSetter. * runtime/JSCell.cpp: * runtime/JSCell.h: De-virtualize and remove silly base implementation. * runtime/Structure.h: (JSC::JSCell::isGetterSetter): Use type info to determine getter-setter-hood. 2011-09-09 Oliver Hunt <oliver@apple.com> Remove support for anonymous storage from jsobjects https://bugs.webkit.org/show_bug.cgi?id=67881 Reviewed by Sam Weinig. Remove all use of anonymous slots, essentially a mechanical change in JavaScriptCore * API/JSCallbackConstructor.h: (JSC::JSCallbackConstructor::createStructure): * API/JSCallbackFunction.h: (JSC::JSCallbackFunction::createStructure): * API/JSCallbackObject.h: (JSC::JSCallbackObject::createStructure): * JavaScriptCore.exp: * debugger/DebuggerActivation.h: (JSC::DebuggerActivation::createStructure): * heap/MarkStack.cpp: (JSC::MarkStack::validateValue): * heap/MarkStack.h: * runtime/Arguments.h: (JSC::Arguments::createStructure): * runtime/ArrayConstructor.h: (JSC::ArrayConstructor::createStructure): * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::finishCreation): * runtime/ArrayPrototype.h: (JSC::ArrayPrototype::createStructure): * runtime/BooleanObject.h: (JSC::BooleanObject::createStructure): * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::BooleanPrototype): * runtime/BooleanPrototype.h: (JSC::BooleanPrototype::createStructure): * runtime/DateConstructor.h: (JSC::DateConstructor::createStructure): * runtime/DateInstance.h: (JSC::DateInstance::createStructure): * runtime/DatePrototype.cpp: (JSC::DatePrototype::DatePrototype): * runtime/DatePrototype.h: (JSC::DatePrototype::createStructure): * runtime/ErrorInstance.h: (JSC::ErrorInstance::createStructure): * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::finishCreation): * runtime/ErrorPrototype.h: (JSC::ErrorPrototype::createStructure): * runtime/ExceptionHelpers.h: (JSC::InterruptedExecutionError::createStructure): (JSC::TerminatedExecutionError::createStructure): * runtime/Executable.h: (JSC::ExecutableBase::createStructure): (JSC::NativeExecutable::createStructure): (JSC::EvalExecutable::createStructure): (JSC::ProgramExecutable::createStructure): (JSC::FunctionExecutable::createStructure): * runtime/FunctionPrototype.h: (JSC::FunctionPrototype::createStructure): * runtime/GetterSetter.h: (JSC::GetterSetter::createStructure): * runtime/InternalFunction.h: (JSC::InternalFunction::createStructure): * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): * runtime/JSActivation.h: (JSC::JSActivation::createStructure): * runtime/JSArray.h: (JSC::JSArray::createStructure): * runtime/JSByteArray.cpp: (JSC::JSByteArray::createStructure): * runtime/JSCell.h: * runtime/JSFunction.h: (JSC::JSFunction::createStructure): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::finishCreation): (JSC::JSGlobalObject::createStructure): * runtime/JSNotAnObject.h: (JSC::JSNotAnObject::createStructure): * runtime/JSONObject.h: (JSC::JSONObject::createStructure): * runtime/JSObject.h: (JSC::JSObject::createStructure): (JSC::JSNonFinalObject::createStructure): (JSC::JSFinalObject::createStructure): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::create): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::createStructure): * runtime/JSStaticScopeObject.h: (JSC::JSStaticScopeObject::createStructure): * runtime/JSString.h: (JSC::RopeBuilder::createStructure): * runtime/JSVariableObject.h: (JSC::JSVariableObject::createStructure): * runtime/JSWrapperObject.h: (JSC::JSWrapperObject::createStructure): * runtime/MathObject.h: (JSC::MathObject::createStructure): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::createStructure): * runtime/NumberConstructor.h: (JSC::NumberConstructor::createStructure): * runtime/NumberObject.h: (JSC::NumberObject::createStructure): * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::NumberPrototype): * runtime/NumberPrototype.h: (JSC::NumberPrototype::createStructure): * runtime/ObjectConstructor.h: (JSC::ObjectConstructor::createStructure): * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::finishCreation): * runtime/ObjectPrototype.h: (JSC::ObjectPrototype::createStructure): * runtime/RegExp.h: (JSC::RegExp::createStructure): * runtime/RegExpConstructor.h: (JSC::RegExpConstructor::createStructure): * runtime/RegExpObject.h: (JSC::RegExpObject::createStructure): * runtime/RegExpPrototype.h: (JSC::RegExpPrototype::createStructure): * runtime/ScopeChain.h: (JSC::ScopeChainNode::createStructure): * runtime/StrictEvalActivation.h: (JSC::StrictEvalActivation::createStructure): * runtime/StringConstructor.h: (JSC::StringConstructor::createStructure): * runtime/StringObject.h: (JSC::StringObject::createStructure): * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::createStructure): * runtime/StringPrototype.cpp: (JSC::StringPrototype::StringPrototype): * runtime/StringPrototype.h: (JSC::StringPrototype::createStructure): * runtime/Structure.cpp: (JSC::Structure::Structure): (JSC::Structure::materializePropertyMap): (JSC::Structure::addPropertyTransitionToExistingStructure): (JSC::Structure::addPropertyTransition): (JSC::Structure::removePropertyTransition): (JSC::Structure::changePrototypeTransition): (JSC::Structure::despecifyFunctionTransition): (JSC::Structure::getterSetterTransition): (JSC::Structure::toDictionaryTransition): (JSC::Structure::preventExtensionsTransition): (JSC::Structure::flattenDictionaryStructure): (JSC::Structure::addPropertyWithoutTransition): (JSC::Structure::removePropertyWithoutTransition): (JSC::Structure::get): (JSC::Structure::putSpecificValue): (JSC::Structure::remove): (JSC::Structure::checkConsistency): * runtime/Structure.h: (JSC::Structure::create): (JSC::Structure::propertyStorageSize): (JSC::Structure::get): * runtime/StructureChain.h: (JSC::StructureChain::createStructure): 2011-09-11 Jarred Nicholls <jarred@sencha.com> [Qt] Win32 build broken due to MachineStackMarker.cpp/.o failing to link against pthreads library https://bugs.webkit.org/show_bug.cgi?id=67864 Qt Win32 is not pthread compatible and cannot participate in multithreaded JSC or it fails to build. Reviewed by Csaba Osztrogonác. * wtf/Platform.h: 2011-09-11 Filip Pizlo <fpizlo@apple.com> ARM and MIPS assemblers still refer to executable pools. https://bugs.webkit.org/show_bug.cgi?id=67903 Reviewed by Csaba Osztrogonác. * assembler/ARMAssembler.cpp: (JSC::ARMAssembler::executableCopy): * assembler/ARMAssembler.h: * assembler/AssemblerBufferWithConstantPool.h: * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::executableCopy): 2011-09-08 Filip Pizlo <fpizlo@apple.com> The executable allocator makes it difficult to free individual chunks of executable memory https://bugs.webkit.org/show_bug.cgi?id=66363 Reviewed by Oliver Hunt. Introduced a best-fit, balanced-tree based allocator. The allocator required a balanced tree that does not allocate memory and that permits the removal of individual nodes directly (as opposed to by key); neither AVLTree nor WebCore's PODRedBlackTree supported this. Changed all references to executable code to use a reference counted handle. * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * assembler/AssemblerBuffer.h: (JSC::AssemblerBuffer::executableCopy): * assembler/LinkBuffer.h: (JSC::LinkBuffer::LinkBuffer): (JSC::LinkBuffer::finalizeCode): (JSC::LinkBuffer::linkCode): * assembler/MacroAssemblerCodeRef.h: (JSC::MacroAssemblerCodeRef::MacroAssemblerCodeRef): (JSC::MacroAssemblerCodeRef::createSelfManagedCodeRef): (JSC::MacroAssemblerCodeRef::executableMemory): (JSC::MacroAssemblerCodeRef::code): (JSC::MacroAssemblerCodeRef::size): (JSC::MacroAssemblerCodeRef::operator!): * assembler/X86Assembler.h: (JSC::X86Assembler::executableCopy): (JSC::X86Assembler::X86InstructionFormatter::executableCopy): * bytecode/CodeBlock.h: * bytecode/Instruction.h: * bytecode/StructureStubInfo.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compile): (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGRepatch.cpp: (JSC::DFG::generateProtoChainAccessStub): (JSC::DFG::tryCacheGetByID): (JSC::DFG::tryBuildGetByIDList): (JSC::DFG::tryBuildGetByIDProtoList): (JSC::DFG::tryCachePutByID): * jit/ExecutableAllocator.cpp: (JSC::ExecutableAllocator::initializeAllocator): (JSC::ExecutableAllocator::ExecutableAllocator): (JSC::ExecutableAllocator::allocate): (JSC::ExecutableAllocator::committedByteCount): (JSC::ExecutableAllocator::dumpProfile): * jit/ExecutableAllocator.h: (JSC::ExecutableAllocator::dumpProfile): * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::ExecutableAllocator::initializeAllocator): (JSC::ExecutableAllocator::ExecutableAllocator): (JSC::ExecutableAllocator::isValid): (JSC::ExecutableAllocator::underMemoryPressure): (JSC::ExecutableAllocator::allocate): (JSC::ExecutableAllocator::committedByteCount): (JSC::ExecutableAllocator::dumpProfile): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JIT.h: (JSC::JIT::compileCTIMachineTrampolines): (JSC::JIT::compileCTINativeCall): * jit/JITCode.h: (JSC::JITCode::operator !): (JSC::JITCode::addressForCall): (JSC::JITCode::offsetOf): (JSC::JITCode::execute): (JSC::JITCode::start): (JSC::JITCode::size): (JSC::JITCode::getExecutableMemory): (JSC::JITCode::HostFunction): (JSC::JITCode::JITCode): * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::privateCompileCTINativeCall): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::privateCompileCTINativeCall): * jit/JITPropertyAccess.cpp: (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::emitSlow_op_get_by_val): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::emitSlow_op_get_by_val): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): * jit/JITStubs.cpp: (JSC::JITThunks::JITThunks): (JSC::DEFINE_STUB_FUNCTION): (JSC::getPolymorphicAccessStructureListSlot): (JSC::JITThunks::ctiStub): (JSC::JITThunks::hostFunctionStub): * jit/JITStubs.h: * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::SpecializedThunkJIT): (JSC::SpecializedThunkJIT::finalize): * jit/ThunkGenerators.cpp: (JSC::charCodeAtThunkGenerator): (JSC::charAtThunkGenerator): (JSC::fromCharCodeThunkGenerator): (JSC::sqrtThunkGenerator): (JSC::floorThunkGenerator): (JSC::ceilThunkGenerator): (JSC::roundThunkGenerator): (JSC::expThunkGenerator): (JSC::logThunkGenerator): (JSC::absThunkGenerator): (JSC::powThunkGenerator): * jit/ThunkGenerators.h: * runtime/Executable.h: (JSC::NativeExecutable::create): * runtime/InitializeThreading.cpp: (JSC::initializeThreadingOnce): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::dumpSampleData): * runtime/JSGlobalData.h: (JSC::JSGlobalData::getCTIStub): * wtf/CMakeLists.txt: * wtf/MetaAllocator.cpp: Added. (WTF::MetaAllocatorHandle::MetaAllocatorHandle): (WTF::MetaAllocatorHandle::~MetaAllocatorHandle): (WTF::MetaAllocatorHandle::shrink): (WTF::MetaAllocator::MetaAllocator): (WTF::MetaAllocator::allocate): (WTF::MetaAllocator::currentStatistics): (WTF::MetaAllocator::findAndRemoveFreeSpace): (WTF::MetaAllocator::addFreeSpaceFromReleasedHandle): (WTF::MetaAllocator::addFreshFreeSpace): (WTF::MetaAllocator::debugFreeSpaceSize): (WTF::MetaAllocator::addFreeSpace): (WTF::MetaAllocator::incrementPageOccupancy): (WTF::MetaAllocator::decrementPageOccupancy): (WTF::MetaAllocator::roundUp): (WTF::MetaAllocator::allocFreeSpaceNode): (WTF::MetaAllocator::freeFreeSpaceNode): (WTF::MetaAllocator::dumpProfile): * wtf/MetaAllocator.h: Added. (WTF::MetaAllocator::bytesAllocated): (WTF::MetaAllocator::bytesReserved): (WTF::MetaAllocator::bytesCommitted): (WTF::MetaAllocator::dumpProfile): (WTF::MetaAllocator::~MetaAllocator): * wtf/MetaAllocatorHandle.h: Added. * wtf/RedBlackTree.h: Added. (WTF::RedBlackTree::Node::Node): (WTF::RedBlackTree::Node::successor): (WTF::RedBlackTree::Node::predecessor): (WTF::RedBlackTree::Node::reset): (WTF::RedBlackTree::Node::parent): (WTF::RedBlackTree::Node::setParent): (WTF::RedBlackTree::Node::left): (WTF::RedBlackTree::Node::setLeft): (WTF::RedBlackTree::Node::right): (WTF::RedBlackTree::Node::setRight): (WTF::RedBlackTree::Node::color): (WTF::RedBlackTree::Node::setColor): (WTF::RedBlackTree::RedBlackTree): (WTF::RedBlackTree::insert): (WTF::RedBlackTree::remove): (WTF::RedBlackTree::findExact): (WTF::RedBlackTree::findLeastGreaterThanOrEqual): (WTF::RedBlackTree::findGreatestLessThanOrEqual): (WTF::RedBlackTree::first): (WTF::RedBlackTree::last): (WTF::RedBlackTree::size): (WTF::RedBlackTree::isEmpty): (WTF::RedBlackTree::treeMinimum): (WTF::RedBlackTree::treeMaximum): (WTF::RedBlackTree::treeInsert): (WTF::RedBlackTree::leftRotate): (WTF::RedBlackTree::rightRotate): (WTF::RedBlackTree::removeFixup): * wtf/wtf.pri: * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::compile): * yarr/YarrJIT.h: (JSC::Yarr::YarrCodeBlock::execute): (JSC::Yarr::YarrCodeBlock::getAddr): 2011-09-10 Sam Weinig <sam@webkit.org> Remove JSC::isZombie() function, it did nothing and was called by no-one. https://bugs.webkit.org/show_bug.cgi?id=67901 Reviewed by Andy Estes. * JavaScriptCore.exp: * runtime/JSCell.cpp: * runtime/JSValue.h: 2011-09-10 Sam Weinig <sam@webkit.org> Add isInterruptedExecutionException and isTerminatedExecutionException predicates https://bugs.webkit.org/show_bug.cgi?id=67892 Reviewed by Andy "First Time Reviewer" Estes. * JavaScriptCore.exp: Add symbols. * interpreter/Interpreter.cpp: (JSC::Interpreter::throwException): Use new predicates. * runtime/ExceptionHelpers.cpp: (JSC::createInterruptedExecutionException): (JSC::isInterruptedExecutionException): (JSC::createTerminatedExecutionException): (JSC::isTerminatedExecutionException): * runtime/ExceptionHelpers.h: (JSC::InterruptedExecutionError::InterruptedExecutionError): Add predicates. 2011-09-10 Filip Pizlo <fpizlo@apple.com> DFG JIT completely undoes speculative compilation even in the case of a partial static speculation failure https://bugs.webkit.org/show_bug.cgi?id=67798 Reviewed by Geoffrey Garen. This is a regression with static speculation, so it is turned off by default. But it is a necessary prerequisite for further work on dynamic speculation. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::clearGenerationInfo): * dfg/DFGJITCodeGenerator.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution): 2011-09-09 Chris Marrin <cmarrin@apple.com> requestAnimationFrame doesn't throttle on Mac https://bugs.webkit.org/show_bug.cgi?id=67171 Reviewed by Simon Fraser. Added WTF_USE_REQUEST_ANIMATION_FRAME_TIMER to allow any platform to run requestAnimationFrame callbacks on a Timer defined in ScriptedAnimationController. Currently only enabled for PLATFORM(MAC) * wtf/Platform.h: 2011-09-09 Geoffrey Garen <ggaren@apple.com> Reviewed by Dan Bernstein. Removed ENABLE(SINGLE_THREADED) support, since it is always false https://bugs.webkit.org/show_bug.cgi?id=67862 Next step toward making the baseline platform assumption that threads exist. * wtf/wtf.pri: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: Removed references to ThreadingNone.cpp, which was only compiled in single-threaded mode. * wtf/Platform.h: * wtf/ThreadSpecific.h: (WTF::::destroy): * wtf/qt/ThreadingQt.cpp: Removed now-dead code. * wtf/ThreadingNone.cpp: Removed. 2011-09-09 Mark Hahnenberg <mhahnenberg@apple.com> Unzip initialization lists and constructors in JSCell hierarchy (5/7) https://bugs.webkit.org/show_bug.cgi?id=67420 Reviewed by Geoffrey Garen. Completed the fifth level of the refactoring to add finishCreation() methods to all classes within the JSCell hierarchy with non-trivial constructor bodies. This primarily consists of pushing the calls to finishCreation() down into the constructors of the subclasses of the second level of the hierarchy as well as pulling the finishCreation() calls out into the class's corresponding create() method if it has one. Doing both simultaneously allows us to maintain the invariant that the finishCreation() method chain is called exactly once during the creation of an object, since calling it any other number of times (0, 2, or more) will cause an assertion failure. * API/JSCallbackConstructor.cpp: (JSC::JSCallbackConstructor::JSCallbackConstructor): * API/JSCallbackConstructor.h: (JSC::JSCallbackConstructor::create): * API/JSCallbackFunction.cpp: (JSC::JSCallbackFunction::JSCallbackFunction): (JSC::JSCallbackFunction::finishCreation): * API/JSCallbackFunction.h: * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::JSCallbackObject): (JSC::::finishCreation): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: * debugger/DebuggerActivation.h: (JSC::DebuggerActivation::create): * jsc.cpp: (GlobalObject::finishCreation): (GlobalObject::GlobalObject): * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::ArrayConstructor): (JSC::ArrayConstructor::finishCreation): * runtime/ArrayConstructor.h: * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::ArrayPrototype): * runtime/ArrayPrototype.h: (JSC::ArrayPrototype::create): * runtime/BooleanConstructor.cpp: (JSC::BooleanConstructor::BooleanConstructor): (JSC::BooleanConstructor::finishCreation): * runtime/BooleanConstructor.h: * runtime/BooleanObject.cpp: (JSC::BooleanObject::BooleanObject): * runtime/BooleanObject.h: (JSC::BooleanObject::create): * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::BooleanPrototype): (JSC::BooleanPrototype::finishCreation): * runtime/BooleanPrototype.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::DateConstructor): (JSC::DateConstructor::finishCreation): * runtime/DateConstructor.h: * runtime/DateInstance.cpp: (JSC::DateInstance::DateInstance): * runtime/DateInstance.h: (JSC::DateInstance::create): * runtime/DatePrototype.cpp: (JSC::DatePrototype::DatePrototype): (JSC::DatePrototype::finishCreation): * runtime/DatePrototype.h: * runtime/Error.cpp: (JSC::StrictModeTypeErrorFunction::StrictModeTypeErrorFunction): * runtime/ErrorConstructor.cpp: (JSC::ErrorConstructor::ErrorConstructor): (JSC::ErrorConstructor::finishCreation): * runtime/ErrorConstructor.h: * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::ErrorPrototype): * runtime/ErrorPrototype.h: (JSC::ErrorPrototype::create): * runtime/FunctionConstructor.cpp: (JSC::FunctionConstructor::FunctionConstructor): (JSC::FunctionConstructor::finishCreation): * runtime/FunctionConstructor.h: * runtime/FunctionPrototype.cpp: (JSC::FunctionPrototype::FunctionPrototype): (JSC::FunctionPrototype::finishCreation): * runtime/FunctionPrototype.h: * runtime/InternalFunction.cpp: (JSC::InternalFunction::InternalFunction): * runtime/InternalFunction.h: * runtime/JSActivation.cpp: (JSC::JSActivation::JSActivation): * runtime/JSActivation.h: (JSC::JSActivation::create): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::create): (JSC::JSGlobalObject::JSGlobalObject): * runtime/JSONObject.cpp: (JSC::JSONObject::JSONObject): * runtime/JSONObject.h: (JSC::JSONObject::create): * runtime/JSStaticScopeObject.h: (JSC::JSStaticScopeObject::create): (JSC::JSStaticScopeObject::JSStaticScopeObject): * runtime/JSString.cpp: (JSC::StringObject::create): * runtime/MathObject.cpp: (JSC::MathObject::MathObject): * runtime/MathObject.h: (JSC::MathObject::create): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::NativeErrorConstructor): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::finishCreation): * runtime/NativeErrorPrototype.cpp: (JSC::NativeErrorPrototype::NativeErrorPrototype): (JSC::NativeErrorPrototype::finishCreation): * runtime/NativeErrorPrototype.h: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::NumberConstructor): (JSC::NumberConstructor::finishCreation): * runtime/NumberConstructor.h: * runtime/NumberObject.cpp: (JSC::NumberObject::NumberObject): * runtime/NumberObject.h: (JSC::NumberObject::create): * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::NumberPrototype): (JSC::NumberPrototype::finishCreation): * runtime/NumberPrototype.h: * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::ObjectConstructor): (JSC::ObjectConstructor::finishCreation): * runtime/ObjectConstructor.h: * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::RegExpConstructor): (JSC::RegExpConstructor::finishCreation): (JSC::RegExpMatchesArray::RegExpMatchesArray): * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::create): * runtime/RegExpObject.cpp: (JSC::RegExpObject::RegExpObject): * runtime/RegExpObject.h: (JSC::RegExpObject::create): * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::RegExpPrototype): * runtime/StringConstructor.cpp: (JSC::StringConstructor::StringConstructor): (JSC::StringConstructor::finishCreation): * runtime/StringConstructor.h: * runtime/StringObject.cpp: (JSC::StringObject::StringObject): * runtime/StringObject.h: (JSC::StringObject::create): * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::StringObjectThatMasqueradesAsUndefined): * runtime/StringPrototype.cpp: (JSC::StringPrototype::StringPrototype): (JSC::StringPrototype::finishCreation): * runtime/StringPrototype.h: 2011-09-09 Geoffrey Garen <ggaren@apple.com> Build fix: Guard against double-#define for something already #defined by the build system. * wtf/Platform.h: 2011-09-09 Geoffrey Garen <ggaren@apple.com> Reviewed by Dan Bernstein. Never #define ENABLE_SINGLE_THREADED, !ENABLE_JSC_MULTIPLE_THREADS, or !ENABLE_WTF_MULTIPLE_THREADS https://bugs.webkit.org/show_bug.cgi?id=67860 First step toward making the baseline platform assumption that threads exist: Never #define ENABLE_SINGLE_THREADED, !ENABLE_JSC_MULTIPLE_THREADS, or !ENABLE_WTF_MULTIPLE_THREADS. * wtf/Platform.h: 2011-09-09 Laszlo Gombos <laszlo.1.gombos@nokia.com> [Qt] Remove common.pri https://bugs.webkit.org/show_bug.cgi?id=67814 Reviewed by Andreas Kling. * JavaScriptCore.pri: 2011-09-08 Mark Hahnenberg <mhahnenberg@apple.com> REGRESSION(r94811): Assertion failure in 2 worker tests https://bugs.webkit.org/show_bug.cgi?id=67829 Reviewed by Sam Weinig. Fixing a couple tests that were broken due to the wrong values being set in the parent class pointers in the ClassInfo structs for TerminatedExecutionError and InterruptedExecutionError. * runtime/ExceptionHelpers.cpp: 2011-09-08 Oliver Hunt <oliver@apple.com> Use bump allocator for initial property storage https://bugs.webkit.org/show_bug.cgi?id=67494 Reviewed by Geoffrey Garen. Use a bump allocator for initial allocation of property storage, and promote to fastMalloc memory only if it survives a GC pass. Comes out as a 1% win on v8, and is a useful step on the way to GC allocation of all property storage. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.xcodeproj/project.pbxproj: * heap/Heap.cpp: (JSC::Heap::collect): * heap/Heap.h: (JSC::Heap::allocatePropertyStorage): (JSC::Heap::inPropertyStorageNursery): * heap/MarkedBlock.h: * heap/NewSpace.cpp: (JSC::NewSpace::NewSpace): * heap/NewSpace.h: (JSC::NewSpace::resetPropertyStorageNursery): (JSC::NewSpace::allocatePropertyStorage): (JSC::NewSpace::inPropertyStorageNursery): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/JSObject.cpp: (JSC::JSObject::allocatePropertyStorage): * runtime/JSObject.h: (JSC::JSObject::isUsingInlineStorage): (JSC::JSObject::JSObject): (JSC::JSObject::propertyStorage): (JSC::JSObject::~JSObject): (JSC::JSObject::putDirectInternal): (JSC::JSObject::putDirectWithoutTransition): (JSC::JSObject::putDirectFunctionWithoutTransition): (JSC::JSObject::transitionTo): (JSC::JSObject::visitChildrenDirect): * runtime/StorageBarrier.h: Added. (JSC::StorageBarrier::StorageBarrier): (JSC::StorageBarrier::set): (JSC::StorageBarrier::operator->): (JSC::StorageBarrier::operator*): (JSC::StorageBarrier::operator[]): (JSC::StorageBarrier::get): 2011-09-08 Sam Weinig <sam@webkit.org> Remove the Completion object from JSC, I have never liked it https://bugs.webkit.org/show_bug.cgi?id=67755 Reviewed by Gavin Barraclough. - Removes the Completion object and replaces its use with out parameter exceptions. - Remove ComplType and virtual exceptionType() function on JSObject. Replace with ClassInfo for InterruptedExecutionError and TerminatedExecutionError. * API/JSBase.cpp: (JSEvaluateScript): (JSCheckScriptSyntax): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * interpreter/Interpreter.cpp: (JSC::Interpreter::throwException): * jsc.cpp: (functionLoad): (functionCheckSyntax): (runWithScripts): (runInteractive): * runtime/Completion.cpp: (JSC::checkSyntax): (JSC::evaluate): * runtime/Completion.h: * runtime/ExceptionHelpers.cpp: (JSC::InterruptedExecutionError::toString): (JSC::TerminatedExecutionError::toString): (JSC::createInterruptedExecutionException): * runtime/ExceptionHelpers.h: (JSC::InterruptedExecutionError::InterruptedExecutionError): (JSC::InterruptedExecutionError::create): (JSC::InterruptedExecutionError::createStructure): (JSC::TerminatedExecutionError::TerminatedExecutionError): (JSC::TerminatedExecutionError::create): (JSC::TerminatedExecutionError::createStructure): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSObject.h: 2011-09-08 Ryosuke Niwa <rniwa@webkit.org> Build fix. * dfg/DFGCapabilities.cpp: 2011-09-08 Filip Pizlo <fpizlo@apple.com> Value profling and execution count profiling is performed even for code that cannot be optimized https://bugs.webkit.org/show_bug.cgi?id=67694 Reviewed by Gavin Barraclough. This is a 2% speed-up on V8 when tiered compilation is enabled. * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.cpp: (JSC::ProgramCodeBlock::canCompileWithDFG): (JSC::EvalCodeBlock::canCompileWithDFG): (JSC::FunctionCodeBlock::canCompileWithDFG): * bytecode/CodeBlock.h: * dfg/DFGCapabilities.cpp: Added. (JSC::DFG::canCompileOpcodes): * dfg/DFGCapabilities.h: Added. (JSC::DFG::mightCompileEval): (JSC::DFG::mightCompileProgram): (JSC::DFG::mightCompileFunctionForCall): (JSC::DFG::mightCompileFunctionForConstruct): (JSC::DFG::canCompileOpcode): (JSC::DFG::canCompileEval): (JSC::DFG::canCompileProgram): (JSC::DFG::canCompileFunctionForCall): (JSC::DFG::canCompileFunctionForConstruct): * jit/JIT.cpp: (JSC::JIT::emitOptimizationCheck): (JSC::JIT::privateCompile): * jit/JIT.h: (JSC::JIT::shouldEmitProfiling): * jit/JITInlineMethods.h: (JSC::JIT::emitValueProfilingSite): 2011-09-08 Filip Pizlo <fpizlo@apple.com> DFG speculative JIT does not initialize integer tags for PredictInt32 temporaries https://bugs.webkit.org/show_bug.cgi?id=67840 Reviewed by Gavin Barraclough. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::initializeVariableTypes): 2011-09-08 Thouraya ANDOLSI <thouraya.andolsi@st.com> https://bugs.webkit.org/show_bug.cgi?id=67771 Fix sequenceGetByIdSlowCaseInstructionSpace, sequenceGetByIdSlowCaseConstantSpace and patchOffsetGetByIdSlowCaseCall and enables DOUBLE_CONVERSION_CORRECT_DOUBLE_OPERATIONS flag for SH4 platforms. Reviewed by Gavin Barraclough. * jit/JIT.h: * wtf/dtoa/utils.h: 2011-09-08 Mark Hahnenberg <mhahnenberg@apple.com> Remove getUInt32 from JSCell https://bugs.webkit.org/show_bug.cgi?id=67691 Reviewed by Oliver Hunt. We don't use JSCell::getUInt32 anymore, so it has been removed. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSCell.cpp: * runtime/JSCell.h: 2011-09-07 Filip Pizlo <fpizlo@apple.com> PPC build fix. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::~CodeBlock): 2011-09-07 Oliver Hunt <oliver@apple.com> Release mode build fix. * API/JSCallbackObject.h: (JSC::JSCallbackObject::create): 2011-09-06 Oliver Hunt <oliver@apple.com> Remove JSObjectWithGlobalObject https://bugs.webkit.org/show_bug.cgi?id=67689 Reviewed by Geoff Garen. Remove JSObjectWithGlobalObject, and update code to stop using anonymous storage to access the global object that a JSObject comes from. Largely mechanical change to remove the use of anonymous storage and JSObjectWithGlobalObject. * API/JSCallbackConstructor.cpp: (JSC::JSCallbackConstructor::JSCallbackConstructor): (JSC::JSCallbackConstructor::finishCreation): * API/JSCallbackConstructor.h: * API/JSCallbackObject.cpp: * API/JSCallbackObject.h: (JSC::JSCallbackObject::create): * API/JSCallbackObjectFunctions.h: (JSC::::JSCallbackObject): (JSC::::finishCreation): (JSC::::staticFunctionGetter): * API/JSClassRef.cpp: (OpaqueJSClass::prototype): * API/JSObjectRef.cpp: (JSObjectMake): (JSObjectGetPrivate): (JSObjectSetPrivate): (JSObjectGetPrivateProperty): (JSObjectSetPrivateProperty): (JSObjectDeletePrivateProperty): * API/JSValueRef.cpp: (JSValueIsObjectOfClass): * API/JSWeakObjectMapRefPrivate.cpp: * JavaScriptCore.exp: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.h: * dfg/DFGRepatch.cpp: (JSC::DFG::dfgRepatchGetMethodFast): (JSC::DFG::tryCacheGetMethod): * jit/JIT.h: * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateJSFunction): * jit/JITPropertyAccess.cpp: (JSC::JIT::patchMethodCallProto): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/DatePrototype.cpp: * runtime/InternalFunction.cpp: (JSC::InternalFunction::InternalFunction): (JSC::InternalFunction::finishCreation): * runtime/InternalFunction.h: * runtime/JSFunction.cpp: (JSC::JSFunction::JSFunction): (JSC::JSFunction::finishCreation): * runtime/JSFunction.h: (JSC::JSFunction::create): (JSC::JSFunction::createStructure): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): * runtime/JSONObject.cpp: (JSC::JSONObject::JSONObject): (JSC::JSONObject::finishCreation): * runtime/JSONObject.h: * runtime/JSObject.h: (JSC::JSObject::globalObject): * runtime/JSObjectWithGlobalObject.cpp: Removed. * runtime/JSObjectWithGlobalObject.h: Removed. * runtime/JSValue.cpp: (JSC::JSValue::isValidCallee): * runtime/Lookup.cpp: (JSC::setUpStaticFunctionSlot): * runtime/Lookup.h: * runtime/MathObject.cpp: (JSC::MathObject::MathObject): (JSC::MathObject::finishCreation): * runtime/MathObject.h: * runtime/NumberPrototype.cpp: * runtime/RegExpObject.cpp: (JSC::RegExpObject::RegExpObject): (JSC::RegExpObject::finishCreation): * runtime/RegExpObject.h: * runtime/Structure.cpp: (JSC::Structure::Structure): * runtime/Structure.h: (JSC::Structure::create): (JSC::Structure::globalObject): 2011-09-07 Gavin Barraclough <barraclough@apple.com> Refactor JIT checks for ObjectType into helper functions. Rubber stamped by Sam Weinig. * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::branchIfNotObject): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JIT.h: * jit/JITCall32_64.cpp: (JSC::JIT::emit_op_ret_object_or_this): * jit/JITInlineMethods.h: (JSC::JIT::emitJumpIfNotObject): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_instanceof): (JSC::JIT::emit_op_ret_object_or_this): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_create_this): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_instanceof): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_create_this): 2011-09-07 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r94627 and r94632. http://trac.webkit.org/changeset/94627 http://trac.webkit.org/changeset/94632 https://bugs.webkit.org/show_bug.cgi?id=67698 It broke tests on GTK and Qt (Requested by Ossy on #webkit). * API/JSCallbackConstructor.cpp: (JSC::JSCallbackConstructor::JSCallbackConstructor): * API/JSCallbackConstructor.h: (JSC::JSCallbackConstructor::create): * API/JSCallbackFunction.cpp: (JSC::JSCallbackFunction::JSCallbackFunction): * API/JSCallbackFunction.h: * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::create): * debugger/DebuggerActivation.h: * jsc.cpp: (GlobalObject::constructorBody): (GlobalObject::GlobalObject): * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::ArrayConstructor): * runtime/ArrayConstructor.h: * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::ArrayPrototype): * runtime/ArrayPrototype.h: (JSC::ArrayPrototype::create): * runtime/BooleanConstructor.cpp: (JSC::BooleanConstructor::BooleanConstructor): * runtime/BooleanConstructor.h: * runtime/BooleanObject.cpp: (JSC::BooleanObject::BooleanObject): * runtime/BooleanObject.h: (JSC::BooleanObject::create): * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::BooleanPrototype): * runtime/BooleanPrototype.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::DateConstructor): * runtime/DateConstructor.h: * runtime/DateInstance.cpp: (JSC::DateInstance::DateInstance): * runtime/DateInstance.h: (JSC::DateInstance::create): * runtime/DatePrototype.cpp: (JSC::DatePrototype::DatePrototype): * runtime/DatePrototype.h: * runtime/Error.cpp: (JSC::StrictModeTypeErrorFunction::StrictModeTypeErrorFunction): * runtime/ErrorConstructor.cpp: (JSC::ErrorConstructor::ErrorConstructor): * runtime/ErrorConstructor.h: (JSC::ErrorConstructor::create): * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::ErrorPrototype): * runtime/ErrorPrototype.h: (JSC::ErrorPrototype::create): * runtime/FunctionConstructor.cpp: (JSC::FunctionConstructor::FunctionConstructor): * runtime/FunctionConstructor.h: * runtime/FunctionPrototype.cpp: (JSC::FunctionPrototype::FunctionPrototype): * runtime/FunctionPrototype.h: * runtime/InternalFunction.cpp: (JSC::InternalFunction::InternalFunction): * runtime/InternalFunction.h: * runtime/JSActivation.cpp: (JSC::JSActivation::JSActivation): * runtime/JSActivation.h: (JSC::JSActivation::create): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::create): (JSC::JSGlobalObject::JSGlobalObject): * runtime/JSONObject.cpp: (JSC::JSONObject::JSONObject): * runtime/JSONObject.h: (JSC::JSONObject::create): * runtime/JSStaticScopeObject.h: (JSC::JSStaticScopeObject::create): (JSC::JSStaticScopeObject::JSStaticScopeObject): * runtime/JSString.cpp: (JSC::StringObject::create): * runtime/MathObject.cpp: (JSC::MathObject::MathObject): * runtime/MathObject.h: (JSC::MathObject::create): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::NativeErrorConstructor): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::constructorBody): * runtime/NativeErrorPrototype.cpp: (JSC::NativeErrorPrototype::NativeErrorPrototype): (JSC::NativeErrorPrototype::constructorBody): * runtime/NativeErrorPrototype.h: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::NumberConstructor): * runtime/NumberConstructor.h: * runtime/NumberObject.cpp: (JSC::NumberObject::NumberObject): * runtime/NumberObject.h: (JSC::NumberObject::create): * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::NumberPrototype): * runtime/NumberPrototype.h: * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::ObjectConstructor): * runtime/ObjectConstructor.h: * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::RegExpConstructor): (JSC::RegExpMatchesArray::RegExpMatchesArray): * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::create): * runtime/RegExpObject.cpp: (JSC::RegExpObject::RegExpObject): * runtime/RegExpObject.h: (JSC::RegExpObject::create): * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::RegExpPrototype): * runtime/StringConstructor.cpp: (JSC::StringConstructor::StringConstructor): * runtime/StringConstructor.h: * runtime/StringObject.cpp: (JSC::StringObject::StringObject): * runtime/StringObject.h: (JSC::StringObject::create): * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::StringObjectThatMasqueradesAsUndefined): * runtime/StringPrototype.cpp: (JSC::StringPrototype::StringPrototype): * runtime/StringPrototype.h: 2011-09-06 Xianzhu Wang <wangxianzhu@chromium.org> Replace usages of Vector<UChar> with existing StringBuilder https://bugs.webkit.org/show_bug.cgi?id=67079 Reviewed by Gavin Barraclough. This is part of work to support 8-bit string buffers. Adds StringBuilder::characters() because the original Vector<UChar>::data() is widely used. Sets the minimum size of buffer to 16 to prevent possible performance regression. Further performance investigation should be done in https://bugs.webkit.org/show_bug.cgi?id=67084. * wtf/Forward.h: * wtf/text/StringBuilder.cpp: (WTF::StringBuilder::appendUninitialized): Sets minimum buffer size to 16 bytes. * wtf/text/StringBuilder.h: (WTF::StringBuilder::operator[]): (WTF::StringBuilder::characters): Added. 2011-09-06 Mark Hahnenberg <mhahnenberg@apple.com> Fix broken snow leopard build https://bugs.webkit.org/show_bug.cgi?id=67693 Reviewed by Daniel Bates. Removed unnecessary symbol export. * JavaScriptCore.exp: 2011-09-06 Filip Pizlo <fpizlo@apple.com> DFG JIT does not optimize booleans https://bugs.webkit.org/show_bug.cgi?id=67670 Reviewed by Gavin Barraclough. This adds boolean value profiling, boolean prediction in the DFG, boolean forward flow propagation in the DFGPropagator, boolean data format in DFG generation info, and comprehensive optimizations based on both boolean prediction and boolean generation info. This is brings the speed-up on v8-richards to 12%, and gives slight speed-ups elsewhere as well. Making this work right required navigating some subtleties in value profiling. Some functions get compiled with insufficient information because some important path of the function never executed. In these cases, we wish to fall back on static speculation. But to do so, we need to ensure that predictions that are inherent in the code (like that GetById almost certainly takes a cell operand) are reflected in predictions that we make in DFGPropagator. Thus, DFGPropagator now does both backward and forward flow, using a both forward and backward fixpoint. The backward flow in DFGPropagator is a separate static analysis, and needs to keep a set of backward flow abstract values for variables, arguments, and globals. To make this easy, this patch factors out DFGGraph's prediction tracking capability into DFGPredictionTracker, which now gets used by both DFGGraph (for forward flow predictions) and DFGPropagator (for backward flow predictions). Backward flow predictions eventually get merged into forward flow ones, but the two are not equivalent: a forward flow prediction is a superset of the backward flow prediction. Debugging these prediction issues required a better understanding of where we fail speculation, and what our value predictions look like. This patch also adds optional verbose speculation failure (so an informative printf fires whenever speculation failure occurs) and slight improvements to the verbosity in other places. * bytecode/ValueProfile.h: (JSC::ValueProfile::numberOfBooleans): (JSC::ValueProfile::probabilityOfBoolean): (JSC::ValueProfile::dump): (JSC::ValueProfile::computeStatistics): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::stronglyPredict): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGenerationInfo.h: (JSC::DFG::dataFormatToString): (JSC::DFG::needDataFormatConversion): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGGraph.h: (JSC::DFG::Graph::Graph): (JSC::DFG::Graph::predictions): (JSC::DFG::Graph::predict): (JSC::DFG::Graph::predictGlobalVar): (JSC::DFG::Graph::getPrediction): (JSC::DFG::Graph::getGlobalVarPrediction): (JSC::DFG::Graph::isBooleanConstant): (JSC::DFG::Graph::valueOfBooleanConstant): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillInteger): (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::isKnownNotInteger): (JSC::DFG::JITCodeGenerator::isKnownBoolean): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompareNull): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeStrictEq): (JSC::DFG::JITCodeGenerator::emitBranch): (JSC::DFG::JITCodeGenerator::speculationCheck): (JSC::DFG::GPRTemporary::GPRTemporary): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::isBooleanConstant): (JSC::DFG::JITCodeGenerator::valueOfBooleanConstant): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::debugCall): (JSC::DFG::JITCompiler::isBooleanConstant): (JSC::DFG::JITCompiler::valueOfBooleanConstant): * dfg/DFGNode.h: (JSC::DFG::isBooleanPrediction): (JSC::DFG::predictionToString): (JSC::DFG::mergePredictions): (JSC::DFG::makePrediction): (JSC::DFG::Node::isBooleanConstant): (JSC::DFG::Node::valueOfBooleanConstant): (JSC::DFG::Node::hasBooleanResult): (JSC::DFG::Node::hasNumericResult): (JSC::DFG::Node::predict): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGPredictionTracker.h: Added. (JSC::DFG::operandIsArgument): (JSC::DFG::PredictionSlot::PredictionSlot): (JSC::DFG::PredictionTracker::PredictionTracker): (JSC::DFG::PredictionTracker::initializeSimilarTo): (JSC::DFG::PredictionTracker::numberOfArguments): (JSC::DFG::PredictionTracker::numberOfVariables): (JSC::DFG::PredictionTracker::argumentOffsetForOperand): (JSC::DFG::PredictionTracker::predictArgument): (JSC::DFG::PredictionTracker::predict): (JSC::DFG::PredictionTracker::predictGlobalVar): (JSC::DFG::PredictionTracker::getArgumentPrediction): (JSC::DFG::PredictionTracker::getPrediction): (JSC::DFG::PredictionTracker::getGlobalVarPrediction): * dfg/DFGPropagator.cpp: (JSC::DFG::Propagator::Propagator): (JSC::DFG::Propagator::fixpoint): (JSC::DFG::Propagator::setPrediction): (JSC::DFG::Propagator::mergeUse): (JSC::DFG::Propagator::mergePrediction): (JSC::DFG::Propagator::propagateNode): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::fillSpeculateBoolean): (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculateBooleanOperand::SpeculateBooleanOperand): (JSC::DFG::SpeculateBooleanOperand::~SpeculateBooleanOperand): (JSC::DFG::SpeculateBooleanOperand::index): (JSC::DFG::SpeculateBooleanOperand::gpr): (JSC::DFG::SpeculateBooleanOperand::use): * runtime/JSGlobalData.h: * runtime/JSValue.cpp: (JSC::JSValue::description): 2011-09-06 Mark Hahnenberg <mhahnenberg@apple.com> Unzip initialization lists and constructors in JSCell hierarchy (5/7) https://bugs.webkit.org/show_bug.cgi?id=67420 Reviewed by Geoffrey Garen. Completed the fifth level of the refactoring to add finishCreation() methods to all classes within the JSCell hierarchy with non-trivial constructor bodies. This primarily consists of pushing the calls to finishCreation() down into the constructors of the subclasses of the second level of the hierarchy as well as pulling the finishCreation() calls out into the class's corresponding create() method if it has one. Doing both simultaneously allows us to maintain the invariant that the finishCreation() method chain is called exactly once during the creation of an object, since calling it any other number of times (0, 2, or more) will cause an assertion failure. * API/JSCallbackConstructor.cpp: (JSC::JSCallbackConstructor::JSCallbackConstructor): * API/JSCallbackConstructor.h: (JSC::JSCallbackConstructor::create): * API/JSCallbackFunction.cpp: (JSC::JSCallbackFunction::JSCallbackFunction): (JSC::JSCallbackFunction::finishCreation): * API/JSCallbackFunction.h: * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: * debugger/DebuggerActivation.h: (JSC::DebuggerActivation::create): * jsc.cpp: (GlobalObject::finishCreation): (GlobalObject::GlobalObject): * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::ArrayConstructor): (JSC::ArrayConstructor::finishCreation): * runtime/ArrayConstructor.h: * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::ArrayPrototype): * runtime/ArrayPrototype.h: (JSC::ArrayPrototype::create): * runtime/BooleanConstructor.cpp: (JSC::BooleanConstructor::BooleanConstructor): (JSC::BooleanConstructor::finishCreation): * runtime/BooleanConstructor.h: * runtime/BooleanObject.cpp: (JSC::BooleanObject::BooleanObject): * runtime/BooleanObject.h: (JSC::BooleanObject::create): * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::BooleanPrototype): (JSC::BooleanPrototype::finishCreation): * runtime/BooleanPrototype.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::DateConstructor): (JSC::DateConstructor::finishCreation): * runtime/DateConstructor.h: * runtime/DateInstance.cpp: (JSC::DateInstance::DateInstance): * runtime/DateInstance.h: (JSC::DateInstance::create): * runtime/DatePrototype.cpp: (JSC::DatePrototype::DatePrototype): (JSC::DatePrototype::finishCreation): * runtime/DatePrototype.h: * runtime/Error.cpp: (JSC::StrictModeTypeErrorFunction::StrictModeTypeErrorFunction): * runtime/ErrorConstructor.cpp: (JSC::ErrorConstructor::ErrorConstructor): (JSC::ErrorConstructor::finishCreation): * runtime/ErrorConstructor.h: * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::ErrorPrototype): * runtime/ErrorPrototype.h: (JSC::ErrorPrototype::create): * runtime/FunctionConstructor.cpp: (JSC::FunctionConstructor::FunctionConstructor): (JSC::FunctionConstructor::finishCreation): * runtime/FunctionConstructor.h: * runtime/FunctionPrototype.cpp: (JSC::FunctionPrototype::FunctionPrototype): (JSC::FunctionPrototype::finishCreation): * runtime/FunctionPrototype.h: * runtime/InternalFunction.cpp: (JSC::InternalFunction::InternalFunction): * runtime/InternalFunction.h: * runtime/JSActivation.cpp: (JSC::JSActivation::JSActivation): * runtime/JSActivation.h: (JSC::JSActivation::create): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::create): (JSC::JSGlobalObject::JSGlobalObject): * runtime/JSONObject.cpp: (JSC::JSONObject::JSONObject): * runtime/JSONObject.h: (JSC::JSONObject::create): * runtime/JSStaticScopeObject.h: (JSC::JSStaticScopeObject::create): (JSC::JSStaticScopeObject::JSStaticScopeObject): * runtime/JSString.cpp: (JSC::StringObject::create): * runtime/MathObject.cpp: (JSC::MathObject::MathObject): * runtime/MathObject.h: (JSC::MathObject::create): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::NativeErrorConstructor): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::finishCreation): * runtime/NativeErrorPrototype.cpp: (JSC::NativeErrorPrototype::NativeErrorPrototype): (JSC::NativeErrorPrototype::finishCreation): * runtime/NativeErrorPrototype.h: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::NumberConstructor): (JSC::NumberConstructor::finishCreation): * runtime/NumberConstructor.h: * runtime/NumberObject.cpp: (JSC::NumberObject::NumberObject): * runtime/NumberObject.h: (JSC::NumberObject::create): * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::NumberPrototype): (JSC::NumberPrototype::finishCreation): * runtime/NumberPrototype.h: * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::ObjectConstructor): (JSC::ObjectConstructor::finishCreation): * runtime/ObjectConstructor.h: * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::RegExpConstructor): (JSC::RegExpConstructor::finishCreation): (JSC::RegExpMatchesArray::RegExpMatchesArray): * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::create): * runtime/RegExpObject.cpp: (JSC::RegExpObject::RegExpObject): * runtime/RegExpObject.h: (JSC::RegExpObject::create): * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::RegExpPrototype): * runtime/StringConstructor.cpp: (JSC::StringConstructor::StringConstructor): (JSC::StringConstructor::finishCreation): * runtime/StringConstructor.h: * runtime/StringObject.cpp: (JSC::StringObject::StringObject): * runtime/StringObject.h: (JSC::StringObject::create): * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::StringObjectThatMasqueradesAsUndefined): * runtime/StringPrototype.cpp: (JSC::StringPrototype::StringPrototype): (JSC::StringPrototype::finishCreation): * runtime/StringPrototype.h: 2011-09-06 Filip Pizlo <fpizlo@apple.com> Accessibility tests crashing in BasicRawSentinelNode code https://bugs.webkit.org/show_bug.cgi?id=67682 Reviewed by Geoffrey Garen. A CodeBlock should ensure that no other CodeBlocks have references to it after it is destroyed. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::~CodeBlock): 2011-09-06 Yong Li <yoli@rim.com> https://bugs.webkit.org/show_bug.cgi?id=67486 This reverts r65993 which gives wrong results for rshift in some corner cases (see the test). Reviewed by Gavin Barraclough. New test: fast/js/floating-point-truncate-rshift.html * assembler/ARMAssembler.h: * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::supportsFloatingPointTruncate): (JSC::MacroAssemblerARM::branchTruncateDoubleToInt32): 2011-09-06 Filip Pizlo <fpizlo@apple.com> Unreviewed build fix for r94559. Marked the relevant parameters as unused if !ENABLE(JIT), and surrounded new out-of-line JIT-specific method definitions with !ENABLE(JIT). * bytecode/CodeBlock.cpp: * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::compileForCallInternal): 2011-09-06 Mark Hahnenberg <mhahnenberg@apple.com> Fix broken PPC build due to new dtoa library https://bugs.webkit.org/show_bug.cgi?id=67654 Reviewed by Dan Bernstein. Added condition for PPC in the new dtoa compatibility check so that building won't fail. * wtf/dtoa/utils.h: 2011-09-05 Oliver Hunt <oliver@apple.com> An object's structure should reference the global object responsible for its creation https://bugs.webkit.org/show_bug.cgi?id=67624 Reviewed by Gavin Barraclough. Add a reference to a GlobalObject to Structure, and update all calls to Structure::create() to pass the global object that is the origin for that structure. For objects where the appropriate global object isn't available at construction time (global object prototypes, etc), or objects that logically don't have a global object (strings, etc) we just pass null. This change is largely mechanical (passing a new globalObject parameter around). * API/JSCallbackConstructor.h: (JSC::JSCallbackConstructor::createStructure): * API/JSCallbackFunction.h: (JSC::JSCallbackFunction::createStructure): * API/JSCallbackObject.h: (JSC::JSCallbackObject::createStructure): * API/JSContextRef.cpp: * JavaScriptCore.exp: * debugger/DebuggerActivation.h: (JSC::DebuggerActivation::createStructure): * runtime/Arguments.h: (JSC::Arguments::createStructure): * runtime/ArrayConstructor.h: (JSC::ArrayConstructor::createStructure): * runtime/ArrayPrototype.h: (JSC::ArrayPrototype::createStructure): * runtime/BooleanObject.h: (JSC::BooleanObject::createStructure): * runtime/BooleanPrototype.h: (JSC::BooleanPrototype::createStructure): * runtime/DateConstructor.h: (JSC::DateConstructor::createStructure): * runtime/DateInstance.h: (JSC::DateInstance::createStructure): * runtime/DatePrototype.h: (JSC::DatePrototype::createStructure): * runtime/ErrorInstance.h: (JSC::ErrorInstance::createStructure): * runtime/ErrorPrototype.h: (JSC::ErrorPrototype::createStructure): * runtime/Executable.h: (JSC::ExecutableBase::createStructure): (JSC::NativeExecutable::createStructure): (JSC::EvalExecutable::createStructure): (JSC::ProgramExecutable::createStructure): (JSC::FunctionExecutable::createStructure): * runtime/FunctionPrototype.h: (JSC::FunctionPrototype::createStructure): * runtime/GetterSetter.h: (JSC::GetterSetter::createStructure): * runtime/InternalFunction.h: (JSC::InternalFunction::createStructure): * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): * runtime/JSActivation.h: (JSC::JSActivation::createStructure): * runtime/JSArray.h: (JSC::JSArray::createStructure): * runtime/JSByteArray.cpp: (JSC::JSByteArray::createStructure): * runtime/JSByteArray.h: * runtime/JSFunction.h: (JSC::JSFunction::createStructure): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::finishCreation): (JSC::JSGlobalObject::createStructure): * runtime/JSNotAnObject.h: (JSC::JSNotAnObject::createStructure): * runtime/JSONObject.h: (JSC::JSONObject::createStructure): * runtime/JSObject.cpp: (JSC::JSObject::createInheritorID): * runtime/JSObject.h: (JSC::JSObject::createStructure): (JSC::JSNonFinalObject::createStructure): (JSC::JSFinalObject::createStructure): (JSC::createEmptyObjectStructure): * runtime/JSObjectWithGlobalObject.h: (JSC::JSObjectWithGlobalObject::createStructure): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::createStructure): * runtime/JSStaticScopeObject.h: (JSC::JSStaticScopeObject::createStructure): * runtime/JSString.h: (JSC::RopeBuilder::createStructure): * runtime/JSVariableObject.h: (JSC::JSVariableObject::createStructure): * runtime/JSWrapperObject.h: (JSC::JSWrapperObject::createStructure): * runtime/MathObject.h: (JSC::MathObject::createStructure): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::createStructure): (JSC::NativeErrorConstructor::constructorBody): * runtime/NumberConstructor.h: (JSC::NumberConstructor::createStructure): * runtime/NumberObject.h: (JSC::NumberObject::createStructure): * runtime/NumberPrototype.h: (JSC::NumberPrototype::createStructure): * runtime/ObjectConstructor.h: (JSC::ObjectConstructor::createStructure): * runtime/ObjectPrototype.h: (JSC::ObjectPrototype::createStructure): * runtime/RegExp.h: (JSC::RegExp::createStructure): * runtime/RegExpConstructor.h: (JSC::RegExpConstructor::createStructure): * runtime/RegExpObject.h: (JSC::RegExpObject::createStructure): * runtime/RegExpPrototype.h: (JSC::RegExpPrototype::createStructure): * runtime/ScopeChain.h: (JSC::ScopeChainNode::createStructure): * runtime/StrictEvalActivation.h: (JSC::StrictEvalActivation::createStructure): * runtime/StringConstructor.h: (JSC::StringConstructor::createStructure): * runtime/StringObject.h: (JSC::StringObject::createStructure): * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::create): (JSC::StringObjectThatMasqueradesAsUndefined::createStructure): * runtime/StringPrototype.h: (JSC::StringPrototype::createStructure): * runtime/Structure.cpp: (JSC::Structure::Structure): (JSC::Structure::visitChildren): * runtime/Structure.h: (JSC::Structure::create): (JSC::Structure::globalObject): (JSC::Structure::setGlobalObject): * runtime/StructureChain.h: (JSC::StructureChain::createStructure): 2011-09-06 Michael Saboff <msaboff@apple.com> Add windows changes for JSC:RegExp functional tests https://bugs.webkit.org/show_bug.cgi?id=67521 Windows build changes for regular expression functional test. Rubber-stamped by Gavin Barraclough. * JavaScriptCore.vcproj/JavaScriptCore.sln: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.vcproj/testRegExp: Added. * JavaScriptCore.vcproj/testRegExp/testRegExp.vcproj: Added. * JavaScriptCore.vcproj/testRegExp/testRegExpCommon.vsprops: Added. * JavaScriptCore.vcproj/testRegExp/testRegExpDebug.vsprops: Added. * JavaScriptCore.vcproj/testRegExp/testRegExpDebugAll.vsprops: Added. * JavaScriptCore.vcproj/testRegExp/testRegExpDebugCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/testRegExp/testRegExpPostBuild.cmd: Added. * JavaScriptCore.vcproj/testRegExp/testRegExpPreBuild.cmd: Added. * JavaScriptCore.vcproj/testRegExp/testRegExpPreLink.cmd: Added. * JavaScriptCore.vcproj/testRegExp/testRegExpProduction.vsprops: Added. * JavaScriptCore.vcproj/testRegExp/testRegExpRelease.vsprops: Added. * JavaScriptCore.vcproj/testRegExp/testRegExpReleaseCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/testRegExp/testRegExpReleasePGO.vsprops: Added. 2011-09-06 Filip Pizlo <fpizlo@apple.com> JavaScriptCore does not have tiered compilation https://bugs.webkit.org/show_bug.cgi?id=67176 Reviewed by Gavin Barraclough. This adds the ability to have multiple CodeBlocks associated with a particular role in an Executable. These are stored in descending order of compiler tier. CodeBlocks are optimized when a counter (m_executeCounter) that is incremented in loops and epilogues becomes positive. Optimizing means that all calls to the old CodeBlock are unlinked. The DFG can now pull in predictions from ValueProfiles, and propagate them along the graph. To support the new phase while maintaing some level of abstraction, a DFGDriver was introduced that encapsulates how to run the DFG compiler. This is turned off by default because it's not yet a performance win on all benchmarks. It speeds up crypto and richards by 10% and 6% respectively, but still does not do as good of a job as it could. Notably, the DFG backend has not changed, and is largely oblivious to the new information being made available to it. When turned off (the default), this patch is performance neutral. * CMakeLists.txt: * GNUmakefile.am: * GNUmakefile.list.am: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: * JavaScriptCore.xcodeproj/project.pbxproj: * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::branchAdd32): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::branchAdd32): * bytecode/CodeBlock.cpp: (JSC::CodeBlock::CodeBlock): (JSC::CodeBlock::~CodeBlock): (JSC::CodeBlock::visitAggregate): (JSC::CallLinkInfo::unlink): (JSC::CodeBlock::unlinkCalls): (JSC::CodeBlock::unlinkIncomingCalls): (JSC::CodeBlock::clearEvalCache): (JSC::replaceExistingEntries): (JSC::CodeBlock::copyDataFromAlternative): (JSC::ProgramCodeBlock::replacement): (JSC::EvalCodeBlock::replacement): (JSC::FunctionCodeBlock::replacement): (JSC::ProgramCodeBlock::compileOptimized): (JSC::EvalCodeBlock::compileOptimized): (JSC::FunctionCodeBlock::compileOptimized): * bytecode/CodeBlock.h: (JSC::GlobalCodeBlock::GlobalCodeBlock): (JSC::ProgramCodeBlock::ProgramCodeBlock): (JSC::EvalCodeBlock::EvalCodeBlock): (JSC::FunctionCodeBlock::FunctionCodeBlock): * bytecode/ValueProfile.h: (JSC::ValueProfile::dump): (JSC::ValueProfile::computeStatistics): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): * bytecompiler/BytecodeGenerator.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::addCall): (JSC::DFG::ByteCodeParser::dynamicallyPredict): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::parse): * dfg/DFGDriver.cpp: Added. (JSC::DFG::compile): (JSC::DFG::tryCompile): (JSC::DFG::tryCompileFunction): * dfg/DFGDriver.h: Added. (JSC::DFG::tryCompile): (JSC::DFG::tryCompileFunction): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGGraph.h: (JSC::DFG::Graph::predict): (JSC::DFG::Graph::predictGlobalVar): (JSC::DFG::Graph::isConstant): (JSC::DFG::Graph::isJSConstant): (JSC::DFG::Graph::isInt32Constant): (JSC::DFG::Graph::isDoubleConstant): (JSC::DFG::Graph::valueOfJSConstant): (JSC::DFG::Graph::valueOfInt32Constant): (JSC::DFG::Graph::valueOfDoubleConstant): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::isConstant): (JSC::DFG::JITCompiler::isJSConstant): (JSC::DFG::JITCompiler::isInt32Constant): (JSC::DFG::JITCompiler::isDoubleConstant): (JSC::DFG::JITCompiler::valueOfJSConstant): (JSC::DFG::JITCompiler::valueOfInt32Constant): (JSC::DFG::JITCompiler::valueOfDoubleConstant): * dfg/DFGNode.h: (JSC::DFG::isCellPrediction): (JSC::DFG::isNumberPrediction): (JSC::DFG::predictionToString): (JSC::DFG::mergePrediction): (JSC::DFG::makePrediction): (JSC::DFG::Node::valueOfJSConstant): (JSC::DFG::Node::isInt32Constant): (JSC::DFG::Node::isDoubleConstant): (JSC::DFG::Node::valueOfInt32Constant): (JSC::DFG::Node::valueOfDoubleConstant): (JSC::DFG::Node::predict): * dfg/DFGPropagation.cpp: Added. (JSC::DFG::Propagator::Propagator): (JSC::DFG::Propagator::fixpoint): (JSC::DFG::Propagator::setPrediction): (JSC::DFG::Propagator::mergePrediction): (JSC::DFG::Propagator::propagateNode): (JSC::DFG::Propagator::propagateForward): (JSC::DFG::Propagator::propagateBackward): (JSC::DFG::propagate): * dfg/DFGPropagation.h: Added. (JSC::DFG::propagate): * dfg/DFGRepatch.cpp: (JSC::DFG::dfgLinkFor): * heap/HandleHeap.h: (JSC::HandleHeap::Node::Node): * jit/JIT.cpp: (JSC::JIT::emitOptimizationCheck): (JSC::JIT::emitTimeoutCheck): (JSC::JIT::privateCompile): (JSC::JIT::linkFor): * jit/JIT.h: (JSC::JIT::emitOptimizationCheck): * jit/JITCall32_64.cpp: (JSC::JIT::emit_op_ret): (JSC::JIT::emit_op_ret_object_or_this): * jit/JITCode.h: (JSC::JITCode::JITCode): (JSC::JITCode::bottomTierJIT): (JSC::JITCode::topTierJIT): (JSC::JITCode::nextTierJIT): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_ret): (JSC::JIT::emit_op_ret_object_or_this): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jit/JITStubs.h: * runtime/Executable.cpp: (JSC::EvalExecutable::compileOptimized): (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileOptimized): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::compileOptimizedForCall): (JSC::FunctionExecutable::compileOptimizedForConstruct): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): * runtime/Executable.h: (JSC::EvalExecutable::compile): (JSC::ProgramExecutable::compile): (JSC::FunctionExecutable::compileForCall): (JSC::FunctionExecutable::compileForConstruct): (JSC::FunctionExecutable::compileOptimizedFor): * wtf/Platform.h: * wtf/SentinelLinkedList.h: (WTF::BasicRawSentinelNode::BasicRawSentinelNode): (WTF::BasicRawSentinelNode::setPrev): (WTF::BasicRawSentinelNode::setNext): (WTF::BasicRawSentinelNode::prev): (WTF::BasicRawSentinelNode::next): (WTF::BasicRawSentinelNode::isOnList): (WTF::::remove): (WTF::::SentinelLinkedList): (WTF::::begin): (WTF::::end): (WTF::::push): 2011-09-05 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r94445 and r94448. http://trac.webkit.org/changeset/94445 http://trac.webkit.org/changeset/94448 https://bugs.webkit.org/show_bug.cgi?id=67595 It broke everything (Requested by ossy on #webkit). * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * heap/Heap.cpp: (JSC::Heap::collect): * heap/Heap.h: * heap/NewSpace.cpp: (JSC::NewSpace::NewSpace): * heap/NewSpace.h: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/JSObject.cpp: (JSC::JSObject::allocatePropertyStorage): * runtime/JSObject.h: (JSC::JSObject::~JSObject): (JSC::JSObject::putDirectInternal): (JSC::JSObject::putDirectWithoutTransition): (JSC::JSObject::putDirectFunctionWithoutTransition): (JSC::JSObject::transitionTo): (JSC::JSObject::visitChildrenDirect): 2011-09-05 Patrick Gansterer <paroga@webkit.org> Unreviewed build fix for r94452. Add config.h as the first header to the cc files as required by the coding style. Reuse macros from Assertions.h instead of adding addional #ifdefs. * wtf/dtoa/bignum-dtoa.cc: * wtf/dtoa/bignum.cc: * wtf/dtoa/cached-powers.cc: * wtf/dtoa/diy-fp.cc: * wtf/dtoa/double-conversion.cc: * wtf/dtoa/fast-dtoa.cc: * wtf/dtoa/fixed-dtoa.cc: * wtf/dtoa/strtod.cc: * wtf/dtoa/utils.h: 2011-09-05 Andras Becsi <andras.becsi@nokia.com> [Qt][WK2] Fix the build Rubber-stamped by Csaba Osztrogonác. * wtf/dtoa/double-conversion.cc: Remove dead variable in file added in r94452. The variable fractional_part is only set but never used. 2011-09-04 Mark Hahnenberg <mhahnenberg@apple.com> REGRESSION (r94452): 20 http/tests tests failing on Qt Linux Release https://bugs.webkit.org/show_bug.cgi?id=67562 Reviewed by Darin Adler. Fixing the build (again which was broken by the dtoa patch. Needed to make sure WTF::double_conversion::initialize() is called for Qt as well as adding a check for WinCE in dtoa/utils.h * runtime/InitializeThreading.cpp: (JSC::initializeThreadingOnce): * wtf/dtoa/cached-powers.cc: * wtf/dtoa/utils.h: 2011-09-03 Filip Pizlo <fpizlo@apple.com> ThunkGenerators does not convert positive double zero into integer zero https://bugs.webkit.org/show_bug.cgi?id=67553 Reviewed by Gavin Barraclough. This is an 0.5% speed-up on V8 and neutral elsewhere. * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::returnDouble): 2011-09-03 Kevin Ollivier <kevino@theolliviers.com> [wx] Unreviewed build fix. Add wtf/dtoa directory to build. * wscript: 2011-09-03 Filip Pizlo <fpizlo@apple.com> DFG variable predictions only work for local variables, not temporaries https://bugs.webkit.org/show_bug.cgi?id=67554 Reviewed by Gavin Barraclough. This appears to be a slight speed-up in Kraken (0.3% but significant) and neutral elsewhere. * dfg/DFGGraph.h: (JSC::DFG::Graph::predict): 2011-09-02 Filip Pizlo <fpizlo@apple.com> DFG JIT speculation failure does recovery of additions in reverse and doesn't rebox https://bugs.webkit.org/show_bug.cgi?id=67551 Reviewed by Sam Weinig. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): 2011-09-02 Filip Pizlo <fpizlo@apple.com> ValueProfile does not make it safe to introspect cell values after garbage collection https://bugs.webkit.org/show_bug.cgi?id=67354 Reviewed by Gavin Barraclough. ValueProfile buckets are now weak references, implemented using a light-weight weak reference mechanism that this patch also adds (the WeakReferenceHarvester). If a cell stored in a ValueProfile bucket is not marked, then the bucket is transformed into a Structure pointer. If the Structure is not marked either, then it is turned into a ClassInfo pointer. * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::~CodeBlock): (JSC::CodeBlock::visitAggregate): (JSC::CodeBlock::visitWeakReferences): * bytecode/CodeBlock.h: * bytecode/ValueProfile.h: (JSC::ValueProfile::ValueProfile): (JSC::ValueProfile::classInfo): (JSC::ValueProfile::numberOfInt32s): (JSC::ValueProfile::numberOfDoubles): (JSC::ValueProfile::numberOfCells): (JSC::ValueProfile::numberOfArrays): (JSC::ValueProfile::probabilityOfArray): (JSC::ValueProfile::WeakBucket::WeakBucket): (JSC::ValueProfile::WeakBucket::operator!): (JSC::ValueProfile::WeakBucket::isEmpty): (JSC::ValueProfile::WeakBucket::isClassInfo): (JSC::ValueProfile::WeakBucket::isStructure): (JSC::ValueProfile::WeakBucket::asStructure): (JSC::ValueProfile::WeakBucket::asClassInfo): (JSC::ValueProfile::WeakBucket::getClassInfo): * heap/Heap.cpp: (JSC::Heap::harvestWeakReferences): (JSC::Heap::markRoots): * heap/Heap.h: * heap/MarkStack.cpp: (JSC::SlotVisitor::drain): (JSC::SlotVisitor::harvestWeakReferences): * heap/MarkStack.h: (JSC::MarkStack::addWeakReferenceHarvester): (JSC::MarkStack::MarkStack): (JSC::MarkStack::appendUnbarrieredPointer): * heap/SlotVisitor.h: * heap/WeakReferenceHarvester.h: Added. (JSC::WeakReferenceHarvester::WeakReferenceHarvester): (JSC::WeakReferenceHarvester::~WeakReferenceHarvester): 2011-09-02 Michael Saboff <msaboff@apple.com> Replace local implementation of string equals() methods with UString versions https://bugs.webkit.org/show_bug.cgi?id=67342 In preparation to allowing StringImpl to be backed by 8 bit characters when appropriate, we need to eliminate or change the usage of StringImpl::characters(). Change the uses of characters() that are used to implement redundant equals() methods. Reviewed by Gavin Barraclough. * runtime/Identifier.cpp: (JSC::Identifier::equal): * runtime/Identifier.h: (JSC::Identifier::equal): * wtf/text/AtomicString.cpp: (WTF::CStringTranslator::equal): Moved an optimized method to here. (WTF::operator==): * wtf/text/StringImpl.cpp: (WTF::equal): * wtf/text/StringImpl.h: 2011-09-02 Michael Saboff <msaboff@apple.com> Add JSC:RegExp functional tests https://bugs.webkit.org/show_bug.cgi?id=67339 Added new test driver program (testRegExp) and corresponding data file along with build scripts changes. Reviewed by Gavin Barraclough. * JavaScriptCore.exp: * JavaScriptCore.xcodeproj/project.pbxproj: * testRegExp.cpp: Added. (Options::Options): (StopWatch::start): (StopWatch::stop): (StopWatch::getElapsedMS): (RegExpTest::RegExpTest): (GlobalObject::create): (GlobalObject::className): (GlobalObject::GlobalObject): (main): (cleanupGlobalData): (testOneRegExp): (scanString): (parseRegExpLine): (parseTestLine): (runFromFiles): (printUsageStatement): (parseArguments): (realMain): * tests/regexp: Added. * tests/regexp/RegExpTest.data: Added. 2011-09-02 Michael Saboff <msaboff@apple.com> Add JSC:RegExp functional test data generator https://bugs.webkit.org/show_bug.cgi?id=67519 Add a data generator for regular expressions. To enable, change the #undef REGEXP_FUNC_TEST_DATA_GEN to #define. Then compile and use regular expressions. The resulting data will be in /tmp/RegExpTestsData. Reviewed by Gavin Barraclough. * runtime/RegExp.cpp: (JSC::regExpFlags): (JSC::RegExpFunctionalTestCollector::clearRegExp): (JSC::RegExpFunctionalTestCollector::get): (JSC::RegExpFunctionalTestCollector::outputOneTest): (JSC::RegExpFunctionalTestCollector::RegExpFunctionalTestCollector): (JSC::RegExpFunctionalTestCollector::~RegExpFunctionalTestCollector): (JSC::RegExpFunctionalTestCollector::outputEscapedUString): (JSC::RegExp::~RegExp): (JSC::RegExp::compile): (JSC::RegExp::match): (JSC::RegExp::matchCompareWithInterpreter): 2011-09-02 Mark Hahnenberg <mhahnenberg@apple.com> Fix the broken build due to dtoa patch https://bugs.webkit.org/show_bug.cgi?id=67534 Reviewed by Oliver Hunt. Fixing the build. * GNUmakefile.list.am: * wtf/dtoa/bignum.cc: * wtf/dtoa/fast-dtoa.cc: * wtf/dtoa/utils.h: 2011-09-02 Oliver Hunt <oliver@apple.com> Remove OldSpace classes https://bugs.webkit.org/show_bug.cgi?id=67533 Reviewed by Gavin Barraclough. Remove the unused OldSpace classes * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * heap/Heap.cpp: (JSC::Heap::writeBarrierSlowCase): * heap/MarkedBlock.h: * heap/OldSpace.cpp: Removed. * heap/OldSpace.h: Removed. 2011-09-02 James Robinson <jamesr@chromium.org> Compile fix for mac build. * wtf/CheckedArithmetic.h: (WTF::operator+): (WTF::operator-): (WTF::operator*): 2011-08-30 Matthew Delaney <mdelaney@apple.com> Read out of bounds in sUnpremultiplyData_RGBA8888 / ImageBufferData::getData https://bugs.webkit.org/show_bug.cgi?id=65352 Reviewed by Simon Fraser. New test: fast/canvas/canvas-getImageData-large-crash.html This patch prevents overflows from happening in getImageData, createImageData, and canvas creation calls that specify widths and heights that end up overflowing the ints that we store those values in as well as derived values such as area and maxX / maxY of the bounding rects involved. Overflow of integer arithmetic is detected via the use of the new Checked type that was introduced in r94207. The change to JSC is just to add a new helper method described below. * wtf/MathExtras.h: (isWithinIntRange): Reports if a float's value is within the range expressible by an int. 2011-09-02 Mark Hahnenberg <mhahnenberg@apple.com> Incorporate newer, faster dtoa library https://bugs.webkit.org/show_bug.cgi?id=66346 Reviewed by Oliver Hunt. Added new dtoa library at http://code.google.com/p/double-conversion/. Replaced old call to dtoa. The new library is much faster than the old one. We still use the old dtoa for some stuff in WebCore as well as the old strtod, but we can phase these out eventually as well. * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * runtime/InitializeThreading.cpp: * runtime/NumberPrototype.cpp: (JSC::numberProtoFuncToExponential): (JSC::numberProtoFuncToFixed): (JSC::numberProtoFuncToPrecision): * runtime/UString.cpp: (JSC::UString::number): * wtf/CMakeLists.txt: * wtf/ThreadingPthreads.cpp: (WTF::initializeThreading): * wtf/ThreadingWin.cpp: (WTF::initializeThreading): * wtf/dtoa.cpp: (WTF::dtoa): * wtf/dtoa.h: * wtf/dtoa/COPYING: Added. * wtf/dtoa/LICENSE: Added. * wtf/dtoa/README: Added. * wtf/dtoa/bignum-dtoa.cc: Added. * wtf/dtoa/bignum-dtoa.h: Added. * wtf/dtoa/bignum.cc: Added. * wtf/dtoa/bignum.h: Added. (WTF::double_conversion::Bignum::Times10): (WTF::double_conversion::Bignum::Equal): (WTF::double_conversion::Bignum::LessEqual): (WTF::double_conversion::Bignum::Less): (WTF::double_conversion::Bignum::PlusEqual): (WTF::double_conversion::Bignum::PlusLessEqual): (WTF::double_conversion::Bignum::PlusLess): (WTF::double_conversion::Bignum::EnsureCapacity): (WTF::double_conversion::Bignum::BigitLength): * wtf/dtoa/cached-powers.cc: Added. * wtf/dtoa/cached-powers.h: Added. * wtf/dtoa/diy-fp.cc: Added. * wtf/dtoa/diy-fp.h: Added. (WTF::double_conversion::DiyFp::DiyFp): (WTF::double_conversion::DiyFp::Subtract): (WTF::double_conversion::DiyFp::Minus): (WTF::double_conversion::DiyFp::Times): (WTF::double_conversion::DiyFp::Normalize): (WTF::double_conversion::DiyFp::f): (WTF::double_conversion::DiyFp::e): (WTF::double_conversion::DiyFp::set_f): (WTF::double_conversion::DiyFp::set_e): * wtf/dtoa/double-conversion.cc: Added. * wtf/dtoa/double-conversion.h: Added. (WTF::double_conversion::DoubleToStringConverter::DoubleToStringConverter): (WTF::double_conversion::StringToDoubleConverter::StringToDoubleConverter): * wtf/dtoa/double.h: Added. (WTF::double_conversion::double_to_uint64): (WTF::double_conversion::uint64_to_double): (WTF::double_conversion::Double::Double): (WTF::double_conversion::Double::AsDiyFp): (WTF::double_conversion::Double::AsNormalizedDiyFp): (WTF::double_conversion::Double::AsUint64): (WTF::double_conversion::Double::NextDouble): (WTF::double_conversion::Double::Exponent): (WTF::double_conversion::Double::Significand): (WTF::double_conversion::Double::IsDenormal): (WTF::double_conversion::Double::IsSpecial): (WTF::double_conversion::Double::IsNan): (WTF::double_conversion::Double::IsInfinite): (WTF::double_conversion::Double::Sign): (WTF::double_conversion::Double::UpperBoundary): (WTF::double_conversion::Double::NormalizedBoundaries): (WTF::double_conversion::Double::value): (WTF::double_conversion::Double::SignificandSizeForOrderOfMagnitude): (WTF::double_conversion::Double::Infinity): (WTF::double_conversion::Double::NaN): (WTF::double_conversion::Double::DiyFpToUint64): * wtf/dtoa/fast-dtoa.cc: Added. * wtf/dtoa/fast-dtoa.h: Added. * wtf/dtoa/fixed-dtoa.cc: Added. * wtf/dtoa/fixed-dtoa.h: Added. * wtf/dtoa/strtod.cc: Added. * wtf/dtoa/strtod.h: Added. * wtf/dtoa/utils.h: Added. (WTF::double_conversion::Max): (WTF::double_conversion::Min): (WTF::double_conversion::StrLength): (WTF::double_conversion::Vector::Vector): (WTF::double_conversion::Vector::SubVector): (WTF::double_conversion::Vector::length): (WTF::double_conversion::Vector::is_empty): (WTF::double_conversion::Vector::start): (WTF::double_conversion::Vector::operator[]): (WTF::double_conversion::Vector::first): (WTF::double_conversion::Vector::last): (WTF::double_conversion::StringBuilder::StringBuilder): (WTF::double_conversion::StringBuilder::~StringBuilder): (WTF::double_conversion::StringBuilder::size): (WTF::double_conversion::StringBuilder::position): (WTF::double_conversion::StringBuilder::Reset): (WTF::double_conversion::StringBuilder::AddCharacter): (WTF::double_conversion::StringBuilder::AddString): (WTF::double_conversion::StringBuilder::AddSubstring): (WTF::double_conversion::StringBuilder::AddPadding): (WTF::double_conversion::StringBuilder::Finalize): (WTF::double_conversion::StringBuilder::is_finalized): (WTF::double_conversion::BitCast): * wtf/wtf.pri: 2011-09-02 Filip Pizlo <fpizlo@apple.com> DFG graph has no way of distinguishing or reconciling between static and dynamic predictions https://bugs.webkit.org/show_bug.cgi?id=67343 Reviewed by Gavin Barraclough. PredictedType now stores the source of the prediction. Merging predictions, which was previously done with a bitwise or, is now done via the mergePredictions (equivalent to |) and mergePrediction (equivalent to |=) functions, which correctly handle combinations of static and dynamic. This is performance-neutral, since all predictions are currently static and so the code has no visible effects. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::set): (JSC::DFG::ByteCodeParser::staticallyPredictArray): (JSC::DFG::ByteCodeParser::staticallyPredictInt32): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.h: (JSC::DFG::Graph::predict): (JSC::DFG::Graph::predictGlobalVar): * dfg/DFGNode.h: (JSC::DFG::isArrayPrediction): (JSC::DFG::isInt32Prediction): (JSC::DFG::isDoublePrediction): (JSC::DFG::isDynamicPrediction): (JSC::DFG::mergePredictions): (JSC::DFG::mergePrediction): (JSC::DFG::makePrediction): (JSC::DFG::Node::predict): 2011-09-02 Oliver Hunt <oliver@apple.com> Fix 32bit build. * heap/NewSpace.h: (JSC::NewSpace::allocatePropertyStorage): (JSC::NewSpace::inPropertyStorageNursery): 2011-09-02 Oliver Hunt <oliver@apple.com> Use bump allocator for initial property storage https://bugs.webkit.org/show_bug.cgi?id=67494 Reviewed by Gavin Barraclough. Switch to a bump allocator for the initial out of line property storage. This gives us slightly faster allocation for short lived objects that need out of line storage at the cost of an additional memcpy when the object survives a GC pass. No performance impact. * JavaScriptCore.exp: * heap/Heap.cpp: (JSC::Heap::collect): * heap/Heap.h: (JSC::Heap::allocatePropertyStorage): (JSC::Heap::inPropertyStorageNursary): * heap/NewSpace.cpp: (JSC::NewSpace::NewSpace): * heap/NewSpace.h: (JSC::NewSpace::resetPropertyStorageNursary): (JSC::NewSpace::allocatePropertyStorage): (JSC::NewSpace::inPropertyStorageNursary): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/JSObject.cpp: (JSC::JSObject::allocatePropertyStorage): * runtime/JSObject.h: (JSC::JSObject::~JSObject): (JSC::JSObject::putDirectInternal): (JSC::JSObject::putDirectWithoutTransition): (JSC::JSObject::putDirectFunctionWithoutTransition): (JSC::JSObject::transitionTo): (JSC::JSObject::visitChildrenDirect): 2011-09-01 Mark Rowe <mrowe@apple.com> Fix the build. * JavaScriptCore.JSVALUE32_64only.exp: * JavaScriptCore.JSVALUE64only.exp: * JavaScriptCore.exp: 2011-09-01 Mark Hahnenberg <mhahnenberg@apple.com> Unzip initialization lists and constructors in JSCell hierarchy (4/7) https://bugs.webkit.org/show_bug.cgi?id=67174 Reviewed by Oliver Hunt. Completed the fourth level of the refactoring to add finishCreation() methods to all classes within the JSCell hierarchy with non-trivial constructor bodies. This primarily consists of pushing the calls to finishCreation() down into the constructors of the subclasses of the second level of the hierarchy as well as pulling the finishCreation() calls out into the class's corresponding create() method if it has one. Doing both simultaneously allows us to maintain the invariant that the finishCreation() method chain is called exactly once during the creation of an object, since calling it any other number of times (0, 2, or more) will cause an assertion failure. * API/JSCallbackConstructor.cpp: (JSC::JSCallbackConstructor::JSCallbackConstructor): (JSC::JSCallbackConstructor::finishCreation): * API/JSCallbackConstructor.h: * API/JSCallbackObject.h: (JSC::JSCallbackObject::create): * API/JSCallbackObjectFunctions.h: (JSC::::JSCallbackObject): (JSC::::finishCreation): * JavaScriptCore.JSVALUE64only.exp: * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::DebuggerActivation): (JSC::DebuggerActivation::create): * debugger/DebuggerActivation.h: * runtime/Arguments.h: (JSC::Arguments::create): (JSC::Arguments::createNoParameters): (JSC::Arguments::Arguments): * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::ArrayPrototype): (JSC::ArrayPrototype::finishCreation): * runtime/ArrayPrototype.h: * runtime/BooleanObject.cpp: (JSC::BooleanObject::BooleanObject): (JSC::BooleanObject::finishCreation): * runtime/BooleanObject.h: * runtime/DateInstance.cpp: (JSC::DateInstance::DateInstance): (JSC::DateInstance::finishCreation): * runtime/DateInstance.h: * runtime/ErrorInstance.cpp: (JSC::ErrorInstance::ErrorInstance): * runtime/ErrorInstance.h: (JSC::ErrorInstance::create): * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::ErrorPrototype): (JSC::ErrorPrototype::finishCreation): * runtime/ErrorPrototype.h: * runtime/ExceptionHelpers.cpp: (JSC::InterruptedExecutionError::InterruptedExecutionError): (JSC::InterruptedExecutionError::create): (JSC::TerminatedExecutionError::TerminatedExecutionError): (JSC::TerminatedExecutionError::create): * runtime/Executable.cpp: (JSC::EvalExecutable::EvalExecutable): (JSC::ProgramExecutable::ProgramExecutable): (JSC::FunctionExecutable::FunctionExecutable): * runtime/Executable.h: (JSC::NativeExecutable::create): (JSC::NativeExecutable::NativeExecutable): (JSC::EvalExecutable::create): (JSC::ProgramExecutable::create): (JSC::FunctionExecutable::create): * runtime/InternalFunction.cpp: (JSC::InternalFunction::InternalFunction): (JSC::InternalFunction::finishCreation): * runtime/InternalFunction.h: * runtime/JSActivation.cpp: (JSC::JSActivation::JSActivation): (JSC::JSActivation::finishCreation): * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::JSArray): * runtime/JSArray.h: (JSC::JSArray::create): * runtime/JSByteArray.cpp: (JSC::JSByteArray::JSByteArray): * runtime/JSByteArray.h: (JSC::JSByteArray::create): * runtime/JSFunction.cpp: (JSC::JSFunction::JSFunction): (JSC::JSFunction::finishCreation): * runtime/JSFunction.h: (JSC::JSFunction::create): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::JSGlobalObject): (JSC::JSGlobalObject::finishCreation): * runtime/JSNotAnObject.h: (JSC::JSNotAnObject::JSNotAnObject): (JSC::JSNotAnObject::create): * runtime/JSONObject.cpp: (JSC::JSONObject::JSONObject): (JSC::JSONObject::finishCreation): * runtime/JSONObject.h: * runtime/JSObjectWithGlobalObject.cpp: (JSC::JSObjectWithGlobalObject::JSObjectWithGlobalObject): * runtime/JSObjectWithGlobalObject.h: * runtime/JSStaticScopeObject.h: (JSC::JSStaticScopeObject::create): (JSC::JSStaticScopeObject::finishCreation): (JSC::JSStaticScopeObject::JSStaticScopeObject): * runtime/JSVariableObject.h: (JSC::JSVariableObject::JSVariableObject): * runtime/JSWrapperObject.h: (JSC::JSWrapperObject::JSWrapperObject): * runtime/MathObject.cpp: (JSC::MathObject::MathObject): (JSC::MathObject::finishCreation): * runtime/MathObject.h: * runtime/NumberObject.cpp: (JSC::NumberObject::NumberObject): (JSC::NumberObject::finishCreation): * runtime/NumberObject.h: * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::ObjectPrototype): * runtime/ObjectPrototype.h: (JSC::ObjectPrototype::create): * runtime/RegExpConstructor.cpp: (JSC::RegExpMatchesArray::RegExpMatchesArray): (JSC::RegExpMatchesArray::finishCreation): * runtime/RegExpMatchesArray.h: * runtime/RegExpObject.cpp: (JSC::RegExpObject::RegExpObject): (JSC::RegExpObject::finishCreation): * runtime/RegExpObject.h: * runtime/StrictEvalActivation.cpp: (JSC::StrictEvalActivation::StrictEvalActivation): * runtime/StrictEvalActivation.h: (JSC::StrictEvalActivation::create): * runtime/StringObject.cpp: (JSC::StringObject::StringObject): (JSC::StringObject::finishCreation): * runtime/StringObject.h: 2011-09-01 Daniel Bates <dbates@rim.com> QNX GCC distribution doesn't support vasprintf() https://bugs.webkit.org/show_bug.cgi?id=67423 Reviewed by Antonio Gomes. * wtf/Platform.h: Don't enable HAVE_VASPRINTF when building with GCC on QNX. 2011-09-01 Michael Saboff <msaboff@apple.com> Remove simple usage of UString::characters() from JavaScriptCore https://bugs.webkit.org/show_bug.cgi?id=67340 In preparation to allowing StringImpl to be backed by 8 bit characters when appropriate, we need to eliminate or change the usage of StringImpl::characters(). Most of the changes below change s->characters()[0] to s[0]. Reviewed by Geoffrey Garen. * bytecompiler/BytecodeGenerator.cpp: (JSC::keyForCharacterSwitch): * bytecompiler/NodesCodegen.cpp: (JSC::processClauseList): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Identifier.cpp: (JSC::Identifier::addSlowCase): * runtime/JSGlobalObjectFunctions.cpp: (JSC::jsToNumber): (JSC::parseFloat): * runtime/JSString.cpp: (JSC::JSString::substringFromRope): * runtime/JSString.h: (JSC::jsSingleCharacterSubstring): (JSC::jsString): (JSC::jsSubstring): (JSC::jsOwnedString): * runtime/RegExp.cpp: (JSC::regExpFlags): * wtf/text/StringBuilder.h: (WTF::StringBuilder::operator[]): 2011-09-01 Ada Chan <adachan@apple.com> Export fastMallocStatistics and Heap::objectTypeCounts for https://bugs.webkit.org/show_bug.cgi?id=67160. Reviewed by Darin Adler. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-09-01 Hao Zheng <zhenghao@chromium.org> Define PTHREAD_KEYS_MAX to fix Android port build. https://bugs.webkit.org/show_bug.cgi?id=67362 Reviewed by Adam Barth. PTHREAD_KEYS_MAX is not defined in bionic, so explicitly define it. * wtf/ThreadIdentifierDataPthreads.cpp: 2011-08-31 Oliver Hunt <oliver@apple.com> Fix build. * wtf/CheckedArithmetic.h: (WTF::Checked::Checked): (WTF::Checked::operator=): 2011-08-31 Oliver Hunt <oliver@apple.com> fast/regex/overflow.html asserts in debug builds https://bugs.webkit.org/show_bug.cgi?id=67326 Reviewed by Gavin Barraclough. The deliberate overflows in these expressions don't interact nicely with Checked<32bit-type> so we just bump up to Checked<int64_t> for the intermediate calculations. * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::generatePatternCharacterFixed): (JSC::Yarr::YarrGenerator::generateCharacterClassFixed): 2011-08-31 Jeff Miller <jeffm@apple.com> REGRESSION(92210): AVFoundation media engine is disabled on OS X https://bugs.webkit.org/show_bug.cgi?id=67316 Move the definition of WTF_USE_AVFOUNDATION on the Mac back to JavaScriptCore/wtf/Platform.h, since WebKit2 doesn't have access to WebCore/config.h on this platform. This reverts the changes that were made in r92210. Reviewed by Darin Adler. * wtf/Platform.h: Added definition of WTF_USE_AVFOUNDATION on the Mac. 2011-08-31 Peter Beverloo <peter@chromium.org> Add Android's platform specification and the right atomic functions. https://bugs.webkit.org/show_bug.cgi?id=66687 Reviewed by Adam Barth. * wtf/Atomics.h: (WTF::atomicIncrement): (WTF::atomicDecrement): * wtf/Platform.h: 2011-08-30 Oliver Hunt <oliver@apple.com> Add support for checked arithmetic https://bugs.webkit.org/show_bug.cgi?id=67095 Reviewed by Sam Weinig. Add a checked arithmetic class Checked<T> that provides overflow-safe arithmetic over all integral types. Checked<T> supports addition, subtraction and multiplication, along with "bool" conversions and equality operators. Checked<> can be used in either CRASH() on overflow or delayed failure modes, although the default is to CRASH(). To ensure the code is actually in use (rather than checking in dead code) I've made a couple of properties in YARR use Checked<int> and Checked<unsigned> instead of raw value arithmetic. This has resulted in a moderate set of changes, to YARR - mostly adding .get() calls, but a couple of casts from unsigned long to unsigned for some uses of sizeof, as Checked<> currently does not support mixed signed-ness of types wider that 32 bits. Happily the increased type safety of Checked<> means that it's not possible to accidentally assign away precision, nor accidentally call integer overload of a function instead of the bool version. No measurable regression in performance, and SunSpider claims this patch to be a progression of 0.3%. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/CheckedArithmetic.h: Added. (WTF::CrashOnOverflow::overflowed): (WTF::CrashOnOverflow::clearOverflow): (WTF::CrashOnOverflow::hasOverflowed): (WTF::RecordOverflow::RecordOverflow): (WTF::RecordOverflow::overflowed): (WTF::RecordOverflow::clearOverflow): (WTF::RecordOverflow::hasOverflowed): (WTF::isInBounds): (WTF::safeAdd): (WTF::safeSub): (WTF::safeMultiply): (WTF::safeEquals): (WTF::workAroundClangBug): (WTF::Checked::Checked): (WTF::Checked::operator=): (WTF::Checked::operator++): (WTF::Checked::operator--): (WTF::Checked::operator!): (WTF::Checked::operator UnspecifiedBoolType*): (WTF::Checked::get): (WTF::Checked::operator+=): (WTF::Checked::operator-=): (WTF::Checked::operator*=): (WTF::Checked::operator==): (WTF::Checked::operator!=): (WTF::operator+): (WTF::operator-): (WTF::operator*): * yarr/YarrInterpreter.cpp: (JSC::Yarr::ByteCompiler::atomPatternCharacter): (JSC::Yarr::ByteCompiler::atomCharacterClass): (JSC::Yarr::ByteCompiler::atomBackReference): (JSC::Yarr::ByteCompiler::atomParentheticalAssertionEnd): (JSC::Yarr::ByteCompiler::atomParenthesesSubpatternEnd): (JSC::Yarr::ByteCompiler::atomParenthesesOnceEnd): (JSC::Yarr::ByteCompiler::atomParenthesesTerminalEnd): * yarr/YarrInterpreter.h: (JSC::Yarr::ByteTerm::ByteTerm): (JSC::Yarr::ByteTerm::CheckInput): (JSC::Yarr::ByteTerm::UncheckInput): * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::generateAssertionEOL): (JSC::Yarr::YarrGenerator::generatePatternCharacterFixed): (JSC::Yarr::YarrGenerator::generatePatternCharacterGreedy): (JSC::Yarr::YarrGenerator::backtrackPatternCharacterNonGreedy): (JSC::Yarr::YarrGenerator::generateCharacterClassOnce): (JSC::Yarr::YarrGenerator::generateCharacterClassFixed): (JSC::Yarr::YarrGenerator::generateCharacterClassGreedy): (JSC::Yarr::YarrGenerator::backtrackCharacterClassNonGreedy): * yarr/YarrPattern.cpp: (JSC::Yarr::YarrPatternConstructor::setupAlternativeOffsets): * yarr/YarrPattern.h: 2011-08-31 Andrei Popescu <andreip@google.com> Investigate current uses of OS(ANDROID) https://bugs.webkit.org/show_bug.cgi?id=66761 Unreviewed, build fix for ARM platforms. * wtf/Platform.h: 2011-08-31 Andrei Popescu <andreip@google.com> Investigate current uses of OS(ANDROID) https://bugs.webkit.org/show_bug.cgi?id=66761 Reviewed by Darin Adler. Remove the last legacy Android code. No new tests needed as the code wasn't tested in the first place. * wtf/Atomics.h: * wtf/Platform.h: * wtf/ThreadingPthreads.cpp: (WTF::createThreadInternal): 2011-08-30 Aaron Colwell <acolwell@chromium.org> Add MediaSource API to HTMLMediaElement https://bugs.webkit.org/show_bug.cgi?id=64731 Reviewed by Eric Carlson. * Configurations/FeatureDefines.xcconfig: 2011-08-30 Oliver Hunt <oliver@apple.com> TypedArrays don't ensure that denormalised values are normalised https://bugs.webkit.org/show_bug.cgi?id=67178 Reviewed by Gavin Barraclough. Add a couple of assertions to jsNumber() to ensure that we block signaling NaNs * runtime/JSValue.h: (JSC::jsDoubleNumber): (JSC::jsNumber): 2011-08-30 Ademar de Souza Reis Jr. <ademar.reis@openbossa.org> [Qt] Do not unconditionally use pkg-config in .pro files https://bugs.webkit.org/show_bug.cgi?id=67055 Reviewed by Andreas Kling. Original patch from Rohan McGovern <rohan.mcgovern@nokia.com> Using the first pkg-config in PATH is prone to errors when cross compiling inside the Qt repository (using Qt's build-system). This patch protect calls for pkg-config with !contains(QT_CONFIG, no-pkg-config). no-pkg-config is added to QT_CONFIG by Qt's 'configure' when cross-compiling on systems without pkg-config. The respective change in Qt's configure has been submited already. No new tests as this is just a build change. * wtf/wtf.pri: protect pkg-config calls 2011-08-29 Daniel Bates <dbates@webkit.org> Add HAVE(VASPRINTF) macro to test for vasprintf() support https://bugs.webkit.org/show_bug.cgi?id=67156 Reviewed by Darin Adler. Encapsulate testing of vasprintf() support in a HAVE macro instead of hardcoding the list of supported/unsupported compilers at the call site. * wtf/Platform.h: 2011-08-29 Mark Hahnenberg <mhahnenberg@apple.com> Unzip initialization lists and constructors in JSCell hierarchy (3/7) https://bugs.webkit.org/show_bug.cgi?id=67064 Reviewed by Darin Adler. Completed the third level of the refactoring to add finishCreation() methods to all classes within the JSCell hierarchy with non-trivial constructor bodies. This primarily consists of pushing the calls to finishCreation() down into the constructors of the subclasses of the second level of the hierarchy as well as pulling the finishCreation() calls out into the class's corresponding create() method if it has one. Doing both simultaneously allows us to maintain the invariant that the finishCreation() method chain is called exactly once during the creation of an object, since calling it any other number of times (0, 2, or more) will cause an assertion failure. * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::DebuggerActivation): (JSC::DebuggerActivation::finishCreation): * debugger/DebuggerActivation.h: (JSC::DebuggerActivation::create): * runtime/Arguments.h: (JSC::Arguments::create): (JSC::Arguments::createNoParameters): (JSC::Arguments::Arguments): (JSC::Arguments::finishCreation): * runtime/ErrorInstance.cpp: (JSC::ErrorInstance::ErrorInstance): * runtime/ErrorInstance.h: (JSC::ErrorInstance::finishCreation): * runtime/ExceptionHelpers.cpp: (JSC::InterruptedExecutionError::InterruptedExecutionError): (JSC::TerminatedExecutionError::TerminatedExecutionError): * runtime/Executable.cpp: (JSC::EvalExecutable::EvalExecutable): (JSC::ProgramExecutable::ProgramExecutable): (JSC::FunctionExecutable::FunctionExecutable): Moved the assignment of m_firstLine and m_lastLine into the FunctionExecutable::finishCreation() method in Executable.h * runtime/Executable.h: (JSC::ScriptExecutable::ScriptExecutable): (JSC::EvalExecutable::create): (JSC::ProgramExecutable::create): (JSC::FunctionExecutable::create): (JSC::FunctionExecutable::finishCreation): * runtime/JSArray.cpp: (JSC::JSArray::JSArray): (JSC::JSArray::finishCreation): * runtime/JSArray.h: * runtime/JSByteArray.cpp: (JSC::JSByteArray::JSByteArray): * runtime/JSByteArray.h: (JSC::JSByteArray::finishCreation): * runtime/JSNotAnObject.h: (JSC::JSNotAnObject::JSNotAnObject): * runtime/JSObject.h: (JSC::JSNonFinalObject::JSNonFinalObject): * runtime/JSObjectWithGlobalObject.cpp: (JSC::JSObjectWithGlobalObject::JSObjectWithGlobalObject): (JSC::JSObjectWithGlobalObject::finishCreation): * runtime/JSObjectWithGlobalObject.h: * runtime/JSVariableObject.h: (JSC::JSVariableObject::JSVariableObject): (JSC::JSVariableObject::finishCreation): * runtime/JSWrapperObject.h: (JSC::JSWrapperObject::JSWrapperObject): * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::ObjectPrototype): (JSC::ObjectPrototype::finishCreation): * runtime/ObjectPrototype.h: * runtime/StrictEvalActivation.cpp: (JSC::StrictEvalActivation::StrictEvalActivation): 2011-08-29 Andreas Kling <kling@webkit.org> Unreviewed build fix after r93990. * wtf/HashTable.h: 2011-08-29 Andreas Kling <kling@webkit.org> Viewing a post on reddit.com wastes a lot of memory on event listeners. https://bugs.webkit.org/show_bug.cgi?id=67133 Reviewed by Darin Adler. Add a minimum table size to the HashTraits, instead of having it hard coded. The default value remains at 64, but can now be specialized. * runtime/StructureTransitionTable.h: * wtf/HashTable.h: (WTF::HashTable::shouldShrink): (WTF::::expand): (WTF::::checkTableConsistencyExceptSize): * wtf/HashTraits.h: 2011-08-28 Jonathan Liu <net147@gmail.com> Fix build error when compiling with MinGW-w64 by disabling JIT on Windows 64-bit https://bugs.webkit.org/show_bug.cgi?id=61235 Reviewed by Gavin Barraclough. The fixed mmap executable allocator for JIT on x86_64 requires sys/mman.h which is not available on Windows. * wtf/Platform.h: 2011-08-27 Filip Pizlo <fpizlo@apple.com> JSC::Executable is inconsistent about using weak handle finalizers and destructors for releasing memory https://bugs.webkit.org/show_bug.cgi?id=67072 Reviewed by Darin Adler. Moved more of the destruction of Executable state into the finalizer, which also resulted in an opportunity to mostly combine this with discardCode(). This also means that the finalizer is now enabled even when the JIT is turned off. This is performance neutral on SunSpider, V8, and Kraken. * runtime/Executable.cpp: (JSC::ExecutableBase::clearCode): (JSC::ExecutableFinalizer::finalize): (JSC::EvalExecutable::clearCode): (JSC::ProgramExecutable::clearCode): (JSC::FunctionExecutable::discardCode): (JSC::FunctionExecutable::clearCode): * runtime/Executable.h: (JSC::ExecutableBase::finishCreation): 2011-08-26 Gavin Barraclough <barraclough@apple.com> DFG JIT - ArithMod may clobber operands. https://bugs.webkit.org/show_bug.cgi?id=67085 Reviewed by Sam Weinig. unboxDouble must be called on a temporary. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillDouble): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::boxDouble): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): 2011-08-26 Mark Hahnenberg <mhahnenberg@apple.com> Unzip initialization lists and constructors in JSCell hierarchy (2/7) https://bugs.webkit.org/show_bug.cgi?id=66957 Reviewed by Darin Adler. Completed the second level of the refactoring to add finishCreation() methods to all classes within the JSCell hierarchy with non-trivial constructor bodies. * runtime/Executable.h: (JSC::ExecutableBase::ExecutableBase): (JSC::ExecutableBase::create): (JSC::NativeExecutable::create): (JSC::NativeExecutable::finishCreation): (JSC::NativeExecutable::NativeExecutable): (JSC::ScriptExecutable::ScriptExecutable): (JSC::ScriptExecutable::finishCreation): * runtime/GetterSetter.h: (JSC::GetterSetter::GetterSetter): (JSC::GetterSetter::create): * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::create): (JSC::JSAPIValueWrapper::JSAPIValueWrapper): * runtime/JSObject.h: (JSC::JSNonFinalObject::JSNonFinalObject): (JSC::JSNonFinalObject::finishCreation): (JSC::JSFinalObject::create): (JSC::JSFinalObject::finishCreation): (JSC::JSFinalObject::JSFinalObject): (JSC::JSObject::JSObject): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::JSPropertyNameIterator): (JSC::JSPropertyNameIterator::create): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::create): * runtime/RegExp.cpp: (JSC::RegExp::RegExp): (JSC::RegExp::createWithoutCaching): * runtime/ScopeChain.h: (JSC::ScopeChainNode::ScopeChainNode): (JSC::ScopeChainNode::create): * runtime/Structure.cpp: (JSC::Structure::Structure): * runtime/Structure.h: (JSC::Structure::create): (JSC::Structure::finishCreation): (JSC::Structure::createStructure): * runtime/StructureChain.cpp: (JSC::StructureChain::StructureChain): * runtime/StructureChain.h: (JSC::StructureChain::create): 2011-08-26 Filip Pizlo <fpizlo@apple.com> The GC does not have a facility for profiling the kinds of objects that occupy the heap https://bugs.webkit.org/show_bug.cgi?id=66849 Reviewed by Geoffrey Garen. Destructor calls and object scans are now optionally counted, per vtable. When the heap is destroyed and profiling is enabled, the counts are dumped, with care taken to print the names of classes (modulo C++ mangling) sorted in descending commonality. * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * heap/Heap.cpp: (JSC::Heap::destroy): * heap/Heap.h: * heap/MarkStack.cpp: (JSC::SlotVisitor::visitChildren): (JSC::SlotVisitor::drain): * heap/MarkStack.h: * heap/MarkedBlock.cpp: (JSC::MarkedBlock::callDestructor): * heap/MarkedBlock.h: * heap/VTableSpectrum.cpp: Added. (JSC::VTableSpectrum::VTableSpectrum): (JSC::VTableSpectrum::~VTableSpectrum): (JSC::VTableSpectrum::countVPtr): (JSC::VTableSpectrum::count): (JSC::VTableAndCount::VTableAndCount): (JSC::VTableAndCount::operator<): (JSC::VTableSpectrum::dump): * heap/VTableSpectrum.h: Added. * wtf/Platform.h: 2011-08-26 Juan C. Montemayor <jmont@apple.com> Update topCallFrame when calling host functions in the JIT https://bugs.webkit.org/show_bug.cgi?id=67010 Reviewed by Oliver Hunt. The topCallFrame is not being updated when a host function is called by the JIT. This causes problems when trying to create a stack trace (https://bugs.webkit.org/show_bug.cgi?id=66994). * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::privateCompileCTINativeCall): 2011-08-26 Alexey Proskuryakov <ap@apple.com> Get rid of frame life support timer https://bugs.webkit.org/show_bug.cgi?id=66874 Reviewed by Geoff Garen. * runtime/JSGlobalObject.h: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: globalExec() no longer needs to be virtual, its only override was in JSDOMWindowBase. 2011-08-26 Chao-ying Fu <fu@mips.com> Fix MIPS patchOffsetGetByIdSlowCaseCall https://bugs.webkit.org/show_bug.cgi?id=67046 Reviewed by Gavin Barraclough. * jit/JIT.h: 2011-08-25 Mark Hahnenberg <mhahnenberg@apple.com> Fixing broken build due to unused variables in release mode https://bugs.webkit.org/show_bug.cgi?id=67004 Unreviewed, release build fix. Fixing broken build due to unused variables in ASSERTs in release build. * runtime/JSObject.h: (JSC::JSObject::finishCreation): * runtime/JSString.h: (JSC::RopeBuilder::finishCreation): * runtime/ScopeChain.h: (JSC::ScopeChainNode::finishCreation): 2011-08-25 Mark Hahnenberg <mhahnenberg@apple.com> Unzip initialization lists and constructors in JSCell hierarchy (1/7) https://bugs.webkit.org/show_bug.cgi?id=66827 Reviewed by Geoffrey Garen. Added finishCreation() methods to all immediately subclasses of JSCell with non-empty constructors. Part of a larger refactoring to "unzip" initialization lists and constructor bodies. Also renamed JSCell's constructorBody() method to finishCreation(). * runtime/Executable.h: (JSC::ExecutableBase::ExecutableBase): (JSC::ExecutableBase::constructorBody): * runtime/GetterSetter.h: (JSC::GetterSetter::GetterSetter): * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::constructorBody): (JSC::JSAPIValueWrapper::JSAPIValueWrapper): * runtime/JSCell.h: (JSC::JSCell::JSCell::JSCell): (JSC::JSCell::JSCell::constructorBody): * runtime/JSObject.h: (JSC::JSObject::constructorBody): (JSC::JSObject::JSObject): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::constructorBody): * runtime/JSString.h: (JSC::RopeBuilder::JSString): (JSC::RopeBuilder::constructorBody): * runtime/RegExp.cpp: (JSC::RegExp::RegExp): (JSC::RegExp::constructorBody): * runtime/RegExp.h: * runtime/ScopeChain.h: (JSC::ScopeChainNode::ScopeChainNode): (JSC::ScopeChainNode::constructorBody): * runtime/Structure.cpp: (JSC::Structure::Structure): * runtime/StructureChain.cpp: (JSC::StructureChain::StructureChain): * runtime/StructureChain.h: (JSC::StructureChain::create): (JSC::StructureChain::constructorBody): 2011-08-25 Gabor Loki <loki@webkit.org> REGRESSION(r93755): It made 14 jsc test and ~500 layout test fail on Qt-ARM bot https://bugs.webkit.org/show_bug.cgi?id=66956 Rebaseline constants for patching GetByIdSlowCaseCall on ARM. Reviewed by Oliver Hunt. * jit/JIT.h: 2011-08-24 Juan C. Montemayor <jmont@apple.com> Keep track of topCallFrame for Stack traces https://bugs.webkit.org/show_bug.cgi?id=66571 Reviewed by Geoffrey Garen. This patch adds a TopCallFrame to JSC in order to have that information when an error is thrown to create a stack trace. The TopCallFrame is updated throughout select points in the Interpreter and the JSC. * interpreter/Interpreter.cpp: (JSC::Interpreter::unwindCallFrame): (JSC::Interpreter::throwException): (JSC::Interpreter::execute): (JSC::Interpreter::executeCall): (JSC::Interpreter::executeConstruct): (JSC::Interpreter::privateExecute): * interpreter/Interpreter.h: (JSC::TopCallFrameSetter::TopCallFrameSetter): (JSC::TopCallFrameSetter::~TopCallFrameSetter): * jit/JIT.h: * jit/JITInlineMethods.h: (JSC::JIT::updateTopCallFrame): * jit/JITStubCall.h: (JSC::JITStubCall::call): * jit/JITStubs.cpp: (JSC::throwExceptionFromOpCall): (JSC::DEFINE_STUB_FUNCTION): (JSC::arityCheckFor): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: 2011-08-24 Filip Pizlo <fpizlo@apple.com> ErrorInstance::create sometimes has two heap object constructions in flight at once https://bugs.webkit.org/show_bug.cgi?id=66845 Reviewed by Darin Adler. The fix is simple since there is already a second create() method that takes a UString. * runtime/ErrorInstance.cpp: (JSC::ErrorInstance::create): 2011-08-24 Filip Pizlo <fpizlo@apple.com> There is no facility for profiling how the write barrier is used https://bugs.webkit.org/show_bug.cgi?id=66747 Reviewed by Geoffrey Garen. Added facilities for the JIT to specify the kind of write barrier being executed. Added code for profiling the number of each kind of barrier encountered. * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::writeBarrier): (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCodeGenerator.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::emitCount): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::emitCount): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGRepatch.cpp: (JSC::DFG::tryCachePutByID): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * heap/Heap.h: (JSC::Heap::writeBarrier): * heap/WriteBarrierSupport.cpp: Added. (JSC::WriteBarrierCounters::initialize): * heap/WriteBarrierSupport.h: Added. (JSC::WriteBarrierCounters::WriteBarrierCounters): (JSC::WriteBarrierCounters::jitCounterFor): (JSC::WriteBarrierCounters::countWriteBarrier): * jit/JIT.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_put_by_id): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::emit_op_put_scoped_var): (JSC::JIT::emit_op_put_global_var): (JSC::JIT::emitWriteBarrier): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_put_by_val): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::emit_op_put_scoped_var): (JSC::JIT::emit_op_put_global_var): (JSC::JIT::emitWriteBarrier): * runtime/InitializeThreading.cpp: (JSC::initializeThreadingOnce): * runtime/WriteBarrier.h: (JSC::WriteBarrierBase::setWithoutWriteBarrier): 2011-08-23 Mark Hahnenberg <mhahnenberg@apple.com> Add checks to ensure allocation does not take place during initialization of GC-managed objects https://bugs.webkit.org/show_bug.cgi?id=65288 Reviewed by Darin Adler. Adding the new validation functionality. In its current state, it will performs checks, but they don't fail unless you do allocation in the arguments to the parent constructor in the initialization list of a class. The allocateCell() method turns on the global flag disallowing any new allocations, and the constructorBody() method in JSCell turns it off. This way, allocation is still allowed in constructor bodies while other refactoring efforts continue. * runtime/JSCell.h: (JSC::JSCell::JSCell::constructorBody): (JSC::JSCell::JSCell::JSCell): (JSC::JSCell::allocateCell): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: (JSC::JSGlobalData::isInitializingObject): (JSC::JSGlobalData::setInitializingObject): * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::create): 2011-08-23 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=55347 "name" and "message" enumerable on *Error.prototype Reviewed by Sam Weinig. The default value of a NativeErrorPrototype's message property is "", not the name of the error. * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::NativeErrorConstructor): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::create): (JSC::NativeErrorConstructor::constructorBody): * runtime/NativeErrorPrototype.cpp: (JSC::NativeErrorPrototype::NativeErrorPrototype): (JSC::NativeErrorPrototype::constructorBody): * runtime/NativeErrorPrototype.h: (JSC::NativeErrorPrototype::create): * runtime/StringPrototype.cpp: (JSC::StringPrototype::StringPrototype): * runtime/StringPrototype.h: (JSC::StringPrototype::create): 2011-08-23 Steve Block <steveblock@google.com> Remove last occurrences of PLATFORM(ANDROID) https://bugs.webkit.org/show_bug.cgi?id=66763 Reviewed by Tony Gentilcore. * wtf/Platform.h: 2011-08-23 Steve Block <steveblock@google.com> Remove all mention of removed Android files from build scripts https://bugs.webkit.org/show_bug.cgi?id=66755 Reviewed by Tony Gentilcore. * JavaScriptCore.gyp/JavaScriptCore.gyp: * JavaScriptCore.gypi: * gyp/JavaScriptCore.gyp: 2011-08-23 Adam Barth <abarth@webkit.org> Remove WebCore/editing/android and other Android-specific directories https://bugs.webkit.org/show_bug.cgi?id=66739 Reviewed by Steve Block. Now that Android shares more code with Chromium, we don't need these Android-specific files. * wtf/android: Removed. * wtf/android/AndroidThreading.h: Removed. * wtf/android/MainThreadAndroid.cpp: Removed. 2011-08-23 Ilya Tikhonovsky <loislo@chromium.org> Unreviewed build fix for compile error on Windows for r93560. * runtime/SamplingCounter.h: 2011-08-22 Filip Pizlo <fpizlo@apple.com> Sampling counter support is in the bytecode directory https://bugs.webkit.org/show_bug.cgi?id=66724 Reviewed by Darin Adler. Moved SamplingCounter to a separate header in runtime/. * GNUmakefile.list.am: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/SamplingTool.cpp: * bytecode/SamplingTool.h: * runtime/SamplingCounter.cpp: Added. (JSC::AbstractSamplingCounter::dump): * runtime/SamplingCounter.h: Added. (JSC::AbstractSamplingCounter::count): (JSC::AbstractSamplingCounter::addressOfCounter): (JSC::AbstractSamplingCounter::init): (JSC::SamplingCounter::SamplingCounter): (JSC::GlobalSamplingCounter::name): (JSC::DeletableSamplingCounter::DeletableSamplingCounter): (JSC::DeletableSamplingCounter::~DeletableSamplingCounter): 2011-08-21 Martin Robinson <mrobinson@igalia.com> Fix 'make dist' for WebKitGTK+. * GNUmakefile.list.am: Add a missing header to the sources list. 2011-08-20 Filip Pizlo <fpizlo@apple.com> JavaScriptCore bytecompiler does not compute scope depth correctly in the case of constant declarations https://bugs.webkit.org/show_bug.cgi?id=66572 Reviewed by Oliver Hunt. Changed the handling of const to add the dynamic scope depth. * bytecompiler/NodesCodegen.cpp: (JSC::ConstDeclNode::emitCodeSingle): 2011-08-19 Daniel Bates <dbates@webkit.org> Only #include <signal.h> and require SA_RESTART when building with JSC_MULTIPLE_THREADS https://bugs.webkit.org/show_bug.cgi?id=66617 Both <signal.h> and SA_RESTART usage are guarded behind ENABLE(JSC_MULTIPLE_THREADS). But we cause a compile error if the platform doesn't support SA_RESTART regardless of whether JSC_MULTIPLE_THREADS is enabled for the port. Instead, we shouldn't require SA_RESTART support unless we are building with JSC_MULTIPLE_THREADS enabled. Reviewed by Antonio Gomes. * heap/MachineStackMarker.cpp: 2011-08-19 Filip Pizlo <fpizlo@apple.com> The JSC JIT currently has no facility to profile and report the types of values https://bugs.webkit.org/show_bug.cgi?id=65901 Reviewed by Gavin Barraclough. Added the ability to profile the values seen at function calls (both arguments and results) and heap loads. This is done with emphasis on performance. A value profiling site consists of: add, and, move, and store; no branching is necessary. Each value profiling site (called a ValueProfile) has a ring buffer of 8 recently-seen values. ValueProfiles are stored in the CodeBlock; there will be one for each argument (excluding this) and each heap load or callsite. Each time a value profiling site executes, it stores the value into a pseudo-random element in the ValueProfile buffer. The point is that for frequently executed code, we will have 8 somewhat recent values in the buffer and will be able to not only figure out what type it is, but also to be able to reason about the actual values if we wish to do so. This feature is currently disabled by default. When enabled, it results in a 3.7% slow-down on SunSpider. * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::~CodeBlock): * bytecode/CodeBlock.h: (JSC::CodeBlock::addValueProfile): (JSC::CodeBlock::numberOfValueProfiles): (JSC::CodeBlock::valueProfile): (JSC::CodeBlock::valueProfileForBytecodeOffset): * bytecode/ValueProfile.h: Added. (JSC::ValueProfile::ValueProfile): (JSC::ValueProfile::numberOfSamples): (JSC::ValueProfile::computeProbability): (JSC::ValueProfile::numberOfInt32s): (JSC::ValueProfile::numberOfDoubles): (JSC::ValueProfile::numberOfCells): (JSC::ValueProfile::probabilityOfInt32): (JSC::ValueProfile::probabilityOfDouble): (JSC::ValueProfile::probabilityOfCell): (JSC::getValueProfileBytecodeOffset): * jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases): (JSC::JIT::privateCompile): * jit/JIT.h: (JSC::JIT::emitValueProfilingSite): * jit/JITCall.cpp: (JSC::JIT::emit_op_call_put_result): * jit/JITInlineMethods.h: (JSC::JIT::emitValueProfilingSite): * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_get_by_val): (JSC::JIT::emitSlow_op_get_by_val): (JSC::JIT::emit_op_method_check): (JSC::JIT::emit_op_get_by_id): (JSC::JIT::emitSlow_op_get_by_id): * jit/JSInterfaceJIT.h: * wtf/Platform.h: * wtf/StdLibExtras.h: (WTF::binarySearch): (WTF::genericBinarySearch): 2011-08-19 Daniel Bates <dbates@webkit.org> Don't include DisallowCType.h when building on QNX https://bugs.webkit.org/show_bug.cgi?id=66616 Reviewed by Antonio Gomes. * config.h: 2011-08-19 Daniel Bates <dbates@webkit.org> Implement ExecutableAllocator::cacheFlush() for QNX https://bugs.webkit.org/show_bug.cgi?id=66611 Reviewed by Antonio Gomes. * jit/ExecutableAllocator.h: (JSC::ExecutableAllocator::cacheFlush): 2011-08-19 Daniel Bates <dbates@webkit.org> Implement WTF::atomic{Increment, Decrement}() for QNX https://bugs.webkit.org/show_bug.cgi?id=66605 Reviewed by Darin Adler. * wtf/Atomics.h: (WTF::atomicIncrement): (WTF::atomicDecrement): 2011-08-19 Beth Dakin <bdakin@apple.com> https://bugs.webkit.org/show_bug.cgi?id=66590 Re-name scrollbar painter types Reviewed by Sam Weinig. WTF_USE_WK_SCROLLBAR_PAINTER is now WTF_USE_SCROLLBAR_PAINTER since WK no longer applies. * wtf/Platform.h: 2011-08-18 Mark Hahnenberg <mhahnenberg@apple.com> Move allocation in constructors into separate constructorBody() methods https://bugs.webkit.org/show_bug.cgi?id=66265 Reviewed by Oliver Hunt. Refactoring to put all allocations that need to be done after the object's initialization list has executed but before the object is ready for use into a separate constructorBody() method. This method is still called by the constructor, so the patch doesn't resolve any potential issues, it's just to set up the code for further refactoring. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * jsc.cpp: (GlobalObject::constructorBody): (GlobalObject::GlobalObject): * runtime/ErrorInstance.cpp: (JSC::ErrorInstance::ErrorInstance): * runtime/ErrorInstance.h: (JSC::ErrorInstance::constructorBody): * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::ErrorPrototype): (JSC::ErrorPrototype::constructorBody): * runtime/ErrorPrototype.h: * runtime/Executable.cpp: (JSC::FunctionExecutable::FunctionExecutable): * runtime/Executable.h: (JSC::FunctionExecutable::constructorBody): * runtime/InternalFunction.cpp: (JSC::InternalFunction::InternalFunction): * runtime/InternalFunction.h: (JSC::InternalFunction::constructorBody): * runtime/JSByteArray.cpp: (JSC::JSByteArray::JSByteArray): * runtime/JSByteArray.h: (JSC::JSByteArray::constructorBody): * runtime/JSFunction.cpp: (JSC::JSFunction::JSFunction): (JSC::JSFunction::constructorBody): * runtime/JSFunction.h: * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::JSGlobalObject): (JSC::JSGlobalObject::constructorBody): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::JSPropertyNameIterator): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::constructorBody): * runtime/JSString.h: (JSC::RopeBuilder::JSString): (JSC::RopeBuilder::constructorBody): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::NativeErrorConstructor): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::constructorBody): * runtime/NativeErrorPrototype.cpp: (JSC::NativeErrorPrototype::NativeErrorPrototype): (JSC::NativeErrorPrototype::constructorBody): * runtime/NativeErrorPrototype.h: * runtime/StringObject.cpp: * runtime/StringObject.h: (JSC::StringObject::create): * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::create): (JSC::StringObjectThatMasqueradesAsUndefined::StringObjectThatMasqueradesAsUndefined): * runtime/StringPrototype.cpp: (JSC::StringPrototype::StringPrototype): * runtime/StringPrototype.h: (JSC::StringPrototype::create): 2011-08-10 Filip Pizlo <fpizlo@apple.com> DFG non-speculative JIT does not inline the double case of ValueAdd https://bugs.webkit.org/show_bug.cgi?id=66025 Reviewed by Gavin Barraclough. This is a 1.3% win on Kraken overall, with >=8% speed-ups on a few benchmarks (imaging-darkroom, stanford-crypto-pbkdf2, stanford-crypto-sha256-iterative). It looks like it might have a speed-up in SunSpider (though not statistically significant or particularly reproducible) and a slight slow-down in V8 (0.14%, not statistically significant). It does slow down v8-crypto by 1.5%. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::isKnownInteger): (JSC::DFG::JITCodeGenerator::isKnownNumeric): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::knownConstantArithOp): (JSC::DFG::NonSpeculativeJIT::basicArithOp): * dfg/DFGOperations.cpp: 2011-08-18 Filip Pizlo <fpizlo@apple.com> [jsfunfuzz] DFG speculative JIT does divide-by-zero checks incorrectly https://bugs.webkit.org/show_bug.cgi?id=66426 Reviewed by Oliver Hunt. Changed the branchTestPtr to branchTest32. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-08-17 Thouraya ANDOLSI <thouraya.andolsi@st.com> https://bugs.webkit.org/show_bug.cgi?id=66379 implements load32WithCompactAddressOffsetPatch function and fixes store32 and moveWithPatch functions for SH4 platforms. Reviewed by Gavin Barraclough. * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::rshift32): (JSC::MacroAssemblerSH4::store32): (JSC::MacroAssemblerSH4::load32WithCompactAddressOffsetPatch): (JSC::MacroAssemblerSH4::moveWithPatch): * assembler/SH4Assembler.h: (JSC::SH4Assembler::movlMemRegCompact): (JSC::SH4Assembler::readPointer): (JSC::SH4Assembler::repatchCompact): * jit/JIT.h: 2011-08-17 Filip Pizlo <fpizlo@apple.com> JSC verbose debugging output sometimes doesn't work as expected. https://bugs.webkit.org/show_bug.cgi?id=66107 Reviewed by Gavin Barraclough. Hardened the CodeBlock::dump() code so that it no longer crashes. Improved the DFG verbose code so that it prints slightly more useful information. * assembler/LinkBuffer.h: (JSC::LinkBuffer::debugSize): * bytecode/CodeBlock.cpp: (JSC::valueToSourceString): (JSC::CodeBlock::dump): * bytecode/CodeBlock.h: (JSC::CodeBlock::numberOfRegExps): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::link): 2011-08-16 Michael Saboff <msaboff@apple.com> Crash in Structure::visitChildren running iAd.js regression test suite under memory pressure https://bugs.webkit.org/show_bug.cgi?id=66351 JIT::privateCompilePutByIdTransition expects that regT0 and regT1 have the basePayload and baseTag respectively. In some cases, we may get to this generated code with one or both of these registers trash. One know case is that regT0 on ARM may be trashed as regT0 (r0) is also arg0 and can be overrun with sp due to calls to JIT::restoreReturnAddress(). This patch uses the values on the stack. A longer term solution is to work out all cases so that the register entry assumptions can assured. While fixing this, also determined that the additional stack offset of sizeof(void*) is not needed for ARM. Reviewed by Gavin Barraclough. * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::privateCompilePutByIdTransition): 2011-08-15 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=66263 DFG JIT does not always zero extend boolean result of DFG operations Reviewed by Sam Weinig. * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: - Change bool return values to a 64-bit type. 2011-08-15 Gavin Barraclough <barraclough@apple.com> Crash accessing static property on sealed object https://bugs.webkit.org/show_bug.cgi?id=66242 Reviewed by Sam Weinig. * runtime/JSObject.h: (JSC::JSObject::putDirectInternal): - should only check isExtensible if checkReadOnly. 2011-08-15 Sam Weinig <sam@webkit.org> Fix release build when building with Clang. Reviewed by Anders Carlsson. * runtime/Identifier.cpp: (JSC::Identifier::checkCurrentIdentifierTable): Add NO_RETURN_DUE_TO_CRASH. 2011-08-15 Oliver Varga <Varga.Oliver@stud.u-szeged.hu> Reviewed by Nikolas Zimmermann. Speed up SVGSMILElement::findInstanceTime. https://bugs.webkit.org/show_bug.cgi?id=61025 Add a new parameter to StdlibExtras.h::binarySerarch function to also handle cases when the array does not contain the key value. This is needed for an svg function. * wtf/StdLibExtras.h: (WTF::binarySearch): 2011-08-13 Sam Weinig <sam@webkit.org> Add back 0xbbadbeef to CRASH to allow for old habits https://bugs.webkit.org/show_bug.cgi?id=66190 Reviewed by David Kilzer. * wtf/Assertions.h: Add back the assignment to the memory address 0xbbadbeef in the CRASH macro, as it does not cause issue in the clang static analyzer and many people use its presence in crash reports to easily identify ASSERTs. 2011-08-13 Sam Weinig <sam@webkit.org> Fix a bunch of minor bugs caught by the clang static analyzer in JavaScriptCore https://bugs.webkit.org/show_bug.cgi?id=66182 Reviewed by Dan Bernstein. Fixes 10 warnings in JavaScriptCore and 2 in testapi. * API/tests/testapi.c: (main): Remove dead variables. * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): Initialize hasPrinted and silence an unused warning by casting to void (Ok here since it is debug code and I want to keep it clear that if other cases are added, the hasPrinted flag would be needed). * wtf/dtoa.cpp: (WTF::d2b): The variable "de" in the else block is always zero, so there is no reason to use it. 2011-08-12 Sam Weinig <sam@webkit.org> Use __builtin_trap() for CRASH when building with clang https://bugs.webkit.org/show_bug.cgi?id=66152 Reviewed by Anders Carlsson. * wtf/Assertions.h: Add Clang specific CRASH macro that calls __builtin_trap() instead of silly techniques to crash. This allows the static analyzer to understand that we are intentionally crashing. As a result, we need to mark some functions as not returning. Also adds a macros that annotates a function as never returning due to ASSERT or CRASH. * wtf/Compiler.h: Add COMPILIER(CLANG) and fix some formatting and spelling mistakes. * wtf/FastMalloc.cpp: (WTF::Internal::fastMallocMatchFailed): Add NO_RETURN_DUE_TO_CRASH. * yarr/YarrParser.h: (JSC::Yarr::Parser::CharacterClassParserDelegate::assertionWordBoundary): (JSC::Yarr::Parser::CharacterClassParserDelegate::atomBackReference): Add NO_RETURN_DUE_TO_ASSERT. 2011-08-12 Filip Pizlo <fpizlo@apple.com> DFG JIT has inconsistent use of boxDouble and unboxDouble, inconsistent use of assertions regarding doubles, and those assertions are not turned on in debug builds https://bugs.webkit.org/show_bug.cgi?id=66160 Reviewed by Gavin Barraclough. JIT assertions are now turned on in debug builds. JIT assertions are now used for boxing and unboxing doubles, and boxing and unboxing no longer involves code duplication. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillDouble): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::boxDouble): (JSC::DFG::JITCodeGenerator::unboxDouble): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::fillNumericToDouble): (JSC::DFG::GeneralizedRegister::moveTo): (JSC::DFG::GeneralizedRegister::swapWith): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::boxDouble): (JSC::DFG::JITCompiler::unboxDouble): * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::knownConstantArithOp): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::convertToDouble): 2011-08-12 Mark Rowe <mrowe@apple.com> Be more forward-looking in the choice of compiler. Rubber-stamped by Jon Honeycutt. * Configurations/CompilerVersion.xcconfig: 2011-08-12 Kalev Lember <kalevlember@gmail.com> [GTK] Fix non-pthreads build after r91906. https://bugs.webkit.org/show_bug.cgi?id=66151 Reviewed by David Levin. r91906 broke the non-pthreads GTK+ build by including a header which doesn't exist. Fix it by including DateMath.h instead of DateMap.h. * wtf/gtk/ThreadingGtk.cpp: 2011-08-12 Mark Rowe <mrowe@apple.com> Update some configuration settings that were missed back in r92432. * Configurations/CompilerVersion.xcconfig: 2011-08-12 Filip Pizlo <fpizlo@apple.com> REGRESSION (r91610?): Bing Maps fail to initialize (InvalidOperation: Matrix3D.invert) https://bugs.webkit.org/show_bug.cgi?id=66038 Reviewed by Gavin Barraclough. Simplest and lowest-impact fix for the case where the spilled format of a DFG node differs from the register format: if the format is converted then indicate that the spilled value is no longer valid ("kill the spill"). * dfg/DFGGenerationInfo.h: (JSC::DFG::GenerationInfo::killSpilled): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillDouble): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): 2011-08-12 Sam Weinig <sam@webkit.org> Move compiler specific macros to their own header https://bugs.webkit.org/show_bug.cgi?id=66119 Reviewed by Anders Carlsson. * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/CMakeLists.txt: Add Compiler.h * wtf/AlwaysInline.h: Move the contents of this file (which no longer was just about ALWAYS_INLINE) to Compiler.h. We can remove this file in a later commit. * wtf/Compiler.h: Added. Put all compiler specific checks and features in this file. * wtf/Platform.h: Move COMPILER macro and definitions (and the odd WARN_UNUSED_RETURN compiler feature) to Compiler.h. Include Compiler.h since it is necessary. 2011-08-11 Filip Pizlo <fpizlo@apple.com> DFG JIT-specific structure stub info code offset fields are signed 8-bit, but it is possible for the offsets to be greater than 127 https://bugs.webkit.org/show_bug.cgi?id=66122 Reviewed by Gavin Barraclough. * bytecode/StructureStubInfo.h: * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::cachedPutById): 2011-08-11 Filip Pizlo <fpizlo@apple.com> DFG JIT speculation failure code sometimes picks the wrong register as a scratch register. https://bugs.webkit.org/show_bug.cgi?id=66104 Reviewed by Gavin Barraclough. Hardened the code with more assertions and fixed the bug. Now a spilled register is only used for scratch if it also isn't being used for shuffling. * dfg/DFGJITCompiler.cpp: (JSC::DFG::ShuffledRegister::handleNonCyclingPermutation): (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): 2011-08-11 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r92880. http://trac.webkit.org/changeset/92880 https://bugs.webkit.org/show_bug.cgi?id=66123 Breaks compile in VS2010 (Requested by jamesr_ on #webkit). * wtf/PassRefPtr.h: 2011-08-11 Mark Rowe <mrowe@apple.com> Don't conditionalize the use of -fomit-frame-pointer on compiler version as all of our supported compilers are now new enough to have the same, sane behavior. Rubber-stamped by Sam Weinig. * Configurations/JavaScriptCore.xcconfig: 2011-08-11 Filip Pizlo <fpizlo@apple.com> DFG JIT verbose mode does not report the generated types of nodes https://bugs.webkit.org/show_bug.cgi?id=65830 Reviewed by Sam Weinig. Added code that prints the type selected for each node's result. * dfg/DFGGenerationInfo.h: (JSC::DFG::dataFormatToString): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-08-11 James Robinson <jamesr@chromium.org> nullptr can't be used for PassRefPtr https://bugs.webkit.org/show_bug.cgi?id=66024 Reviewed by Anders Carlsson. * wtf/PassRefPtr.h: (WTF::PassRefPtr::PassRefPtr): 2011-08-11 Daniel Bates <dbates@rim.com> Removed unused variable in StackBounds::initialize() to resolve compiler warning when building on QNX. https://bugs.webkit.org/show_bug.cgi?id=66072 Reviewed by Antonio Gomes. * wtf/StackBounds.cpp: (WTF::StackBounds::initialize): 2011-08-11 Devdatta Deshpande <pwjd73@motorola.com> Implementation of monotonically increasing clock on GTK https://bugs.webkit.org/show_bug.cgi?id=62175 Reviewed by Martin Robinson. * wtf/CurrentTime.cpp: (WTF::monotonicallyIncreasingTime): The default implementation of monotonicallyIncreasingTime only guarantees the result to be non-decreasing. If the system time is changed to past then default implementation will still fail and WebCore timers will not fire. 2011-08-10 Geoffrey Garen <ggaren@apple.com> Removed some incorrect code that was dead. Reviewed by Oliver Hunt. clearSingleTransition() wasn't resetting m_data. Luckily, no one cares, because its caller was unused. Removed both. * runtime/Structure.cpp: * runtime/StructureTransitionTable.h: (JSC::StructureTransitionTable::~StructureTransitionTable): 2011-08-10 Filip Pizlo <fpizlo@apple.com> REGRESSION(r92670-r92744): WebKit crashes when opening Gmail https://bugs.webkit.org/show_bug.cgi?id=66010 Reviewed by Oliver Hunt. Made sure that Construct calls use() on the this argument. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::emitCall): 2011-08-10 Mark Hahnenberg <mhahnenberg@apple.com> JSC should always throw when function arg list is too long https://bugs.webkit.org/show_bug.cgi?id=65869 Reviewed by Oliver Hunt. Changed the behavior of the interpreter and JIT to throw an exception when too many arguments are passed rather than truncating the list. Added a new method to create a "Too many arguments." exception used by this new functionality. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ExceptionHelpers.cpp: (JSC::createTooManyParamsError): * runtime/ExceptionHelpers.h: 2011-08-10 Oliver Hunt <oliver@apple.com> Make GC checks more aggressive in release builds https://bugs.webkit.org/show_bug.cgi?id=66001 Reviewed by Gavin Barraclough. * heap/HandleHeap.cpp: (JSC::HandleHeap::visitStrongHandles): (JSC::HandleHeap::visitWeakHandles): (JSC::HandleHeap::finalizeWeakHandles): (JSC::HandleHeap::writeBarrier): (JSC::HandleHeap::isLiveNode): (JSC::HandleHeap::isValidWeakNode): Increase handle heap validation logic, and make some of the crashes trigger in release builds as well as debug. * heap/HandleHeap.h: (JSC::HandleHeap::allocate): (JSC::HandleHeap::makeWeak): Ditto * runtime/JSGlobalData.cpp: (WTF::Recompiler::operator()): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::visitChildren): Fix GC bugs found while testing this patch 2011-08-10 Oliver Hunt <oliver@apple.com> JSEvaluteScript does not return the correct object when given JSONP data https://bugs.webkit.org/show_bug.cgi?id=66003 Reviewed by Gavin Barraclough. Make sure we propagate the result of the function call rather than the argument. * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): 2011-08-10 Filip Pizlo <fpizlo@apple.com> DFG JIT heap prediction causes regressions when combined with aggressive integer prediction https://bugs.webkit.org/show_bug.cgi?id=65954 Reviewed by Gavin Barraclough. Disabled heap prediction, but did not remove the capability. This improves V8 crypto performance by 20%. * dfg/DFGGraph.h: (JSC::DFG::Graph::predict): 2011-08-09 Filip Pizlo <fpizlo@apple.com> DFG JIT does not speculative integers as aggressively as it should https://bugs.webkit.org/show_bug.cgi?id=65949 Reviewed by Gavin Barraclough. Added a tree walk to propagate integer predictions through arithmetic expressions. This is a 71% speed-up on Kraken's imaging-gaussian-blur, which translates to a 19% speed-up on Kraken overall. It's neutral on other benchmarks. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::predictInt32): 2011-08-09 Filip Pizlo <fpizlo@apple.com> DFG JIT has no way of propagating predictions to loads and calls https://bugs.webkit.org/show_bug.cgi?id=65883 Reviewed by Gavin Barraclough. This introduces the capability to store predictions on graph nodes. To save space while being somewhat consistent, the prediction is always stored in the second OpInfo slot (since a GetById will use the first one for the identifier). This change is a natural extension of r92593 (global variable prediction). This is a 1.5% win on V8 in the arithmetic mean, and a 0.6% win on V8 in the geometric mean. It is neutral on SunSpider and Kraken. Interestingly, on V8 it regresses crypto by 3% while progressing deltablue and richards by 2.6% and 4.3%, respectively. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::addToGraph): (JSC::DFG::ByteCodeParser::addCall): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::Graph::predict): (JSC::DFG::Graph::getPrediction): * dfg/DFGNode.h: (JSC::DFG::isCellPrediction): (JSC::DFG::isArrayPrediction): (JSC::DFG::isInt32Prediction): (JSC::DFG::isDoublePrediction): (JSC::DFG::isNumberPrediction): (JSC::DFG::predictionToString): (JSC::DFG::Node::Node): (JSC::DFG::Node::hasPrediction): (JSC::DFG::Node::getPrediction): (JSC::DFG::Node::predict): 2011-08-09 Filip Pizlo <fpizlo@apple.com> DFG JIT passes the this argument to constructors even though it's not necessary https://bugs.webkit.org/show_bug.cgi?id=65943 Reviewed by Gavin Barraclough. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::emitCall): 2011-08-09 Chao-ying Fu <fu@mips.com> Fix one MIPS instruction to call JITStubThunked_##op https://bugs.webkit.org/show_bug.cgi?id=65942 Reviewed by Gavin Barraclough. Changed "bal" to "jalr" for a possible processor mode change from MIPS32 to MIPS16. * jit/JITStubs.cpp: 2011-08-09 Filip Pizlo <fpizlo@apple.com> DFG JIT failure loading web site https://bugs.webkit.org/show_bug.cgi?id=65930 Reviewed by Oliver Hunt. Put the use() call after the fpr()/gpr() calls, since doing otherwise breaks the register allocator. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): 2011-08-09 Mark Hahnenberg <mhahnenberg@apple.com> Add ParentClass typedef in all JSC classes https://bugs.webkit.org/show_bug.cgi?id=65731 Reviewed by Oliver Hunt. Just added the Base typedefs in all the classes that are a subclass of JSCell to point at their parent classes. This is a change to support future changes to the way constructors and destructors are implemented in JS objects, among other things. * API/JSCallbackConstructor.h: * API/JSCallbackFunction.h: * API/JSCallbackObject.h: (JSC::JSCallbackObject::createStructure): (JSC::JSCallbackObject::visitChildren): * API/JSCallbackObjectFunctions.h: (JSC::::asCallbackObject): (JSC::::JSCallbackObject): (JSC::::init): (JSC::::className): (JSC::::getOwnPropertySlot): (JSC::::getOwnPropertyDescriptor): (JSC::::put): (JSC::::deleteProperty): (JSC::::getConstructData): (JSC::::construct): (JSC::::hasInstance): (JSC::::getCallData): (JSC::::call): (JSC::::getOwnPropertyNames): (JSC::::toNumber): (JSC::::toString): (JSC::::setPrivate): (JSC::::getPrivate): (JSC::::inherits): (JSC::::getStaticValue): (JSC::::staticFunctionGetter): (JSC::::callbackGetter): * debugger/DebuggerActivation.h: * jsc.cpp: * runtime/Arguments.h: * runtime/ArrayConstructor.h: * runtime/ArrayPrototype.h: * runtime/BooleanConstructor.h: * runtime/BooleanObject.h: * runtime/BooleanPrototype.h: * runtime/DateConstructor.h: * runtime/DateInstance.h: * runtime/DatePrototype.h: * runtime/Error.cpp: * runtime/ErrorConstructor.h: * runtime/ErrorInstance.h: * runtime/ErrorPrototype.h: * runtime/ExceptionHelpers.cpp: * runtime/Executable.h: * runtime/FunctionConstructor.h: * runtime/FunctionPrototype.h: * runtime/GetterSetter.h: * runtime/InternalFunction.h: * runtime/JSAPIValueWrapper.h: * runtime/JSActivation.h: * runtime/JSArray.h: * runtime/JSFunction.h: * runtime/JSGlobalObject.h: * runtime/JSNotAnObject.h: * runtime/JSONObject.h: * runtime/JSObject.h: * runtime/JSPropertyNameIterator.h: * runtime/JSStaticScopeObject.h: * runtime/JSString.h: * runtime/JSVariableObject.h: * runtime/JSWrapperObject.h: * runtime/MathObject.h: * runtime/NativeErrorConstructor.h: * runtime/NativeErrorPrototype.h: * runtime/NumberConstructor.h: * runtime/NumberObject.h: * runtime/NumberPrototype.h: * runtime/ObjectConstructor.h: * runtime/ObjectPrototype.h: * runtime/RegExp.h: * runtime/RegExpConstructor.h: * runtime/RegExpMatchesArray.h: * runtime/RegExpObject.h: (JSC::RegExpObject::create): * runtime/RegExpPrototype.h: * runtime/ScopeChain.h: * runtime/StrictEvalActivation.h: * runtime/StringConstructor.h: * runtime/StringObject.h: * runtime/StringObjectThatMasqueradesAsUndefined.h: * runtime/StringPrototype.h: * runtime/Structure.h: * runtime/StructureChain.h: 2011-08-08 Oliver Hunt <oliver@apple.com> Using mprotect to create guard pages breaks our use of madvise to release executable memory https://bugs.webkit.org/show_bug.cgi?id=65870 Reviewed by Gavin Barraclough. Use mmap rather than mprotect to clear guard page permissions. * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveAndCommit): 2011-08-08 Oliver Hunt <oliver@apple.com> Non-extensibility does not prevent mutating [[Prototype]] https://bugs.webkit.org/show_bug.cgi?id=65832 Reviewed by Gavin Barraclough. Disallow mutation of __proto__ on objects that are not extensible. * runtime/JSObject.cpp: (JSC::JSObject::put): 2011-08-08 Filip Pizlo <fpizlo@apple.com> DFG JIT does not track speculation decisions for global variables https://bugs.webkit.org/show_bug.cgi?id=65825 Reviewed by Gavin Barraclough. Added the capability to track predictions for global variables, and ensured that code can abstract over the source of prediction (local versus global variable) wherever it is appropriate to do so. Also cleaned up the code in SpeculativeJIT that decides how to speculate based on recorded predictions (for example instead of using isInteger, which makes sense for local predictions where the GetLocal would return an integer value, we now tend to use shouldSpeculateInteger, which checks if the value is either already an integer or should be speculated to be an integer). This is an 0.8% win on SunSpider, almost entirely thanks to a 25% win on controlflow-recursive. It's also a 4.8% win on v8-crypto. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::predictArray): (JSC::DFG::ByteCodeParser::predictInt32): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::Graph::predictGlobalVar): (JSC::DFG::Graph::predict): (JSC::DFG::Graph::getGlobalVarPrediction): (JSC::DFG::Graph::getPrediction): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::shouldSpeculateInteger): (JSC::DFG::SpeculativeJIT::shouldSpeculateDouble): 2011-08-07 Martin Robinson <mrobinson@igalia.com> Distribution fix for GTK+. * GNUmakefile.list.am: Strip removed files from the source list. 2011-08-06 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=65821 Don't form identifiers the first time a string is used as a property name. Reviewed by Oliver Hunt. This is a 1% win on SunSpider. * dfg/DFGOperations.cpp: - Use fastGetOwnProperty. * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): - Use fastGetOwnProperty. * runtime/JSCell.h: * runtime/JSObject.h: (JSC::JSCell::fastGetOwnProperty): - Fast call to get a property without creating an identifier the first time. * runtime/PropertyMapHashTable.h: (JSC::PropertyTable::find): (JSC::PropertyTable::findWithString): - Add interface to look up by either strinsg or identifiers. * runtime/Structure.h: (JSC::Structure::get): - Add a get() call that takes a UString, not an Identifier. * wtf/text/StringImpl.h: (WTF::StringImpl::hasHash): - Add a call to check if the has has been set (to detect the first use as a property name). 2011-08-06 Aron Rosenberg <arosenberg@logitech.com> Reviewed by Benjamin Poulain. [Qt] Fix build with Intel compiler on Windows https://bugs.webkit.org/show_bug.cgi?id=65088 Intel compiler needs .lib suffixes instead of .a Intel compiler doesn't support nullptr Intel compiler supports unsized arrays * JavaScriptCore.pri: * jsc.cpp: * wtf/ByteArray.h: * wtf/NullPtr.h: 2011-08-05 Gavin Barraclough <barraclough@apple.com> String replace with the empty string means string removal https://bugs.webkit.org/show_bug.cgi?id=65799 Reviewed by Sam Weinig. Optimization for String.prototype.replace([RegExp], ""), this improves v8-regexp by ~3%. * runtime/StringPrototype.cpp: (JSC::jsSpliceSubstrings): (JSC::stringProtoFuncReplace): 2011-08-05 Noel Gordon <noel.gordon@gmail.com> [Chromium] Remove JSZombie references from gyp project files. https://bugs.webkit.org/show_bug.cgi?id=65798 JSC runtime/JSZombie.{cpp,h} were removed in r92046. Remove references to these file names from the gyp projects. Reviewed by Darin Adler. * JavaScriptCore.gypi: zombies be gone. 2011-08-05 Mark Rowe <mrowe@apple.com> <http://webkit.org/b/65785> ThreadRestrictionVerifier needs a mode where an object is tied to a particular dispatch queue A RefCounted object can be opted in to this mode by calling setDispatchQueueForVerifier with the dispatch queue it will be tied to. This will cause ThreadRestrictionVerifier to ensure that all operations are performed on the given dispatch queue. Reviewed by Anders Carlsson. * wtf/RefCounted.h: (WTF::RefCountedBase::setDispatchQueueForVerifier): * wtf/ThreadRestrictionVerifier.h: (WTF::ThreadRestrictionVerifier::ThreadRestrictionVerifier): (WTF::ThreadRestrictionVerifier::~ThreadRestrictionVerifier): (WTF::ThreadRestrictionVerifier::setDispatchQueueMode): (WTF::ThreadRestrictionVerifier::setShared): (WTF::ThreadRestrictionVerifier::isSafeToUse): 2011-08-05 Oliver Hunt <oliver@apple.com> Inline allocation of function objects https://bugs.webkit.org/show_bug.cgi?id=65779 Reviewed by Gavin Barraclough. Inline allocation and initilisation of function objects in generated code. This ended up being a 60-70% improvement in function allocation performance. This improvement shows up as a ~2% improvement in 32bit sunspider and V8, but is a wash on 64-bit. We currently don't inline the allocation of named function expressions, as that requires being able to gc allocate a variable object. * jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases): * jit/JIT.h: (JSC::JIT::emitStoreCell): * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateBasicJSObject): (JSC::JIT::emitAllocateJSFinalObject): (JSC::JIT::emitAllocateJSFunction): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_new_func): (JSC::JIT::emitSlow_op_new_func): (JSC::JIT::emit_op_new_func_exp): (JSC::JIT::emitSlow_op_new_func_exp): * jit/JITOpcodes32_64.cpp: Removed duplicate implementation of op_new_func and op_new_func_exp * runtime/JSFunction.h: (JSC::JSFunction::offsetOfScopeChain): (JSC::JSFunction::offsetOfExecutable): 2011-08-04 David Levin <levin@chromium.org> CStringBuffer should have thread safety checks turned on. https://bugs.webkit.org/show_bug.cgi?id=58093 Reviewed by Dmitry Titov. * wtf/text/CString.h: (WTF::CStringBuffer::CStringBuffer): Removed the ifdef that turned this off for Chromium. 2011-08-04 Mark Rowe <mrowe@apple.com> Future-proof Xcode configuration settings. * Configurations/Base.xcconfig: * Configurations/DebugRelease.xcconfig: * Configurations/JavaScriptCore.xcconfig: * Configurations/Version.xcconfig: 2011-08-04 Mark Hahnenberg <mhahnenberg@apple.com> Interpreter can potentially GC in the middle of initializing a structure chain https://bugs.webkit.org/show_bug.cgi?id=65638 Reviewed by Oliver Hunt. Moved the allocation of a prototype StructureChain before the initialization of the structure chain within the interpreter that was causing intermittent GC crashes. * interpreter/Interpreter.cpp: (JSC::Interpreter::tryCachePutByID): * wtf/Platform.h: 2011-08-04 Filip Pizlo <fpizlo@apple.com> Eval handling attempts literal parsing even when the eval string is in the cache https://bugs.webkit.org/show_bug.cgi?id=65675 Reviewed by Oliver Hunt. This is a 25% speed-up on date-format-tofte and a 1.5% speed-up overall in SunSpider. It's neutral on V8. * bytecode/EvalCodeCache.h: (JSC::EvalCodeCache::tryGet): (JSC::EvalCodeCache::getSlow): (JSC::EvalCodeCache::get): * interpreter/Interpreter.cpp: (JSC::Interpreter::callEval): 2011-08-03 Mark Rowe <mrowe@apple.com> Bring some order to FeatureDefines.xcconfig to make it easier to follow. Reviewed by Sam Weinig. * Configurations/FeatureDefines.xcconfig: 2011-08-03 Mark Rowe <mrowe@apple.com> Clean up FeatureDefines.xcconfig to remove some unnecessary conditional settings Reviewed by Dave Kilzer. * Configurations/FeatureDefines.xcconfig: 2011-08-03 Filip Pizlo <fpizlo@apple.com> JSC GC heap size improvement breaks build on some platforms due to unused parameter https://bugs.webkit.org/show_bug.cgi?id=65641 Reviewed by Darin Adler. Fix build on non-x86 platforms, by ensuring that the relevant parameter always appears to be used even when it isn't. * heap/Heap.cpp: 2011-08-03 Carlos Garcia Campos <cgarcia@igalia.com> [GTK] Reorganize pkg-config files https://bugs.webkit.org/show_bug.cgi?id=65548 Reviewed by Martin Robinson. * GNUmakefile.am: * javascriptcoregtk.pc.in: Renamed from Source/WebKit/gtk/javascriptcoregtk.pc.in. 2011-08-01 David Levin <levin@chromium.org> Add asserts to RefCounted to make sure ref/deref happens on the right thread. https://bugs.webkit.org/show_bug.cgi?id=31639 Reviewed by Dmitry Titov. * GNUmakefile.list.am: Added new files to the build. * JavaScriptCore.gypi: Ditto. * JavaScriptCore.vcproj/WTF/WTF.vcproj: Ditto. * JavaScriptCore.xcodeproj/project.pbxproj: Ditto. * jit/ExecutableAllocator.h: (JSC::ExecutablePool::ExecutablePool): Turned off checks for this due to not being able to figure out what was guarding it (bug 58091). * parser/SourceProvider.h: (JSC::SourceProvider::SourceProvider): Ditto. * wtf/CMakeLists.txt: Added new files to the build. * wtf/ThreadRestrictionVerifier.h: Added. Everything is done in the header to avoid the issue with exports that are only useful in debug but still needing to export them. * wtf/RefCounted.h: (WTF::RefCountedBase::ref): Added checks using the non thread safe verifier. and filed bug 58171 about making it stricter. (WTF::RefCountedBase::hasOneRef): Ditto. (WTF::RefCountedBase::refCount): Ditto. (WTF::RefCountedBase::setMutexForVerifier): Expose a way to change the checks to be based on a mutex. This is in the header to avoid adding more exports from JavaScriptCore. (WTF::RefCountedBase::deprecatedTurnOffVerifier): Temporary way to turn off verification. Filed bug 58174 to remove this method. (WTF::RefCountedBase::derefBase): * wtf/SizeLimits.cpp: Adjusted the debug size check for RefCounted. * wtf/text/CString.h: (WTF::CStringBuffer::CStringBuffer): Turned off checks for this while a fix is being done in Chromium (bug 58093). 2011-08-02 Filip Pizlo <fpizlo@apple.com> JSC GC may not be able to reuse partially-free blocks after a full collection https://bugs.webkit.org/show_bug.cgi?id=65585 Reviewed by Darin Adler. This fixes the linked list management bug. This fix is performance neutral on SunSpider. * heap/NewSpace.cpp: (JSC::NewSpace::removeBlock): 2011-07-30 Oliver Hunt <oliver@apple.com> Simplify JSFunction creation for functions written in JS https://bugs.webkit.org/show_bug.cgi?id=65422 Reviewed by Gavin Barraclough. Remove hash lookups used to write name property and transition function structure by caching the resultant structure and property offset in JSGlobalObject. This doesn't impact performance, but we can use this change to make other improvements later. * runtime/Executable.cpp: (JSC::FunctionExecutable::FunctionExecutable): * runtime/Executable.h: (JSC::ScriptExecutable::ScriptExecutable): (JSC::FunctionExecutable::jsName): * runtime/JSFunction.cpp: (JSC::JSFunction::JSFunction): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::namedFunctionStructure): (JSC::JSGlobalObject::functionNameOffset): 2011-08-02 Filip Pizlo <fpizlo@apple.com> JSC GC uses dummy cells to avoid having to remember which cells it has already destroyed https://bugs.webkit.org/show_bug.cgi?id=65556 Reviewed by Oliver Hunt. This gets rid of dummy cells, and ensures that it's not necessary to invoke a destructor on cells that have already been swept. In the common case, a block knows that either all of its free cells still need to have destructors called, or none of them do, which minimizes the amount of branching that needs to happen per cell when performing a sweep. This is performance neutral on SunSpider and V8. It is meant as a stepping stone to simplify the implementation of more sophisticated sweeping algorithms. * heap/Heap.cpp: (JSC::CountFunctor::ClearMarks::operator()): * heap/MarkedBlock.cpp: (JSC::MarkedBlock::initForCellSize): (JSC::MarkedBlock::callDestructor): (JSC::MarkedBlock::specializedReset): (JSC::MarkedBlock::reset): (JSC::MarkedBlock::specializedSweep): (JSC::MarkedBlock::sweep): (JSC::MarkedBlock::produceFreeList): (JSC::MarkedBlock::lazySweep): (JSC::MarkedBlock::blessNewBlockForFastPath): (JSC::MarkedBlock::blessNewBlockForSlowPath): (JSC::MarkedBlock::canonicalizeBlock): * heap/MarkedBlock.h: (JSC::MarkedBlock::FreeCell::setNoObject): (JSC::MarkedBlock::setDestructorState): (JSC::MarkedBlock::destructorState): (JSC::MarkedBlock::notifyMayHaveFreshFreeCells): * runtime/JSCell.cpp: * runtime/JSCell.h: (JSC::JSCell::JSCell::JSCell): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::clearBuiltinStructures): * runtime/JSGlobalData.h: * runtime/Structure.h: 2011-08-01 Michael Saboff <msaboff@apple.com> Virtual copying of FastMalloc allocated memory causes madvise MADV_FREE_REUSABLE errors https://bugs.webkit.org/show_bug.cgi?id=65502 Reviewed by Anders Carlsson. With the fix of the issues causing madvise MADV_FREE_REUSABLE to fail, added an assert to the return code of madvise to catch any regressions. * wtf/TCSystemAlloc.cpp: (TCMalloc_SystemRelease): 2011-08-02 Anders Carlsson <andersca@apple.com> Fix Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-08-02 Anders Carlsson <andersca@apple.com> Fix a Windows build error. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-08-02 Filip Pizlo <fpizlo@apple.com> JSC GC is far too conservative about growing the heap size, particularly on desktop platforms https://bugs.webkit.org/show_bug.cgi?id=65438 Reviewed by Oliver Hunt. The minimum heap size is now 16MB instead of 512KB, provided all of the following are true: a) ENABLE(LARGE_HEAP) is set, which currently only happens on x86 targets, but could reasonably happen on any platform that is known to have a decent amount of RAM. b) JSGlobalData is initialized with HeapSize = LargeHeap, which currently only happens when it's the JSDOMWindowBase in WebCore or in the jsc command-line tool. This is a 4.1% speed-up on SunSpider. * JavaScriptCore.exp: * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::collect): * heap/Heap.h: * jsc.cpp: (main): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::createContextGroup): (JSC::JSGlobalData::create): (JSC::JSGlobalData::createLeaked): (JSC::JSGlobalData::sharedInstance): * runtime/JSGlobalData.h: * wtf/Platform.h: 2011-08-02 Filip Pizlo <fpizlo@apple.com> JSC does a GC even when the heap still has free pages https://bugs.webkit.org/show_bug.cgi?id=65445 Reviewed by Oliver Hunt. If the high watermark is not reached, then we allocate new blocks as before. If the current watermark does reach (or exceed) the high watermark, then we check if there is a block on the free block pool. If there is, we simply allocation from it. If there isn't, we invoke a collectin as before. This effectively couples the elastic scavenging to the collector's decision function. That is, if an application rapidly varies its heap usage (sometimes using more and sometimes less) then the collector will not thrash as it used to. But if heap usage drops and stays low then the scavenger thread and the GC will eventually reach a kind of consensus: the GC will set the watermark low because of low heap usage, and the scavenger thread will steadily eliminate pages from the free page pool, until the size of the free pool is below the high watermark. On command-line, this is neutral on SunSpider and Kraken and a 3% win on V8. In browser, this is a 1% win on V8 and neutral on the other two. * heap/Heap.cpp: (JSC::Heap::allocateSlowCase): (JSC::Heap::allocateBlock): * heap/Heap.h: 2011-08-02 Jeff Miller <jeffm@apple.com> Move WTF_USE_AVFOUNDATION from JavaScriptCore/wtf/platform.h to WebCore/config.h https://bugs.webkit.org/show_bug.cgi?id=65552 Since this is a WebCore feature, there's no need to define it in JavaScriptCore/wtf/platform.h. Reviewed by Adam Roben. * wtf/Platform.h: Removed WTF_USE_AVFOUNDATION. 2011-08-01 Jean-luc Brouillet <jeanluc@chromium.org> Removing old source files in gyp files that slow build https://bugs.webkit.org/show_bug.cgi?id=65503 Reviewed by Adam Barth. A number of stale files are listed in the gyp files. These slow the build on Visual Studio 2010. Removing them. * JavaScriptCore.gypi: 2011-07-14 David Levin <levin@chromium.org> currentThread is too slow! https://bugs.webkit.org/show_bug.cgi?id=64577 Reviewed by Darin Adler and Dmitry Titov. The problem is that currentThread results in a pthread_once call which always takes a lock. With this change, currentThread is 10% faster than isMainThread in release mode and only 5% slower than isMainThread in debug. * wtf/ThreadIdentifierDataPthreads.cpp: (WTF::ThreadIdentifierData::initializeOnce): Remove the pthread once stuff which is no longer needed because this is called from initializeThreading(). (WTF::ThreadIdentifierData::identifier): Remove the initializeKeyOnce call because intialization of the pthread key should already be done. (WTF::ThreadIdentifierData::initialize): Ditto. * wtf/ThreadIdentifierDataPthreads.h: * wtf/ThreadingPthreads.cpp: (WTF::initializeThreading): Acquire the pthread key here. 2011-08-01 Filip Pizlo <fpizlo@apple.com> DFG JIT sometimes creates speculation check data structures that have invalid information about the format of a register https://bugs.webkit.org/show_bug.cgi?id=65490 Reviewed by Gavin Barraclough. The code now makes sure to (1) always have correct and up-to-date information about register format at the time that a speculation check is emitted, (2) assert that speculation data is correct inside the speculation check implementation, and (3) avoid creating speculation data altogether if compilation has already failed, since at that point the format data is almost guaranteed to be bogus. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::EntryLocation::EntryLocation): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculationCheck::SpeculationCheck): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::speculationCheck): 2011-08-01 Filip Pizlo <fpizlo@apple.com> REGRESSION(r92092): Build fails on 64 bit https://bugs.webkit.org/show_bug.cgi?id=65458 Reviewed by Oliver Hunt. The build was broken because some compilers were smart enough to see an array index out of bounds due to the decision fuction for when to go from precise size classes to imprecise size classes being broken: it would assume that sizes in the range 97..128 belonged to a precise size class when in fact they belonged to an imprecise one. In fact, the code would have run correctly, by way of a fluke, because though the 4th precise size class (for 97..128) didn't exist, the next array over from m_preciseSizeClasses was m_impreciseSizeClasses, and its first entry would have been a size class that is appropriate for allocations in the range 97..128. However, this relies on specific ordering of fields in NewSpace, so it's still a bug. This fixes the bug by ensuring that allocations larger than 96 use the imprecise size classes. * heap/NewSpace.h: (JSC::NewSpace::sizeClassFor): 2011-07-31 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64679 Fix bugs in Array.prototype this handling. Unreviewed - rolling out r91290. Looks like the wild wild web isn't ready for this yet. This change broke http://slides.html5rocks.com/#landing-slide. Interestingly, this might only be due to our lack of bind support - it looks like this site is calling Array.prototype.slice as a part of its bind implementation. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncJoin): (JSC::arrayProtoFuncConcat): (JSC::arrayProtoFuncPop): (JSC::arrayProtoFuncPush): (JSC::arrayProtoFuncReverse): (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSlice): (JSC::arrayProtoFuncSort): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): (JSC::arrayProtoFuncFilter): (JSC::arrayProtoFuncMap): (JSC::arrayProtoFuncEvery): (JSC::arrayProtoFuncForEach): (JSC::arrayProtoFuncSome): (JSC::arrayProtoFuncReduce): (JSC::arrayProtoFuncReduceRight): (JSC::arrayProtoFuncIndexOf): (JSC::arrayProtoFuncLastIndexOf): 2011-07-31 Filip Pizlo <fpizlo@apple.com> JSC GC lays out size classes under wrong assumptions about expected object size. https://bugs.webkit.org/show_bug.cgi?id=65437 Reviewed by Oliver Hunt. Changed the atom size - which is both the smallest allocation size and the smallest possible stepping unit for size class spacing - from 8 bytes to 4 pointer-size words. This is a 1% win on SunSpider. * heap/MarkedBlock.h: 2011-07-31 Filip Pizlo <fpizlo@apple.com> DFG non-speculative JIT does not optimize PutByVal https://bugs.webkit.org/show_bug.cgi?id=65424 Reviewed by Gavin Barraclough. Added code to emit PutByVal inline fast path. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): 2011-07-31 Filip Pizlo <fpizlo@apple.com> The JSC garbage collector returns memory to the operating system too eagerly. https://bugs.webkit.org/show_bug.cgi?id=65382 Reviewed by Oliver Hunt. This introduces a memory reuse model similar to the one in FastMalloc. A periodic scavenger thread runs in the background and returns half the free memory to the OS on each timer fire. New block allocations first attempt to get the memory from the collector's internal pool, reverting to OS allocation only when this pool is empty. * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::~Heap): (JSC::Heap::destroy): (JSC::Heap::waitForRelativeTimeWhileHoldingLock): (JSC::Heap::waitForRelativeTime): (JSC::Heap::blockFreeingThreadStartFunc): (JSC::Heap::blockFreeingThreadMain): (JSC::Heap::allocateBlock): (JSC::Heap::freeBlocks): (JSC::Heap::releaseFreeBlocks): * heap/Heap.h: * heap/MarkedBlock.cpp: (JSC::MarkedBlock::destroy): (JSC::MarkedBlock::MarkedBlock): (JSC::MarkedBlock::initForCellSize): (JSC::MarkedBlock::reset): * heap/MarkedBlock.h: * wtf/Platform.h: 2011-07-30 Filip Pizlo <fpizlo@apple.com> DFG JIT speculation failure pass sometimes forgets to emit code to move certain registers. https://bugs.webkit.org/show_bug.cgi?id=65421 Reviewed by Oliver Hunt. Restructured the offending loops (for gprs and fprs). It's once again possible to use spreadsheets on docs.google.com. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): 2011-07-30 Patrick Gansterer <paroga@webkit.org> Remove inclusion of MainThread.h from Threading.h https://bugs.webkit.org/show_bug.cgi?id=65081 Reviewed by Darin Adler. Add missing and remove unneeded include statements for MainThread. * wtf/CryptographicallyRandomNumber.cpp: * wtf/Threading.h: * wtf/ThreadingPthreads.cpp: * wtf/text/StringStatics.cpp: 2011-07-30 Oliver Hunt <oliver@apple.com> Reduce the size of JSGlobalObject slightly https://bugs.webkit.org/show_bug.cgi?id=65417 Reviewed by Dan Bernstein. Push a few members that either aren't commonly used, or aren't frequently accessed into a separate struct. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::init): (JSC::JSGlobalObject::WeakMapsFinalizer::finalize): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::JSGlobalObjectRareData::JSGlobalObjectRareData): (JSC::JSGlobalObject::createRareDataIfNeeded): (JSC::JSGlobalObject::setProfileGroup): (JSC::JSGlobalObject::profileGroup): (JSC::JSGlobalObject::registerWeakMap): (JSC::JSGlobalObject::deregisterWeakMap): 2011-07-30 Balazs Kelemen <kbalazs@webkit.org> MessageQueue::waitForMessageFilteredWithTimeout can triggers an assertion https://bugs.webkit.org/show_bug.cgi?id=65263 Reviewed by Dmitry Titov. * wtf/Deque.h: (WTF::::operator): Don't check the validity of an iterator that will be reassigned right now. * wtf/MessageQueue.h: (WTF::::removeIf): Revert r51198 as I beleave this is the better solution for the problem that was solved by that. 2011-07-29 Filip Pizlo <fpizlo@apple.com> JSC GC zombie support no longer works, and is likely no longer needed. https://bugs.webkit.org/show_bug.cgi?id=65404 Reviewed by Darin Adler. This removes zombies, because they no longer work, are not tested, are probably not needed, and are getting in the way of GC optimization work. * JavaScriptCore.xcodeproj/project.pbxproj: * heap/Handle.h: (JSC::HandleConverter::operator->): (JSC::HandleConverter::operator*): * heap/HandleHeap.cpp: (JSC::HandleHeap::isValidWeakNode): * heap/Heap.cpp: (JSC::Heap::destroy): (JSC::Heap::collect): * heap/MarkedBlock.cpp: (JSC::MarkedBlock::sweep): * heap/MarkedBlock.h: (JSC::MarkedBlock::clearMarks): * interpreter/Register.h: (JSC::Register::Register): (JSC::Register::operator=): * runtime/ArgList.h: (JSC::MarkedArgumentBuffer::append): (JSC::ArgList::ArgList): * runtime/JSCell.cpp: (JSC::isZombie): * runtime/JSCell.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::clearBuiltinStructures): * runtime/JSGlobalData.h: * runtime/JSValue.h: * runtime/JSValueInlineMethods.h: (JSC::JSValue::JSValue): * runtime/JSZombie.cpp: Removed. * runtime/JSZombie.h: Removed. * runtime/WriteBarrier.h: (JSC::WriteBarrierBase::setEarlyValue): (JSC::WriteBarrierBase::operator*): (JSC::WriteBarrierBase::setWithoutWriteBarrier): * wtf/Platform.h: 2011-07-29 Filip Pizlo <fpizlo@apple.com> DFG JIT verbose mode provides no details about predictions https://bugs.webkit.org/show_bug.cgi?id=65389 Reviewed by Darin Adler. Added a print-out of the predictions to the IR dump, with names as follows: "p-bottom" = the parser made no predictions "p-int32" = the parser predicted int32 ... (same for array, cell, double, number) "p-top" = the parser made conflicting predictions which will be ignored. * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::predictionToString): 2011-07-29 Filip Pizlo <fpizlo@apple.com> DFG JIT does not have any way of undoing double speculation. https://bugs.webkit.org/show_bug.cgi?id=65334 Reviewed by Gavin Barraclough. This adds code to do a branchConvertDoubleToInt on specualtion failure. This is performance-neutral on most benchmarks but does result in a slight improvement in Kraken. * dfg/DFGJITCompiler.cpp: (JSC::DFG::GeneralizedRegister::moveTo): (JSC::DFG::GeneralizedRegister::swapWith): (JSC::DFG::ShuffledRegister::handleNonCyclingPermutation): (JSC::DFG::ShuffledRegister::handleCyclingPermutation): (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): 2011-07-29 Filip Pizlo <fpizlo@apple.com> Crash when opening docs.google.com https://bugs.webkit.org/show_bug.cgi?id=65327 Reviewed by Gavin Barraclough. The speculative JIT was only checking whether a value is an array when we had already checked that it was, rather then when we hadn't. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-07-28 Oliver Hunt <oliver@apple.com> *_list instructions are only used in one place, where the code is wrong. https://bugs.webkit.org/show_bug.cgi?id=65348 Reviewed by Darin Adler. Simply remove the instructions and all users. Speeds up the interpreter slightly due to code motion, but otherwise has no effect (because none of the _list instructions are ever used). * bytecode/CodeBlock.cpp: (JSC::isPropertyAccess): (JSC::CodeBlock::dump): (JSC::CodeBlock::visitStructures): * bytecode/Instruction.h: * bytecode/Opcode.h: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): 2011-07-28 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=65325 Performance tweak to parseInt Reviewed by Oliver Hunt. * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncParseInt): - This change may an existing optimization redundant, cleanup from Darin's comments, plus fix existing bugs. 2011-07-28 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=65325 Performance tweak to parseInt Reviewed by Oliver Hunt. * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncParseInt): - parseInt applied to small positive numbers = floor. 2011-07-28 Dan Bernstein <mitz@apple.com> Build fix. * runtime/Executable.cpp: (JSC::FunctionExecutable::compileForCallInternal): 2011-07-28 Kent Tamura <tkent@chromium.org> Improve StringImpl::stripWhiteSpace() and simplifyWhiteSpace(). https://bugs.webkit.org/show_bug.cgi?id=65300 Reviewed by Darin Adler. r91837 had performance regression of StringImpl::stripWhiteSpace() and simplifyWhiteSpace(). This changes the code so that compilers generates code equivalent to r91836 or piror. * wtf/text/StringImpl.cpp: (WTF::StringImpl::stripMatchedCharacters): A template member function for stripWhiteSpace(). This function takes a functor. (WTF::UCharPredicate): A functor for generic predicate for single UChar argument. (WTF::SpaceOrNewlinePredicate): A special functor for isSpaceOrNewline(). (WTF::StringImpl::stripWhiteSpace): Use stripmatchedCharacters(). (WTF::StringImpl::simplifyMatchedCharactersToSpace): A template member function for simplifyWhiteSpace(). (WTF::StringImpl::simplifyWhiteSpace): Use simplifyMatchedCharactersToSpace(). * wtf/text/StringImpl.h: 2011-07-27 Dmitry Lomov <dslomov@google.com> [chromium] Turn on WTF_MULTIPLE_THREADS. https://bugs.webkit.org/show_bug.cgi?id=61017 The patch turns on WTF_MULTIPLE_THREADS in chromium and pushes some relevant initializations from JSC::initializeThreading to WTF::initializeThreading. Reviewed by David Levin. * runtime/InitializeThreading.cpp: (JSC::initializeThreadingOnce): * wtf/FastMalloc.cpp: (WTF::isForbidden): (WTF::fastMallocForbid): (WTF::fastMallocAllow): * wtf/Platform.h: * wtf/ThreadingPthreads.cpp: (WTF::initializeThreading): * wtf/ThreadingWin.cpp: (WTF::initializeThreading): * wtf/gtk/ThreadingGtk.cpp: (WTF::initializeThreading): * wtf/qt/ThreadingQt.cpp: (WTF::initializeThreading): 2011-07-27 Mark Hahnenberg <mhahnenberg@apple.com> Remove operator new from JSCell https://bugs.webkit.org/show_bug.cgi?id=64999 Reviewed by Oliver Hunt. Removed the implementation of operator new in JSCell, so any further uses will not successfully link. Also removed any remaining uses of operator new. * API/JSContextRef.cpp: * debugger/DebuggerActivation.h: (JSC::DebuggerActivation::create): * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): (JSC::Interpreter::createExceptionScope): (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/JSCell.h: * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::create): * runtime/JSStaticScopeObject.h: (JSC::JSStaticScopeObject::create): (JSC::JSStaticScopeObject::JSStaticScopeObject): * runtime/StrictEvalActivation.h: (JSC::StrictEvalActivation::create): 2011-07-27 Filip Pizlo <fpizlo@apple.com> DFG graph has no notion of double prediction. https://bugs.webkit.org/show_bug.cgi?id=65234 Reviewed by Gavin Barraclough. Added the notion of PredictDouble, and PredictNumber, which is the least upper bound of PredictInt32 and PredictDouble. Least upper bound is defined as the bitwise-or of two predictions. Bottom is defined as 0, and Top is defined as all bits being set. Added the ability to explicitly distinguish between a node having had a prediction associated with it, and that prediction still being valid (i.e. no conflicting predictions have also been added). Used this to guard the speculative JIT from speculating Int32 in cases where the graph knows that the value is double, which currently only happens for GetLocal nodes on arguments which were double at compile-time. * dfg/DFGGraph.cpp: (JSC::DFG::Graph::predictArgumentTypes): * dfg/DFGGraph.h: (JSC::DFG::isCellPrediction): (JSC::DFG::isArrayPrediction): (JSC::DFG::isInt32Prediction): (JSC::DFG::isDoublePrediction): (JSC::DFG::isNumberPrediction): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::initializeVariableTypes): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isRegisterDataFormatDouble): 2011-07-27 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=65294 DFG JIT - may speculate based on wrong arguments. Reviewed by Oliver Hunt In the case of a DFG compiled function calling to and compiling a second function that also compiles through the DFG JIT (i.e. compilation triggered with DFGOperations.cpp), we call compileFor passing the caller functions exec state, rather than the callee's. This may lead to mis-optimization, since the DFG compiler will example the exec state's arguments on the assumption that these will be passed to the callee - it is wanting the callee exec state, not the caller's exec state. Fixing this for all cases of compilation is tricksy, due to the way the numeric sort function is compiled, & the structure of the calls in the Interpreter::execute methods. Only fix for compilation from the JIT, in other calls don't speculate based on arguments for now. * dfg/DFGOperations.cpp: * runtime/Executable.cpp: (JSC::tryDFGCompile): (JSC::tryDFGCompileFunction): (JSC::FunctionExecutable::compileForCallInternal): * runtime/Executable.h: (JSC::FunctionExecutable::compileForCall): (JSC::FunctionExecutable::compileFor): 2011-07-27 Oliver Hunt <oliver@apple.com> Handle callback oriented JSONP https://bugs.webkit.org/show_bug.cgi?id=65271 Reviewed by Gavin Barraclough. Handle the callback oriented versions of JSONP. The Literal parser now handles <Identifier> (. <Identifier>)* (jsonData). * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): * runtime/LiteralParser.cpp: (JSC::LiteralParser::tryJSONPParse): (JSC::LiteralParser::Lexer::lex): * runtime/LiteralParser.h: 2011-07-27 Stephanie Lewis <slewis@apple.com> Revert http://trac.webkit.org/changeset/90415. Caused a 5% sunspider regression in-browser. Unreviewed rollout. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): * heap/Heap.cpp: (JSC::Heap::collectAllGarbage): * heap/MarkStack.h: (JSC::MarkStack::MarkStack): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::releaseExecutableMemory): * runtime/RegExp.cpp: (JSC::RegExp::compile): (JSC::RegExp::invalidateCode): * runtime/RegExp.h: 2011-07-27 Shinya Kawanaka <shinyak@google.com> Added an interface to take IsWhiteSpaceFunctionPtr. https://bugs.webkit.org/show_bug.cgi?id=57746 Reviewed by Kent Tamura. * wtf/text/StringImpl.cpp: (WTF::StringImpl::stripWhiteSpace): Added an interface to take IsWhiteSpaceFunctionPtr. (WTF::StringImpl::simplifyWhiteSpace): ditto. * wtf/text/StringImpl.h: * wtf/text/WTFString.cpp: (WTF::String::stripWhiteSpace): ditto. (WTF::String::simplifyWhiteSpace): ditto. * wtf/text/WTFString.h: 2011-07-27 Filip Pizlo <fpizlo@apple.com> DFG JIT speculation failure code performs incorrect conversions in the case where two registers need to be swapped. https://bugs.webkit.org/show_bug.cgi?id=65233 Reviewed by Gavin Barraclough. * dfg/DFGJITCompiler.cpp: (JSC::DFG::GeneralizedRegister::swapWith): 2011-07-26 Mark Hahnenberg <mhahnenberg@apple.com> reduce and reduceRight bind callback's this to null rather than undefined https://bugs.webkit.org/show_bug.cgi?id=62264 Reviewed by Oliver Hunt. Fixed Array.prototype.reduce and Array.prototype.reduceRight so that they behave correctly when calling the callback function without an argument for this, which means it should be undefined according to ES 15.4.4.21 and 15.4.4.22. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncReduce): (JSC::arrayProtoFuncReduceRight): 2011-07-26 Filip Pizlo <fpizlo@apple.com> JSC command-line tool does not come with any facility for measuring time precisely. https://bugs.webkit.org/show_bug.cgi?id=65223 Reviewed by Gavin Barraclough. Exposed WTF::currentTime() as currentTimePrecise(). * jsc.cpp: (GlobalObject::GlobalObject): (functionPreciseTime): 2011-07-26 Filip Pizlo <fpizlo@apple.com> DFG speculative JIT never emits inline double comparisons, even when it would be obvious more efficient to do so. https://bugs.webkit.org/show_bug.cgi?id=65212 Reviewed by Gavin Barraclough. This handles the obvious case of inlining double comparisons: it only addresses the speculative JIT, and only for fused compare/branch sequences. But it does handle the case where both operands are double (and there is no slow path), or where one operand is double and the other is unknown type (in which case it attempts to unbox the double, otherwise taking slow path). This is an 0.8% speed-up on SunSpider. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::convertToDouble): (JSC::DFG::SpeculativeJIT::compilePeepHoleDoubleBranch): (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isRegisterDataFormatDouble): (JSC::DFG::SpeculativeJIT::shouldSpeculateInteger): 2011-07-26 Filip Pizlo <fpizlo@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64969 DFG JIT generates inefficient code for speculation failures. Reviewed by Gavin Barraclough. This implements a speculation failure strategy where (1) values spilled on non-speculative but not spilled on speculative are spilled, (2) values that are in registers on both paths are rearranged without ever touching memory, and (3) values spilled on speculative but not spilled on non-speculative are filled. The register shuffling is the most interesting part of this patch. It constructs a permutation graph for registers. Each node represents a register, and each directed edge corresponds to the register's value having to be moved to a different register as part of the shuffling. This is a directed graph where each node may only have 0 or 1 incoming edges, and 0 or 1 outgoing edges. The algorithm then first finds maximal non-cyclic subgraphs where all nodes in the subgraph are reachable from a start node. Such subgraphs always resemble linked lists, and correspond to simply moving the value in the second-to-last register into the last register, and then moving the value in the third-to-last register into the second-to-last register, and so on. Once these subgraphs are taken care of, the remaining subgraphs are cycles, and are handled using either (a) conversion or no-op if the cycle involves one node, (b) swap if it involves two nodes, or (c) a cyclic shuffle involving a scratch register if there are three or more nodes. * dfg/DFGGenerationInfo.h: (JSC::DFG::needDataFormatConversion): * dfg/DFGJITCompiler.cpp: (JSC::DFG::GeneralizedRegister::GeneralizedRegister): (JSC::DFG::GeneralizedRegister::createGPR): (JSC::DFG::GeneralizedRegister::createFPR): (JSC::DFG::GeneralizedRegister::dump): (JSC::DFG::GeneralizedRegister::findInSpeculationCheck): (JSC::DFG::GeneralizedRegister::findInEntryLocation): (JSC::DFG::GeneralizedRegister::previousDataFormat): (JSC::DFG::GeneralizedRegister::nextDataFormat): (JSC::DFG::GeneralizedRegister::convert): (JSC::DFG::GeneralizedRegister::moveTo): (JSC::DFG::GeneralizedRegister::swapWith): (JSC::DFG::ShuffledRegister::ShuffledRegister): (JSC::DFG::ShuffledRegister::isEndOfNonCyclingPermutation): (JSC::DFG::ShuffledRegister::handleNonCyclingPermutation): (JSC::DFG::ShuffledRegister::handleCyclingPermutation): (JSC::DFG::ShuffledRegister::lookup): (JSC::DFG::lookupForRegister): (JSC::DFG::NodeToRegisterMap::Tuple::Tuple): (JSC::DFG::NodeToRegisterMap::NodeToRegisterMap): (JSC::DFG::NodeToRegisterMap::set): (JSC::DFG::NodeToRegisterMap::end): (JSC::DFG::NodeToRegisterMap::find): (JSC::DFG::NodeToRegisterMap::clear): (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): (JSC::DFG::JITCompiler::linkSpeculationChecks): * dfg/DFGJITCompiler.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::EntryLocation::EntryLocation): * dfg/DFGNonSpeculativeJIT.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculationCheck::SpeculationCheck): * dfg/DFGSpeculativeJIT.h: 2011-07-26 Oliver Hunt <oliver@apple.com> Buffer overflow creating error messages for JSON.parse https://bugs.webkit.org/show_bug.cgi?id=65211 Reviewed by Darin Adler. Parse string length to the UString constructor. * runtime/LiteralParser.cpp: (JSC::LiteralParser::parse): 2011-07-26 Mark Hahnenberg <mhahnenberg@apple.com> Refactor automatically generated JS DOM bindings to replace operator new with static create methods https://bugs.webkit.org/show_bug.cgi?id=64732 Reviewed by Oliver Hunt. Replacing the public constructors in the automatically generated JS DOM bindings with static create methods. JSByteArray is used by several of these bindings in WebCore. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/JSByteArray.cpp: (JSC::JSByteArray::create): * runtime/JSByteArray.h: 2011-07-26 Alexis Menard <alexis.menard@openbossa.org> Unreviewed build fix for Qt/Linux. On platforms with no glib and gstreamer we should not build javascriptcore with the Glib support. This is related to http://trac.webkit.org/changeset/91752. * wtf/wtf.pri: 2011-07-26 Juan C. Montemayor <jmont@apple.com> JSON errors should be informative https://bugs.webkit.org/show_bug.cgi?id=63339 Added error messages to the JSON Parser. Reviewed by Oliver Hunt. * runtime/JSONObject.cpp: (JSC::JSONProtoFuncParse): * runtime/LiteralParser.cpp: (JSC::LiteralParser::Lexer::lex): (JSC::LiteralParser::Lexer::lexString): (JSC::LiteralParser::Lexer::lexNumber): (JSC::LiteralParser::parse): * runtime/LiteralParser.h: (JSC::LiteralParser::getErrorMessage): (JSC::LiteralParser::Lexer::sawError): (JSC::LiteralParser::Lexer::getErrorMessage): 2011-07-26 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r91746. http://trac.webkit.org/changeset/91746 https://bugs.webkit.org/show_bug.cgi?id=65180 It broke SL build (Requested by Ossy on #webkit). * wtf/text/StringImpl.cpp: (WTF::StringImpl::stripWhiteSpace): (WTF::StringImpl::simplifyWhiteSpace): * wtf/text/StringImpl.h: * wtf/text/WTFString.cpp: * wtf/text/WTFString.h: 2011-07-26 Alexis Menard <alexis.menard@openbossa.org> Reviewed by Andreas Kling. [Qt] Change default backend to use GStreamer on Linux and QuickTime on Mac. https://bugs.webkit.org/show_bug.cgi?id=63472 Enable the bits needed for GStreamer only when QtMultimedia is not used. * wtf/wtf.pri: 2011-07-26 Shinya Kawanaka <shinyak@google.com> Added an interface to take IsWhiteSpaceFunctionPtr. https://bugs.webkit.org/show_bug.cgi?id=57746 Reviewed by Kent Tamura. * wtf/text/StringImpl.cpp: (WTF::StringImpl::stripWhiteSpace): Added an interface to take IsWhiteSpaceFunctionPtr. (WTF::StringImpl::simplifyWhiteSpace): ditto. * wtf/text/StringImpl.h: * wtf/text/WTFString.cpp: (WTF::String::stripWhiteSpace): ditto. (WTF::String::simplifyWhiteSpace): ditto. * wtf/text/WTFString.h: 2011-07-25 Filip Pizlo <fpizlo@apple.com> DFG non-speculative JIT emits inefficient code for arithmetic involving two registers https://bugs.webkit.org/show_bug.cgi?id=65160 Reviewed by Gavin Barraclough. The non-speculative JIT now emits inline code for double arithmetic, but still attempts integer arithmetic first. This is a speed-up on SunSpider (albeit a small one), and a large speed-up on Kraken. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::basicArithOp): 2011-07-25 Ryuan Choi <ryuan.choi@samsung.com> [EFL] Build break with --debug after r89153. https://bugs.webkit.org/show_bug.cgi?id=65150 Unreviewed build fix. * wtf/CMakeListsEfl.txt: Add missing libraries. 2011-07-25 Filip Pizlo <fpizlo@apple.com> DFG non-speculative JIT emits obviously inefficient code for arithmetic where one operand is a constant. https://bugs.webkit.org/show_bug.cgi?id=65146 Reviewed by Gavin Barraclough. Changed the code to emit double arithmetic inline. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::knownConstantArithOp): 2011-07-25 Filip Pizlo <fpizlo@apple.com> DFG JIT bytecode parser misuses pointers into objects allocated as part of a WTF::Vector. https://bugs.webkit.org/show_bug.cgi?id=65128 Reviewed by Gavin Barraclough. The bytecode parser code seems to be right to have a DFGNode& phiNode reference into the graph, since this makes the code greatly more readable. This patch thus makes the minimal change necessary to make the code right: it uses a pointer (to disambiguate between reloading the pointer and performing a copy from one location of the vector to another) and reloads it after the calls to addToGraph(). * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::processPhiStack): 2011-07-25 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r91686. http://trac.webkit.org/changeset/91686 https://bugs.webkit.org/show_bug.cgi?id=65144 1.5% regression in JSC (Requested by jmontemayor on #webkit). * runtime/JSONObject.cpp: (JSC::JSONProtoFuncParse): * runtime/LiteralParser.cpp: (JSC::LiteralParser::Lexer::lex): (JSC::LiteralParser::Lexer::lexString): (JSC::LiteralParser::Lexer::lexNumber): (JSC::LiteralParser::parse): * runtime/LiteralParser.h: 2011-07-25 Jon Lee <jonlee@apple.com> Assertion called in ExecutableBase::generatedJITCodeForCall() when JIT is not available https://bugs.webkit.org/show_bug.cgi?id=65132 <rdar://problem/9836297> Reviewed by Oliver Hunt. Make sure the JIT is available to use before running the following calls: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::unlinkCalls): Added check, return early if JIT is not available. * bytecode/CodeBlock.h: (JSC::CodeBlock::addMethodCallLinkInfos): Added assertion. 2011-07-25 Juan C. Montemayor <jmont@apple.com> JSON errors should be informative https://bugs.webkit.org/show_bug.cgi?id=63339 Added error messages to the JSON Parser. Reviewed by Oliver Hunt. * runtime/JSONObject.cpp: (JSC::JSONProtoFuncParse): * runtime/LiteralParser.cpp: (JSC::LiteralParser::Lexer::lex): (JSC::LiteralParser::Lexer::lexString): (JSC::LiteralParser::Lexer::lexNumber): (JSC::LiteralParser::parse): * runtime/LiteralParser.h: (JSC::LiteralParser::getErrorMessage): (JSC::LiteralParser::Lexer::sawError): (JSC::LiteralParser::Lexer::getErrorMessage): 2011-07-25 Filip Pizlo <fpizlo@apple.com> X86-64 assembler emits three instructions instead of two for certain loads and stores. https://bugs.webkit.org/show_bug.cgi?id=65095 Reviewed by Gavin Barraclough. Simply made these four methods in the assembler use the scratch register, which they were previously avoiding. It still optimizes for the case where an absolute address memory accesses is using EAX. This results in a slight performance improvement. * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::load32): (JSC::MacroAssemblerX86_64::store32): (JSC::MacroAssemblerX86_64::loadPtr): (JSC::MacroAssemblerX86_64::storePtr): 2011-07-25 Ryuan Choi <ryuan.choi@samsung.com> [EFL] Implement EFL-specific current time and monotonicallyIncreasingTime. https://bugs.webkit.org/show_bug.cgi?id=64354 Use ecore_time_unix_get which returns unix time as double type for currentTime and ecore_time_get which uses monotonic clock for monotonicallyIncreasingTime. Reviewed by Kent Tamura. * wtf/CurrentTime.cpp: (WTF::currentTime): (WTF::monotonicallyIncreasingTime): 2011-07-22 Sommer Panage <panage@apple.com> Reviewed by Oliver Hunt. export JSContextCreateBacktrace as SPI in JSContextRefPrivate.h https://bugs.webkit.org/show_bug.cgi?id=64981 UIAutomation for iOS would like to support a Javascript backtrace in our error logs. Currently, the C API does not provide the tools to do this. However, the private API does expose the necessary functionality to get a backtrace (via Interpreter::retrieveLastCaller). We recognize this information may result in failure in the cases of programs run by 'eval', stack frames beneath host function call frames, and in programs run from other programs. Thus, we propose exporting our JSContextCreateBacktrace in JSContextRefPrivate.h. This will provide us with the tools we need while not advertising an API that isn't really ready for full use. * API/JSContextRef.cpp: * API/JSContextRefPrivate.h: * JavaScriptCore.exp: 2011-07-22 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=65051 DFG JIT - Enable by default for mac platform on x86-64. Rubber Stamped by Geoff Garen. This is now a performance progression. * wtf/Platform.h: - Removed definition of ENABLE_DFG_JIT_RESTRICTIONS. 2011-07-22 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=65047 DFG JIT - Add support for op_resolve/op_resolve_base Reviewed by Sam Weinig. These are necessary for any significant eval code coverage (and as such increase LayoutTest coverage). * dfg/DFGAliasTracker.h: (JSC::DFG::AliasTracker::recordResolve): - Conservatively blow aliasing optimizations for now. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): - Add support for op_resolve/op_resolve_base. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::callOperation): - Add call with exec, identifer aguments. * dfg/DFGNode.h: - Add new node types. (JSC::DFG::Node::hasIdentifier): - Resolve nodes have identifiers, too! * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): - Add generation for new Nodes. * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: - Added new operations. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): - Add generation for new Nodes. 2011-07-22 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=65036 Messing with the register allocation within flow control = badness. Reviewed by Sam Weinig. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): - Fix register allocation. 2011-07-22 Mark Hahnenberg <mhahnenberg@apple.com> Date.prototype.toISOString doesn't handle negative years or years > 9999 correctly. https://bugs.webkit.org/show_bug.cgi?id=63986 Reviewed by Geoffrey Garen. Changed the implementation of Date.prototype.toISOString() to use the extended year format (+/-yyyyyy) for years outside of [0,9999] to be in compliance with ES 15.9.1.15.1. * runtime/DatePrototype.cpp: (JSC::dateProtoFuncToISOString): 2011-07-21 Gavin Barraclough <barraclough@apple.com> Windows build fix * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-07-21 Ryosuke Niwa <rniwa@webkit.org> Build fix after r91555. * JavaScriptCore.exp: 2011-07-21 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=19271 eliminate PIC branches by changing NaN handling in JSValue::toNumber Reviewed by Sam Weinig. Moving the non-numeric cases out of line seems to be a consistent win on SunSpider for me, to the order of about 0.5%. * runtime/JSCell.h: (JSC::JSCell::JSValue::toNumber): - Changed to only handle values that are already numbers, moce non-numeric cases out of line. * runtime/JSValue.cpp: (JSC::JSValue::toNumberSlowCase): - Added toNumberSlowCase, handling non-numeric cases. * runtime/JSValue.h: - Add declaration of toNumberSlowCase. 2011-07-21 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64875 Use of `yield` keyword is broken Reviewed by Sam Weinig. * parser/Lexer.cpp: (JSC::Lexer::parseIdentifier): - The bug here is that a successful match of a RESERVED_IF_STRICT token from parseKeyword is being nullified back to IDENT. The problem is that in the case of IDENT matches parseKeyword should not move the lexer's input position, but in the case of RESERVED_IF_STRICT it has done so. 2011-07-21 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64900 Function.prototype.apply should accept an array-like object as its second argument Reviewed by Sam Weinig. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/FunctionPrototype.cpp: (JSC::functionProtoFuncApply): - Remove the type error if object is not an array. 2011-07-21 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64964 DFG JIT - Enable support for eval code Reviewed by Sam Weinig. This is basically the same as program code, to the JIT! * bytecode/Opcode.cpp: * bytecode/Opcode.h: - Enable opcodeNames in !NDEBUG builds. * dfg/DFGOperations.cpp: - Fix a bug exposed by eval support, throw correct type error for new. * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): - Enable DFG JIT for eval code. 2011-07-20 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r91380. http://trac.webkit.org/changeset/91380 https://bugs.webkit.org/show_bug.cgi?id=64924 Caused assertion failures in Chromium's IndexedDB tests (Requested by rniwa on #webkit). * wtf/ThreadIdentifierDataPthreads.cpp: (WTF::ThreadIdentifierData::identifier): (WTF::ThreadIdentifierData::initialize): (WTF::ThreadIdentifierData::initializeKeyOnceHelper): (WTF::ThreadIdentifierData::initializeKeyOnce): * wtf/ThreadIdentifierDataPthreads.h: * wtf/ThreadingPthreads.cpp: (WTF::initializeThreading): 2011-07-20 Filip Pizlo <fpizlo@apple.com> DFG non-speculative JIT does not use() the aliased GetByVal, resulting in bloated use counts. https://bugs.webkit.org/show_bug.cgi?id=64911 Reviewed by Gavin Barraclough. Inserted a call to use() for the aliased GetByVal. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): 2011-07-20 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64909 DFG JIT - Missing ToInt32 conversions for double constants. Reviewed by Sam Weinig. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::toInt32): - We cannot trivially omit ToInt32 conversions on double constants. 2011-07-20 Filip Pizlo <fpizlo@apple.com> DFG speculative JIT sometimes claims to use compare operands twice, leading to use count corruption. https://bugs.webkit.org/show_bug.cgi?id=64903 Reviewed by Gavin Barraclough. Move the calls to use() in SpeculativeJIT::compare() so that they only happen if the JITCodeGenerator's helper method (which also calls use()) is not called. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compare): 2011-07-20 Oliver Hunt <oliver@apple.com> Don't throw away code when JSGarbageCollect API is called https://bugs.webkit.org/show_bug.cgi?id=64894 Reviewed by Sam Weinig. Just call collectAllGarbage. That will clean up all unneeded code without causing any pathological recompilation problems. * API/JSBase.cpp: (JSGarbageCollect): 2011-07-20 Oliver Hunt <oliver@apple.com> Codeblock doesn't visit cached structures in global resolve instructions https://bugs.webkit.org/show_bug.cgi?id=64889 Reviewed by Sam Weinig. Visit the global resolve instructions. This fixes a couple of random crashes seen in the jquery tests when using the interpreter. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): 2011-07-20 James Robinson <jamesr@chromium.org> Revert worker and WebKit2 runloops to use currentTime() for scheduling instead of the monotonic clock https://bugs.webkit.org/show_bug.cgi?id=64841 Reviewed by Mark Rowe. http://trac.webkit.org/changeset/91206 converted most of WebKit's deferred work scheduling to using the monotonic clock instead of WTF::currentTime(). This broke many plugin tests on WebKit2 for reasons that are unclear. This reverts everything except for WebCore::ThreadTimers back to the previous behavior. * wtf/ThreadingPthreads.cpp: (WTF::ThreadCondition::timedWait): * wtf/ThreadingWin.cpp: (WTF::absoluteTimeToWaitTimeoutInterval): * wtf/gtk/ThreadingGtk.cpp: (WTF::ThreadCondition::timedWait): * wtf/qt/ThreadingQt.cpp: (WTF::ThreadCondition::timedWait): 2011-07-14 David Levin <levin@chromium.org> currentThread is too slow! https://bugs.webkit.org/show_bug.cgi?id=64577 Reviewed by Darin Adler and Dmitry Titov. The problem is that currentThread results in a pthread_once call which always takes a lock. With this change, currentThread is 10% faster than isMainThread in release mode and only 5% slower than isMainThread in debug. * wtf/ThreadIdentifierDataPthreads.cpp: (WTF::ThreadIdentifierData::initializeOnce): Remove the pthread once stuff which is no longer needed because this is called from initializeThreading(). (WTF::ThreadIdentifierData::identifier): Remove the initializeKeyOnce call because intialization of the pthread key should already be done. (WTF::ThreadIdentifierData::initialize): Ditto. * wtf/ThreadIdentifierDataPthreads.h: * wtf/ThreadingPthreads.cpp: (WTF::initializeThreading): Acquire the pthread key here. 2011-07-20 Mark Rowe <mrowe@apple.com> Fix the 32-bit build. * runtime/ObjectPrototype.cpp: (JSC::objectProtoFuncToString): 2011-07-19 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64678 Fix bugs in Object.prototype this handling. Reviewed by Darin Adler. Fix ES5.1 correctness issues identified by Mads Ager. * runtime/ObjectPrototype.cpp: (JSC::objectProtoFuncToString): - ES5.1 expects toString of undefined/null to produce "[object Undefined]"/"[object Null]". 2011-07-19 Mark Hahnenberg <mhahnenberg@apple.com> [JSC] WebKit allocates gigabytes of memory when doing repeated string concatenation https://bugs.webkit.org/show_bug.cgi?id=63918 Reviewed by Darin Adler. When allocating JSStrings during concatenation, we needed to call the Heap's reportExtraMemoryCost method due to additional string copying within several of the constructors when dealing with UStrings. This has been added to the UString version of the appendStringInConstruct method within the JSString class. * runtime/JSString.h: (JSC::RopeBuilder::JSString): (JSC::RopeBuilder::appendStringInConstruct): 2011-07-19 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64679 Fix bugs in Array.prototype this handling. Reviewed by Oliver Hunt. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncJoin): (JSC::arrayProtoFuncConcat): (JSC::arrayProtoFuncPop): (JSC::arrayProtoFuncPush): (JSC::arrayProtoFuncReverse): (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSlice): (JSC::arrayProtoFuncSort): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): (JSC::arrayProtoFuncFilter): (JSC::arrayProtoFuncMap): (JSC::arrayProtoFuncEvery): (JSC::arrayProtoFuncForEach): (JSC::arrayProtoFuncSome): (JSC::arrayProtoFuncReduce): (JSC::arrayProtoFuncReduceRight): (JSC::arrayProtoFuncIndexOf): (JSC::arrayProtoFuncLastIndexOf): - These methods should throw if this value is undefined. 2011-07-19 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64677 Fix bugs in String.prototype this handling. Reviewed by Oliver Hunt. undefined/null this values should throw TypeErrors, not convert to the global object, and primitive values should not be converted via object types. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncReplace): (JSC::stringProtoFuncCharAt): (JSC::stringProtoFuncCharCodeAt): (JSC::stringProtoFuncIndexOf): (JSC::stringProtoFuncLastIndexOf): (JSC::stringProtoFuncMatch): (JSC::stringProtoFuncSearch): (JSC::stringProtoFuncSlice): (JSC::stringProtoFuncSplit): (JSC::stringProtoFuncSubstr): (JSC::stringProtoFuncSubstring): (JSC::stringProtoFuncToLowerCase): (JSC::stringProtoFuncToUpperCase): (JSC::stringProtoFuncLocaleCompare): (JSC::stringProtoFuncBig): (JSC::stringProtoFuncSmall): (JSC::stringProtoFuncBlink): (JSC::stringProtoFuncBold): (JSC::stringProtoFuncFixed): (JSC::stringProtoFuncItalics): (JSC::stringProtoFuncStrike): (JSC::stringProtoFuncSub): (JSC::stringProtoFuncSup): (JSC::stringProtoFuncFontcolor): (JSC::stringProtoFuncFontsize): (JSC::stringProtoFuncAnchor): (JSC::stringProtoFuncLink): (JSC::trimString): - These methods should throw if this value is undefined, convert ToString directly, not via ToObject. 2011-07-19 Filip Pizlo <fpizlo@apple.com> DFG JIT sometimes emits spill code even when the respective values are never needed. https://bugs.webkit.org/show_bug.cgi?id=64774 Reviewed by Gavin Barraclough. The main high-level change is that it is now easier to call use() on a virtual register. JSValueOperand and its other-typed relatives now have a handy use() method, and jsValueResult() and friends now make it easier to pass UseChildrenCalledExplicitly. The rest of this patch hoists the call to use() as high as possible for all of those cases where either flushRegisters() or silentSpillAllRegisters() may be called. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::cachedGetMethod): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): (JSC::DFG::JITCodeGenerator::nonSpeculativeCompare): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeStrictEq): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeStrictEq): (JSC::DFG::JITCodeGenerator::nonSpeculativeStrictEq): (JSC::DFG::JITCodeGenerator::emitBranch): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::use): (JSC::DFG::JITCodeGenerator::integerResult): (JSC::DFG::JITCodeGenerator::jsValueResult): (JSC::DFG::IntegerOperand::use): (JSC::DFG::DoubleOperand::use): (JSC::DFG::JSValueOperand::use): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::valueToNumber): (JSC::DFG::NonSpeculativeJIT::valueToInt32): (JSC::DFG::NonSpeculativeJIT::knownConstantArithOp): (JSC::DFG::NonSpeculativeJIT::basicArithOp): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculateStrictInt32Operand::use): (JSC::DFG::SpeculateCellOperand::use): 2011-07-19 Xan Lopez <xlopez@igalia.com> ARMv7 backend broken, lacks 3 parameter rshift32 method https://bugs.webkit.org/show_bug.cgi?id=64571 Reviewed by Zoltan Herczeg. * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::rshift32): add missing rshift32 method. 2011-07-18 Filip Pizlo <fpizlo@apple.com> DFG JIT does not optimize strict equality as effectively as the old JIT does. https://bugs.webkit.org/show_bug.cgi?id=64759 Reviewed by Gavin Barraclough. This adds a more complete set of strict equality optimizations. If either operand is known numeric, then the code reverts to the old style of optimizing (first try integer comparison). Otherwise it uses the old JIT's trick of first simultaneously checking if both operands are either numbers or cells; if not then a fast path is taken. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeStrictEq): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeStrictEq): (JSC::DFG::JITCodeGenerator::nonSpeculativeStrictEq): * dfg/DFGJITCodeGenerator.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-07-18 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64760 DFG JIT - Should be able to compile program code. Reviewed by Geoff Garen. Add support for op_end, hooks to compile program code in Executable.cpp. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): - Add support for op_end * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileEntry): (JSC::DFG::JITCompiler::compileBody): (JSC::DFG::JITCompiler::link): - Added, separate out steps of compileFunction. (JSC::DFG::JITCompiler::compile): - Added, compile program code. (JSC::DFG::JITCompiler::compileFunction): - Sections separated out to helper functions. * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::JITCompiler): - Added m_exceptionCheckCount. * runtime/Executable.cpp: (JSC::tryDFGCompile): (JSC::tryDFGCompileFunction): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::compileForCallInternal): - Renamed tryDFGCompile to tryDFGCompileFunction, added tryDFGCompile to compile program code. 2011-07-18 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64678 Fix bugs in Object.prototype this handling. Reviewed by Oliver Hunt. undefined/null this values should throw TypeErrors, not convert to the global object, also, to toLocaleString should be calling the ToObject & invoking the object's toString function, even for values that are already strings. * runtime/ObjectPrototype.cpp: (JSC::objectProtoFuncValueOf): (JSC::objectProtoFuncHasOwnProperty): (JSC::objectProtoFuncIsPrototypeOf): (JSC::objectProtoFuncPropertyIsEnumerable): (JSC::objectProtoFuncToLocaleString): (JSC::objectProtoFuncToString): 2011-07-18 Filip Pizlo <fpizlo@apple.com> JSC GC lazy sweep does not inline the common cases of cell destruction. https://bugs.webkit.org/show_bug.cgi?id=64745 Reviewed by Oliver Hunt. This inlines the case of JSFinalObject destruction. * heap/MarkedBlock.cpp: (JSC::MarkedBlock::lazySweep): 2011-07-18 Oliver Hunt <oliver@apple.com> Interpreter build-fix * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): 2011-07-18 Filip Pizlo <fpizlo@apple.com> DFG JIT does not optimize equal-null comparisons and branches. https://bugs.webkit.org/show_bug.cgi?id=64659 Reviewed by Gavin Barraclough. Added a peephole-aware compare-to-null implementation to JITCodeGenerator, which is used by both the speculative and non-speculative JIT. Through the use of the new isNullConstant helper, the two JITs invoke the nonSpecualtiveCompareNull() helper instead of their regular comparison helpers when compiling CompareEq. Through the use of the new isKnownCell helper, the compare-null code will skip the is-a-cell check if the speculative JIT had been speculating cell. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::isKnownCell): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompareNull): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranchNull): (JSC::DFG::JITCodeGenerator::nonSpeculativeCompareNull): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::isNullConstant): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-07-18 James Robinson <jamesr@chromium.org> Timer scheduling should be based off the monotonic clock https://bugs.webkit.org/show_bug.cgi?id=64544 Reviewed by Darin Adler. Switches ThreadCondition::timedWait and related utility functions from currentTime() to monotonicallyIncreasingTime(). Add WTF::monotonicallyIncreasingTime() to list of exported functions so it can be accessed from WebCore/WebKit. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/ThreadingPthreads.cpp: (WTF::ThreadCondition::timedWait): * wtf/ThreadingWin.cpp: (WTF::absoluteTimeToWaitTimeoutInterval): * wtf/gtk/ThreadingGtk.cpp: (WTF::ThreadCondition::timedWait): * wtf/qt/ThreadingQt.cpp: (WTF::ThreadCondition::timedWait): 2011-07-18 Filip Pizlo <fpizlo@apple.com> JSC JIT does not inline GC allocation fast paths https://bugs.webkit.org/show_bug.cgi?id=64582 Reviewed by Oliver Hunt. This addresses inlining allocation for the easiest-to-allocate cases: op_new_object and op_create_this. Inlining GC allocation fast paths required three changes. First, the JSGlobalData now saves the vtable pointer of JSFinalObject, since that's what op_new_object and op_create_this allocate. Second, the Heap exposes a reference to the appropriate SizeClass, so that the JIT may inline accesses directly to the SizeClass for JSFinalObject allocations. And third, the JIT is extended with code to emit inline fast paths for GC allocation. A stub call is emitted in the case where the inline fast path fails. * heap/Heap.h: (JSC::Heap::sizeClassFor): (JSC::Heap::allocate): * jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases): * jit/JIT.h: * jit/JITInlineMethods.h: (JSC::JIT::emitAllocateJSFinalObject): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_new_object): (JSC::JIT::emitSlow_op_new_object): (JSC::JIT::emit_op_create_this): (JSC::JIT::emitSlow_op_create_this): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_new_object): (JSC::JIT::emitSlow_op_new_object): (JSC::JIT::emit_op_create_this): (JSC::JIT::emitSlow_op_create_this): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::storeVPtrs): * runtime/JSGlobalData.h: * runtime/JSObject.h: (JSC::JSFinalObject::JSFinalObject): (JSC::JSObject::offsetOfInheritorID): 2011-07-18 Mark Hahnenberg <mhahnenberg@apple.com> Refactor JSC to replace JSCell::operator new with static create method https://bugs.webkit.org/show_bug.cgi?id=64466 Reviewed by Oliver Hunt (oliver@apple.com) and Darin Adler (darin@apple.com). First step in a longer refactoring process to remove the use of operator new overloading in order to allocate GC objects and to replace this method with static create methods for each individual type of heap-allocated JS object. This particular patch only deals with replacing uses of operator new within JSC proper. Future patches will remove it from the parts that interface with the DOM. Due to the DOM's continued dependence on it, operator new has not actually been removed from JSCell. * API/JSCallbackConstructor.h: (JSC::JSCallbackConstructor::create): * API/JSCallbackFunction.h: (JSC::JSCallbackFunction::create): * API/JSCallbackObject.h: (JSC::JSCallbackObject::operator new): (JSC::JSCallbackObject::create): * API/JSCallbackObjectFunctions.h: (JSC::::staticFunctionGetter): * API/JSClassRef.cpp: (OpaqueJSClass::prototype): * API/JSContextRef.cpp: * API/JSObjectRef.cpp: (JSObjectMake): (JSObjectMakeFunctionWithCallback): (JSObjectMakeConstructor): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::createActivation): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): * bytecompiler/BytecodeGenerator.h: (JSC::BytecodeGenerator::makeFunction): * bytecompiler/NodesCodegen.cpp: (JSC::RegExpNode::emitBytecode): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): (JSC::Interpreter::retrieveArguments): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jsc.cpp: (GlobalObject::create): (GlobalObject::GlobalObject): (functionRun): (jscmain): * runtime/Arguments.h: (JSC::Arguments::create): (JSC::Arguments::createNoParameters): * runtime/ArrayConstructor.cpp: (JSC::constructArrayWithSizeQuirk): * runtime/ArrayConstructor.h: (JSC::ArrayConstructor::create): * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncSplice): * runtime/ArrayPrototype.h: (JSC::ArrayPrototype::create): * runtime/BooleanConstructor.cpp: (JSC::constructBoolean): (JSC::constructBooleanFromImmediateBoolean): * runtime/BooleanConstructor.h: (JSC::BooleanConstructor::create): * runtime/BooleanObject.h: (JSC::BooleanObject::create): * runtime/BooleanPrototype.h: (JSC::BooleanPrototype::create): * runtime/DateConstructor.cpp: (JSC::constructDate): * runtime/DateConstructor.h: (JSC::DateConstructor::create): * runtime/DateInstance.h: (JSC::DateInstance::create): * runtime/DatePrototype.h: (JSC::DatePrototype::create): * runtime/Error.cpp: (JSC::createError): (JSC::createEvalError): (JSC::createRangeError): (JSC::createReferenceError): (JSC::createSyntaxError): (JSC::createTypeError): (JSC::createURIError): (JSC::StrictModeTypeErrorFunction::create): (JSC::createTypeErrorFunction): * runtime/ErrorConstructor.h: (JSC::ErrorConstructor::create): * runtime/ErrorInstance.cpp: (JSC::ErrorInstance::ErrorInstance): (JSC::ErrorInstance::create): * runtime/ErrorInstance.h: * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::ErrorPrototype): * runtime/ErrorPrototype.h: (JSC::ErrorPrototype::create): * runtime/ExceptionHelpers.cpp: (JSC::InterruptedExecutionError::InterruptedExecutionError): (JSC::InterruptedExecutionError::create): (JSC::createInterruptedExecutionException): (JSC::TerminatedExecutionError::TerminatedExecutionError): (JSC::TerminatedExecutionError::create): (JSC::createTerminatedExecutionException): * runtime/Executable.cpp: (JSC::FunctionExecutable::FunctionExecutable): (JSC::FunctionExecutable::fromGlobalCode): * runtime/Executable.h: (JSC::ExecutableBase::create): (JSC::NativeExecutable::create): (JSC::ScriptExecutable::ScriptExecutable): (JSC::EvalExecutable::create): (JSC::ProgramExecutable::create): (JSC::FunctionExecutable::create): (JSC::FunctionExecutable::make): * runtime/FunctionConstructor.cpp: (JSC::constructFunctionSkippingEvalEnabledCheck): * runtime/FunctionConstructor.h: (JSC::FunctionConstructor::create): * runtime/FunctionPrototype.cpp: (JSC::FunctionPrototype::addFunctionProperties): * runtime/FunctionPrototype.h: (JSC::FunctionPrototype::create): * runtime/GetterSetter.h: (JSC::GetterSetter::create): * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::create): (JSC::jsAPIValueWrapper): * runtime/JSActivation.cpp: (JSC::JSActivation::argumentsGetter): * runtime/JSActivation.h: (JSC::JSActivation::create): * runtime/JSArray.h: (JSC::JSArray::create): * runtime/JSCell.h: (JSC::JSCell::allocateCell): * runtime/JSFunction.h: (JSC::JSFunction::create): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::init): (JSC::JSGlobalObject::reset): * runtime/JSGlobalObject.h: (JSC::constructEmptyArray): (JSC::constructArray): * runtime/JSNotAnObject.h: (JSC::JSNotAnObject::create): * runtime/JSONObject.h: (JSC::JSONObject::create): * runtime/JSObject.cpp: (JSC::JSObject::defineGetter): (JSC::JSObject::defineSetter): (JSC::putDescriptor): * runtime/JSObject.h: (JSC::JSFinalObject::create): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::create): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::create): * runtime/JSString.cpp: (JSC::JSString::substringFromRope): (JSC::JSString::replaceCharacter): (JSC::StringObject::create): * runtime/JSString.h: (JSC::RopeBuilder::JSString): (JSC::RopeBuilder::create): (JSC::RopeBuilder::createHasOtherOwner): (JSC::jsSingleCharacterString): (JSC::jsSingleCharacterSubstring): (JSC::jsNontrivialString): (JSC::jsString): (JSC::jsSubstring): (JSC::jsOwnedString): * runtime/JSValue.cpp: (JSC::JSValue::toObjectSlowCase): (JSC::JSValue::synthesizeObject): (JSC::JSValue::synthesizePrototype): * runtime/Lookup.cpp: (JSC::setUpStaticFunctionSlot): * runtime/MathObject.h: (JSC::MathObject::create): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::NativeErrorConstructor): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::create): * runtime/NativeErrorPrototype.h: (JSC::NativeErrorPrototype::create): * runtime/NumberConstructor.cpp: (JSC::constructWithNumberConstructor): * runtime/NumberConstructor.h: (JSC::NumberConstructor::create): * runtime/NumberObject.cpp: (JSC::constructNumber): * runtime/NumberObject.h: (JSC::NumberObject::create): * runtime/NumberPrototype.h: (JSC::NumberPrototype::create): * runtime/ObjectConstructor.h: (JSC::ObjectConstructor::create): * runtime/ObjectPrototype.h: (JSC::ObjectPrototype::create): * runtime/Operations.h: (JSC::jsString): * runtime/RegExp.cpp: (JSC::RegExp::RegExp): (JSC::RegExp::createWithoutCaching): (JSC::RegExp::create): * runtime/RegExp.h: * runtime/RegExpCache.cpp: (JSC::RegExpCache::lookupOrCreate): * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::arrayOfMatches): (JSC::constructRegExp): * runtime/RegExpConstructor.h: (JSC::RegExpConstructor::create): * runtime/RegExpMatchesArray.h: (JSC::RegExpMatchesArray::create): * runtime/RegExpObject.h: (JSC::RegExpObject::create): * runtime/RegExpPrototype.cpp: (JSC::regExpProtoFuncCompile): * runtime/RegExpPrototype.h: (JSC::RegExpPrototype::create): * runtime/ScopeChain.h: (JSC::ScopeChainNode::create): (JSC::ScopeChainNode::push): * runtime/SmallStrings.cpp: (JSC::SmallStrings::createEmptyString): (JSC::SmallStrings::createSingleCharacterString): * runtime/StringConstructor.cpp: (JSC::constructWithStringConstructor): * runtime/StringConstructor.h: (JSC::StringConstructor::create): * runtime/StringObject.h: (JSC::StringObject::create): * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::create): * runtime/StringPrototype.cpp: (JSC::stringProtoFuncMatch): (JSC::stringProtoFuncSearch): * runtime/StringPrototype.h: (JSC::StringPrototype::create): * runtime/Structure.h: (JSC::Structure::create): (JSC::Structure::createStructure): * runtime/StructureChain.h: (JSC::StructureChain::create): 2011-07-17 Ryuan Choi <ryuan.choi@samsung.com> [EFL] Refactor scheduleDispatchFunctionsOnMainThread to fix crash. https://bugs.webkit.org/show_bug.cgi?id=64337 Replace ecore_timer_add to Ecore_Pipe. This is needed because ecore_timer should not be called in a child thread, but in the main thread. Reviewed by Antonio Gomes. * wtf/efl/MainThreadEfl.cpp: (WTF::pipeObject): (WTF::monitorDispatchFunctions): (WTF::initializeMainThreadPlatform): (WTF::scheduleDispatchFunctionsOnMainThread): 2011-07-17 Filip Pizlo <fpizlo@apple.com> DFG JIT operationCompareEqual does not inline JSValue::equalSlowCaseInline. https://bugs.webkit.org/show_bug.cgi?id=64637 Reviewed by Gavin Barraclough. * dfg/DFGOperations.cpp: 2011-07-16 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64657 Converted this value not preserved when accessed via direct eval. Reviewed by Oliver Hunt. Upon entry into a non-strict function, primitive this values should be boxed as Object types (or substituted with the global object) - which is done by op_convert_this. However we only do so where this is used lexically within the function (we omit the conversion op if not). The problem comes if a direct eval (running within the function's scope) accesses the this value. We are safe in the case of a single eval, since the this object will be converted within callEval, however the converted value is not preserved, and a new wrapper object is allocated each time eval is invoked. This is inefficient and incorrect, since any changes to the wrapper object will be lost between eval statements. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): - If a function uses eval, we always need to convert this. * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): - Don't convert primitive values here - this is too late! (JSC::Interpreter::privateExecute): - Changed op_convert_this to call new isPrimitive method. * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): - Changed op_convert_this to call new isPrimitive method. * runtime/JSCell.h: (JSC::JSCell::JSValue::isPrimitive): - Added JSValue::isPrimitive. * runtime/JSValue.h: - Added JSValue::isPrimitive. 2011-07-16 Filip Pizlo <fpizlo@apple.com> DFG JIT compare/branch code emits is-integer tests even when a value is definitely not an integer. https://bugs.webkit.org/show_bug.cgi?id=64654 Reviewed by Gavin Barraclough. Added the isKnownNotInteger() method, which returns true if a node is definitely not an integer and will always fail any is-integer test. Then modified the compare and branch code to use this method; if it returns true then is-int tests are omitted and the compiler always emits a slow call. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::isKnownNotInteger): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativeNonPeepholeCompare): (JSC::DFG::JITCodeGenerator::nonSpeculativeCompare): * dfg/DFGJITCodeGenerator.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compare): 2011-07-16 Filip Pizlo <fpizlo@apple.com> DFG speculative JIT has dead code for slow calls for branches. https://bugs.webkit.org/show_bug.cgi?id=64653 Reviewed by Gavin Barraclough. Removed SpeculativeJIT::compilePeepHoleCall. * dfg/DFGSpeculativeJIT.cpp: * dfg/DFGSpeculativeJIT.h: 2011-07-15 Mark Rowe <mrowe@apple.com> Fix the build. * dfg/DFGGraph.h: 2011-07-15 Gavin Barraclough <barraclough@apple.com> NativeError.prototype objects have [[Class]] of "Object" but should be "Error" https://bugs.webkit.org/show_bug.cgi?id=55346 Reviewed by Sam Weinig. * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::ErrorPrototype): - Switch to putDirect since we're not the only ones tranitioning this Structure now. * runtime/NativeErrorPrototype.cpp: (JSC::NativeErrorPrototype::NativeErrorPrototype): * runtime/NativeErrorPrototype.h: - Switch base class to ErrorPrototype. 2011-07-15 Gavin Barraclough <barraclough@apple.com> DFG JIT - Where arguments passed are integers, speculate this. https://bugs.webkit.org/show_bug.cgi?id=64630 Reviewed by Sam Weinig. Presently the DFG JIT is overly aggressively predicting double. Use a bit of dynamic information, and curtail this a little. * dfg/DFGGraph.cpp: (JSC::DFG::Graph::predictArgumentTypes): - Check for integer arguments. * dfg/DFGGraph.h: - Function declaration. * runtime/Executable.cpp: (JSC::tryDFGCompile): (JSC::FunctionExecutable::compileForCallInternal): - Add call to predictArgumentTypes. 2011-07-15 Filip Pizlo <fpizlo@apple.com> DFG JIT is inconsistent about fusing branches and speculating integer comparisons for branches. https://bugs.webkit.org/show_bug.cgi?id=64573 Reviewed by Gavin Barraclough. This patch moves some of NonSpeculativeJIT's functionality up into the JITCodeGenerator superclass so that it can be used from both JITs. Now, in cases where the speculative JIT doesn't want to speculate but still wants to emit good code, it can reliably emit the same code sequence as the non-speculative JIT. This patch also extends the non-speculative JIT's compare optimizations to include compare/branch fusing, and extends the speculative JIT's compare optimizations to cover StrictEqual. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::isKnownInteger): (JSC::DFG::JITCodeGenerator::isKnownNumeric): (JSC::DFG::JITCodeGenerator::nonSpeculativePeepholeBranch): (JSC::DFG::JITCodeGenerator::nonSpeculativeCompare): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::detectPeepHoleBranch): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGNonSpeculativeJIT.h: * dfg/DFGOperations.cpp: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: * wtf/Platform.h: 2011-07-14 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64250 Global strict mode function leaking global object as "this". Reviewed by Oliver Hunt. The root problem here is that we pass the wrong values into calls, and then try to fix them up in the callee. Correct behaviour per the spec is to pass in the value undefined, as this unless either (1) the function call is based on an explicit property access or (2) the base of the call comes directly from a 'with'. This change does away with the need for this conversion of objects (non strict code should only box primitives), and does away with all this conversion for strict functions. This patch may have web compatibility ramifications, and may require some advocacy. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): - Removed op_convert_this_strict, added op_resolve_with_this. * bytecode/Opcode.h: - Removed op_convert_this_strict, added op_resolve_with_this. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): (JSC::BytecodeGenerator::emitResolveWithThis): - Removed op_convert_this_strict, added op_resolve_with_this. * bytecompiler/BytecodeGenerator.h: - Removed op_convert_this_strict, added op_resolve_with_this. * bytecompiler/NodesCodegen.cpp: (JSC::EvalFunctionCallNode::emitBytecode): (JSC::FunctionCallResolveNode::emitBytecode): - Removed op_convert_this_strict, added op_resolve_with_this. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): - Change NeedsThisConversion check to test for JSString's vptr (objects no longer need conversion). * interpreter/Interpreter.cpp: (JSC::Interpreter::resolveThisAndProperty): - Based on resolveBaseAndProperty, but produce correct this value. (JSC::Interpreter::privateExecute): - Removed op_convert_this_strict, added op_resolve_with_this. * interpreter/Interpreter.h: * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompileSlowCases): - Removed op_convert_this_strict, added op_resolve_with_this. * jit/JIT.h: * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_resolve_with_this): - Removed op_convert_this_strict, added op_resolve_with_this. (JSC::JIT::emit_op_convert_this): (JSC::JIT::emitSlow_op_convert_this): - Change NeedsThisConversion check to test for JSString's vptr (objects no longer need conversion). * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_resolve_with_this): - Removed op_convert_this_strict, added op_resolve_with_this. (JSC::JIT::emit_op_convert_this): (JSC::JIT::emitSlow_op_convert_this): - Change NeedsThisConversion check to test for JSString's vptr (objects no longer need conversion). * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): - Removed op_convert_this_strict, added op_resolve_with_this. * jit/JITStubs.h: - Removed op_convert_this_strict, added op_resolve_with_this. * runtime/JSActivation.h: - removed NeedsThisConversion flag, added IsEnvironmentRecord. * runtime/JSStaticScopeObject.h: - removed NeedsThisConversion flag, added IsEnvironmentRecord. * runtime/JSString.h: (JSC::RopeBuilder::createStructure): - removed NeedsThisConversion. * runtime/JSTypeInfo.h: (JSC::TypeInfo::isEnvironmentRecord): (JSC::TypeInfo::overridesHasInstance): - removed NeedsThisConversion flag, added IsEnvironmentRecord. * runtime/JSValue.h: - removed NeedsThisConversion. * runtime/JSVariableObject.h: - Corrected StructureFlags inheritance. * runtime/StrictEvalActivation.h: (JSC::StrictEvalActivation::createStructure): - Added IsEnvironmentRecord to StructureFlags, addded createStructure. * runtime/Structure.h: - removed NeedsThisConversion. * tests/mozilla/ecma/String/15.5.4.6-2.js: (getTestCases): - Removed invalid test case. 2011-07-15 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r91082, r91087, and r91089. http://trac.webkit.org/changeset/91082 http://trac.webkit.org/changeset/91087 http://trac.webkit.org/changeset/91089 https://bugs.webkit.org/show_bug.cgi?id=64616 gtk tests are failing a lot after this change. (Requested by dave_levin on #webkit). * wtf/ThreadIdentifierDataPthreads.cpp: (WTF::ThreadIdentifierData::identifier): (WTF::ThreadIdentifierData::initialize): (WTF::ThreadIdentifierData::initializeKeyOnceHelper): (WTF::ThreadIdentifierData::initializeKeyOnce): * wtf/ThreadIdentifierDataPthreads.h: * wtf/ThreadingPthreads.cpp: (WTF::initializeThreading): 2011-07-15 David Levin <levin@chromium.org> Another attempted build fix. * wtf/ThreadIdentifierDataPthreads.cpp: Add include to pick up the definition of PTHREAD_KEYS_MAX. 2011-07-15 David Levin <levin@chromium.org> Chromium build fix. * wtf/ThreadIdentifierDataPthreads.cpp: Add include to pick up the definition of PTHREAD_KEYS_MAX. 2011-07-14 David Levin <levin@chromium.org> currentThread is too slow! https://bugs.webkit.org/show_bug.cgi?id=64577 Reviewed by Darin Adler and Dmitry Titov. The problem is that currentThread results in a pthread_once call which always takes a lock. With this change, currentThread is 10% faster than isMainThread in release mode and only 5% slower than isMainThread in debug. * wtf/ThreadIdentifierDataPthreads.cpp: (WTF::ThreadIdentifierData::initializeOnce): Remove the pthread once stuff which is no longer needed because this is called from initializeThreading(). (WTF::ThreadIdentifierData::identifier): Remove the initializeKeyOnce call because intialization of the pthread key should already be done. (WTF::ThreadIdentifierData::initialize): Ditto. * wtf/ThreadIdentifierDataPthreads.h: * wtf/ThreadingPthreads.cpp: (WTF::initializeThreading): Acquire the pthread key here. 2011-07-14 Filip Pizlo <fpizlo@apple.com> DFG JIT does not optimize Branch as well as it could. https://bugs.webkit.org/show_bug.cgi?id=64574 Reviewed by Gavin Barraclough. This creates a common code path for emitting unfused branches, which does no speculation, and only performs a slow call if absolutely necessary. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::emitBranch): * dfg/DFGJITCodeGenerator.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-07-14 Filip Pizlo <fpizlo@apple.com> GC allocation fast path has too many operations. https://bugs.webkit.org/show_bug.cgi?id=64493 Reviewed by Darin Adler. Changed the timing of the lazy sweep so that it occurs when we land on a previously-unsweeped block, rather than whenever we land on an unsweeped cell. After the per-block lazy sweep occurs, the block is turned into a singly linked list of free cells. The allocation fast path is now just a load-branch-store to remove a cell from the head of the list. Additionally, this changes the way new blocks are allocated. Previously, they would be populated with dummy cells. With this patch, they are turned into a free list, which means that there will never be destructor calls for allocations in fresh blocks. These changes result in a 1.9% speed-up on V8, and a 0.6% speed-up on SunSpider. There are no observed statistically significant slow-downs on any individual benchmark. * JavaScriptCore.exp: * heap/Heap.cpp: (JSC::Heap::allocateSlowCase): (JSC::Heap::collect): (JSC::Heap::canonicalizeBlocks): (JSC::Heap::resetAllocator): * heap/Heap.h: (JSC::Heap::forEachProtectedCell): (JSC::Heap::forEachCell): (JSC::Heap::forEachBlock): (JSC::Heap::allocate): * heap/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): (JSC::MarkedBlock::lazySweep): (JSC::MarkedBlock::blessNewBlockForFastPath): (JSC::MarkedBlock::blessNewBlockForSlowPath): (JSC::MarkedBlock::canonicalizeBlock): * heap/MarkedBlock.h: * heap/NewSpace.cpp: (JSC::NewSpace::addBlock): (JSC::NewSpace::canonicalizeBlocks): * heap/NewSpace.h: (JSC::NewSpace::allocate): (JSC::NewSpace::SizeClass::SizeClass): (JSC::NewSpace::SizeClass::canonicalizeBlock): * heap/OldSpace.cpp: (JSC::OldSpace::addBlock): 2011-07-14 Filip Pizlo <fpizlo@apple.com> DFG JIT crashes on host constructor calls in debug mode. https://bugs.webkit.org/show_bug.cgi?id=64562 Reviewed by Gavin Barraclough. Fixed the relevant ASSERT. * dfg/DFGOperations.cpp: 2011-07-14 Filip Pizlo <fpizlo@apple.com> DFG speculative JIT contains a FIXME for rewinding speculative code generation that has already been fixed. https://bugs.webkit.org/show_bug.cgi?id=64022 Reviewed by Gavin Barraclough. * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution): 2011-07-14 Ryuan Choi <ryuan.choi@samsung.com> [EFL] Add OwnPtr specialization for Ecore_Pipe. https://bugs.webkit.org/show_bug.cgi?id=64515 Add an overload for deleteOwnedPtr(Ecore_Pipe*) on EFL port. Reviewed by Xan Lopez. * wtf/OwnPtrCommon.h: * wtf/efl/OwnPtrEfl.cpp: (WTF::deleteOwnedPtr): 2011-07-14 Filip Pizlo <fpizlo@apple.com> DFG JIT unnecessarily boxes and unboxes values during silent spilling. https://bugs.webkit.org/show_bug.cgi?id=64068 Reviewed by Gavin Barraclough. Silent spilling and filling of registers is done during slow-path C function calls. The silent spill/fill logic does not affect register allocation on paths that don't involve the C function call. This changes the silent spilling code to spill in unboxed form. The silent fill will refill in whatever form the register was spilled in. For example, the silent spill code may choose not to spill the register because it was already spilled previously, which would imply that it was spilled in boxed form. The filling code detects this and either unboxes, or not, depending on what is appropriate. This change also results in a simplification of the silent spill/fill API: silent spilling no longer needs to know about the set of registers that cannot be trampled, since it never does boxing and hence does not need a temporary register. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentSpillGPR): (JSC::DFG::JITCodeGenerator::silentSpillFPR): (JSC::DFG::JITCodeGenerator::silentFillFPR): (JSC::DFG::JITCodeGenerator::silentSpillAllRegisters): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::valueToNumber): (JSC::DFG::NonSpeculativeJIT::valueToInt32): (JSC::DFG::NonSpeculativeJIT::knownConstantArithOp): (JSC::DFG::NonSpeculativeJIT::basicArithOp): (JSC::DFG::NonSpeculativeJIT::compare): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-07-13 Michael Saboff <msaboff@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64202 Enh: Improve handling of RegExp in the form of /.*blah.*/ Reviewed by Gavin Barraclough. Added code to both the Yarr interpreter and JIT to handle these expressions a little differently. First off, the terms in between the leading and trailing .*'s cannot capture and also this enhancement is limited to single alternative expressions. If an expression is of the right form with the aforementioned restrictions, we process the inner terms and then look for the beginning of the string and end of the string. There is handling for multiline expressions to allow the beginning and end to be right after and right before newlines. This enhancement speeds up expressions of this type 12x on a MacBookPro. Cleaned up 'case' statement indentation. A new set of tests was added as LayoutTests/fast/regex/dotstar.html * yarr/YarrInterpreter.cpp: (JSC::Yarr::Interpreter::InputStream::end): (JSC::Yarr::Interpreter::matchDotStarEnclosure): (JSC::Yarr::Interpreter::matchDisjunction): (JSC::Yarr::ByteCompiler::assertionDotStarEnclosure): (JSC::Yarr::ByteCompiler::emitDisjunction): * yarr/YarrInterpreter.h: (JSC::Yarr::ByteTerm::DotStarEnclosure): * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::generateDotStarEnclosure): (JSC::Yarr::YarrGenerator::backtrackDotStarEnclosure): (JSC::Yarr::YarrGenerator::generateTerm): (JSC::Yarr::YarrGenerator::backtrackTerm): * yarr/YarrPattern.cpp: (JSC::Yarr::YarrPatternConstructor::setupAlternativeOffsets): (JSC::Yarr::YarrPatternConstructor::containsCapturingTerms): (JSC::Yarr::YarrPatternConstructor::optimizeDotStarWrappedExpressions): (JSC::Yarr::YarrPattern::compile): * yarr/YarrPattern.h: (JSC::Yarr::PatternTerm::PatternTerm): 2011-07-13 Xan Lopez <xlopez@igalia.com> [GTK] Fix distcheck Reviewed by Martin Robinson. * GNUmakefile.list.am: add missing files. 2011-07-13 Filip Pizlo <fpizlo@apple.com> DFG JIT does not implement prototype chain or list caching for get_by_id. https://bugs.webkit.org/show_bug.cgi?id=64147 Reviewed by Gavin Barraclough. This implements unified support for prototype caching, prototype chain caching, and polymorphic (i.e. list) prototype and prototype chain caching. This is done by creating common code for emitting prototype or chain access stubs, and having it factored out into generateProtoChainAccessStub(). This function is called by tryCacheGetByID once the latter determines that some form of prototype access caching is necessary (i.e. the slot being accessed is not on the base value but on some other object). Direct prototype list, and prototype chain list, caching is implemented by linking the slow path to operationGetByIdProtoBuildList(), which uses the same helper function (generateProtoChainAccessStub()) as tryCacheGetByID. This change required ensuring that the value in the scratchGPR field in StructureStubInfo is preserved even after the stub info is in the chain, or proto_list, states. Hence scratchGPR was moved out of the union and into the top-level of StructureStubInfo. * bytecode/StructureStubInfo.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGRepatch.cpp: (JSC::DFG::emitRestoreScratch): (JSC::DFG::linkRestoreScratch): (JSC::DFG::generateProtoChainAccessStub): (JSC::DFG::tryCacheGetByID): (JSC::DFG::tryBuildGetByIDProtoList): (JSC::DFG::dfgBuildGetByIDProtoList): (JSC::DFG::tryCachePutByID): * dfg/DFGRepatch.h: 2011-07-12 Brent Fulgham <bfulgham@webkit.org> Standardize WinCairo conditionalized code under PLATFORM macro. https://bugs.webkit.org/show_bug.cgi?id=64377 Reviewed by Maciej Stachowiak. * wtf/Platform.h: Update to use PLATFORM(WIN_CAIRO) for tests. 2011-07-13 David Levin <levin@chromium.org> Possible race condition in ThreadIdentifierData::initializeKeyOnce and shouldCallRealDebugger. https://bugs.webkit.org/show_bug.cgi?id=64465 Reviewed by Dmitry Titov. There isn't a good way to test this as it is very highly unlikely to occur. * wtf/ThreadIdentifierDataPthreads.cpp: (WTF::ThreadIdentifierData::initializeKeyOnce): Since scoped static initialization isn't thread-safe, change the initialization to be global. 2011-07-12 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64424 Our direct eval behaviour deviates slightly from the spec. Reviewed by Oliver Hunt. The ES5 spec defines a concept of 'Direct Call to Eval' (see section 15.1.2.1.1), where behaviour will differ from that of an indirect call (e.g. " { eval: window.eval }.eval();" or "var a = eval; a();" are indirect calls), particularly in non-strict scopes variables may be introduced into the caller's environment. ES5 direct calls are any call where the callee function is provided by a reference, a base of that Reference is an EnvironmentRecord (this corresponds to all productions "PrimaryExpression: Identifier", see 10.2.2.1 GetIdentifierReference), and where the name of the reference is "eval". This means any expression of the form "eval(...)", and that calls the standard built in eval method from on the Global Object, is considered to be direct. In JavaScriptCore we are currently overly restrictive. We also check that the EnvironmentRecord that is the base of the reference is the Declaractive Environment Record at the root of the scope chain, corresponding to the Global Object - an "eval(..)" statement that hits a var eval in a nested scope is not considered to be direct. This behaviour does not emanate from the spec, and is incorrect. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): - Fixed direct eval check in op_call_eval. * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): - Fixed direct eval check in op_call_eval. * runtime/Executable.h: (JSC::isHostFunction): - Added check for host function with specific NativeFunction. 2011-07-13 Ademar de Souza Reis Jr. <ademar.reis@openbossa.org> Reviewed by Andreas Kling. Broken build on QNX https://bugs.webkit.org/show_bug.cgi?id=63717 QNX doesn't support pthread's SA_RESTART (required by JSC_MULTIPLE_THREADS), JIT is broken at runtime and there a few minor compilation errors here and there. Original patch by Ritt Konstantin <ritt.ks@gmail.com>, also tested by him on QNX v6.5 (x86) * wtf/DateMath.cpp: fix usage of abs/labs * wtf/Platform.h: Disable JIT and JSC_MULTIPLE_THREADS * wtf/StackBounds.cpp: Add a couple of missing includes (and sort them) 2011-07-12 Anders Carlsson <andersca@apple.com> If a compiler has nullptr support, include <cstddef> to get the nullptr_t definition https://bugs.webkit.org/show_bug.cgi?id=64429 Include the cstddef which has the nullptr_t typedef according to the C++0x standard. * wtf/NullPtr.h: 2011-07-13 MORITA Hajime <morrita@google.com> Refactoring: Ignored ExceptionCode value should be less annoying. https://bugs.webkit.org/show_bug.cgi?id=63688 Added ASSERT_AT macro. Reviewed by Darin Adler. * wtf/Assertions.h: 2011-07-12 Filip Pizlo <fpizlo@apple.com> DFG JIT does not implement op_construct. https://bugs.webkit.org/show_bug.cgi?id=64066 Reviewed by Gavin Barraclough. This is a fixed implementation of op_construct. Constructor calls are implemented by reusing almost all of the code for Call, with care taken to make sure that where the are differences (like selecting different code blocks), those differences are respected. The two fixes over the last patch are: (1) make sure the CodeBlock::unlinkCalls respects differences between Call and Construct, and (2) make sure that virtualFor() in DFGOperations respects the CodeSpecializationKind (either CodeForCall or CodeForConstruct) when invoking the compiler. * dfg/DFGAliasTracker.h: (JSC::DFG::AliasTracker::recordConstruct): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::addCall): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::emitCall): * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGRepatch.cpp: (JSC::DFG::dfgLinkFor): * dfg/DFGRepatch.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/CodeBlock.cpp: (JSC::CodeBlock::unlinkCalls): 2011-07-12 Oliver Hunt <oliver@apple.com> Overzealous type validation in method_check https://bugs.webkit.org/show_bug.cgi?id=64415 Reviewed by Gavin Barraclough. method_check is essentially just a value look up optimisation, but it internally stores the value as a JSFunction, even though it never relies on this fact. Under GC validation however we end up trying to enforce that assumption. The fix is simply to store the value as a correct supertype. * bytecode/CodeBlock.h: * dfg/DFGRepatch.cpp: (JSC::DFG::dfgRepatchGetMethodFast): (JSC::DFG::tryCacheGetMethod): * jit/JIT.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::patchMethodCallProto): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): 2011-07-12 Filip Pizlo <fpizlo@apple.com> COLLECT_ON_EVERY_ALLOCATION no longer works. https://bugs.webkit.org/show_bug.cgi?id=64388 Reviewed by Oliver Hunt. Added a flag to Heap that determines if it's safe to collect (which for now means that JSGlobalObject has actually been initialized, but it should work for other things, too). This allows JSGlobalObject to allocate even if the allocator wants to GC; instead of GCing it just grows the heap, if necessary. Then changed Heap::allocate() to not recurse ad infinitum when COLLECT_ON_EVERY_ALLOCATION is set. This also makes the allocator generally more resilient against bugs; this change allowed me to put in handy assertions, such as that an allocation must succeed after either a collection or after a new block was added. * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::tryAllocate): (JSC::Heap::allocate): (JSC::Heap::collectAllGarbage): (JSC::Heap::collect): * heap/Heap.h: (JSC::Heap::notifyIsSafeToCollect): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): 2011-07-12 Filip Pizlo <fpizlo@apple.com> DFG JIT put_by_id transition caching does not inform the GC about the structure and prototype chain that it is referencing. https://bugs.webkit.org/show_bug.cgi?id=64387 Reviewed by Gavin Barraclough. Fixed the relevant code in DFGRepatch to call StructureStubInfo::initPutByIdTransition(). * dfg/DFGRepatch.cpp: (JSC::DFG::tryCachePutByID): 2011-07-12 Adam Roben <aroben@apple.com> Ensure no intermediate WTF::Strings are created when concatenating with string literals Fixes <http://webkit.org/b/63330> Concatenating string literals and WTF::Strings using operator+ is suboptimal Reviewed by Darin Adler. * wtf/text/StringConcatenate.h: (WTF::StringTypeAdapter<String>::writeTo): Added a macro that can be used for testing how many WTF::Strings get copied while evaluating an operator+ expression. * wtf/text/StringOperators.h: (WTF::operator+): Changed the overload that takes a StringAppend to take it on the left-hand side, since operator+ is left-associative. Having the StringAppend on the right-hand side was causing us to make intermediate WTF::Strings when evaluating expressions that contained multiple calls to operator+. Added some more overloads for that take a left-hand side of const char* to resolve overload ambiguity for certain expressions. Added overloads that take a left-hand side of const UChar* (matching the const char* overloads) so that wide string literals don't first have to be converted to a WTF::String in operator+ expressions. 2011-07-12 Adam Roben <aroben@apple.com> Unreviewed, rolling out r90811. http://trac.webkit.org/changeset/90811 https://bugs.webkit.org/show_bug.cgi?id=61025 Several svg tests failing assertions beneath SVGSMILElement::findInstanceTime * wtf/StdLibExtras.h: (WTF::binarySearch): 2011-07-12 Oliver Varga <Varga.Oliver@stud.u-szeged.hu> Reviewed by Nikolas Zimmermann. Speed up SVGSMILElement::findInstanceTime. https://bugs.webkit.org/show_bug.cgi?id=61025 Add a new parameter to StdlibExtras.h::binarySerarch function to also handle cases when the array does not contain the key value. This is needed for an svg function. * wtf/StdLibExtras.h: (WTF::binarySearch): 2011-07-11 Filip Pizlo <fpizlo@apple.com> DFG speculative JIT does not guard itself against floating point speculation failures on non-floating-point constants. https://bugs.webkit.org/show_bug.cgi?id=64330 Reviewed by Gavin Barraclough. Made fillSpeculateDouble immediate invoke terminateSpeculativeExecution() as soon as it notices that it's speculating on something that is a non-numeric JSConstant. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): 2011-07-11 Filip Pizlo <fpizlo@apple.com> DFG Speculative JIT does not always insert speculation checks when speculating arrays. https://bugs.webkit.org/show_bug.cgi?id=64254 Reviewed by Gavin Barraclough. Changed the SetLocal instruction to always validate that the value being stored into the local variable is an array, if that variable was marked PredictArray. This is necessary since uses of arrays assume that if a PredictArray value is in a local variable then the speculation check validating that the value is an array was already performed. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-07-11 Gabor Loki <loki@webkit.org> Fix the condition of the optimized code in doubleTransfer https://bugs.webkit.org/show_bug.cgi?id=64261 Reviewed by Zoltan Herczeg. The condition of the optimized code in doubleTransfer is wrong. The data transfer should be executed with four bytes aligned address. VFP cannot perform unaligned memory access. Reported by Jacob Bramley. * assembler/ARMAssembler.cpp: (JSC::ARMAssembler::doubleTransfer): 2011-07-11 Gabor Loki <loki@webkit.org> Signed arithmetic bug in dataTransfer32. https://bugs.webkit.org/show_bug.cgi?id=64257 Reviewed by Zoltan Herczeg. An arithmetic bug is fixed. If the offset of dataTransfer is half of the addressable memory space on a 32-bit machine (-2147483648 = 0x80000000) a load instruction is emitted with a wrong zero offset. Inspired by Jacob Bramley's patch from JaegerMonkey. * assembler/ARMAssembler.cpp: (JSC::ARMAssembler::dataTransfer32): 2011-07-09 Thouraya Andolsi <thouraya.andolsi@st.com> Fix unaligned userspace access for SH4 platforms. https://bugs.webkit.org/show_bug.cgi?id=62993 * wtf/Platform.h: 2011-07-09 Chao-ying Fu <fu@mips.com> Fix MIPS build due to readInt32 and readPointer https://bugs.webkit.org/show_bug.cgi?id=63962 * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::readInt32): (JSC::MIPSAssembler::readPointer): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::rshift32): 2011-07-08 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=64181 REGRESSION (r90602): Gmail doesn't load Rolling out r90601, r90602. * dfg/DFGAliasTracker.h: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::addVarArgChild): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::emitCall): * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGRepatch.cpp: (JSC::DFG::tryCacheGetByID): (JSC::DFG::dfgLinkCall): * dfg/DFGRepatch.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * runtime/JSObject.h: (JSC::JSObject::isUsingInlineStorage): 2011-07-08 Kalev Lember <kalev@smartlink.ee> Reviewed by Adam Roben. Add missing _WIN32_WINNT and WINVER definitions https://bugs.webkit.org/show_bug.cgi?id=59702 Moved _WIN32_WINNT and WINVER definitions to config.h so that they are available for all source files. In particular, wtf/FastMalloc.cpp uses CreateTimerQueueTimer and DeleteTimerQueueTimer which are both guarded by #if (_WIN32_WINNT >= 0x0500) in MinGW headers. * config.h: * wtf/Assertions.cpp: 2011-07-08 Chang Shu <cshu@webkit.org> Rename "makeSecure" to "fill" and remove the support for displaying last character to avoid layering violatation. https://bugs.webkit.org/show_bug.cgi?id=59114 Reviewed by Alexey Proskuryakov. * JavaScriptCore.exp: * JavaScriptCore.order: * wtf/text/StringImpl.cpp: (WTF::StringImpl::fill): * wtf/text/StringImpl.h: * wtf/text/WTFString.h: (WTF::String::fill): 2011-07-08 Benjamin Poulain <benjamin@webkit.org> [WK2] Do not forward touch events to the web process when it does not need them https://bugs.webkit.org/show_bug.cgi?id=64164 Reviewed by Kenneth Rohde Christiansen. Add a convenience function to obtain a reference to the last element of a Deque. * wtf/Deque.h: (WTF::Deque::last): 2011-07-07 Filip Pizlo <fpizlo@apple.com> DFG JIT does not implement op_construct. https://bugs.webkit.org/show_bug.cgi?id=64066 Reviewed by Gavin Barraclough. * dfg/DFGAliasTracker.h: (JSC::DFG::AliasTracker::recordConstruct): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::addCall): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::emitCall): * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGRepatch.cpp: (JSC::DFG::dfgLinkFor): * dfg/DFGRepatch.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-07-07 Filip Pizlo <fpizlo@apple.com> DFG JIT does not implement get_by_id prototype caching. https://bugs.webkit.org/show_bug.cgi?id=64077 Reviewed by Gavin Barraclough. * dfg/DFGRepatch.cpp: (JSC::DFG::emitRestoreScratch): (JSC::DFG::linkRestoreScratch): (JSC::DFG::tryCacheGetByID): * runtime/JSObject.h: (JSC::JSObject::addressOfPropertyAtOffset): 2011-07-07 Filip Pizlo <fpizlo@apple.com> DFG JIT method_check implementation does not link to optimized get_by_id slow path. https://bugs.webkit.org/show_bug.cgi?id=64073 Reviewed by Gavin Barraclough. * dfg/DFGRepatch.cpp: (JSC::DFG::dfgRepatchGetMethodFast): 2011-07-07 Oliver Hunt <oliver@apple.com> Encode jump and link sizes into the appropriate enums https://bugs.webkit.org/show_bug.cgi?id=64123 Reviewed by Sam Weinig. Finally kill off the out of line jump and link size arrays, so we can avoid icky loads and constant fold the linking arithmetic. * assembler/ARMv7Assembler.cpp: * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::jumpSizeDelta): (JSC::ARMv7Assembler::computeJumpType): 2011-07-06 Juan C. Montemayor <jmont@apple.com> ASSERT_NOT_REACHED running test 262 https://bugs.webkit.org/show_bug.cgi?id=63951 Added a case to the switch statement where the code was failing. Fixed some logic as well that gave faulty error messages. Reviewed by Gavin Barraclough. * parser/JSParser.cpp: (JSC::JSParser::getTokenName): (JSC::JSParser::updateErrorMessageSpecialCase): (JSC::JSParser::updateErrorMessage): 2011-07-06 Filip Pizlo <fpizlo@apple.com> DFG JIT implementation of op_call results in regressions on sunspider controlflow-recursive. https://bugs.webkit.org/show_bug.cgi?id=64039 Reviewed by Gavin Barraclough. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::isSmallInt32Constant): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isInteger): 2011-07-06 Filip Pizlo <fpizlo@apple.com> DFG JIT does not support method_check https://bugs.webkit.org/show_bug.cgi?id=63972 Reviewed by Gavin Barraclough. * assembler/CodeLocation.h: (JSC::CodeLocationPossiblyNearCall::CodeLocationPossiblyNearCall): * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): * bytecode/CodeBlock.h: (JSC::MethodCallLinkInfo::MethodCallLinkInfo): (JSC::MethodCallLinkInfo::seenOnce): (JSC::MethodCallLinkInfo::setSeen): * dfg/DFGAliasTracker.h: (JSC::DFG::AliasTracker::recordGetMethod): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::cachedGetMethod): * dfg/DFGJITCodeGenerator.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::addMethodGet): (JSC::DFG::JITCompiler::MethodGetRecord::MethodGetRecord): * dfg/DFGNode.h: (JSC::DFG::Node::hasIdentifier): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGRepatch.cpp: (JSC::DFG::dfgRepatchGetMethodFast): (JSC::DFG::tryCacheGetMethod): (JSC::DFG::dfgRepatchGetMethod): * dfg/DFGRepatch.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JITWriteBarrier.h: (JSC::JITWriteBarrier::set): 2011-07-06 Filip Pizlo <fpizlo@apple.com> DFG JIT op_call implementation will flush registers even when those registers are dead https://bugs.webkit.org/show_bug.cgi?id=64023 Reviewed by Gavin Barraclough. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::emitCall): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::integerResult): (JSC::DFG::JITCodeGenerator::noResult): (JSC::DFG::JITCodeGenerator::cellResult): (JSC::DFG::JITCodeGenerator::jsValueResult): (JSC::DFG::JITCodeGenerator::doubleResult): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-07-06 Filip Pizlo <fpizlo@apple.com> DFG speculative JIT may crash when speculating int on a non-int JSConstant. https://bugs.webkit.org/show_bug.cgi?id=64017 Reviewed by Gavin Barraclough. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::compile): 2011-07-06 Dmitriy Vyukov <dvyukov@google.com> Reviewed by David Levin. Allow substitution of dynamic annotations and prevent identical code folding by the linker. https://bugs.webkit.org/show_bug.cgi?id=62443 * wtf/DynamicAnnotations.cpp: (WTFAnnotateBenignRaceSized): (WTFAnnotateHappensBefore): (WTFAnnotateHappensAfter): 2011-07-06 Zoltan Herczeg <zherczeg@inf.u-szeged.hu> Calls on 32 bit machines are failed after r90423 https://bugs.webkit.org/show_bug.cgi?id=63980 Reviewed by Gavin Barraclough. Copy the necessary lines from JITCall.cpp. * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCall): 2011-07-05 Filip Pizlo <fpizlo@apple.com> DFG JIT virtual call implementation is inefficient. https://bugs.webkit.org/show_bug.cgi?id=63974 Reviewed by Gavin Barraclough. * dfg/DFGOperations.cpp: * runtime/Executable.h: (JSC::ExecutableBase::generatedJITCodeForCallWithArityCheck): (JSC::ExecutableBase::generatedJITCodeForConstructWithArityCheck): (JSC::ExecutableBase::generatedJITCodeWithArityCheckFor): (JSC::ExecutableBase::hasJITCodeForCall): (JSC::ExecutableBase::hasJITCodeForConstruct): (JSC::ExecutableBase::hasJITCodeFor): * runtime/JSFunction.h: (JSC::JSFunction::scopeUnchecked): 2011-07-05 Oliver Hunt <oliver@apple.com> Force inlining of simple functions that show up as not being inlined https://bugs.webkit.org/show_bug.cgi?id=63964 Reviewed by Gavin Barraclough. Looking at profile data indicates the gcc is failing to inline a number of trivial functions. This patch hits the ones that show up in profiles with the ALWAYS_INLINE hammer. We also replace the memcpy() call in linking with a manual loop. Apparently memcpy() is almost never faster than an inlined loop. * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::add): (JSC::ARMv7Assembler::add_S): (JSC::ARMv7Assembler::ARM_and): (JSC::ARMv7Assembler::asr): (JSC::ARMv7Assembler::b): (JSC::ARMv7Assembler::blx): (JSC::ARMv7Assembler::bx): (JSC::ARMv7Assembler::clz): (JSC::ARMv7Assembler::cmn): (JSC::ARMv7Assembler::cmp): (JSC::ARMv7Assembler::eor): (JSC::ARMv7Assembler::it): (JSC::ARMv7Assembler::ldr): (JSC::ARMv7Assembler::ldrCompact): (JSC::ARMv7Assembler::ldrh): (JSC::ARMv7Assembler::ldrb): (JSC::ARMv7Assembler::lsl): (JSC::ARMv7Assembler::lsr): (JSC::ARMv7Assembler::movT3): (JSC::ARMv7Assembler::mov): (JSC::ARMv7Assembler::movt): (JSC::ARMv7Assembler::mvn): (JSC::ARMv7Assembler::neg): (JSC::ARMv7Assembler::orr): (JSC::ARMv7Assembler::orr_S): (JSC::ARMv7Assembler::ror): (JSC::ARMv7Assembler::smull): (JSC::ARMv7Assembler::str): (JSC::ARMv7Assembler::sub): (JSC::ARMv7Assembler::sub_S): (JSC::ARMv7Assembler::tst): (JSC::ARMv7Assembler::linkRecordSourceComparator): (JSC::ARMv7Assembler::link): (JSC::ARMv7Assembler::ARMInstructionFormatter::oneWordOp5Reg3Imm8): (JSC::ARMv7Assembler::ARMInstructionFormatter::oneWordOp5Imm5Reg3Reg3): (JSC::ARMv7Assembler::ARMInstructionFormatter::oneWordOp7Reg3Reg3Reg3): (JSC::ARMv7Assembler::ARMInstructionFormatter::oneWordOp8Imm8): (JSC::ARMv7Assembler::ARMInstructionFormatter::oneWordOp8RegReg143): (JSC::ARMv7Assembler::ARMInstructionFormatter::oneWordOp9Imm7): (JSC::ARMv7Assembler::ARMInstructionFormatter::oneWordOp10Reg3Reg3): (JSC::ARMv7Assembler::ARMInstructionFormatter::twoWordOp12Reg4FourFours): (JSC::ARMv7Assembler::ARMInstructionFormatter::twoWordOp16FourFours): (JSC::ARMv7Assembler::ARMInstructionFormatter::twoWordOp16Op16): (JSC::ARMv7Assembler::ARMInstructionFormatter::twoWordOp5i6Imm4Reg4EncodedImm): (JSC::ARMv7Assembler::ARMInstructionFormatter::twoWordOp12Reg4Reg4Imm12): (JSC::ARMv7Assembler::ARMInstructionFormatter::vfpOp): (JSC::ARMv7Assembler::ARMInstructionFormatter::vfpMemOp): * assembler/LinkBuffer.h: (JSC::LinkBuffer::linkCode): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::nearCall): (JSC::MacroAssemblerARMv7::call): (JSC::MacroAssemblerARMv7::ret): (JSC::MacroAssemblerARMv7::moveWithPatch): (JSC::MacroAssemblerARMv7::branchPtrWithPatch): (JSC::MacroAssemblerARMv7::storePtrWithPatch): (JSC::MacroAssemblerARMv7::tailRecursiveCall): (JSC::MacroAssemblerARMv7::makeTailRecursiveCall): (JSC::MacroAssemblerARMv7::jump): (JSC::MacroAssemblerARMv7::makeBranch): 2011-07-05 Zoltan Herczeg <zherczeg@inf.u-szeged.hu> Make "Add optimised paths for a few maths functions" work on Qt https://bugs.webkit.org/show_bug.cgi?id=63893 Reviewed by Oliver Hunt. Move the generated code to the .text section instead of .data section. Fix alignment for the 32 bit thunk code. * jit/ThunkGenerators.cpp: 2011-07-05 Filip Pizlo <fpizlo@apple.com> DFG JIT does not implement op_call. https://bugs.webkit.org/show_bug.cgi?id=63858 Reviewed by Gavin Barraclough. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::unlinkCalls): * bytecode/CodeBlock.h: (JSC::CodeBlock::setNumberOfCallLinkInfos): (JSC::CodeBlock::numberOfCallLinkInfos): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitCall): (JSC::BytecodeGenerator::emitConstruct): * dfg/DFGAliasTracker.h: (JSC::DFG::AliasTracker::lookupGetByVal): (JSC::DFG::AliasTracker::recordCall): (JSC::DFG::AliasTracker::equalIgnoringLaterNumericConversion): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::getLocal): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::toInt32): (JSC::DFG::ByteCodeParser::addToGraph): (JSC::DFG::ByteCodeParser::addVarArgChild): (JSC::DFG::ByteCodeParser::predictInt32): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::processPhiStack): (JSC::DFG::ByteCodeParser::allocateVirtualRegisters): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::opName): (JSC::DFG::Graph::dump): (JSC::DFG::Graph::refChildren): * dfg/DFGGraph.h: * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::useChildren): (JSC::DFG::JITCodeGenerator::emitCall): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::addressOfCallData): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGJITCompiler.h: (JSC::DFG::CallRecord::CallRecord): (JSC::DFG::JITCompiler::notifyCall): (JSC::DFG::JITCompiler::appendCallWithFastExceptionCheck): (JSC::DFG::JITCompiler::addJSCall): (JSC::DFG::JITCompiler::PropertyAccessRecord::PropertyAccessRecord): (JSC::DFG::JITCompiler::JSCallRecord::JSCallRecord): * dfg/DFGNode.h: (JSC::DFG::Node::Node): (JSC::DFG::Node::child1): (JSC::DFG::Node::child2): (JSC::DFG::Node::child3): (JSC::DFG::Node::firstChild): (JSC::DFG::Node::numChildren): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::basicArithOp): (JSC::DFG::NonSpeculativeJIT::compare): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGRepatch.cpp: (JSC::DFG::dfgLinkCall): * dfg/DFGRepatch.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleIntegerBranch): (JSC::DFG::SpeculativeJIT::compilePeepHoleCall): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::detectPeepHoleBranch): * interpreter/CallFrame.h: (JSC::ExecState::calleeAsValue): * jit/JIT.cpp: (JSC::JIT::JIT): (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompileSlowCases): (JSC::JIT::privateCompile): (JSC::JIT::linkCall): (JSC::JIT::linkConstruct): * jit/JITCall.cpp: (JSC::JIT::compileOpCall): * jit/JITCode.h: (JSC::JITCode::JITCode): (JSC::JITCode::jitType): (JSC::JITCode::HostFunction): * runtime/JSFunction.h: * runtime/JSGlobalData.h: 2011-07-05 Oliver Hunt <oliver@apple.com> Initialize new MarkStack member * heap/MarkStack.h: (JSC::MarkStack::MarkStack): 2011-07-05 Oliver Hunt <oliver@apple.com> Don't throw out compiled code repeatedly https://bugs.webkit.org/show_bug.cgi?id=63960 Reviewed by Gavin Barraclough. Stop throwing away all compiled code every time we're told to do a full GC. Instead unlink all callsites during such GC passes to maximise the number of collectable functions, but otherwise leave compiled functions alone. * API/JSBase.cpp: (JSGarbageCollect): * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): * heap/Heap.cpp: (JSC::Heap::collectAllGarbage): * heap/MarkStack.h: (JSC::MarkStack::shouldUnlinkCalls): (JSC::MarkStack::setShouldUnlinkCalls): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::recompileAllJSFunctions): (JSC::JSGlobalData::releaseExecutableMemory): * runtime/RegExp.cpp: (JSC::RegExp::compile): (JSC::RegExp::invalidateCode): * runtime/RegExp.h: 2011-07-05 Filip Pizlo <fpizlo@apple.com> JSC JIT has code duplication for the handling of call and construct https://bugs.webkit.org/show_bug.cgi?id=63957 Reviewed by Gavin Barraclough. * jit/JIT.cpp: (JSC::JIT::linkFor): * jit/JIT.h: * jit/JITStubs.cpp: (JSC::jitCompileFor): (JSC::DEFINE_STUB_FUNCTION): (JSC::arityCheckFor): (JSC::lazyLinkFor): * runtime/Executable.h: (JSC::ExecutableBase::generatedJITCodeFor): (JSC::FunctionExecutable::compileFor): (JSC::FunctionExecutable::isGeneratedFor): (JSC::FunctionExecutable::generatedBytecodeFor): (JSC::FunctionExecutable::generatedJITCodeWithArityCheckFor): 2011-07-05 Gavin Barraclough <barraclough@apple.com> Build fix following last patch. * runtime/JSFunction.cpp: (JSC::createPrototypeProperty): 2011-07-05 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=63947 ASSERT running Object.preventExtensions(Math.sin) Reviewed by Oliver Hunt. This is due to calling scope() on a hostFunction as a part of calling createPrototypeProperty to reify the prototype property. But host functions don't have a prototype property anyway! Prevent callling createPrototypeProperty on a host function. * runtime/JSFunction.cpp: (JSC::JSFunction::createPrototypeProperty): (JSC::JSFunction::preventExtensions): 2011-07-04 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=63880 Evaluation order of conversions of operands to >, >= incorrect. Reviewed by Sam Weinig. Add 'leftFirst' parameter to jsLess, jsLessEq matching that described in the ES5 spec. This allows these methods to be reused to perform >, >= relational compares with correct ordering of type conversions. * dfg/DFGOperations.cpp: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Operations.h: (JSC::jsLess): (JSC::jsLessEq): 2011-07-04 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=16652 Firefox and JavaScriptCore differ in Number.toString(integer) Our arbitrary radix (2..36) toString conversion is inaccurate. This is partly because it uses doubles to perform math that requires higher accuracy, and partly becasue it does not attempt to correctly detect where to terminate, instead relying on a simple 'epsilon'. * runtime/NumberPrototype.cpp: (JSC::decomposeDouble): - helper function to extract sign, exponent, mantissa from IEEE doubles. (JSC::Uint16WithFraction::Uint16WithFraction): - helper class, u16int with infinite precision fraction, used to convert the fractional part of the number to a string. (JSC::Uint16WithFraction::operator*=): - Multiply by a uint16. (JSC::Uint16WithFraction::operator<): - Compare two Uint16WithFractions. (JSC::Uint16WithFraction::floorAndSubtract): - Extract the integer portion of the number, and subtract it (clears the integer portion). (JSC::Uint16WithFraction::comparePoint5): - Compare to 0.5. (JSC::Uint16WithFraction::sumGreaterThanOne): - Passed a second Uint16WithFraction, returns true if the result of adding the two values would be greater than one. (JSC::Uint16WithFraction::isNormalized): - Used by ASSERTs to consistency check internal representation. (JSC::BigInteger::BigInteger): - helper class, unbounded integer value, used to convert the integer part of the number to a string. (JSC::BigInteger::divide): - Divide this value through by a uint32. (JSC::BigInteger::operator!): - test for zero. (JSC::toStringWithRadix): - Performs number to string conversion, with the given radix (2..36). (JSC::numberProtoFuncToString): - Changed to use toStringWithRadix. 2011-07-04 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=63881 Need separate bytecodes for handling >, >= comparisons. Reviewed by Oliver Hunt. This clears the way to fix Bug#63880. We currently handle greater-than comparisons as being using the corresponding op_less, etc opcodes. This is incorrect with respect to evaluation ordering of the implicit conversions performed on operands - we should be calling ToPrimitive on the LHS and RHS operands to the greater than, but instead convert RHS then LHS. This patch adds opcodes for greater-than comparisons mirroring existing ones used for less-than. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): * bytecode/Opcode.h: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitJumpIfTrue): (JSC::BytecodeGenerator::emitJumpIfFalse): * bytecompiler/NodesCodegen.cpp: * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compare): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGNonSpeculativeJIT.h: * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compare): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompileSlowCases): * jit/JIT.h: (JSC::JIT::emit_op_loop_if_greater): (JSC::JIT::emitSlow_op_loop_if_greater): (JSC::JIT::emit_op_loop_if_greatereq): (JSC::JIT::emitSlow_op_loop_if_greatereq): * jit/JITArithmetic.cpp: (JSC::JIT::emit_op_jgreater): (JSC::JIT::emit_op_jgreatereq): (JSC::JIT::emit_op_jngreater): (JSC::JIT::emit_op_jngreatereq): (JSC::JIT::emitSlow_op_jgreater): (JSC::JIT::emitSlow_op_jgreatereq): (JSC::JIT::emitSlow_op_jngreater): (JSC::JIT::emitSlow_op_jngreatereq): (JSC::JIT::emit_compareAndJumpSlow): * jit/JITArithmetic32_64.cpp: (JSC::JIT::emitBinaryDoubleOp): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jit/JITStubs.h: * parser/NodeConstructors.h: (JSC::GreaterNode::GreaterNode): (JSC::GreaterEqNode::GreaterEqNode): * parser/Nodes.h: 2011-07-03 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=63879 Reduce code duplication for op_jless, op_jlesseq, op_jnless, op_jnlesseq. Reviewed by Sam Weinig. There is a lot of copy & paste code here; we can reduce duplication by making a shared implementation. * assembler/MacroAssembler.h: (JSC::MacroAssembler::branch32): (JSC::MacroAssembler::commute): - Make these function platform agnostic. * assembler/MacroAssemblerX86Common.h: - Moved branch32/commute up to MacroAssembler. * jit/JIT.h: (JSC::JIT::emit_op_loop_if_lesseq): (JSC::JIT::emitSlow_op_loop_if_lesseq): - Add an implementation matching that for op_loop_if_less, which just calls op_jless. * jit/JITArithmetic.cpp: (JSC::JIT::emit_op_jless): (JSC::JIT::emit_op_jlesseq): (JSC::JIT::emit_op_jnless): (JSC::JIT::emit_op_jnlesseq): (JSC::JIT::emitSlow_op_jless): (JSC::JIT::emitSlow_op_jlesseq): (JSC::JIT::emitSlow_op_jnless): (JSC::JIT::emitSlow_op_jnlesseq): - Common implmentations of these methods for JSVALUE64 & JSVALUE32_64. (JSC::JIT::emit_compareAndJump): (JSC::JIT::emit_compareAndJumpSlow): - Internal implmementation of jless etc for JSVALUE64. * jit/JITArithmetic32_64.cpp: (JSC::JIT::emit_compareAndJump): (JSC::JIT::emit_compareAndJumpSlow): - Internal implmementation of jless etc for JSVALUE32_64. * jit/JITOpcodes.cpp: * jit/JITOpcodes32_64.cpp: * jit/JITStubs.cpp: * jit/JITStubs.h: - Remove old implementation of emit_op_loop_if_lesseq. 2011-07-03 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r90347. http://trac.webkit.org/changeset/90347 https://bugs.webkit.org/show_bug.cgi?id=63886 Build breaks on Leopard, Chromium-win, WinCairo, and WinCE. (Requested by tkent on #webkit). * JavaScriptCore.xcodeproj/project.pbxproj: * runtime/BigInteger.h: Removed. * runtime/NumberPrototype.cpp: (JSC::numberProtoFuncToPrecision): (JSC::numberProtoFuncToString): * runtime/Uint16WithFraction.h: Removed. * wtf/MathExtras.h: 2011-06-30 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=16652 Firefox and JavaScriptCore differ in Number.toString(integer) Our arbitrary radix (2..36) toString conversion is inaccurate. This is partly because it uses doubles to perform math that requires higher accuracy, and partly becasue it does not attempt to correctly detect where to terminate, instead relying on a simple 'epsilon'. * runtime/NumberPrototype.cpp: (JSC::decomposeDouble): - helper function to extract sign, exponent, mantissa from IEEE doubles. (JSC::Uint16WithFraction::Uint16WithFraction): - helper class, u16int with infinite precision fraction, used to convert the fractional part of the number to a string. (JSC::Uint16WithFraction::operator*=): - Multiply by a uint16. (JSC::Uint16WithFraction::operator<): - Compare two Uint16WithFractions. (JSC::Uint16WithFraction::floorAndSubtract): - Extract the integer portion of the number, and subtract it (clears the integer portion). (JSC::Uint16WithFraction::comparePoint5): - Compare to 0.5. (JSC::Uint16WithFraction::sumGreaterThanOne): - Passed a second Uint16WithFraction, returns true if the result of adding the two values would be greater than one. (JSC::Uint16WithFraction::isNormalized): - Used by ASSERTs to consistency check internal representation. (JSC::BigInteger::BigInteger): - helper class, unbounded integer value, used to convert the integer part of the number to a string. (JSC::BigInteger::divide): - Divide this value through by a uint32. (JSC::BigInteger::operator!): - test for zero. (JSC::toStringWithRadix): - Performs number to string conversion, with the given radix (2..36). (JSC::numberProtoFuncToString): - Changed to use toStringWithRadix. 2011-07-02 Gavin Barraclough <barraclough@apple.com> https://bugs.webkit.org/show_bug.cgi?id=63866 DFG JIT - implement instanceof Reviewed by Sam Weinig. Add ops CheckHasInstance & InstanceOf to implement bytecodes op_check_has_instance & op_instanceof. This is an initial functional implementation, performance is a wash. We can follow up with changes to fuse the InstanceOf node with a subsequant branch, as we do with other comparisons. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::jitAssertIsCell): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::jitAssertIsCell): * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-07-01 Oliver Hunt <oliver@apple.com> IE Web Workers demo crashes in JSC::SlotVisitor::visitChildren() https://bugs.webkit.org/show_bug.cgi?id=63732 Reviewed by Gavin Barraclough. Initialise the memory at the head of the new storage so that GC is safe if triggered by reportExtraMemoryCost. * runtime/JSArray.cpp: (JSC::JSArray::increaseVectorPrefixLength): 2011-07-01 Oliver Hunt <oliver@apple.com> GC sweep can occur before an object is completely initialised https://bugs.webkit.org/show_bug.cgi?id=63836 Reviewed by Gavin Barraclough. In rare cases it's possible for a GC sweep to occur while a live, but not completely initialised object is on the stack. In such a case we may incorrectly choose to mark it, even though it has no children that need marking. We resolve this by always zeroing out the structure of any value returned from JSCell::operator new(), and making the markstack tolerant of a null structure. * runtime/JSCell.h: (JSC::JSCell::JSCell::~JSCell): (JSC::JSCell::JSCell::operator new): * runtime/Structure.h: (JSC::MarkStack::internalAppend): 2011-07-01 Filip Pizlo <fpizlo@apple.com> Reviewed by Gavin Barraclough. DFG non-speculative JIT always performs slow C calls for div and mod. https://bugs.webkit.org/show_bug.cgi?id=63684 * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): 2011-07-01 Juan C. Montemayor <jmont@apple.com> Reviewed by Oliver Hunt. Lexer error messages are currently appalling https://bugs.webkit.org/show_bug.cgi?id=63340 Added error messages for the Lexer. These messages will be displayed instead of the lexer error messages from the parser that are currently shown. * parser/Lexer.cpp: (JSC::Lexer::getInvalidCharMessage): (JSC::Lexer::setCode): (JSC::Lexer::parseString): (JSC::Lexer::lex): (JSC::Lexer::clear): * parser/Lexer.h: (JSC::Lexer::getErrorMessage): (JSC::Lexer::setOffset): * parser/Parser.cpp: (JSC::Parser::parse): 2011-07-01 Jungshik Shin <jshin@chromium.org> Reviewed by Alexey Proskuryakov. Add ScriptCodesFromICU.h to wtf/unicode and make necessary changes in build files for ports not using ICU. Add icu/unicode/uscript.h for ports using ICU. It's taken from ICU 3.6 (the version used on Mac OS 10.5) http://bugs.webkit.org/show_bug.cgi?id=20797 * GNUmakefile.list.am: * JavaScriptCore.gypi: * icu/unicode/uscript.h: Added for UScriptCode enum. * wtf/unicode/ScriptCodesFromICU.h: UScriptCode enum added. * wtf/unicode/icu/UnicodeIcu.h: * wtf/unicode/brew/UnicodeBrew.h: * wtf/unicode/glib/UnicodeGLib.h: * wtf/unicode/qt4/UnicodeQt4.h: * wtf/unicode/wince/UnicodeWinCE.h: 2011-07-01 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=63819 Escaping of forwardslashes in strings incorrect if multiple exist. The bug is in the parameters passed to a substring - should be start & length, but we're passing start & end indices! * runtime/RegExpObject.cpp: (JSC::regExpObjectSource): 2011-07-01 Adam Roben <aroben@apple.com> Roll out r90194 http://trac.webkit.org/changeset/90194 https://bugs.webkit.org/show_bug.cgi?id=63778 Fixes <http://webkit.org/b/63812> REGRESSION (r90194): Multiple tests intermittently failing assertions in WriteBarrierBase<JSC::Structure>::get * runtime/JSCell.h: (JSC::JSCell::JSCell::~JSCell): 2011-06-30 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Add optimised paths for a few maths functions https://bugs.webkit.org/show_bug.cgi?id=63757 Relanding as a Mac only patch. This adds specialised thunks for Math.abs, Math.round, Math.ceil, Math.floor, Math.log, and Math.exp as they are apparently more important in real web content than we thought, which is somewhat mind-boggling. On average doubles the performance of the common cases (eg. actually passing numbers in). They're not as efficient as they could be, but this way gives them the most portability. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::supportsDoubleBitops): (JSC::MacroAssemblerARM::andnotDouble): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::supportsDoubleBitops): (JSC::MacroAssemblerARMv7::andnotDouble): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::andnotDouble): (JSC::MacroAssemblerMIPS::supportsDoubleBitops): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::supportsDoubleBitops): (JSC::MacroAssemblerSH4::andnotDouble): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::supportsDoubleBitops): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::andnotDouble): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::supportsDoubleBitops): * assembler/X86Assembler.h: (JSC::X86Assembler::andnpd_rr): * create_hash_table: * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::finalize): (JSC::SpecializedThunkJIT::callDoubleToDouble): * jit/ThunkGenerators.cpp: (JSC::floorThunkGenerator): (JSC::ceilThunkGenerator): (JSC::roundThunkGenerator): (JSC::expThunkGenerator): (JSC::logThunkGenerator): (JSC::absThunkGenerator): * jit/ThunkGenerators.h: 2011-07-01 David Kilzer <ddkilzer@apple.com> <http://webkit.org/b/63814> Fix clang build error in JITOpcodes32_64.cpp Fixes the following build error in clang: JavaScriptCore/jit/JITOpcodes32_64.cpp:741:36:{741:9-741:35}: error: operator '?:' has lower precedence than '+'; '+' will be evaluated first [-Werror,-Wparentheses,3] map(m_bytecodeOffset + dynamic ? OPCODE_LENGTH(op_resolve_global_dynamic) : OPCODE_LENGTH(op_resolve_global), dst, regT1, regT0); ~~~~~~~~~~~~~~~~~~~~~~~~~~ ^ JavaScriptCore/jit/JITOpcodes32_64.cpp:741:36: note: place parentheses around the '+' expression to silence this warning [3] map(m_bytecodeOffset + dynamic ? OPCODE_LENGTH(op_resolve_global_dynamic) : OPCODE_LENGTH(op_resolve_global), dst, regT1, regT0); ^ ( ) fix-it:"JavaScriptCore/jit/JITOpcodes32_64.cpp":{741:9-741:9}:"(" fix-it:"JavaScriptCore/jit/JITOpcodes32_64.cpp":{741:35-741:35}:")" JavaScriptCore/jit/JITOpcodes32_64.cpp:741:36:{741:28-741:94}: note: place parentheses around the '?:' expression to evaluate it first [3] map(m_bytecodeOffset + dynamic ? OPCODE_LENGTH(op_resolve_global_dynamic) : OPCODE_LENGTH(op_resolve_global), dst, regT1, regT0); ~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 1 error generated. * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_resolve_global): Add parenthesis to make the tertiary expression evaluate first. 2011-07-01 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r90177 and r90179. http://trac.webkit.org/changeset/90177 http://trac.webkit.org/changeset/90179 https://bugs.webkit.org/show_bug.cgi?id=63790 It caused crashes on Qt in debug mode (Requested by Ossy on #webkit). * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::rshift32): (JSC::MacroAssemblerARM::supportsFloatingPointSqrt): (JSC::MacroAssemblerARM::sqrtDouble): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::supportsFloatingPointSqrt): (JSC::MacroAssemblerARMv7::sqrtDouble): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::sqrtDouble): (JSC::MacroAssemblerMIPS::supportsFloatingPointSqrt): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::sqrtDouble): * assembler/MacroAssemblerX86.h: * assembler/MacroAssemblerX86Common.h: * assembler/MacroAssemblerX86_64.h: * assembler/X86Assembler.h: * create_hash_table: * jit/JSInterfaceJIT.h: (JSC::JSInterfaceJIT::emitLoadDouble): * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::finalize): * jit/ThunkGenerators.cpp: * jit/ThunkGenerators.h: 2011-06-30 Oliver Hunt <oliver@apple.com> Reviewed by Beth Dakin. Make GC validation clear cell structure on destruction https://bugs.webkit.org/show_bug.cgi?id=63778 * runtime/JSCell.h: (JSC::JSCell::JSCell::~JSCell): 2011-06-30 Geoffrey Garen <ggaren@apple.com> Reviewed by Gavin Barraclough. Added write barrier that was missing from put_by_id_transition https://bugs.webkit.org/show_bug.cgi?id=63775 * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::writeBarrier): Made this static with a MacroAssembler& argument so our patching functions could use it. (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCodeGenerator.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): Updated for signature change. * dfg/DFGRepatch.cpp: (JSC::DFG::tryCachePutByID): Missing barrier! * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): Updated for signature change. * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompilePutByIdTransition): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::privateCompilePutByIdTransition): * jit/JSInterfaceJIT.h: Same game here. Removed storePtrWithWriteBarrier because its meaning isn't clear -- maybe in the future we'll have a clear way to pass all stores through a common function that guarantees a write barrier, but that's not the case right now. 2011-06-30 Filip Pizlo <fpizlo@apple.com> Reviewed by Gavin Barraclough. DFG non-speculative JIT does not reuse registers when compiling comparisons. https://bugs.webkit.org/show_bug.cgi?id=63565 * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::knownConstantArithOp): (JSC::DFG::NonSpeculativeJIT::basicArithOp): (JSC::DFG::NonSpeculativeJIT::compare): 2011-06-30 Geoffrey Garen <ggaren@apple.com> Reviewed by Gavin Barraclough. Added empty write barrier stubs in all the right places in the DFG JIT https://bugs.webkit.org/show_bug.cgi?id=63764 SunSpider thinks this might be a 0.5% speedup. Meh. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::writeBarrier): Le stub. (JSC::DFG::JITCodeGenerator::cachedPutById): Don't do anything special for the case where base == scratch, since we now require base and scratch to be not equal, for the sake of the write barrier. * dfg/DFGJITCodeGenerator.h: Le stub. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): Don't reuse the base register as the scratch register, since that's incompatible with the write barrier, which needs a distinct base and scratch. Do put the global object into a register before loading its var storage, since it needs to be in a register for the write barrier to operate on it. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JITPropertyAccess.cpp: (JSC::JIT::emitWriteBarrier): Second verse, same as the first. * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_get_scoped_var): (JSC::JIT::emit_op_put_scoped_var): (JSC::JIT::emit_op_put_global_var): Deployed offsetOfRegisters() to more places. (JSC::JIT::emitWriteBarrier): Added a teeny tiny ASSERT so this function is a little more than meaningless. * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_get_scoped_var): (JSC::JIT::emit_op_put_scoped_var): (JSC::JIT::emit_op_put_global_var): Deployed offsetOfRegisters() to more places. (JSC::JIT::emitWriteBarrier): Added a teeny tiny ASSERT so this function is a little more than meaningless. * runtime/JSVariableObject.h: (JSC::JSVariableObject::offsetOfRegisters): Now used by the JIT, since we put the global object in a register and only then load its var storage by offset. (JSC::JIT::emitWriteBarrier): 2011-06-30 Oliver Hunt <oliver@apple.com> Fix ARMv6 build * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::rshift32): 2011-06-30 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Add optimised paths for a few maths functions https://bugs.webkit.org/show_bug.cgi?id=63757 This adds specialised thunks for Math.abs, Math.round, Math.ceil, Math.floor, Math.log, and Math.exp as they are apparently more important in real web content than we thought, which is somewhat mind-boggling. On average doubles the performance of the common cases (eg. actually passing numbers in). They're not as efficient as they could be, but this way gives them the most portability. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::supportsDoubleBitops): (JSC::MacroAssemblerARM::andnotDouble): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::supportsDoubleBitops): (JSC::MacroAssemblerARMv7::andnotDouble): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::andnotDouble): (JSC::MacroAssemblerMIPS::supportsDoubleBitops): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::supportsDoubleBitops): (JSC::MacroAssemblerSH4::andnotDouble): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::supportsDoubleBitops): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::andnotDouble): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::supportsDoubleBitops): * assembler/X86Assembler.h: (JSC::X86Assembler::andnpd_rr): * create_hash_table: * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::finalize): (JSC::SpecializedThunkJIT::callDoubleToDouble): * jit/ThunkGenerators.cpp: (JSC::floorThunkGenerator): (JSC::ceilThunkGenerator): (JSC::roundThunkGenerator): (JSC::expThunkGenerator): (JSC::logThunkGenerator): (JSC::absThunkGenerator): * jit/ThunkGenerators.h: 2011-06-30 Cary Clark <caryclark@google.com> Reviewed by James Robinson. Use Skia if Skia on Mac Chrome is enabled https://bugs.webkit.org/show_bug.cgi?id=62999 * wtf/Platform.h: Add switch to use Skia if, externally, Skia has been enabled by a gyp define. 2011-06-30 Juan C. Montemayor <jmont@apple.com> Reviewed by Geoffrey Garen. Web Inspector fails to display source for eval with syntax error https://bugs.webkit.org/show_bug.cgi?id=63583 Web Inspector now displays a link to an eval statement that contains a syntax error. * parser/Parser.h: (JSC::isEvalNode): (JSC::EvalNode): (JSC::Parser::parse): 2011-06-30 Filip Pizlo <fpizlo@apple.com> Reviewed by Gavin Barraclough. X86Assembler does not encode byte registers in 64-bit mode correctly. https://bugs.webkit.org/show_bug.cgi?id=63665 * assembler/X86Assembler.h: (JSC::X86Assembler::testb_rr): (JSC::X86Assembler::X86InstructionFormatter::oneByteOp8): 2011-06-30 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r90102. http://trac.webkit.org/changeset/90102 https://bugs.webkit.org/show_bug.cgi?id=63714 Lots of tests asserting beneath SVGSMILElement::findInstanceTime (Requested by aroben on #webkit). * wtf/StdLibExtras.h: (WTF::binarySearch): 2011-06-30 Oliver Varga <Varga.Oliver@stud.u-szeged.hu> Reviewed by Nikolas Zimmermann. Speed up SVGSMILElement::findInstanceTime. https://bugs.webkit.org/show_bug.cgi?id=61025 Add a new parameter to StdlibExtras.h::binarySerarch function to also handle cases when the array does not contain the key value. This is needed for an svg function. * wtf/StdLibExtras.h: (WTF::binarySearch): 2011-06-29 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen. https://bugs.webkit.org/show_bug.cgi?id=63669 DFG JIT - fix spectral-norm regression The problem is a mis-speculation leading to us falling off the speculative path. Make the speculation logic slightly smarter, don't predict int if one of the operands is already loaded as a double (we use this logic already for compares). * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::shouldSpeculateInteger): 2011-06-29 Filip Pizlo <fpizlo@apple.com> Reviewed by Gavin Barraclough. DFG JIT does not do put_by_id transition caching. https://bugs.webkit.org/show_bug.cgi?id=63662 * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::addPropertyAccess): * dfg/DFGRepatch.cpp: (JSC::DFG::testPrototype): (JSC::DFG::tryCachePutByID): 2011-06-29 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Added a dummy write barrier emitting function in all the right places in the old JIT https://bugs.webkit.org/show_bug.cgi?id=63667 SunSpider reports no change. * jit/JIT.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_put_by_id): (JSC::JIT::emit_op_put_scoped_var): Do it. (JSC::JIT::emit_op_put_global_var): Global object needs to be in a register for the sake of the write barrier. (JSC::JIT::emitWriteBarrier): Empty for now. Not for long! * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_put_by_val): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::emit_op_put_scoped_var): Do it. (JSC::JIT::emit_op_put_global_var): Global object needs to be in a register for the sake of the write barrier. (JSC::JIT::emitWriteBarrier): Empty for now. Not for long! 2011-06-29 Filip Pizlo <fpizlo@apple.com> Reviewed by Gavin Barraclough. DFG JIT does not perform get_by_id self list caching. https://bugs.webkit.org/show_bug.cgi?id=63605 * bytecode/StructureStubInfo.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGRepatch.cpp: (JSC::DFG::tryCacheGetByID): (JSC::DFG::tryBuildGetByIDList): (JSC::DFG::dfgBuildGetByIDList): * dfg/DFGRepatch.h: 2011-06-28 Filip Pizlo <fpizlo@apple.com> Reviewed by Gavin Barraclough. DFG JIT lacks array.length caching. https://bugs.webkit.org/show_bug.cgi?id=63505 * bytecode/StructureStubInfo.h: * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::cachedGetById): (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::tryAllocate): (JSC::DFG::JITCodeGenerator::selectScratchGPR): (JSC::DFG::JITCodeGenerator::silentSpillAllRegisters): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::addPropertyAccess): (JSC::DFG::JITCompiler::PropertyAccessRecord::PropertyAccessRecord): * dfg/DFGRegisterBank.h: (JSC::DFG::RegisterBank::tryAllocate): * dfg/DFGRepatch.cpp: (JSC::DFG::tryCacheGetByID): 2011-06-28 Pierre Rossi <pierre.rossi@gmail.com> Reviewed by Eric Seidel. Warnings in JSC's JIT on 32 bit https://bugs.webkit.org/show_bug.cgi?id=63259 Fairly straightforward, just use ASSERT_JIT_OFFSET_UNUSED when it applies. * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_method_check): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::emit_op_put_by_id): 2011-06-28 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r89968. http://trac.webkit.org/changeset/89968 https://bugs.webkit.org/show_bug.cgi?id=63581 Broke chromium windows compile (Requested by jamesr on #webkit). * wtf/Platform.h: 2011-06-28 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Fix sampling build https://bugs.webkit.org/show_bug.cgi?id=63579 Gets opcode sampling building again, doesn't seem to work alas * bytecode/SamplingTool.cpp: (JSC::SamplingTool::notifyOfScope): * bytecode/SamplingTool.h: (JSC::SamplingTool::SamplingTool): * interpreter/Interpreter.cpp: (JSC::Interpreter::enableSampler): * runtime/Executable.h: (JSC::ScriptExecutable::ScriptExecutable): 2011-06-28 Cary Clark <caryclark@google.com> Reviewed by James Robinson. Use Skia if Skia on Mac Chrome is enabled https://bugs.webkit.org/show_bug.cgi?id=62999 * wtf/Platform.h: Add switch to use Skia if, externally, Skia has been enabled by a gyp define. 2011-06-28 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. ASSERT when launching debug builds with interpreter and jit enabled https://bugs.webkit.org/show_bug.cgi?id=63566 Add appropriate guards to the various Executable's memory reporting logic. * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): 2011-06-28 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=63563 DFG JIT - add support for double arith to speculative path Add integer support for div & mod, add double support for div, mod, add, sub & mul, dynamically selecting based on operand types. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::FPRTemporary::FPRTemporary): * dfg/DFGJITCodeGenerator.h: * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::assembler): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateDouble): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculateDoubleOperand::SpeculateDoubleOperand): (JSC::DFG::SpeculateDoubleOperand::~SpeculateDoubleOperand): (JSC::DFG::SpeculateDoubleOperand::index): (JSC::DFG::SpeculateDoubleOperand::fpr): 2011-06-28 Oliver Hunt <oliver@apple.com> Fix interpreter build. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): 2011-06-28 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=63561 DFG JIT - don't always assume integer in relational compare If neither operand is known integer, or either is in double representation, then at least use a function call (don't bail off the speculative path). * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleCall): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isDataFormatDouble): (JSC::DFG::SpeculativeJIT::compareIsInteger): 2011-06-28 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Make constant array optimisation less strict about what constitutes a constant https://bugs.webkit.org/show_bug.cgi?id=63554 Now allow string constants in array literals to actually be considered constant, and so avoid codegen in array literals with strings in them. * bytecode/CodeBlock.h: (JSC::CodeBlock::addConstantBuffer): (JSC::CodeBlock::constantBuffer): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::addConstantBuffer): (JSC::BytecodeGenerator::addStringConstant): (JSC::BytecodeGenerator::emitNewArray): * bytecompiler/BytecodeGenerator.h: * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): 2011-06-28 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=63560 DFG_JIT allow allocation of specific machine registers This allow us to allocate the registers necessary to perform x86 idiv instructions for div/mod, and may be useful for shifts, too. * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::GPRTemporary::GPRTemporary): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::allocate): (JSC::DFG::GPRResult::GPRResult): * dfg/DFGRegisterBank.h: (JSC::DFG::RegisterBank::allocateSpecific): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isInteger): 2011-06-28 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=55040 RegExp constructor returns the argument regexp instead of a new object Per 15.10.3.1, our current behaviour is correct if called as a function, but incorrect when called as a constructor. * runtime/RegExpConstructor.cpp: (JSC::constructRegExp): (JSC::constructWithRegExpConstructor): * runtime/RegExpConstructor.h: 2011-06-28 Luke Macpherson <macpherson@chromium.org> Reviewed by Darin Adler. Clean up integer clamping functions in MathExtras.h and support arbitrary numeric types and limits. https://bugs.webkit.org/show_bug.cgi?id=63469 * wtf/MathExtras.h: (defaultMinimumForClamp): Version of std::numeric_limits::min() that returns the largest negative value for floating point types. (defaultMaximumForClamp): Symmetric alias for std::numeric_limits::max() (clampTo): New templated clamping function that supports arbitrary output types. (clampToInteger): Use new clampTo template. (clampToFloat): Use new clampTo template. (clampToPositiveInteger): Use new clampTo template. 2011-06-28 Adam Roben <aroben@apple.com> Windows Debug build fix after r89885 * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Exported JSGlobalData::releaseExecutableMemory for jsc.exe's benefit. 2011-06-28 Shinya Kawanaka <shinyak@google.com> Reviewed by Kent Tamura. Add const to show() method in WTFString and AtomicString. https://bugs.webkit.org/show_bug.cgi?id=63515 The lack of const in show() method is painful when doing something like printf-debug. * wtf/text/AtomicString.cpp: (WTF::AtomicString::show): * wtf/text/AtomicString.h: * wtf/text/WTFString.cpp: (String::show): * wtf/text/WTFString.h: 2011-06-27 Ryosuke Niwa <rniwa@webkit.org> Build fix attempt after r89885. * JavaScriptCore.exp: * jsc.cpp: 2011-06-27 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Support throwing away non-running code even while other code is running https://bugs.webkit.org/show_bug.cgi?id=63485 Add a function to CodeBlock to support unlinking direct linked callsites, and then with that in place add logic to discard code from any function that is not currently on the stack. The unlinking completely reverts any optimized call sites, such that they may be relinked again in future. * JavaScriptCore.exp: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::unlinkCalls): (JSC::CodeBlock::clearEvalCache): * bytecode/CodeBlock.h: (JSC::CallLinkInfo::CallLinkInfo): (JSC::CallLinkInfo::unlink): * bytecode/EvalCodeCache.h: (JSC::EvalCodeCache::clear): * heap/Heap.cpp: (JSC::Heap::getConservativeRegisterRoots): * heap/Heap.h: * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JIT.h: * jit/JITCall.cpp: (JSC::JIT::compileOpCall): * jit/JITWriteBarrier.h: (JSC::JITWriteBarrierBase::clear): * jsc.cpp: (GlobalObject::GlobalObject): (functionReleaseExecutableMemory): * runtime/Executable.cpp: (JSC::EvalExecutable::unlinkCalls): (JSC::ProgramExecutable::unlinkCalls): (JSC::FunctionExecutable::discardCode): (JSC::FunctionExecutable::unlinkCalls): * runtime/Executable.h: * runtime/JSGlobalData.cpp: (JSC::SafeRecompiler::returnValue): (JSC::SafeRecompiler::operator()): (JSC::JSGlobalData::releaseExecutableMemory): 2011-06-27 Gavin Barraclough <barraclough@apple.com> Reviewed by Darin Adler & Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=50554 RegExp.prototype.toString does not escape slashes The problem here is that we don't escape forwards slashes when converting a RegExp to a string. This means that RegExp("/").toString() is "///", which is not a valid RegExp literal. Also, we return an invalid literal for RegExp.prototype.toString() ("//", which is an empty single-line comment). From ES5: "NOTE: The returned String has the form of a RegularExpressionLiteral that evaluates to another RegExp object with the same behaviour as this object." * runtime/RegExpObject.cpp: (JSC::regExpObjectSource): - Escape forward slashes when getting the source of a RegExp. * runtime/RegExpPrototype.cpp: (JSC::regExpProtoFuncToString): - Remove unnecessary and erroneous hack to return "//" as the string representation of RegExp.prototype. This is not a valid RegExp literal (it is an empty single-line comment). 2011-06-27 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=63497 Add DEBUG_WITH_BREAKPOINT support to the DFG JIT. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-06-27 Juan C. Montemayor <jmont@apple.com> Reviewed by Mark Rowe. Indirectly including TextPosition.h and XPathGrammar.h causes compile errors https://bugs.webkit.org/show_bug.cgi?id=63392 When both TextPosition.h and XPathGrammar.h are included a compile-error is caused, since XPathGrammar.h defines a macro called NUMBER and TextPosition has a typedef named NUMBER. * wtf/text/TextPosition.h: (WTF::TextPosition::TextPosition): (WTF::TextPosition::minimumPosition): (WTF::TextPosition::belowRangePosition): 2011-06-27 Filip Pizlo <fpizlo@apple.com> Reviewed by Gavin Barraclough. DFG JIT does not perform put_by_id caching. https://bugs.webkit.org/show_bug.cgi?id=63409 * bytecode/StructureStubInfo.h: * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::cachedPutById): * dfg/DFGJITCodeGenerator.h: * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::addPropertyAccess): (JSC::DFG::JITCompiler::PropertyAccessRecord::PropertyAccessRecord): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGOperations.cpp: * dfg/DFGOperations.h: * dfg/DFGRepatch.cpp: (JSC::DFG::dfgRepatchByIdSelfAccess): (JSC::DFG::tryCacheGetByID): (JSC::DFG::appropriatePutByIdFunction): (JSC::DFG::tryCachePutByID): (JSC::DFG::dfgRepatchPutByID): * dfg/DFGRepatch.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-06-27 Gustavo Noronha Silva <gns@gnome.org> Unreviewed build fix. One more filed missing during distcheck, for the MIPS build. * GNUmakefile.list.am: 2011-06-26 Filip Pizlo <fpizlo@apple.com> Reviewed by Gavin Barraclough. DFG non-speculative JIT has potentially harmful speculations with respect to arithmetic operations. https://bugs.webkit.org/show_bug.cgi?id=63347 * dfg/DFGNonSpeculativeJIT.cpp: - Changed arithmetic operations to speculate in favor of integers. (JSC::DFG::NonSpeculativeJIT::valueToNumber): (JSC::DFG::NonSpeculativeJIT::knownConstantArithOp): (JSC::DFG::NonSpeculativeJIT::basicArithOp): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGNonSpeculativeJIT.h: * dfg/DFGOperations.cpp: - Added slow-path routines for arithmetic that perform no speculation; the non-speculative JIT will generate calls to these in cases where its speculation fails. * dfg/DFGOperations.h: 2011-06-24 Nikolas Zimmermann <nzimmermann@rim.com> Reviewed by Rob Buis. Integrate SVG Fonts within GlyphPage concept, removing the special SVG code paths from Font, making it possible to reuse the simple text code path for SVG Fonts https://bugs.webkit.org/show_bug.cgi?id=59085 * wtf/Platform.h: Force Qt-EWS into a full rebuild, otherwhise this patch breaks the EWS. 2011-06-24 Michael Saboff <msaboff@apple.com> Reviewed by Gavin Barraclough. Arm Assembler, Immediate stack offset values truncated to 8 bits for add & sub https://bugs.webkit.org/show_bug.cgi?id=63345 The methods ARMThumbImmediate::getUInt9 and ARMThumbImmediate::getUInt10 return 9 and 10 bit quantities, therefore changed their return type from uint8_t to uint16_t. Also casted the places where they are used as they are currently shifted and used as 7 or 8 bit values. These methods are currently used for literals for stack offsets, including creating and destroying stack frames. The prior truncation of the upper bits caused stack frames to be too small, thus allowing a JIT'ed function to access and overwrite stack space outside of the incorrectly sized stack frame. * assembler/ARMv7Assembler.h: (JSC::ARMThumbImmediate::getUInt9): (JSC::ARMThumbImmediate::getUInt10): (JSC::ARMv7Assembler::add): (JSC::ARMv7Assembler::ldr): (JSC::ARMv7Assembler::str): (JSC::ARMv7Assembler::sub): (JSC::ARMv7Assembler::sub_S): 2011-06-24 Michael Saboff <msaboff@apple.com> Reviewed by Geoffrey Garen. releaseFastMallocFreeMemory doesn't adjust free counts for scavenger https://bugs.webkit.org/show_bug.cgi?id=63015 Added code to adjust class TCMalloc_PageHeap variables free_committed_pages_ and min_free_committed_pages_since_last_scavenge_ in ReleaseFreeList(). These adjustments are a bug. These need to reflect the pages that are released in ReleaseFreeLsit so that scavenge doesn't try to free that many pages as well. Made ReleaseFreeList a member of TCMalloc_PageHeap in the process. Updated Check() and helper method CheckList() to check the number of actual free pages with free_committed_pages_. The symptom of the problem of the existing code is that the scavenger may run unneccesarily without any real work to do, i.e. pages on the free lists. The scanvenger would also end up freeing too many pages, that is going below the current 528 target free pages. Note that the style of the changes was kept consistent with the existing style. * wtf/FastMalloc.cpp: (WTF::TCMalloc_PageHeap::Check): (WTF::TCMalloc_PageHeap::CheckList): (WTF::TCMalloc_PageHeap::ReleaseFreeList): 2011-06-24 Abhishek Arya <inferno@chromium.org> Reviewed by Darin Adler. Match other clampTo* functions in style with clampToInteger(float) function. https://bugs.webkit.org/show_bug.cgi?id=53449 * wtf/MathExtras.h: (clampToInteger): (clampToFloat): (clampToPositiveInteger): 2011-06-24 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r89594. http://trac.webkit.org/changeset/89594 https://bugs.webkit.org/show_bug.cgi?id=63316 It broke 5 tests on the Qt bot (Requested by Ossy_DC on #webkit). * GNUmakefile.list.am: * JavaScriptCore.gypi: * icu/unicode/uscript.h: Removed. * wtf/unicode/ScriptCodesFromICU.h: Removed. * wtf/unicode/brew/UnicodeBrew.h: * wtf/unicode/glib/UnicodeGLib.h: * wtf/unicode/icu/UnicodeIcu.h: * wtf/unicode/qt4/UnicodeQt4.h: * wtf/unicode/wince/UnicodeWinCE.h: 2011-06-23 Filip Pizlo <fpizlo@apple.com> Reviewed by Gavin Barraclough. DFG non-speculative JIT should have obvious optimizations for GetById and GetByVal https://bugs.webkit.org/show_bug.cgi?id=63173 * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::cachedGetById): * dfg/DFGJITCodeGenerator.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-06-23 Oliver Hunt <oliver@apple.com> Fix Qt again. * assembler/ARMAssembler.h: (JSC::ARMAssembler::readPointer): 2011-06-23 Oliver Hunt <oliver@apple.com> Fix Qt Build * assembler/ARMAssembler.h: (JSC::ARMAssembler::readPointer): 2011-06-23 Stephanie Lewis <slewis@apple.com> Reviewed by Darin Adler. https://bugs.webkit.org/show_bug.cgi?id=63298 Replace Malloc with FastMalloc to match the rest of wtf. * wtf/BlockStack.h: (WTF::::~BlockStack): (WTF::::grow): (WTF::::shrink): 2011-06-23 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Add the ability to dynamically modify linked call sites https://bugs.webkit.org/show_bug.cgi?id=63291 Add JITWriteBarrier as a writebarrier class that allows reading and writing directly into the code stream. This required adding logic to all the assemblers to allow us to read values back out of the instruction stream. * JavaScriptCore.xcodeproj/project.pbxproj: * assembler/ARMAssembler.h: (JSC::ARMAssembler::readPointer): * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::readPointer): (JSC::ARMv7Assembler::readInt32): (JSC::ARMv7Assembler::decodeTwoWordOp5i6Imm4Reg4EncodedImmFirst): (JSC::ARMv7Assembler::decodeTwoWordOp5i6Imm4Reg4EncodedImmSecond): * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::readPointer): * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::readInt32): (JSC::MIPSAssembler::readPointer): * assembler/MacroAssemblerCodeRef.h: (JSC::MacroAssemblerCodePtr::operator!): * assembler/SH4Assembler.h: (JSC::SH4Assembler::readPCrelativeAddress): (JSC::SH4Assembler::readPointer): (JSC::SH4Assembler::readInt32): * assembler/X86Assembler.h: (JSC::X86Assembler::readPointer): * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): * bytecode/CodeBlock.h: (JSC::MethodCallLinkInfo::seenOnce): (JSC::MethodCallLinkInfo::setSeen): * heap/MarkStack.h: * jit/JIT.cpp: (JSC::JIT::privateCompile): (JSC::JIT::linkCall): (JSC::JIT::linkConstruct): * jit/JITPropertyAccess.cpp: (JSC::JIT::patchMethodCallProto): * jit/JITPropertyAccess32_64.cpp: * jit/JITWriteBarrier.h: Added. (JSC::JITWriteBarrierBase::operator UnspecifiedBoolType*): (JSC::JITWriteBarrierBase::operator!): (JSC::JITWriteBarrierBase::setFlagOnBarrier): (JSC::JITWriteBarrierBase::isFlagged): (JSC::JITWriteBarrierBase::setLocation): (JSC::JITWriteBarrierBase::location): (JSC::JITWriteBarrierBase::JITWriteBarrierBase): (JSC::JITWriteBarrierBase::set): (JSC::JITWriteBarrierBase::get): (JSC::JITWriteBarrier::JITWriteBarrier): (JSC::JITWriteBarrier::set): (JSC::JITWriteBarrier::get): (JSC::MarkStack::append): 2011-06-23 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=61585 Crash running regexp /(?:(?=g))|(?:m).{2147483648,}/ This is due to use of int instead of unsigned, bad math around the 2^31 boundary. * yarr/YarrInterpreter.cpp: (JSC::Yarr::ByteCompiler::emitDisjunction): - Change some uses of int to unsigned, refactor compare logic to restrict to the range 0..2^32-1 (rather than -2^32-1..2^32-1). * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::generate): (JSC::Yarr::YarrGenerator::backtrack): - Ditto. 2011-06-22 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=63218 DFG JIT - remove machine type guarantees from graph The DFG JIT currently makes assumptions about the types of machine registers that certain nodes will be loaded into. This will be broken as we generate nodes to produce both integer and double code paths. Remove int<->double conversions nodes. This design decision also gave rise to multiple types of constant nodes, requiring separate handling for each type. Merge these back into JSConstant. * dfg/DFGAliasTracker.h: (JSC::DFG::AliasTracker::equalIgnoringLaterNumericConversion): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getToInt32): (JSC::DFG::ByteCodeParser::getToNumber): (JSC::DFG::ByteCodeParser::toInt32): (JSC::DFG::ByteCodeParser::toNumber): (JSC::DFG::ByteCodeParser::isInt32Constant): (JSC::DFG::ByteCodeParser::isDoubleConstant): (JSC::DFG::ByteCodeParser::valueOfInt32Constant): (JSC::DFG::ByteCodeParser::valueOfDoubleConstant): (JSC::DFG::ByteCodeParser::one): (JSC::DFG::ByteCodeParser::predictInt32): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentFillGPR): (JSC::DFG::JITCodeGenerator::silentFillFPR): (JSC::DFG::JITCodeGenerator::isJSConstant): (JSC::DFG::JITCodeGenerator::isDoubleConstant): (JSC::DFG::JITCodeGenerator::valueOfJSConstantAsImmPtr): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::fillNumericToDouble): (JSC::DFG::JITCompiler::fillInt32ToInteger): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::isJSConstant): (JSC::DFG::JITCompiler::isInt32Constant): (JSC::DFG::JITCompiler::isDoubleConstant): (JSC::DFG::JITCompiler::valueOfJSConstant): (JSC::DFG::JITCompiler::valueOfInt32Constant): (JSC::DFG::JITCompiler::valueOfDoubleConstant): * dfg/DFGNode.h: (JSC::DFG::Node::Node): (JSC::DFG::Node::isConstant): (JSC::DFG::Node::notTakenBytecodeOffset): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::isKnownInteger): (JSC::DFG::NonSpeculativeJIT::isKnownNumeric): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::compilePeepHoleIntegerBranch): (JSC::DFG::SpeculativeJIT::compile): 2011-06-23 Jungshik Shin <jshin@chromium.org> Reviewed by Alexey Proskuryakov. Add ScriptCodesFromICU.h to wtf/unicode and make necessary changes in build files for ports not using ICU. Add icu/unicode/uscript.h for ports using ICU. It's taken from ICU 3.6 (the version used on Mac OS 10.5) http://bugs.webkit.org/show_bug.cgi?id=20797 * GNUmakefile.list.am: * JavaScriptCore.gypi: * icu/unicode/uscript.h: Added for UScriptCode enum. * wtf/unicode/ScriptCodesFromICU.h: UScriptCode enum added. * wtf/unicode/icu/UnicodeIcu.h: * wtf/unicode/brew/UnicodeBrew.h: * wtf/unicode/glib/UnicodeGLib.h: * wtf/unicode/qt4/UnicodeQt4.h: * wtf/unicode/wince/UnicodeWinCE.h: 2011-06-23 Ryuan Choi <ryuan.choi@samsung.com> Reviewed by Andreas Kling. [EFL][WK2] Add PLATFORM(EFL) to use UNIX_DOMAIN_SOCKETS. https://bugs.webkit.org/show_bug.cgi?id=63228 * wtf/Platform.h: Add PLATFORM(EFL) guard. 2011-06-23 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r89547. http://trac.webkit.org/changeset/89547 https://bugs.webkit.org/show_bug.cgi?id=63252 "Chrmium crash on start" (Requested by yurys on #webkit). * wtf/DynamicAnnotations.cpp: (WTFAnnotateBenignRaceSized): (WTFAnnotateHappensBefore): (WTFAnnotateHappensAfter): * wtf/DynamicAnnotations.h: 2011-06-23 Timur Iskhodzhanov <timurrrr@google.com> Reviewed by David Levin. Make dynamic annotations weak symbols and prevent identical code folding by the linker https://bugs.webkit.org/show_bug.cgi?id=62443 * wtf/DynamicAnnotations.cpp: (WTFAnnotateBenignRaceSized): (WTFAnnotateHappensBefore): (WTFAnnotateHappensAfter): * wtf/DynamicAnnotations.h: 2011-06-22 Yael Aharon <yael.aharon@nokia.com> Reviewed by Andreas Kling. [Qt] Add a build flag for building with libxml2 and libxslt. https://bugs.webkit.org/show_bug.cgi?id=63113 * wtf/Platform.h: 2011-06-22 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r89489. http://trac.webkit.org/changeset/89489 https://bugs.webkit.org/show_bug.cgi?id=63203 Broke chromium mac build on build.webkit.org (Requested by abarth on #webkit). * wtf/Platform.h: 2011-06-22 Cary Clark <caryclark@google.com> Reviewed by Darin Fisher. Use Skia if Skia on Mac Chrome is enabled https://bugs.webkit.org/show_bug.cgi?id=62999 * wtf/Platform.h: Add switch to use Skia if, externally, Skia has been enabled by a gyp define. 2011-06-22 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. * interpreter/RegisterFile.h: Removed unnecessary #include <stdio.h>. 2011-06-22 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed the conceit that global variables are local variables when running global code https://bugs.webkit.org/show_bug.cgi?id=63106 This is required for write barrier correctness. SunSpider reports about a 0.5% regression, mostly from bitops-bitwise-and.js. I was able to reduce the regression with a tiny peephole optimization in the bytecompiler, but not eliminate it. I'm committing this assuming that turning on generational GC will win back at least 0.5%. (FWIW, the DFG JIT can easily eliminate any regression by sharing loads of the global object's var storage. I considered doing the same kind of optimization in the existing JIT, but it seemed like moving in the wrong direction.) * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::addGlobalVar): (JSC::BytecodeGenerator::BytecodeGenerator): Don't give global variables negative indices, since they're no longer negatively offset from the current stack frame. Do give global variables monotonically increasing positive indices, since that's much easier to work with. Don't limit the number of optimizable global variables, since it's no longer limited by the register file, since they're no longer stored in the register file. (JSC::BytecodeGenerator::registerFor): Global code never has any local registers because a var in global code is actually a property of the global object. (JSC::BytecodeGenerator::constRegisterFor): Ditto. (JSC::BytecodeGenerator::emitResolve): Did a tiny bit of constant propagation and dead code elimination to speed up our compiles and reduce WTFs / minute. * bytecompiler/BytecodeGenerator.h: (JSC::BytecodeGenerator::registerFor): Removed special handling of globals. (JSC::BytecodeGenerator::shouldOptimizeLocals): Don't optimize locals in global code, since there are none. (JSC::BytecodeGenerator::canOptimizeNonLocals): Do optimize non-locals in global code (i.e., global vars), since there are some. * interpreter/Interpreter.cpp: (JSC::Interpreter::callEval): (JSC::Interpreter::Interpreter): (JSC::Interpreter::dumpRegisters): (JSC::Interpreter::execute): * interpreter/Interpreter.h: Updated for deleted / renamed code. * interpreter/RegisterFile.cpp: (JSC::RegisterFile::gatherConservativeRoots): (JSC::RegisterFile::releaseExcessCapacity): Updated for deleted / renamed data members. * interpreter/RegisterFile.h: (JSC::RegisterFile::begin): (JSC::RegisterFile::size): (JSC::RegisterFile::RegisterFile): (JSC::RegisterFile::shrink): Removed all code and comments dealing with global variables stored in the register file. (JSC::RegisterFile::grow): Updated for same. Also, a slight correctness fix: Test the VM commit end, and not just the in-use end, when checking for stack overflow. In theory, it's invalid to commit past the end of your allocation, even if you never touch that memory. This makes the usable size of the stack slightly smaller. No test because we don't know of any case in practice where this crashes. * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): Updated for changes above. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::resizeRegisters): (JSC::JSGlobalObject::addStaticGlobals): * runtime/JSGlobalObject.h: Simplified globals to have monotonically increasing indexes, always located in our external storage. 2011-06-21 MORITA Hajime <morrita@google.com> Unreviewed, rolling out r89401 and r89403. http://trac.webkit.org/changeset/89401 http://trac.webkit.org/changeset/89403 https://bugs.webkit.org/show_bug.cgi?id=62970 Breaks mac build and mistakenly enables the spellcheck API * Configurations/FeatureDefines.xcconfig: * JavaScriptCore.xcodeproj/project.pbxproj: 2011-06-21 Kent Tamura <tkent@chromium.org> [Mac] Sort Xcode project files. * JavaScriptCore.xcodeproj/project.pbxproj: 2011-06-20 MORITA Hajime <morrita@google.com> Reviewed by Kent Tamura. Spellcheck API should be build-able. https://bugs.webkit.org/show_bug.cgi?id=62970 No new tests, changing only build related files * Configurations/FeatureDefines.xcconfig: 2011-06-21 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Moved 'const' off the global-variable-as-local-variable crack pipe https://bugs.webkit.org/show_bug.cgi?id=63105 This is necessary for moving the rest of the code off of same. Many problems remain in our handling of const. I have fixed none of them. * bytecompiler/BytecodeGenerator.h: (JSC::BytecodeGenerator::scopeChain): New accessor, needed to enable const to directly implement its unique scoping rules. * bytecompiler/NodesCodegen.cpp: (JSC::PrefixResolveNode::emitBytecode): Do specify that our resolve is for writing, so we don't overwrite const variables. (JSC::ConstDeclNode::emitCodeSingle): Don't assume that all declared const variables are available as local variables, since this won't be the case once global variables are not available as local variables. Instead, use put_scoped_var in the case where there is no local variable. Like a local variable, put_scoped_var succeeds even though const properties are read-only, since put_scoped_var skips read-only checks. (Yay?) 2011-06-21 Oliver Hunt <oliver@apple.com> Reviewed by Alexey Proskuryakov. REGRESSION(r89257): It broke 2 jscore tests (Requested by Ossy_away on #webkit). https://bugs.webkit.org/show_bug.cgi?id=63052 Release mode only failure, the stack overflow guards were getting there error handling inlined, so that they were essentially causing their own demise. * parser/JSParser.cpp: (JSC::JSParser::updateErrorMessage): (JSC::JSParser::updateErrorWithNameAndMessage): 2011-06-20 Kenneth Russell <kbr@google.com> Unreviewed. Rolled out r89233 and r89235 because of crashes in http/tests/misc/acid3.html on Snow Leopard and other platforms https://bugs.webkit.org/show_bug.cgi?id=63022 * wtf/Platform.h: 2011-06-18 Anders Carlsson <andersca@apple.com> Reviewed by Darin Adler. Disallow assigning into PassOwnArrayPtr, PassOwnPtr and PassRefPtr https://bugs.webkit.org/show_bug.cgi?id=62940 Remove clear() and all assignment operators except one which now has a COMPILE_ASSERT. * wtf/PassOwnArrayPtr.h: (WTF::PassOwnArrayPtr::operator=): * wtf/PassOwnPtr.h: (WTF::PassOwnPtr::operator=): * wtf/PassRefPtr.h: (WTF::PassRefPtr::operator=): (WTF::NonNullPassRefPtr::operator=): 2011-06-20 Oliver Hunt <oliver@apple.com> Reviewed by Darin Adler. REGRESSION (r79060): Searching for a flight at united.com fails https://bugs.webkit.org/show_bug.cgi?id=63003 This original change also broke Twitter, and we attempted to refine the fix to address that problem (http://trac.webkit.org/changeset/80542), but since it still breaks United, we need to revert the change until we understand the problem better. * wtf/DateMath.cpp: (WTF::parseDateFromNullTerminatedCharacters): 2011-06-20 Juan C. Montemayor <jmont@apple.com> Reviewed by Oliver Hunt. No context for javascript parse errors. https://bugs.webkit.org/show_bug.cgi?id=62613 Parse errors now show more details like: "Unexpected token: ]" or "Expected token: while" For reserved names, numbers, indentifiers, strings, lexer errors, and EOFs, the following error messages are printed: "Use of reserved word: super" "Unexpected number: 42" "Unexpected identifier: " "Unexpected string: "foobar"" "Invalid token character sequence: \u4023" "Unexpected EOF" * parser/JSParser.cpp: (JSC::JSParser::consume): (JSC::JSParser::getToken): (JSC::JSParser::getTokenName): (JSC::JSParser::updateErrorMessageSpecialCase): (JSC::JSParser::updateErrorMessage): (JSC::JSParser::updateErrorWithNameAndMessage): (JSC::jsParse): (JSC::JSParser::JSParser): (JSC::JSParser::parseProgram): (JSC::JSParser::parseVarDeclarationList): (JSC::JSParser::parseForStatement): (JSC::JSParser::parseBreakStatement): (JSC::JSParser::parseContinueStatement): (JSC::JSParser::parseWithStatement): (JSC::JSParser::parseTryStatement): (JSC::JSParser::parseStatement): (JSC::JSParser::parseFormalParameters): (JSC::JSParser::parseFunctionInfo): (JSC::JSParser::parseAssignmentExpression): (JSC::JSParser::parsePrimaryExpression): (JSC::JSParser::parseMemberExpression): (JSC::JSParser::parseUnaryExpression): * parser/JSParser.h: * parser/Lexer.cpp: (JSC::Lexer::lex): * parser/Parser.cpp: (JSC::Parser::parse): 2011-06-20 Nikolas Zimmermann <nzimmermann@rim.com> Reviewed by Rob Buis. Integrate SVG Fonts within GlyphPage concept, removing the special SVG code paths from Font, making it possible to reuse the simple text code path for SVG Fonts https://bugs.webkit.org/show_bug.cgi?id=59085 * wtf/Platform.h: Force Qt-EWS into a full rebuild, otherwhise this patch breaks the EWS. 2011-06-19 Oliver Hunt <oliver@apple.com> Reviewed by Sam Weinig. Correct logic for putting errors on the correct line when handling JSONP https://bugs.webkit.org/show_bug.cgi?id=62962 Minor fix for the minor fix. *sigh* * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): 2011-06-19 Oliver Hunt <oliver@apple.com> Minor fix to correct layout test results. * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): 2011-06-17 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. JSONP is unnecessarily slow https://bugs.webkit.org/show_bug.cgi?id=62920 JSONP has unfortunately become a fairly common idiom online, yet it triggers very poor performance in JSC as we end up doing codegen for a large number of property accesses that will * only be run once, so the vast amount of logic we dump to handle caching of accesses is unnecessary. * We are doing codegen that is directly proportional to just creating the object in the first place. This patch extends the use of the literal parser to JSONP-like structures in global code, handling a number of different forms I have seen online. In an extreme case this improves performance of JSONP by more than 2x due to removal of code generation and execution time, and a few optimisations that I made to the parser itself. * API/JSValueRef.cpp: (JSValueMakeFromJSONString): * interpreter/Interpreter.cpp: (JSC::Interpreter::callEval): (JSC::Interpreter::execute): * parser/Lexer.cpp: (JSC::Lexer::isKeyword): * parser/Lexer.h: * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncEval): * runtime/JSONObject.cpp: (JSC::JSONProtoFuncParse): * runtime/LiteralParser.cpp: (JSC::LiteralParser::tryJSONPParse): (JSC::LiteralParser::makeIdentifier): (JSC::LiteralParser::Lexer::lex): (JSC::LiteralParser::Lexer::next): (JSC::isSafeStringCharacter): (JSC::LiteralParser::Lexer::lexString): (JSC::LiteralParser::Lexer::lexNumber): (JSC::LiteralParser::parse): * runtime/LiteralParser.h: (JSC::LiteralParser::LiteralParser): (JSC::LiteralParser::tryLiteralParse): (JSC::LiteralParser::Lexer::Lexer): 2011-06-18 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r89184. http://trac.webkit.org/changeset/89184 https://bugs.webkit.org/show_bug.cgi?id=62927 It broke 22 tests on all bot (Requested by Ossy_weekend on #webkit). * API/JSValueRef.cpp: (JSValueMakeFromJSONString): * interpreter/Interpreter.cpp: (JSC::Interpreter::callEval): (JSC::Interpreter::execute): * parser/Lexer.cpp: * parser/Lexer.h: * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncEval): * runtime/JSONObject.cpp: (JSC::JSONProtoFuncParse): * runtime/LiteralParser.cpp: (JSC::LiteralParser::Lexer::lex): (JSC::isSafeStringCharacter): (JSC::LiteralParser::Lexer::lexString): (JSC::LiteralParser::Lexer::lexNumber): (JSC::LiteralParser::parse): * runtime/LiteralParser.h: (JSC::LiteralParser::LiteralParser): (JSC::LiteralParser::tryLiteralParse): (JSC::LiteralParser::Lexer::Lexer): (JSC::LiteralParser::Lexer::next): 2011-06-17 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. JSONP is unnecessarily slow https://bugs.webkit.org/show_bug.cgi?id=62920 JSONP has unfortunately become a fairly common idiom online, yet it triggers very poor performance in JSC as we end up doing codegen for a large number of property accesses that will * only be run once, so the vast amount of logic we dump to handle caching of accesses is unnecessary. * We are doing codegen that is directly proportional to just creating the object in the first place. This patch extends the use of the literal parser to JSONP-like structures in global code, handling a number of different forms I have seen online. In an extreme case this improves performance of JSONP by more than 2x due to removal of code generation and execution time, and a few optimisations that I made to the parser itself. * API/JSValueRef.cpp: (JSValueMakeFromJSONString): * interpreter/Interpreter.cpp: (JSC::Interpreter::callEval): (JSC::Interpreter::execute): * parser/Lexer.cpp: (JSC::Lexer::isKeyword): * parser/Lexer.h: * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncEval): * runtime/JSONObject.cpp: (JSC::JSONProtoFuncParse): * runtime/LiteralParser.cpp: (JSC::LiteralParser::tryJSONPParse): (JSC::LiteralParser::makeIdentifier): (JSC::LiteralParser::Lexer::lex): (JSC::LiteralParser::Lexer::next): (JSC::isSafeStringCharacter): (JSC::LiteralParser::Lexer::lexString): (JSC::LiteralParser::Lexer::lexNumber): (JSC::LiteralParser::parse): * runtime/LiteralParser.h: (JSC::LiteralParser::LiteralParser): (JSC::LiteralParser::tryLiteralParse): (JSC::LiteralParser::Lexer::Lexer): 2011-06-17 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Moved some property access JIT code into property access JIT files https://bugs.webkit.org/show_bug.cgi?id=62906 * jit/JITOpcodes.cpp: * jit/JITOpcodes32_64.cpp: * jit/JITPropertyAccess.cpp: (JSC::JIT::emitSlow_op_put_by_val): (JSC::JIT::emit_op_get_scoped_var): (JSC::JIT::emit_op_put_scoped_var): (JSC::JIT::emit_op_get_global_var): (JSC::JIT::emit_op_put_global_var): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_get_scoped_var): (JSC::JIT::emit_op_put_scoped_var): (JSC::JIT::emit_op_get_global_var): (JSC::JIT::emit_op_put_global_var): 2011-06-17 Anders Carlsson <andersca@apple.com> Build fix. * JavaScriptCore.xcodeproj/project.pbxproj: 2011-06-17 Geoffrey Garen <ggaren@apple.com> Try to fix the Leopard build? * JavaScriptCore.xcodeproj/project.pbxproj: 2011-06-16 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Added some write barrier action, compiled out by default https://bugs.webkit.org/show_bug.cgi?id=62844 * JavaScriptCore.exp: Build! * JavaScriptCore.xcodeproj/project.pbxproj: Fixed an incremental build issue with Heap.cpp. * heap/Heap.cpp: (JSC::Heap::writeBarrierSlowCase): * heap/Heap.h: (JSC::Heap::writeBarrier): * heap/MarkedBlock.h: (JSC::MarkedBlock::isAtomAligned): (JSC::MarkedBlock::blockFor): (JSC::MarkedBlock::atomNumber): (JSC::MarkedBlock::ownerSetNumber): (JSC::MarkedBlock::addOldSpaceOwner): (JSC::MarkedBlock::OwnerSet::OwnerSet): (JSC::MarkedBlock::OwnerSet::add): (JSC::MarkedBlock::OwnerSet::clear): (JSC::MarkedBlock::OwnerSet::size): (JSC::MarkedBlock::OwnerSet::didOverflow): (JSC::MarkedBlock::OwnerSet::owners): Added a basic write barrier that tracks owners for regions within blocks. Currently unused. 2011-06-17 Raphael Kubo da Costa <kubo@profusion.mobi> Reviewed by Eric Seidel. [EFL] Add some OwnPtr specializations for EFL types. For now there are specializations for Ecore_Evas and Evas_Object. https://bugs.webkit.org/show_bug.cgi?id=62877 * wtf/CMakeListsEfl.txt: * wtf/OwnPtrCommon.h: * wtf/efl/OwnPtrEfl.cpp: Added. (WTF::deleteOwnedPtr): 2011-06-17 Joone Hur <joone.hur@collabora.co.uk> Reviewed by Martin Robinson. [GTK] Replace GdkRectangle by cairo_rectangle_int_t https://bugs.webkit.org/show_bug.cgi?id=60687 Replace GdkRectangle by cairo_rectangle_int_t. * wtf/gobject/GTypedefs.h: Replace GdkRectangle by cairo_rectangle_int_t. 2011-06-16 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=53014 ES5 strict mode keyword restrictions aren't implemented The following are future restricted words is strict mode code: implements, interface, let, package, private, protected, public, static, yield * parser/JSParser.h: - Add RESERVED_IF_STRICT token. * parser/Keywords.table: - Add new future restricted words. * parser/Lexer.cpp: (JSC::Lexer::parseIdentifier): - Check for RESERVED_IF_STRICT; in nonstrict code this is converted to IDENT. (JSC::Lexer::lex): - Pass strictMode flag to parseIdentifier. * parser/Lexer.h: - parseIdentifier needs a strictMode flag. * runtime/CommonIdentifiers.h: - Add identifiers for new reserved words. 2011-06-16 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=23611 Multiline Javascript comments cause incorrect parsing of following script. From the spec: "A MultiLineComment [is] simply discarded if it contains no line terminator, but if a MultiLineComment contains one or more line terminators, then it is replaced with a single line terminator, which becomes part of the stream of inputs for the syntactic grammar." This may result in behavioural changes, due to automatic semicolon insertion. * parser/Lexer.cpp: (JSC::Lexer::parseMultilineComment): - Set m_terminator is we see a line terminator in a multiline comment. 2011-06-16 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=62824 DFG JIT - add support for branch-fusion of compareEq, JSValue comparisons in SpeculativeJIT CompareEq of non-integer values is the most common cause of speculation failure. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleIntegerBranch): - Support Equals. (JSC::DFG::SpeculativeJIT::compilePeepHoleEq): - new! - peephole optimized Eq of JSValues. (JSC::DFG::SpeculativeJIT::compile): - Add peephole optimization for CompareEq. * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::detectPeepHoleBranch): - Add support for dead nodes between compare & branch. (JSC::DFG::SpeculativeJIT::isInteger): - Added to determine which form of peephole to do in CompareEq. 2011-06-16 Geoffrey Garen <ggaren@apple.com> Try to fix the Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Export another symbol. * bytecode/EvalCodeCache.h: * heap/HandleHeap.h: * heap/HeapRootVisitor.h: * heap/NewSpace.h: * runtime/ArgList.h: * runtime/ScopeChain.h: * runtime/SmallStrings.h: * runtime/Structure.h: Stop forward-declaring things that don't really exist anymore. 2011-06-16 Geoffrey Garen <ggaren@apple.com> Try to fix the Mac build: Removed and re-added SlotVisitor.h to the Xcode project while crossing my fingers and facing west. * JavaScriptCore.xcodeproj/project.pbxproj: 2011-06-16 Geoffrey Garen <ggaren@apple.com> Build fix: Removed an incorrect symbol on Windows. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-06-16 Geoffrey Garen <ggaren@apple.com> Build fix: Removed an accidental commit from the future. * CMakeLists.txt: 2011-06-16 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Introduced SlotVisitor into the project https://bugs.webkit.org/show_bug.cgi?id=62820 This resolves a class vs typedef forward declaration issue, and gives all exported symbols the correct names. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.xcodeproj/project.pbxproj: Build! * bytecode/EvalCodeCache.h: * heap/HandleHeap.h: * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::markRoots): * heap/Heap.h: * heap/HeapRootVisitor.h: Replaced MarkStack with SlotVisitor. Now no clients operate on a MarkStack. * heap/MarkStack.cpp: (JSC::SlotVisitor::visitChildren): (JSC::SlotVisitor::drain): * heap/SlotVisitor.h: Added. (JSC::SlotVisitor::SlotVisitor): Used 'protected' and a little cheesy inheritance to give SlotVisitor all the attributes of MarkStack without making this change giant. Over time, we will move more behavior into SlotVisitor and its subclasses. * heap/MarkStack.h: * heap/NewSpace.h: Replaced MarkStack with SlotVisitor. Now no clients operate on a MarkStack. * runtime/ArgList.h: * runtime/JSCell.h: * runtime/JSObject.h: * runtime/ScopeChain.h: * runtime/SmallStrings.h: * runtime/Structure.h: Replaced MarkStack with SlotVisitor. Now no clients operate on a MarkStack. 2011-06-15 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Reduce memory usage of resolve_global https://bugs.webkit.org/show_bug.cgi?id=62765 If we have a large number of resolve_globals in a single block start planting plain resolve instructions instead whenever we aren't in a loop. This allows us to reduce the code size for extremely large functions without losing the performance benefits of op_resolve_global. * bytecode/CodeBlock.h: (JSC::CodeBlock::globalResolveInfoCount): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::shouldAvoidResolveGlobal): (JSC::BytecodeGenerator::emitResolve): (JSC::BytecodeGenerator::emitResolveWithBase): * bytecompiler/BytecodeGenerator.h: 2011-06-16 Qi Zhang <qi.2.zhang@nokia.com> Reviewed by Laszlo Gombos. [Qt] Fix building with CONFIG(use_system_icu) https://bugs.webkit.org/show_bug.cgi?id=62744 Do not define WTF_USE_QT4_UNICODE if WTF_USE_ICU_UNICODE is set. * wtf/Platform.h: 2011-06-15 Darin Adler <darin@apple.com> Reviewed by Adam Barth. Remove obsolete LOOSE_OWN_PTR code https://bugs.webkit.org/show_bug.cgi?id=59909 The internal Apple dependency on this is gone now. * wtf/OwnArrayPtr.h: Removed constructor that takes a raw pointer, set function that takes a raw pointer. * wtf/OwnPtr.h: Removed constructor that takes a raw pointer, set functino that takes a raw pointer. * wtf/PassOwnArrayPtr.h: Made constructor that takes a nullptr and assignment operator that takes a nullptr unconditional. Made constructor that takes a raw pointer private and explicit, and removed assignment operator that takes a raw pointer. * wtf/PassOwnPtr.h: Made assignment operator that takes a nullptr unconditional. Made constructor that takes a raw pointer private and explicit, and removed assignment operator that takes a raw pointer. 2011-06-15 Sam Weinig <sam@webkit.org> Reviewed by Geoffrey Garen and Gavin Barraclough. Make access-nseive ~9x faster on the non-speculative path by adding special casing for doubles that can lossless-ly be converted to a uint32_t in getByVal and putByVal. This avoids calls to stringification and the hash lookup. Long term, we should try and get property of a getByVal and putByVal to be an integer immediate even in the non-speculative path. * dfg/DFGOperations.cpp: (JSC::DFG::putByVal): (JSC::DFG::operationPutByValInternal): 2011-06-15 Oliver Hunt <oliver@apple.com> Reviewed by Darin Adler. REGRESSION (r88719): 5by5.tv schedule is not visible https://bugs.webkit.org/show_bug.cgi?id=62720 Problem here is that the lexer wasn't considering '$' to be a valid character in an identifier. * parser/Lexer.h: (JSC::Lexer::lexExpectIdentifier): 2011-06-15 Oliver Hunt <oliver@apple.com> Reviewed by Sam Weinig. Reduce the size of global_resolve https://bugs.webkit.org/show_bug.cgi?id=62738 Reduce the code size of global_resolve in the JIT by replacing multiple pointer loads with a single pointer move + two offset loads. * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_resolve_global): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_resolve_global): 2011-06-14 Geoffrey Garen <ggaren@apple.com> Reviewed by Dan Bernstein. Fixed an inavlid ASSERT I found while investigating <rdar://problem/9580254> Crash in JSC::HandleHeap::finalizeWeakHandles + 92 https://bugs.webkit.org/show_bug.cgi?id=62699 No test since we don't know of a way to get WebCore to deallocate the next-to-finalize handle, which is also the last handle in the list, while finalizing the second-to-last handle in the list. * heap/HandleHeap.h: (JSC::HandleHeap::deallocate): Don't ASSERT that m_nextToFinalize has a non-0 next() after updating it, since it is valid to update m_nextToFinalize to point to the tail sentinel. Do ASSERT that m_nextToFinalize has a non-0 next() before updating it, since it is not valid to update m_nextToFinalize to point past the tail sentinel. Also, use m_nextToFinalize consistently for clarity. 2011-06-14 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=43841 SegmentedVector::operator== typo * wtf/SegmentedVector.h: (WTF::SegmentedVectorIterator::operator==): (WTF::SegmentedVectorIterator::operator!=): 2011-06-14 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Constant array literals result in unnecessarily large amounts of code https://bugs.webkit.org/show_bug.cgi?id=62658 Add a new version of op_new_array that simply copies values from a buffer we hang off of the CodeBlock, rather than generating code to place each entry into the registerfile, and then copying it from the registerfile into the array. This is a slight improvement on some sunspider tests, but no measurable overall change. That's okay though as our goal was to reduce code size without hurting performance. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): * bytecode/CodeBlock.h: (JSC::CodeBlock::addImmediateBuffer): (JSC::CodeBlock::immediateBuffer): * bytecode/Opcode.h: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::addImmediateBuffer): (JSC::BytecodeGenerator::emitNewArray): * bytecompiler/BytecodeGenerator.h: * bytecompiler/NodesCodegen.cpp: (JSC::ArrayNode::emitBytecode): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): * jit/JIT.h: * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_new_array): (JSC::JIT::emit_op_new_array_buffer): * jit/JITOpcodes32_64.cpp: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jit/JITStubs.h: 2011-06-14 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r88841. http://trac.webkit.org/changeset/88841 https://bugs.webkit.org/show_bug.cgi?id=62672 Caused many tests to crash (Requested by rniwa on #webkit). * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): * bytecode/CodeBlock.h: * bytecode/Opcode.h: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitNewArray): * bytecompiler/BytecodeGenerator.h: * bytecompiler/NodesCodegen.cpp: (JSC::ArrayNode::emitBytecode): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): * jit/JIT.h: * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_new_array): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_new_array): * jit/JITStubs.cpp: * jit/JITStubs.h: 2011-06-14 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Constant array literals result in unnecessarily large amounts of code https://bugs.webkit.org/show_bug.cgi?id=62658 Add a new version of op_new_array that simply copies values from a buffer we hang off of the CodeBlock, rather than generating code to place each entry into the registerfile, and then copying it from the registerfile into the array. This is a slight improvement on some sunspider tests, but no measurable overall change. That's okay though as our goal was to reduce code size without hurting performance. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): * bytecode/CodeBlock.h: (JSC::CodeBlock::addImmediateBuffer): (JSC::CodeBlock::immediateBuffer): * bytecode/Opcode.h: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::addImmediateBuffer): (JSC::BytecodeGenerator::emitNewArray): * bytecompiler/BytecodeGenerator.h: * bytecompiler/NodesCodegen.cpp: (JSC::ArrayNode::emitBytecode): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): * jit/JIT.h: * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_new_array): (JSC::JIT::emit_op_new_array_buffer): * jit/JITOpcodes32_64.cpp: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jit/JITStubs.h: 2011-06-14 Stephanie Lewis <slewis@apple.com> Rubber stamped by Oliver Hunt. <rdar://problem/9511169> Update order files. * JavaScriptCore.order: 2011-06-14 Sam Weinig <sam@webkit.org> Reviewed by Geoffrey Garen. Fix dumping of constants to have the correct constant number. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): 2011-06-14 Benjamin Poulain <benjamin@webkit.org> Reviewed by Eric Seidel. KeywordLookupGenerator's Trie does not work with Python 3 https://bugs.webkit.org/show_bug.cgi?id=62635 With Python 3, dict.items() return an iterator. Since the iterator protocol changed between Python 2 and 3, the easiest way to get the values is to have something that use the iterator implicitely, like a for() loop. * KeywordLookupGenerator.py: 2011-06-13 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Fix llocp and lvalp names in the lexer to something more meaningful https://bugs.webkit.org/show_bug.cgi?id=62605 A simple rename * parser/Lexer.cpp: (JSC::Lexer::parseIdentifier): (JSC::Lexer::parseString): (JSC::Lexer::lex): * parser/Lexer.h: (JSC::Lexer::lexExpectIdentifier): 2011-06-13 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Make it possible to inline the common case of identifier lexing https://bugs.webkit.org/show_bug.cgi?id=62600 Add a lexing function that expects to lex an "normal" alpha numeric identifier (that ignores keywords) so it's possible to inline the common parsing cases. This comes out as a reasonable parsing speed boost. * parser/JSParser.cpp: (JSC::JSParser::nextExpectIdentifier): (JSC::JSParser::parseProperty): (JSC::JSParser::parseMemberExpression): * parser/Lexer.cpp: * parser/Lexer.h: (JSC::Lexer::makeIdentifier): (JSC::Lexer::lexExpectIdentifier): 2011-06-13 Xan Lopez <xlopez@igalia.com> Reviewed by Martin Robinson. Distcheck fixes. * GNUmakefile.am: * GNUmakefile.list.am: 2011-06-13 Oliver Hunt <oliver@apple.com> Reviewed by Simon Fraser. Make it possible to inline Identifier::equal https://bugs.webkit.org/show_bug.cgi?id=62584 Move Identifier::equal to the Identifier header file. * runtime/Identifier.cpp: * runtime/Identifier.h: (JSC::Identifier::equal): 2011-06-13 Tony Chang <tony@chromium.org> Reviewed by Dimitri Glazkov. rename ENABLE_NEW_FLEXBOX to ENABLE_CSS3_FLEXBOX https://bugs.webkit.org/show_bug.cgi?id=62578 * Configurations/FeatureDefines.xcconfig: 2011-06-13 Tony Chang <tony@chromium.org> Reviewed by Adam Barth. rename ENABLE_FLEXBOX to ENABLE_NEW_FLEXBOX https://bugs.webkit.org/show_bug.cgi?id=62545 * Configurations/FeatureDefines.xcconfig: 2011-06-12 Patrick Gansterer <paroga@webkit.org> Unreviewed. Build fix for !ENABLE(JIT) after r88604. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): 2011-06-11 Gavin Barraclough <barraclough@apple.com> Reviewed by Darin Adler. https://bugs.webkit.org/show_bug.cgi?id=16777 Remove #define NaN per Darin's comments. * runtime/JSGlobalObjectFunctions.cpp: (JSC::parseIntOverflow): (JSC::parseInt): (JSC::jsStrDecimalLiteral): (JSC::jsToNumber): (JSC::parseFloat): * wtf/DateMath.cpp: (WTF::equivalentYearForDST): (WTF::parseES5DateFromNullTerminatedCharacters): (WTF::parseDateFromNullTerminatedCharacters): (WTF::timeClip): (JSC::parseDateFromNullTerminatedCharacters): 2011-06-11 Gavin Barraclough <barraclough@apple.com> Rubber stamped by Geoff Garen. https://bugs.webkit.org/show_bug.cgi?id=62503 Remove JIT_OPTIMIZE_* switches The alternative code paths are untested, and not well maintained. These were useful when there was more churn in the JIT, but now are a maintenance overhead. Time to move on, removing. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): * jit/JIT.cpp: (JSC::JIT::privateCompileSlowCases): (JSC::JIT::privateCompile): (JSC::JIT::linkConstruct): * jit/JIT.h: * jit/JITCall.cpp: * jit/JITCall32_64.cpp: * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::privateCompileCTINativeCall): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::privateCompileCTINativeCall): (JSC::JIT::softModulo): * jit/JITPropertyAccess.cpp: * jit/JITPropertyAccess32_64.cpp: * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Lookup.cpp: (JSC::setUpStaticFunctionSlot): * runtime/Lookup.h: * wtf/Platform.h: 2011-06-10 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=16777 Eliminate JSC::NaN and JSC::Inf There's no good reason for -K-J-S- JSC to have its own NAN and infinity constants. The ones in std::numeric_limits are perfectly good. Remove JSC::Inf, JSC::NaN, switch some cases of (isnan || isinf) to !isfinite. * API/JSCallbackObjectFunctions.h: (JSC::::toNumber): * API/JSValueRef.cpp: (JSValueMakeNumber): (JSValueToNumber): * JavaScriptCore.exp: * runtime/CachedTranscendentalFunction.h: (JSC::CachedTranscendentalFunction::initialize): * runtime/DateConstructor.cpp: (JSC::constructDate): * runtime/DateInstanceCache.h: (JSC::DateInstanceData::DateInstanceData): (JSC::DateInstanceCache::reset): * runtime/JSCell.cpp: * runtime/JSCell.h: (JSC::JSCell::JSValue::getPrimitiveNumber): (JSC::JSCell::JSValue::toNumber): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::resetDateCache): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncParseInt): (JSC::globalFuncIsFinite): * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::toNumber): * runtime/JSValue.cpp: * runtime/JSValue.h: * runtime/JSValueInlineMethods.h: (JSC::jsNaN): * runtime/MathObject.cpp: (JSC::mathProtoFuncMax): (JSC::mathProtoFuncMin): * runtime/NumberConstructor.cpp: (JSC::numberConstructorNegInfinity): (JSC::numberConstructorPosInfinity): * runtime/NumberPrototype.cpp: (JSC::numberProtoFuncToExponential): (JSC::numberProtoFuncToFixed): (JSC::numberProtoFuncToPrecision): (JSC::numberProtoFuncToString): * runtime/UString.cpp: * wtf/DecimalNumber.h: (WTF::DecimalNumber::DecimalNumber): * wtf/dtoa.cpp: (WTF::dtoa): 2011-06-10 Tony Chang <tony@chromium.org> Reviewed by Ojan Vafai. add a compile guard ENABLE(FLEXBOX) https://bugs.webkit.org/show_bug.cgi?id=62049 * Configurations/FeatureDefines.xcconfig: 2011-06-10 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=55347 "name" and "message" enumerable on *Error.prototype This arises from chapter 15 of the spec: "Every other property described in this clause has the attributes { [[Writable]]: true, [[Enumerable]]: false, [[Configurable]]: true } unless otherwise specified." Standardized properties are not enumerable. * runtime/ErrorInstance.cpp: (JSC::ErrorInstance::ErrorInstance): * runtime/NativeErrorPrototype.cpp: (JSC::NativeErrorPrototype::NativeErrorPrototype): 2011-06-09 Geoffrey Garen <ggaren@apple.com> Build fix: Corrected header spelling. * heap/OldSpace.h: 2011-06-09 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Added OldSpace to the project https://bugs.webkit.org/show_bug.cgi?id=62417 Currently unused. Added OldSpace, the ability to iterate NewSpace vs OldSpace, and a per-block flag for testing whether you're in NewSpace vs OldSpace. * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: Build! * heap/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): * heap/MarkedBlock.h: (JSC::MarkedBlock::inNewSpace): (JSC::MarkedBlock::setInNewSpace): Added inNewSpace flag, for use in write barrier. * heap/NewSpace.cpp: (JSC::NewSpace::addBlock): (JSC::NewSpace::removeBlock): * heap/NewSpace.h: (JSC::NewSpace::forEachBlock): Added forEachBlock, to use for NewSpace-specific operations. * heap/OldSpace.cpp: Added. (JSC::OldSpace::OldSpace): (JSC::OldSpace::addBlock): (JSC::OldSpace::removeBlock): * heap/OldSpace.h: Added. (JSC::OldSpace::forEachBlock): New class for holding promoted blocks. Not in use yet. 2011-06-09 Hyowon Kim <hw1008.kim@samsung.com> Reviewed by Antonio Gomes. [EFL] Make accelerated compositing build in Webkit-EFL https://bugs.webkit.org/show_bug.cgi?id=62361 Add PLATFORM(EFL) to enable ACCELERATED_COMPOSITING on EFL port. * wtf/Platform.h: 2011-06-09 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen. Bug 62405 - Fix integer overflow in Array.prototype.push Fix geoff's review comments re static_cast. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncPush): 2011-06-09 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Factored MarkedBlock set management into a helper class with a fast case Bloom filter https://bugs.webkit.org/show_bug.cgi?id=62413 SunSpider reports a small speedup. This is in preparation for having ConservativeSet operate on arbitrary sets of MarkedBlocks, and in preparation for conservative scanning becoming proportionally more important than other GC activities. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.xcodeproj/project.pbxproj: Build-o. * heap/ConservativeRoots.cpp: (JSC::ConservativeRoots::add): * heap/ConservativeRoots.h: (JSC::ConservativeRoots::ConservativeRoots): Operate on a MarkedBlockSet directly, instead of a Heap, so we can operate on subsets of the Heap instead. Use a TinyBloomFilter for single-cycle exclusion of most pointers. This is particularly important since we expect not to find our subject pointer in the MarkedBlock hash, and hash misses are more expensive than typical hash lookups because they have high collision rates. No need for single-pointer add() to be public anymore, since nobody uses it. * heap/Heap.cpp: (JSC::Heap::markRoots): * heap/Heap.h: (JSC::Heap::forEachCell): (JSC::Heap::forEachBlock): Use MarkedBlockSet since that's what ConservativeRoots relies on. Nixed contains(), since nobody uses it anymore. * heap/MarkedBlock.h: (WTF::MarkedBlockHash::hash): Added a faster hash taking advantage of the VM layout properties of MarkedBlocks. * heap/MarkedBlockSet.h: Added. (JSC::MarkedBlockSet::add): (JSC::MarkedBlockSet::remove): (JSC::MarkedBlockSet::recomputeFilter): (JSC::MarkedBlockSet::filter): (JSC::MarkedBlockSet::set): * heap/TinyBloomFilter.h: Added. (JSC::TinyBloomFilter::TinyBloomFilter): (JSC::TinyBloomFilter::add): (JSC::TinyBloomFilter::ruleOut): New helper class, used above. * interpreter/RegisterFile.cpp: (JSC::RegisterFile::gatherConservativeRoots): No need to specifically exclude values by tag -- the tiny bloom filter is already a register-register compare, so adding another "rule out" factor just slows things down. 2011-06-09 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. Bug 62405 - Fix integer overflow in Array.prototype.push There are three integer overflows here, leading to safe (not a security risk) but incorrect (non-spec-compliant) behaviour. Two overflows occur when calculating the new length after pushing (one in the fast version of push in JSArray, one in the generic version in ArrayPrototype). The other occurs calculating indices to write to when multiple items are pushed. These errors result in three test-262 failures. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncPush): * runtime/JSArray.cpp: (JSC::JSArray::put): (JSC::JSArray::push): 2011-06-09 Dan Bernstein <mitz@apple.com> Reviewed by Anders Carlsson. Add Vector::reverse() https://bugs.webkit.org/show_bug.cgi?id=62393 * wtf/Vector.h: (WTF::Vector::reverse): Added 2011-06-08 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Factored a bunch of Heap functionality into stand-alone functors https://bugs.webkit.org/show_bug.cgi?id=62337 This is in preparation for making these functors operate on arbitrary sets of MarkedBlocks. * JavaScriptCore.exp: This file is a small tragedy. * debugger/Debugger.cpp: (JSC::Debugger::recompileAllJSFunctions): Updated for type change and rename. * heap/HandleHeap.h: (JSC::HandleHeap::forEachStrongHandle): New function for iterating all strong handles, so we can play along in the functor game. * heap/Heap.cpp: (JSC::CountFunctor::CountFunctor::CountFunctor): (JSC::CountFunctor::CountFunctor::count): (JSC::CountFunctor::CountFunctor::returnValue): (JSC::CountFunctor::ClearMarks::operator()): (JSC::CountFunctor::ResetAllocator::operator()): (JSC::CountFunctor::Sweep::operator()): (JSC::CountFunctor::MarkCount::operator()): (JSC::CountFunctor::Size::operator()): (JSC::CountFunctor::Capacity::operator()): (JSC::CountFunctor::Count::operator()): (JSC::CountFunctor::CountIfGlobalObject::operator()): (JSC::CountFunctor::TakeIfEmpty::TakeIfEmpty): (JSC::CountFunctor::TakeIfEmpty::operator()): (JSC::CountFunctor::TakeIfEmpty::returnValue): (JSC::CountFunctor::RecordType::RecordType): (JSC::CountFunctor::RecordType::typeName): (JSC::CountFunctor::RecordType::operator()): (JSC::CountFunctor::RecordType::returnValue): These functors factor out behavior that used to be in the functions below. (JSC::Heap::clearMarks): (JSC::Heap::sweep): (JSC::Heap::objectCount): (JSC::Heap::size): (JSC::Heap::capacity): (JSC::Heap::protectedGlobalObjectCount): (JSC::Heap::protectedObjectCount): (JSC::Heap::protectedObjectTypeCounts): (JSC::Heap::objectTypeCounts): (JSC::Heap::resetAllocator): (JSC::Heap::freeBlocks): (JSC::Heap::shrink): Factored out behavior into the functors above. * heap/Heap.h: (JSC::Heap::forEachProtectedCell): (JSC::Heap::forEachCell): (JSC::Heap::forEachBlock): Added forEach* iteration templates. I chose functor-based templates instead of plain iterators because they're simpler to implement in this case and they require a lot less code at the call site. * heap/MarkedBlock.h: (JSC::MarkedBlock::VoidFunctor::returnValue): Default parent class for trivial functors. (JSC::MarkedBlock::forEachCell): Renamed forEach to forEachCell because we have a few different kind of "for each" now. * runtime/JSGlobalData.cpp: (WTF::Recompile::operator()): (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::recompileAllJSFunctions): Updated for type change and rename. * runtime/JSGlobalData.h: Removed globalObjectCount because it was unused. 2011-06-08 Mikołaj Małecki <m.malecki@samsung.com> Reviewed by Pavel Feldman. Web Inspector: Crash by buffer overrun crash when serializing inspector object tree. https://bugs.webkit.org/show_bug.cgi?id=52791 No new tests. The problem can be reproduced by trying to create InspectorValue from 1.0e-100 and call ->toJSONString() on this. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: export 2 functions DecimalNumber::bufferLengthForStringExponential and DecimalNumber::toStringExponential. 2011-06-08 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r88404. http://trac.webkit.org/changeset/88404 https://bugs.webkit.org/show_bug.cgi?id=62342 broke win and mac build (Requested by tony^work on #webkit). * JavaScriptCore.gyp/JavaScriptCore.gyp: 2011-06-08 Evan Martin <evan@chromium.org> Reviewed by Adam Barth. [chromium] use gyp 'settings' type for settings target https://bugs.webkit.org/show_bug.cgi?id=62323 The 'settings' gyp target type is for targets that exist solely for their settings (no build rules). The comment above this target says it's for this, but it incorrectly uses 'none'. * JavaScriptCore.gyp/JavaScriptCore.gyp: 2011-06-08 Sailesh Agrawal <sail@chromium.org> Reviewed by Mihai Parparita. Chromium Mac: Enable overlay scrollbars https://bugs.webkit.org/show_bug.cgi?id=59756 Enable WTF_USE_WK_SCROLLBAR_PAINTER for Chromium Mac. This allows us to use overlay scrollbars on future versions of Mac OS X. * wtf/Platform.h: 2011-06-08 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Add faster lookup cache for multi character identifiers https://bugs.webkit.org/show_bug.cgi?id=62327 Add a non-hash lookup for mutiple character identifiers. This saves us from adding repeated identifiers to the ParserArena's identifier list as people tend to not start all their variables and properties with the same character and happily identifier locality works in our favour. * parser/ParserArena.h: (JSC::IdentifierArena::isEmpty): (JSC::IdentifierArena::clear): (JSC::IdentifierArena::makeIdentifier): 2011-06-08 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Took some responsibilities away from NewSpace https://bugs.webkit.org/show_bug.cgi?id=62325 NewSpace is basically just an allocator now. Heap acts as a controller, responsible for managing the set of all MarkedBlocks. This is in preparation for moving parts of the controller logic into separate helper classes that can act on arbitrary sets of MarkedBlocks that may or may not be in NewSpace. * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::destroy): (JSC::Heap::allocate): (JSC::Heap::markRoots): (JSC::Heap::clearMarks): (JSC::Heap::sweep): (JSC::Heap::objectCount): (JSC::Heap::size): (JSC::Heap::capacity): (JSC::Heap::collect): (JSC::Heap::resetAllocator): (JSC::Heap::allocateBlock): (JSC::Heap::freeBlocks): (JSC::Heap::shrink): Moved the set of MarkedBlocks from NewSpace to Heap, along with all functions that operate on the set of MarkedBlocks. Also moved responsibility for deciding whether to allocate a new MarkedBlock, and for allocating it. * heap/Heap.h: (JSC::Heap::contains): (JSC::Heap::forEach): Ditto. * heap/NewSpace.cpp: (JSC::NewSpace::addBlock): (JSC::NewSpace::removeBlock): (JSC::NewSpace::resetAllocator): * heap/NewSpace.h: (JSC::NewSpace::waterMark): (JSC::NewSpace::allocate): Ditto. 2011-06-08 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Some more MarkedSpace => NewSpace renaming https://bugs.webkit.org/show_bug.cgi?id=62305 * JavaScriptCore.exp: * JavaScriptCore.order: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * heap/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::destroy): (JSC::Heap::reportExtraMemoryCostSlowCase): (JSC::Heap::allocate): (JSC::Heap::markRoots): (JSC::Heap::objectCount): (JSC::Heap::size): (JSC::Heap::capacity): (JSC::Heap::collect): (JSC::Heap::isValidAllocation): * heap/Heap.h: (JSC::Heap::markedSpace): (JSC::Heap::contains): (JSC::Heap::forEach): (JSC::Heap::allocate): * runtime/JSCell.h: 2011-06-08 Kevin Ollivier <kevino@theolliviers.com> Reviewed by Eric Seidel. Add export macros to profiler headers. https://bugs.webkit.org/show_bug.cgi?id=27551 * profiler/Profiler.h: 2011-06-08 Kevin Ollivier <kevino@theolliviers.com> Reviewed by Eric Seidel. Add export symbols to parser headers. https://bugs.webkit.org/show_bug.cgi?id=27551 * parser/SourceProviderCache.h: 2011-06-08 Kevin Ollivier <kevino@theolliviers.com> Reviewed by Eric Seidel. Add export symbols to interpreter headers. https://bugs.webkit.org/show_bug.cgi?id=27551 * interpreter/Interpreter.h: 2011-06-08 Kevin Ollivier <kevino@theolliviers.com> Reviewed by Eric Seidel. Add export symbols to debugger headers. https://bugs.webkit.org/show_bug.cgi?id=27551 * debugger/Debugger.h: * debugger/DebuggerCallFrame.h: 2011-06-08 Geoffrey Garen <ggaren@apple.com> Reviewed by Darin Adler. Moved MarkedSpace.* to NewSpace.* in preparation for more renaming https://bugs.webkit.org/show_bug.cgi?id=62268 * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * heap/Heap.h: * heap/MarkedBlock.h: * heap/MarkedSpace.cpp: Removed. * heap/MarkedSpace.h: Removed. * heap/NewSpace.cpp: Copied from Source/JavaScriptCore/heap/MarkedSpace.cpp. * heap/NewSpace.h: Copied from Source/JavaScriptCore/heap/MarkedSpace.h. 2011-06-08 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r88365. http://trac.webkit.org/changeset/88365 https://bugs.webkit.org/show_bug.cgi?id=62301 windows bots broken (Requested by loislo_ on #webkit). * JavaScriptCore.exp: 2011-06-08 Ryan Sleevi <rsleevi@chromium.org> Reviewed by Tony Chang. Suppress C++0x compat warnings when compiling Chromium port with GCC 4.6 Compiling Chromium port under GCC 4.6 produces warnings about nullptr https://bugs.webkit.org/show_bug.cgi?id=62242 * JavaScriptCore.gyp/JavaScriptCore.gyp: 2011-06-08 Ademar de Souza Reis Jr. <ademar.reis@openbossa.org> Reviewed by Andreas Kling. Webkit on SPARC Solaris has wrong endian https://bugs.webkit.org/show_bug.cgi?id=29407 Bug 57256 fixed one crash on misaligned reads on sparc/solaris, but there are more ocurrences of the same code pattern in webkit. This patch includes the check on these other parts of the code. This is a speculative fix, I don't have a sparc machine to test and don't know which kind of test would trigger a crash (but it's quite obvious that it's the same code duplicated in different files). * runtime/UString.h: (JSC::UStringHash::equal): * wtf/text/StringHash.h: (WTF::StringHash::equal): 2011-06-08 Yael Aharon <yael.aharon@nokia.com> Reviewed by Andreas Kling. [Qt] Build fix for building QtWebKit inside of Qt. https://bugs.webkit.org/show_bug.cgi?id=62280 Remove CONFIG=staticlib, because it causes the configure script to add -ljavascriptcore into QtWebKit.prl. No new tests, as this is just a build fix. * JavaScriptCore.pri: 2011-06-07 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Split 'reset' into 'collect' and 'resetAllocator' https://bugs.webkit.org/show_bug.cgi?id=62267 * heap/Heap.cpp: (JSC::Heap::allocate): (JSC::Heap::collectAllGarbage): (JSC::Heap::collect): * heap/Heap.h: * heap/MarkedBlock.h: (JSC::MarkedBlock::resetAllocator): * heap/MarkedSpace.cpp: (JSC::MarkedSpace::resetAllocator): * heap/MarkedSpace.h: (JSC::MarkedSpace::SizeClass::resetAllocator): 2011-06-07 Geoffrey Garen <ggaren@apple.com> Reviewed by Sam Weinig. Renamed some more marks to visits https://bugs.webkit.org/show_bug.cgi?id=62254 * heap/HandleHeap.cpp: (JSC::HandleHeap::visitStrongHandles): (JSC::HandleHeap::visitWeakHandles): * heap/HandleHeap.h: * heap/HandleStack.cpp: (JSC::HandleStack::visit): * heap/HandleStack.h: * heap/Heap.cpp: (JSC::Heap::markProtectedObjects): (JSC::Heap::markTempSortVectors): (JSC::Heap::markRoots): * heap/HeapRootVisitor.h: (JSC::HeapRootVisitor::visit): * runtime/ArgList.cpp: (JSC::MarkedArgumentBuffer::markLists): 2011-06-07 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig https://bugs.webkit.org/show_bug.cgi?id=55537 Functions claim to have 'callee' which they actually don't (and shouldn't) * JavaScriptCore.xcodeproj/project.pbxproj: * runtime/JSFunction.cpp: (JSC::JSFunction::getOwnPropertyNames): 2011-06-07 Juan C. Montemayor <jmont@apple.com> Reviewed by Darin Adler. Make JSStaticFunction and JSStaticValue less "const" https://bugs.webkit.org/show_bug.cgi?id=62222 * API/JSObjectRef.h: * API/tests/testapi.c: (checkConstnessInJSObjectNames): (main): * JavaScriptCore.xcodeproj/project.pbxproj: 2011-06-07 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=62240 DFG JIT - add support for for-loop array initialization. Support put by val beyond vector length. Add a operationPutByValBeyondArrayBounds operation, make PutValVal call this if the vector length check fails. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentSpillGPR): (JSC::DFG::JITCodeGenerator::silentFillGPR): (JSC::DFG::JITCodeGenerator::silentSpillAllRegisters): (JSC::DFG::JITCodeGenerator::isDoubleConstantWithInt32Value): (JSC::DFG::JITCodeGenerator::isJSConstantWithInt32Value): (JSC::DFG::JITCodeGenerator::isIntegerConstant): (JSC::DFG::JITCodeGenerator::valueOfIntegerConstant): * dfg/DFGOperations.cpp: (JSC::DFG::operationPutByValInternal): * dfg/DFGOperations.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: 2011-06-06 James Simonsen <simonjam@chromium.org> Reviewed by James Robinson. Add monotonicallyIncreasingTime() to get monotonically increasing time https://bugs.webkit.org/show_bug.cgi?id=37743 * wtf/CurrentTime.cpp: Add monotonicallyIncreasingTime() for mac and a fallback implementation that just wraps currentTime(). (WTF::monotonicallyIncreasingTime): * wtf/CurrentTime.h: Add monotonicallyIncreasingTime(). 2011-06-06 Alexandru Chiculita <achicu@adobe.com> Reviewed by Kent Tamura. Add ENABLE_CSS_EXCLUSIONS support for build-webkit script https://bugs.webkit.org/show_bug.cgi?id=61628 * Configurations/FeatureDefines.xcconfig: 2011-06-06 Mihnea Ovidenie <mihnea@adobe.com> Reviewed by Kent Tamura. Add ENABLE(CSS_REGIONS) guard for CSS Regions support https://bugs.webkit.org/show_bug.cgi?id=61631 * Configurations/FeatureDefines.xcconfig: 2011-06-06 Carlos Garcia Campos <cgarcia@igalia.com> Unreviewed. Fix the GTK+ build. * GNUmakefile.am: Add javascriptcore_cflags variable. 2011-06-04 Kevin Ollivier <kevino@theolliviers.com> [wx] Unreviewed build fix. Restore the PPC build and allow users to specify architectures to build on Mac. * wtf/Platform.h: 2011-06-04 Gustavo Noronha Silva <gns@gnome.org> Unreviewed, MIPS build fix. WebKitGTK+ tarball fails to build on MIPS. https://buildd.debian.org/status/fetch.php?pkg=webkitgtk%2B&arch=mips&ver=1.4.0-1&stamp=1304786691 * GNUmakefile.list.am: Add missing MIPS-related file to the list of files that are added to the tarball on make dist, and fix sorting. 2011-06-04 Sam Weinig <sam@webkit.org> Reviewed by Darin Adler. Fix formatting of the output generated by KeywordLookupGenerator.py https://bugs.webkit.org/show_bug.cgi?id=62083 - Uses correct year for copyright. - Puts ending brace on same line as "else if" - Puts starting brace of function on its own line. - Adds some tasteful whitespace. - Adds comments to make clear that scopes are ending - Make macros actually split on two lines. * KeywordLookupGenerator.py: 2011-06-04 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. KeywordLookupGenerator.py spams stdout in Chromium Linux build https://bugs.webkit.org/show_bug.cgi?id=62087 This action does not appear to be needed. * JavaScriptCore.gyp/JavaScriptCore.gyp: 2011-06-03 Oliver Hunt <oliver@apple.com> Reviewed by Maciej Stachowiak. Lexer needs to provide Identifier for reserved words https://bugs.webkit.org/show_bug.cgi?id=62086 Alas it is necessary to provide an Identifier reference for keywords so that we can do the right thing when they're used in object literals. We now keep Identifiers for all reserved words in the CommonIdentifiers structure so that we can access them without a hash lookup. * KeywordLookupGenerator.py: * parser/Lexer.cpp: (JSC::Lexer::parseIdentifier): * parser/Lexer.h: * runtime/CommonIdentifiers.cpp: (JSC::CommonIdentifiers::CommonIdentifiers): * runtime/CommonIdentifiers.h: 2011-06-03 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Add debug code to break on speculation failures. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGNode.h: 2011-06-03 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=62082 DFG JIT - bug passing arguments that need swap This is really just a typo. When setting up the arguments for a call out to a C operation, we'll fail to swap arguments where this is necessary. For example, in the case of 2 arg calls, where the first argument is in %rdx & the second is in %rsi we should swap (exec will be passed in %rdi), but we don't. This can also affect function calls passing three arguments. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::setupTwoStubArgs): - Call swap with the correct arguments. 2011-06-03 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Force inlining of some hot lexer functions https://bugs.webkit.org/show_bug.cgi?id=62079 Fix more GCC stupidity * parser/Lexer.h: (JSC::Lexer::isWhiteSpace): (JSC::Lexer::isLineTerminator): 2011-06-03 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. GCC not inlining some functions that it really should be https://bugs.webkit.org/show_bug.cgi?id=62075 Add ALWAYS_INLINE to a number of parsing and lexing functions that should always be inlined. This gets us ~1.4% on my ad hoc parser test. * KeywordLookupGenerator.py: * parser/JSParser.cpp: (JSC::JSParser::next): (JSC::JSParser::nextTokenIsColon): (JSC::JSParser::consume): (JSC::JSParser::match): (JSC::JSParser::tokenStart): (JSC::JSParser::tokenLine): (JSC::JSParser::tokenEnd): * parser/Lexer.cpp: (JSC::isIdentPart): 2011-06-03 Oliver Hunt <oliver@apple.com> Whoops, fix last minute bug. * parser/Lexer.cpp: (JSC::Lexer::parseIdentifier): 2011-06-03 Martin Robinson <mrobinson@igalia.com> Try to fix the GTK+ build. * GNUmakefile.am: Clean up some spaces that should be tabs. * GNUmakefile.list.am: Add KeywordLookup.h to the source list and clean up some spaces that should be tabs. 2011-06-03 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Improve keyword lookup https://bugs.webkit.org/show_bug.cgi?id=61913 Rather than doing multiple hash lookups as we currently do when trying to identify keywords we now use an automatically generated decision tree (essentially it's a hard coded patricia trie). We still use the regular lookup table for the last few characters of an input as this allows us to completely skip all bounds checks. * CMakeLists.txt: * DerivedSources.make: * DerivedSources.pro: * GNUmakefile.am: * JavaScriptCore.gyp/JavaScriptCore.gyp: * JavaScriptCore.xcodeproj/project.pbxproj: * KeywordLookupGenerator.py: Added. * make-generated-sources.sh: * parser/Lexer.cpp: (JSC::Lexer::internalShift): (JSC::Lexer::shift): (JSC::Lexer::parseIdentifier): * parser/Lexer.h: 2011-06-03 Siddharth Mathur <siddharth.mathur@nokia.com> Reviewed by Benjamin Poulain. [Qt] Build flag for experimental ICU library support https://bugs.webkit.org/show_bug.cgi?id=60786 Adds a build-time flag (CONFIG+=use_system_icu) that enables experimental ICU powered Unicode support. * JavaScriptCore.pri: Support for use_system_icu CONFIG flag. * wtf/unicode/qt4/UnicodeQt4.h: Guard an include file with USE(ICU_UNICODE). 2011-06-03 Alexis Menard <alexis.menard@openbossa.org> Reviewed by Benjamin Poulain. [Qt] Build fix of QtWebKit 2.2 when inside Qt tree with GCC 4.6. https://bugs.webkit.org/show_bug.cgi?id=61957 When building inside the Qt source tree, qmake always append the mkspecs defines after ours. We have to workaround and make sure that we append our flags after the qmake variable used inside Qt. This workaround was provided by our qmake folks. We need to append in both case because qmake behave differently when called with -spec or via SUBDIR+=. This patch unbreak r87950 on Mac for Qt port. * JavaScriptCore.pro: 2011-06-02 Jay Civelli <jcivelli@chromium.org> Reviewed by Adam Barth. Added a method to generate RFC 2822 compliant date strings. https://bugs.webkit.org/show_bug.cgi?id=7169 * wtf/DateMath.cpp: (WTF::twoDigitStringFromNumber): (WTF::makeRFC2822DateString): * wtf/DateMath.h: 2011-06-02 Alexis Menard <alexis.menard@openbossa.org> Reviewed by Andreas Kling. [Qt] Build fix of QtWebKit 2.2 when inside Qt tree with GCC 4.6. https://bugs.webkit.org/show_bug.cgi?id=61957 When building inside the Qt source tree, qmake always append the mkspecs defines after ours. We have to workaround and make sure that we append our flags after the qmake variable used inside Qt. This workaround was provided by our qmake folks. * JavaScriptCore.pro: 2011-06-01 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Add single character lookup cache to IdentifierArena https://bugs.webkit.org/show_bug.cgi?id=61879 Add a simple lookup cache for single ascii character identifiers. Produces around a 2% improvement in parse time for my adhoc parser test. * parser/ParserArena.h: (JSC::IdentifierArena::IdentifierArena): (JSC::IdentifierArena::clear): (JSC::IdentifierArena::makeIdentifier): 2011-05-31 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Freezing a function and its prototype causes browser to crash. https://bugs.webkit.org/show_bug.cgi?id=61758 Make JSObject::preventExtensions virtual so that we can override it and instantiate all lazy * JavaScriptCore.exp: * runtime/JSFunction.cpp: (JSC::createPrototypeProperty): (JSC::JSFunction::preventExtensions): (JSC::JSFunction::getOwnPropertySlot): * runtime/JSFunction.h: * runtime/JSObject.h: * runtime/JSObject.cpp: (JSC::JSObject::seal): (JSC::JSObject::seal): 2011-06-01 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r87788. http://trac.webkit.org/changeset/87788 https://bugs.webkit.org/show_bug.cgi?id=61856 breaks windows chromium canary (Requested by jknotten on #webkit). * wtf/DateMath.cpp: (WTF::timeClip): * wtf/DateMath.h: 2011-06-01 Jay Civelli <jcivelli@chromium.org> Reviewed by Adam Barth. Added a method to generate RFC 2822 compliant date strings. https://bugs.webkit.org/show_bug.cgi?id=7169 * wtf/DateMath.cpp: (WTF::twoDigitStringFromNumber): (WTF::makeRFC2822DateString): * wtf/DateMath.h: 2011-05-31 Yong Li <yoli@rim.com> Reviewed by Eric Seidel. https://bugs.webkit.org/show_bug.cgi?id=54807 We have been assuming plain bitfields (like "int a : 31") are always signed integers. However some compilers can treat them as unsigned. For example, RVCT 4.0 states plain bitfields (declared without either signed or unsigned qualifiers) are treats as unsigned. http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.dui0348c/Babjddhe.html Although we can use "--signed-bitfields" flag to make RVCT 4.0 behave as most other compilers, always using "signed"/"unsigned" qualifier to declare integral type bitfields is still a good rule we should have in order to make our code independent from compilers and compiler flags. No new test added because this change is not known to fix any issue. * bytecode/StructureStubInfo.h: 2011-05-30 Hojong Han <hojong.han@samsung.com> Reviewed by Geoffrey Garen. [JSC] malfunction during arithmetic condition check with negative number (-2147483648) https://bugs.webkit.org/show_bug.cgi?id=61416 * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::branch32): * tests/mozilla/ecma/Expressions/11.12-1.js: (getTestCases): 2011-05-29 Geoffrey Garen <ggaren@apple.com> Reviewed by Sam Weinig. Some heap refactoring https://bugs.webkit.org/show_bug.cgi?id=61704 SunSpider says no change. * JavaScriptCore.exp: Export! * heap/Heap.cpp: COLLECT_ON_EVERY_ALLOCATION can actually do so now. (JSC::Heap::Heap): Changed Heap sub-objects to point to the heap. (JSC::Heap::allocate): Changed inline allocation code to only select the size class, since this can be optimized out at compile time -- everything else is now inlined into this out-of-line function. No need to duplicate ASSERTs made in our caller. * heap/Heap.h: (JSC::Heap::heap): (JSC::Heap::isMarked): (JSC::Heap::testAndSetMarked): (JSC::Heap::testAndClearMarked): (JSC::Heap::setMarked): Call directly into MarkedBlock instead of adding a layer of indirection through MarkedSpace. (JSC::Heap::allocate): See above. * heap/MarkedBlock.cpp: (JSC::MarkedBlock::create): (JSC::MarkedBlock::MarkedBlock): * heap/MarkedBlock.h: Changed Heap sub-objects to point to the heap. * heap/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::allocateBlock): * heap/MarkedSpace.h: (JSC::MarkedSpace::allocate): Updated to match changes above. 2011-05-28 David Kilzer <ddkilzer@apple.com> BUILD FIX when building only the interpreter Fixes the following compiler warning: JavaScriptCore/runtime/JSGlobalData.cpp:462:6: error: no previous prototype for function 'releaseExecutableMemory' [-Werror,-Wmissing-prototypes,3] void releaseExecutableMemory(JSGlobalData& globalData) ^ * jit/ExecutableAllocator.h: Moved declaration of JSC::releaseExecutableMemory(). 2011-05-28 David Kilzer <ddkilzer@apple.com> BUILD FIX after r87527 with ENABLE(BRANCH_COMPACTION) * assembler/LinkBuffer.h: (JSC::LinkBuffer::linkCode): Added missing argument. 2011-05-27 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. JS API is too aggressive about throwing exceptions for NULL get or set operations https://bugs.webkit.org/show_bug.cgi?id=61678 * API/JSCallbackObject.h: Changed our staticValueGetter to a regular function that returns a JSValue, so it can fail and still forward to normal property lookup. * API/JSCallbackObjectFunctions.h: (JSC::::getOwnPropertySlot): Don't throw an exception when failing to access a static property -- just forward the access. This allows objects to observe get/set operations but still let the JS object manage lifetime. (JSC::::put): Ditto. (JSC::::getStaticValue): Same as JSCallbackObject.h. * API/tests/testapi.c: (MyObject_set_nullGetForwardSet): * API/tests/testapi.js: Updated tests to reflect slightly less strict behavior, which matches headerdoc claims. 2011-05-27 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Property caching is too aggressive for API objects https://bugs.webkit.org/show_bug.cgi?id=61677 * API/JSCallbackObject.h: Opt in to ProhibitsPropertyCaching, since our callback APIs allow the client to change its mind about our propertis at any time. * API/tests/testapi.c: (PropertyCatchalls_getProperty): (PropertyCatchalls_setProperty): (PropertyCatchalls_getPropertyNames): (PropertyCatchalls_class): (main): * API/tests/testapi.js: Some tests for dynamic API objects. * interpreter/Interpreter.cpp: (JSC::Interpreter::tryCachePutByID): (JSC::Interpreter::tryCacheGetByID): * jit/JITStubs.cpp: (JSC::JITThunks::tryCachePutByID): (JSC::JITThunks::tryCacheGetByID): (JSC::DEFINE_STUB_FUNCTION): Opt out of property caching if the client requires it. * runtime/JSTypeInfo.h: (JSC::TypeInfo::TypeInfo): (JSC::TypeInfo::isFinal): (JSC::TypeInfo::prohibitsPropertyCaching): (JSC::TypeInfo::flags): Added a flag to track opting out of property caching. Fixed an "&&" vs "&" typo that was previously harmless, but is now harmful since m_flags2 can have more than one bit set. 2011-05-27 Stephanie Lewis <slewis@apple.com> Unreviewed. Fix a typo in the order_file flag. * Configurations/Base.xcconfig: 2011-05-27 Patrick Gansterer <paroga@webkit.org> Unreviewed. Build fix for !ENABLE(ASSEMBLER) after r87527. * runtime/JSGlobalData.cpp: (JSGlobalData::JSGlobalData): 2011-05-27 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Add a few validity assertions to JSCallbackObject https://bugs.webkit.org/show_bug.cgi?id=61659 * API/JSCallbackObject.h: (JSC::JSCallbackObject::visitChildren): 2011-05-27 Oliver Hunt <oliver@apple.com> Build fix * runtime/RegExpCache.cpp: (JSC::RegExpCache::invalidateCode): 2011-05-27 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Try to release unused executable memory when the FixedVMPool allocator is under pressure https://bugs.webkit.org/show_bug.cgi?id=61651 Rather than crashing when full the FixedVMPool allocator now returns a null allocation. We replace the code that used to CRASH() on null allocations with logic that asks the provided globalData to release any executable memory that it can. Currently this just means throwing away all regexp code, but in future we'll try to be more aggressive. * assembler/ARMAssembler.cpp: (JSC::ARMAssembler::executableCopy): * assembler/ARMAssembler.h: * assembler/AssemblerBuffer.h: (JSC::AssemblerBuffer::executableCopy): * assembler/AssemblerBufferWithConstantPool.h: * assembler/LinkBuffer.h: (JSC::LinkBuffer::LinkBuffer): (JSC::LinkBuffer::linkCode): * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::executableCopy): * assembler/SH4Assembler.h: (JSC::SH4Assembler::executableCopy): * assembler/X86Assembler.h: (JSC::X86Assembler::executableCopy): (JSC::X86Assembler::X86InstructionFormatter::executableCopy): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * jit/ExecutableAllocator.h: (JSC::ExecutablePool::create): (JSC::ExecutablePool::alloc): (JSC::ExecutableAllocator::ExecutableAllocator): (JSC::ExecutableAllocator::poolForSize): (JSC::ExecutablePool::ExecutablePool): (JSC::ExecutablePool::poolAllocate): * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::FixedVMPoolAllocator::alloc): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::privateCompileCTINativeCall): * jit/JITPropertyAccess.cpp: (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::finalize): * jit/ThunkGenerators.cpp: (JSC::charCodeAtThunkGenerator): (JSC::charAtThunkGenerator): (JSC::fromCharCodeThunkGenerator): (JSC::sqrtThunkGenerator): (JSC::powThunkGenerator): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::releaseExecutableMemory): (JSC::releaseExecutableMemory): * runtime/JSGlobalData.h: * runtime/RegExpCache.cpp: (JSC::RegExpCache::invalidateCode): * runtime/RegExpCache.h: * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::compile): 2011-05-26 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Optimized ConservativeSet to avoid double-visiting objects https://bugs.webkit.org/show_bug.cgi?id=61592 SunSpider thinks this might be a 1% speedup * heap/ConservativeRoots.h: (JSC::ConservativeRoots::add): Use testAndClearMarked to avoid double-visiting an object. * heap/Heap.h: (JSC::Heap::isMarked): (JSC::Heap::testAndSetMarked): (JSC::Heap::testAndClearMarked): (JSC::Heap::setMarked): Added testAndClearMarked. Changed argument type to void*, since clients want to ask questions about arbitrary pointers into the heap, even when they aren't known to be JSCells. * heap/MarkedBlock.h: (JSC::MarkedBlock::testAndClearMarked): * heap/MarkedSpace.h: (JSC::MarkedSpace::isMarked): (JSC::MarkedSpace::testAndSetMarked): (JSC::MarkedSpace::testAndClearMarked): (JSC::MarkedSpace::setMarked): (JSC::MarkedSpace::contains): Ditto. * wtf/Bitmap.h: (WTF::::testAndClear): New function for ConservativeRoots's inverted marking pass. 2011-05-27 Stephanie Lewis <slewis@apple.com> Rubber Stamped by Adam Roben. Update Order Files. Use -order_file flag since it can order more of the binary. * Configurations/Base.xcconfig: * JavaScriptCore.order: 2011-05-26 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Renamed heapRootMarker to heapRootVisitor to match its class name https://bugs.webkit.org/show_bug.cgi?id=61584 * heap/Heap.cpp: (JSC::Heap::markProtectedObjects): (JSC::Heap::markTempSortVectors): (JSC::Heap::markRoots): 2011-05-26 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed some interdependency between Heap and SmallStrings by simplifying the SmallStrings lifetime model https://bugs.webkit.org/show_bug.cgi?id=61579 SunSpider reports no change. Using Weak<T> could accomplish this too, but we're not sure it will give us the performance we need. This is a first step, and it accomplishes most of the value of using Weak<T>. * heap/Heap.cpp: (JSC::Heap::destroy): (JSC::Heap::markRoots): (JSC::Heap::reset): Finalize small strings just like other weak handles. * runtime/SmallStrings.cpp: (JSC::finalize): (JSC::SmallStrings::finalizeSmallStrings): * runtime/SmallStrings.h: Make all small strings trivially weak, instead of having an "all for one, one for all" memory model. 2011-05-26 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make RegExpCache a weak map https://bugs.webkit.org/show_bug.cgi?id=61554 Switch to a weak map for the regexp cache, and hide that behaviour behind RegExp::create. When a RegExp is compiled it attempts to add itself to the "strong" cache. This cache is a simple round-robin buffer as was the old strong cache. Happily this can be smaller than the old strong cache as RegExps are only added when they're compiled so it is under less pressure to evict. * bytecompiler/NodesCodegen.cpp: (JSC::RegExpNode::emitBytecode): * runtime/RegExp.cpp: (JSC::RegExp::RegExp): (JSC::RegExp::create): (JSC::RegExp::match): * runtime/RegExp.h: (JSC::RegExp::gcShouldInvalidateCode): (JSC::RegExp::hasCode): (JSC::RegExp::key): * runtime/RegExpCache.cpp: (JSC::RegExpCache::lookupOrCreate): (JSC::RegExpCache::RegExpCache): (JSC::RegExpCache::isReachableFromOpaqueRoots): (JSC::RegExpCache::finalize): * runtime/RegExpCache.h: * runtime/RegExpConstructor.cpp: (JSC::constructRegExp): * runtime/RegExpPrototype.cpp: (JSC::regExpProtoFuncCompile): * runtime/StringPrototype.cpp: (JSC::stringProtoFuncMatch): (JSC::stringProtoFuncSearch): 2011-05-26 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Moved Heap-related functions out of JSCell.h and into respective header files https://bugs.webkit.org/show_bug.cgi?id=61567 * heap/Heap.h: (JSC::Heap::allocate): (JSC::Heap::heap): * heap/MarkedBlock.h: (JSC::MarkedBlock::allocate): * heap/MarkedSpace.h: (JSC::MarkedSpace::sizeClassFor): (JSC::MarkedSpace::allocate): * runtime/JSCell.h: (JSC::JSCell::destructor): 2011-05-26 Geoffrey Garen <ggaren@apple.com> Try to fix Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-05-26 Ryosuke Niwa <rniwa@webkit.org> Reviewed by Eric Seidel. [debug feature] WTFString should have show() method https://bugs.webkit.org/show_bug.cgi?id=61149 Added String::show and AtomicString::show in NDEBUG. * wtf/text/AtomicString.cpp: (WTF::AtomicString::show): * wtf/text/AtomicString.h: * wtf/text/WTFString.cpp: (String::show): * wtf/text/WTFString.h: 2011-05-26 Geoffrey Garen <ggaren@apple.com> Reviewed by Geoffrey Garen. Factored out some Heap ASSERTs https://bugs.webkit.org/show_bug.cgi?id=61565 * JavaScriptCore.exp: * heap/Heap.cpp: (JSC::isValidSharedInstanceThreadState): (JSC::isValidThreadState): (JSC::Heap::markRoots): (JSC::Heap::isValidAllocation): * heap/Heap.h: * runtime/JSCell.h: (JSC::JSCell::Heap::allocate): 2011-05-26 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen. https://bugs.webkit.org/show_bug.cgi?id=61508 DFG JIT - Add support for get by id self caching. Change the call out to be an unexpected call (using silent spill/fill functions), add a structure check & compact load to the JIT code, and add repatching mechanisms. Since DFGOperations may want to be be implemented in asm, make these symbols be extern "C". Add an asm wrapper to pass the return address to the optimizing get-by-id operation, so that it can look up its StructureStubInfo. * JavaScriptCore.xcodeproj/project.pbxproj: - Added new files. * bytecode/StructureStubInfo.h: - Added 'unset' entries to union. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::appendCallWithExceptionCheck): - Return the call, we need this to populate the StructureStubInfo. * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): - Populate the CodebBlock's StructureStubInfo Vector. * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::appendCallWithExceptionCheck): - Return the call, we need this to populate the StructureStubInfo. (JSC::DFG::JITCompiler::addPropertyAccess): (JSC::DFG::JITCompiler::PropertyAccessRecord::PropertyAccessRecord): - Add structures to record property access info during compilation. * dfg/DFGOperations.cpp: - Made all external methods extern "C". (JSC::DFG::operationPutByValInternal): - Moved outside of the extern "C" block. * dfg/DFGOperations.h: - Made all external methods extern "C". * dfg/DFGRepatch.cpp: Added. (JSC::DFG::dfgRepatchCall): - repatch a call to link to a new callee function. (JSC::DFG::dfgRepatchGetByIdSelf): - Modify the JIT code to optimize self accesses. (JSC::DFG::tryCacheGetByID): - Internal implementation of dfgRepatchGetByID (factor out failing cases). (JSC::DFG::dfgRepatchGetByID): - Used to optimize 'operationGetByIdOptimize' - repatches to 'operationGetById', and tries to optimize self accesses! * dfg/DFGRepatch.h: Added. - Expose dfgRepatchGetByID. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): - Changed implementation of GetById ops. 2011-05-26 Geoffrey Garen <ggaren@apple.com> Rolled back in http://trac.webkit.org/changeset/87408 with Windows build fixed. * heap/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): * heap/MarkedBlock.h: * wtf/DoublyLinkedList.h: (WTF::::DoublyLinkedListNode): (WTF::::setPrev): (WTF::::setNext): (WTF::::prev): (WTF::::next): (WTF::::DoublyLinkedList): (WTF::::isEmpty): (WTF::::size): (WTF::::clear): (WTF::::head): (WTF::::append): (WTF::::remove): (WTF::::removeHead): 2011-05-26 Geoffrey Garen <ggaren@apple.com> Rolled out http://trac.webkit.org/changeset/87408 because it broke the Windows build. * heap/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): * heap/MarkedBlock.h: (JSC::MarkedBlock::setPrev): (JSC::MarkedBlock::setNext): (JSC::MarkedBlock::prev): (JSC::MarkedBlock::next): * wtf/DoublyLinkedList.h: (WTF::::DoublyLinkedList): (WTF::::isEmpty): (WTF::::head): (WTF::::append): (WTF::::remove): 2011-05-26 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Provide a real owner when copying a property table, for the sake of write barriers. https://bugs.webkit.org/show_bug.cgi?id=61547 No test because we can't enable the writeBarrier() ASSERT just yet. * runtime/Structure.cpp: (JSC::Structure::addPropertyTransition): 2011-05-26 Adam Roben <aroben@apple.com> Windows build fix after r87346 * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Fixed up exports to match reality. 2011-05-26 Patrick Gansterer <paroga@webkit.org> Reviewed by Adam Barth. ASSERT(isMainThread()) when using single threaded jsc executable https://bugs.webkit.org/show_bug.cgi?id=60846 Remove the ASSERT since we do not have the concept of MainThread in JSC. * wtf/CryptographicallyRandomNumber.cpp: (WTF::ARC4Stream::ARC4RandomNumberGenerator::randomNumber): (WTF::ARC4Stream::ARC4RandomNumberGenerator::randomValues): 2011-05-25 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=61506 Move the silent spill/fill methods in the DFG JIT to the JITCodeGenerator so that they are available to the SpeculativeJIT. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::silentSpillGPR): (JSC::DFG::JITCodeGenerator::silentSpillFPR): (JSC::DFG::JITCodeGenerator::silentFillGPR): (JSC::DFG::JITCodeGenerator::silentFillFPR): (JSC::DFG::JITCodeGenerator::silentSpillAllRegisters): (JSC::DFG::JITCodeGenerator::silentFillAllRegisters): * dfg/DFGNonSpeculativeJIT.h: 2011-05-25 Ryosuke Niwa <rniwa@webkit.org> An attempt to revive Windows bots. * runtime/RegExp.cpp: * runtime/RegExp.h: 2011-05-25 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 61503 - Move population of CodeBlock::m_structureStubInfos into JIT This data structure, used at runtime by the JIT, is currently unnecessarily populated with default entries during byte compilation. Aside from meaning that there is JIT specific code in the bytecompiler, this also ties us to one entry per corresponding bytecode op, which may be undesirable. Instead, populate this array from the JIT. The type StructureStubInfo has two unused states, one for gets & one for puts. Unify these, so that the class can have a default constructor (and to simply switch statements in code walking over the table). This change has ramification for the DFG JIT, in that the DFG JIT used this datastructure to check for functions containing property access. Instead do so in the DFGByteCodeParser. * bytecode/CodeBlock.cpp: (JSC::printStructureStubInfo): * bytecode/CodeBlock.h: (JSC::CodeBlock::setNumberOfStructureStubInfos): (JSC::CodeBlock::numberOfStructureStubInfos): * bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::deref): (JSC::StructureStubInfo::visitAggregate): * bytecode/StructureStubInfo.h: (JSC::StructureStubInfo::StructureStubInfo): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitGetById): (JSC::BytecodeGenerator::emitPutById): (JSC::BytecodeGenerator::emitDirectPutById): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parseBlock): * jit/JIT.cpp: (JSC::JIT::JIT): (JSC::JIT::privateCompileMainPass): (JSC::JIT::privateCompileSlowCases): (JSC::JIT::privateCompile): * jit/JIT.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_get_by_id): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::emit_op_method_check): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::compileGetByIdSlowCase): (JSC::JIT::emitSlow_op_put_by_id): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_get_by_id): (JSC::JIT::emitSlow_op_get_by_id): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::emitSlow_op_put_by_id): (JSC::JIT::emit_op_method_check): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::compileGetByIdSlowCase): * runtime/Executable.cpp: (JSC::tryDFGCompile): 2011-05-25 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 61501 - Unify AbstractMacroAssembler::differenceBetween methods. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::Call::Call): (JSC::AbstractMacroAssembler::Call::fromTailJump): (JSC::AbstractMacroAssembler::Jump::Jump): (JSC::AbstractMacroAssembler::Jump::link): (JSC::AbstractMacroAssembler::Jump::linkTo): (JSC::AbstractMacroAssembler::Jump::isSet): (JSC::AbstractMacroAssembler::differenceBetween): (JSC::AbstractMacroAssembler::linkJump): (JSC::AbstractMacroAssembler::getLinkerCallReturnOffset): * assembler/LinkBuffer.h: (JSC::LinkBuffer::link): (JSC::LinkBuffer::locationOf): (JSC::LinkBuffer::locationOfNearCall): (JSC::LinkBuffer::returnAddressOffset): * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::linkCall): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::linkCall): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::linkCall): * assembler/MacroAssemblerSH4.cpp: (JSC::MacroAssemblerSH4::linkCall): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::linkCall): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::linkCall): 2011-05-25 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=61500 Add JSObject::offsetOfPropertyStorage * jit/JITPropertyAccess.cpp: (JSC::JIT::compileGetDirectOffset): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::compilePutDirectOffset): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::compilePutDirectOffset): (JSC::JIT::compileGetDirectOffset): * runtime/JSObject.h: (JSC::JSObject::offsetOfPropertyStorage): 2011-05-25 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make RegExp GC allocated https://bugs.webkit.org/show_bug.cgi?id=61490 Make RegExp GC allocated. Basically mechanical change to replace most use of [Pass]RefPtr<RegExp> with RegExp* or WriteBarrier<RegExp> where actual ownership happens. Made the RegExpCache use Strong<> references currently to avoid any changes in behaviour. * JavaScriptCore.exp: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): * bytecode/CodeBlock.h: (JSC::CodeBlock::addRegExp): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::addRegExp): (JSC::BytecodeGenerator::emitNewRegExp): * bytecompiler/BytecodeGenerator.h: * runtime/JSCell.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::clearBuiltinStructures): (JSC::JSGlobalData::addRegExpToTrace): * runtime/JSGlobalData.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): * runtime/RegExp.cpp: (JSC::RegExp::RegExp): (JSC::RegExp::create): (JSC::RegExp::invalidateCode): * runtime/RegExp.h: (JSC::RegExp::createStructure): * runtime/RegExpCache.cpp: (JSC::RegExpCache::lookupOrCreate): (JSC::RegExpCache::create): * runtime/RegExpCache.h: * runtime/RegExpConstructor.cpp: (JSC::constructRegExp): * runtime/RegExpObject.cpp: (JSC::RegExpObject::RegExpObject): (JSC::RegExpObject::visitChildren): * runtime/RegExpObject.h: (JSC::RegExpObject::setRegExp): (JSC::RegExpObject::RegExpObjectData::RegExpObjectData): * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::RegExpPrototype): (JSC::regExpProtoFuncCompile): * runtime/RegExpPrototype.h: * runtime/StringPrototype.cpp: (JSC::stringProtoFuncMatch): (JSC::stringProtoFuncSearch): 2011-05-25 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Generate regexp code lazily https://bugs.webkit.org/show_bug.cgi?id=61476 RegExp construction now simply validates the RegExp, it does not perform actual codegen. * runtime/RegExp.cpp: (JSC::RegExp::RegExp): (JSC::RegExp::recompile): (JSC::RegExp::compile): (JSC::RegExp::match): * runtime/RegExp.h: (JSC::RegExp::recompileIfNecessary): * runtime/RegExpConstructor.h: (JSC::RegExpConstructor::performMatch): * runtime/RegExpObject.cpp: (JSC::RegExpObject::match): * runtime/StringPrototype.cpp: (JSC::stringProtoFuncReplace): (JSC::stringProtoFuncMatch): (JSC::stringProtoFuncSearch): (JSC::stringProtoFuncSplit): 2011-05-24 Geoffrey Garen <ggaren@apple.com> Reviewed by Geoffrey Garen. Removed MarkSetProperties because it was unused https://bugs.webkit.org/show_bug.cgi?id=61418 * heap/MarkStack.h: (JSC::MarkSet::MarkSet): (JSC::MarkStack::append): * runtime/JSActivation.cpp: (JSC::JSActivation::visitChildren): * runtime/JSArray.h: (JSC::JSArray::visitChildrenDirect): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::visitChildren): * runtime/WriteBarrier.h: (JSC::MarkStack::appendValues): 2011-05-25 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make allocations with guard pages ensure that the allocation succeeded https://bugs.webkit.org/show_bug.cgi?id=61453 Add null checks, and make PageBlock's operator bool() use the realbase, rather than the start of usable memory. * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveAndCommit): * wtf/PageBlock.h: (WTF::PageBlock::operator bool): (WTF::PageBlock::PageBlock): 2011-04-10 Kevin Ollivier <kevino@theolliviers.com> Reviewed by Eric Seidel. Add JS_EXPORT_PRIVATE macro for exported methods in bytecompiler headers. https://bugs.webkit.org/show_bug.cgi?id=27551 * bytecompiler/BytecodeGenerator.h: 2011-05-24 Keishi Hattori <keishi@webkit.org> Reviewed by Kent Tamura. Disable textfield implementation of <input type=color>. Add INPUT_COLOR feature flag. Add input color sanitizer. https://bugs.webkit.org/show_bug.cgi?id=61273 * Configurations/FeatureDefines.xcconfig: Added COLOR_INPUT feature flag. 2011-05-24 Kevin Ollivier <kevino@theolliviers.com> Reviewed by Eric Seidel. Add export macros to WTFString.h. https://bugs.webkit.org/show_bug.cgi?id=27551 * wtf/text/WTFString.h: (WTF::String::String): (WTF::String::findIgnoringCase): (WTF::String::isHashTableDeletedValue): 2011-05-24 Geoffrey Garen <ggaren@apple.com> Maybe fix the Mac build now? * JavaScriptCore.xcodeproj/project.pbxproj: 2011-05-24 Geoffrey Garen <ggaren@apple.com> Maybe fix the Mac build? * JavaScriptCore.xcodeproj/project.pbxproj: 2011-05-24 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Split HeapRootVisitor into its own class https://bugs.webkit.org/show_bug.cgi?id=61399 * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * heap/HandleHeap.cpp: * heap/HandleStack.cpp: * heap/Heap.cpp: * heap/HeapRootVisitor.h: Copied from Source/JavaScriptCore/heap/MarkStack.h. * heap/MarkStack.h: * runtime/ArgList.cpp: * runtime/SmallStrings.cpp: 2011-05-24 Jay Civelli <jcivelli@chromium.org> Rubberstamped by David Kilzer. Updated some files that I forgot in my previous MHTML CL. * Configurations/FeatureDefines.xcconfig: 2011-05-24 Geoffrey Garen <ggaren@apple.com> Fix the Mac build: Yes, please do remove these files, svn. * JavaScriptCore.xcodeproj/project.pbxproj: 2011-05-24 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Let's just have one way to get the system page size, bokay? https://bugs.webkit.org/show_bug.cgi?id=61384 * CMakeListsEfl.txt: * CMakeListsWinCE.txt: * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: MarkStack[Platform].cpp is gone completely now, since it only existed to provide a duplicate way to access the system page size. * heap/MarkStack.cpp: (JSC::MarkStack::reset): * heap/MarkStack.h: (JSC::::MarkStackArray): (JSC::::shrinkAllocation): Use WTF::pageSize. * heap/MarkStackPosix.cpp: * heap/MarkStackSymbian.cpp: * heap/MarkStackWin.cpp: Removed now-empty files. * jit/ExecutableAllocator.cpp: (JSC::ExecutableAllocator::reprotectRegion): * jit/ExecutableAllocator.h: (JSC::ExecutableAllocator::ExecutableAllocator): (JSC::ExecutablePool::ExecutablePool): (JSC::ExecutablePool::poolAllocate): * jit/ExecutableAllocatorFixedVMPool.cpp: Use WTF::pageSize. * wscript: Removed now-empty files. * wtf/PageBlock.cpp: (WTF::systemPageSize): Integrated questionable Symbian page size rule from ExecutableAllocator, because that seems like what the original author should have done. 2011-05-24 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Interpreter crashes with gc validation enabled due to failure to mark initial cache structure https://bugs.webkit.org/show_bug.cgi?id=61385 The interpreter uses the structure slot of get_by_id and put_by_id to hold the initial structure it encountered so that it can identify whether a given access is stable. When marking though we only visit the slot when we've decided to cache, and so this value could die. This was "safe" as the value was only used for a pointer compare, but it was incorrect. We now just mark the slot like we should have been doing already. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitStructures): 2011-05-24 Adam Roben <aroben@apple.com> Windows build fix * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Removed now-inline functions. 2011-05-24 Geoffrey Garen <ggaren@apple.com> Windows build fix: update the #if OS(WINDOWS) section to match my last patch. * heap/MarkStack.h: (JSC::::shrinkAllocation): 2011-05-24 Geoffrey Garen <ggaren@apple.com> Rubber-stamped by Oliver Hunt. Split out function definitions and class definitions from class declarations in MarkStack.h, for readability. * heap/MarkStack.h: (JSC::MarkStack::MarkStack): (JSC::MarkStack::~MarkStack): (JSC::MarkStack::addOpaqueRoot): (JSC::MarkStack::containsOpaqueRoot): (JSC::MarkStack::opaqueRootCount): (JSC::MarkSet::MarkSet): (JSC::MarkStack::allocateStack): (JSC::MarkStack::releaseStack): (JSC::MarkStack::pageSize): (JSC::::MarkStackArray): (JSC::::~MarkStackArray): (JSC::::expand): (JSC::::append): (JSC::::removeLast): (JSC::::last): (JSC::::isEmpty): (JSC::::size): (JSC::::shrinkAllocation): 2011-05-24 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Avoid creating unnecessary identifiers and strings in the syntax checker https://bugs.webkit.org/show_bug.cgi?id=61378 Selectively tell the lexer that there are some places it does not need to do the real work of creating Identifiers for IDENT and STRING tokens. Make parseString and parseIdentifier templatized on whether they should do real work, or merely validate the tokens. SunSpider --parse-only reports ~5-8% win depending on hardware. * parser/ASTBuilder.h: (JSC::ASTBuilder::createDotAccess): * parser/JSParser.cpp: (JSC::JSParser::next): (JSC::JSParser::consume): (JSC::JSParser::parseVarDeclarationList): (JSC::JSParser::parseConstDeclarationList): (JSC::JSParser::parseExpression): (JSC::JSParser::parseAssignmentExpression): (JSC::JSParser::parseConditionalExpression): (JSC::JSParser::parseBinaryExpression): (JSC::JSParser::parseProperty): (JSC::JSParser::parseObjectLiteral): (JSC::JSParser::parseArrayLiteral): (JSC::JSParser::parseArguments): (JSC::JSParser::parseMemberExpression): * parser/Lexer.cpp: (JSC::Lexer::parseIdentifier): (JSC::Lexer::parseString): (JSC::Lexer::lex): * parser/Lexer.h: * parser/SyntaxChecker.h: (JSC::SyntaxChecker::createDotAccess): (JSC::SyntaxChecker::createProperty): 2011-05-23 Michael Saboff <msaboff@apple.com> Reviewed by Mark Rowe. Safari often freezes when clicking "Return free memory" in Caches dialog https://bugs.webkit.org/show_bug.cgi?id=61325 There are two fixes and improvement in instrumentation code used to find one of the problems. Changed ReleaseFreeList() to set the "decommitted" bit when releasing pages to the system and moving Spans from the normal list to the returned list. Added a "not making forward progress" check to TCMalloc_PageHeap::scavenge to eliminate an infinite loop if we can't meet the pagesToRelease target. Added a check for the decommitted bit being set properly in TCMalloc_PageHeap::CheckList. * wtf/FastMalloc.cpp: (WTF::TCMalloc_PageHeap::scavenge): (WTF::TCMalloc_PageHeap::Check): (WTF::TCMalloc_PageHeap::CheckList): (WTF::ReleaseFreeList): 2011-05-23 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen. https://bugs.webkit.org/show_bug.cgi?id=61306 The begin characters optimization currently has issues (#61129), and does not appear to still be a performance win. The prudent next step seems to be to disable while we ascertain whether this is still a useful performance optimization. * yarr/YarrInterpreter.cpp: (JSC::Yarr::Interpreter::matchDisjunction): (JSC::Yarr::Interpreter::interpret): * yarr/YarrInterpreter.h: (JSC::Yarr::BytecodePattern::BytecodePattern): * yarr/YarrPattern.cpp: (JSC::Yarr::YarrPatternConstructor::YarrPatternConstructor): (JSC::Yarr::YarrPattern::compile): (JSC::Yarr::YarrPattern::YarrPattern): * yarr/YarrPattern.h: (JSC::Yarr::YarrPattern::reset): 2011-05-23 Matthew Delaney <mdelaney@apple.com> Reviewed by Simon Fraser. Remove safeFloatToInt() in FloatRect.cpp and replace with working version of clampToInteger() https://bugs.webkit.org/show_bug.cgi?id=58216 * wtf/MathExtras.h: (clampToInteger): (clampToPositiveInteger): 2011-05-23 Ruben <chromium@hybridsource.org> Reviewed by Tony Chang. Chromium gyp patch to use new POSIX defines toolkit_uses_gtk and os_posix https://bugs.webkit.org/show_bug.cgi?id=61219 * JavaScriptCore.gyp/JavaScriptCore.gyp: 2011-05-23 Thouraya ANDOLSI <thouraya.andolsi@st.com> Reviewed by Gavin Barraclough. [SH4] AssemblerLabel does not name a type https://bugs.webkit.org/show_bug.cgi?id=59927 SH4Assembler.h file shoold be included before AbstractMacroAssembler.h. * assembler/MacroAssemblerSH4.h: 2011-05-23 Ryuan Choi <ryuan.choi@samsung.com> Rubber stamped by Eric Seidel. [CMAKE] Refactoring wtf related code. https://bugs.webkit.org/show_bug.cgi?id=60146 Move wtf-files to Source/JavaScriptCore/wtf/CMakeLists.txt. * CMakeLists.txt: * CMakeListsEfl.txt: * wtf/CMakeLists.txt: * wtf/CMakeListsEfl.txt: 2011-05-22 Adam Barth <abarth@webkit.org> Enable strict PassOwnPtr for everyone. I expect this patch will need some followups to make the GTK and EFL bots green again. * wtf/PassOwnPtr.h: 2011-05-20 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Reduce size of inline cache path of get_by_id on ARMv7 https://bugs.webkit.org/show_bug.cgi?id=61221 This reduces the code size of get_by_id by 20 bytes * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::ldrCompact): (JSC::ARMv7Assembler::repatchCompact): (JSC::ARMv7Assembler::setUInt7ForLoad): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::load32WithCompactAddressOffsetPatch): * jit/JIT.h: 2011-05-20 Zoltan Herczeg <zherczeg@inf.u-szeged.hu> Reviewed by Oliver Hunt. Zombies should "live" forever https://bugs.webkit.org/show_bug.cgi?id=61170 Reusing zombie cells could still hide garbage collected cell related bugs. * JavaScriptCore.pro: * heap/MarkedBlock.cpp: (JSC::MarkedBlock::clearMarks): * heap/MarkedBlock.h: * heap/MarkedSpace.cpp: (JSC::MarkedSpace::destroy): * runtime/JSCell.h: (JSC::JSCell::JSValue::isZombie): * runtime/JSZombie.h: (JSC::JSZombie::~JSZombie): * runtime/WriteBarrier.h: (JSC::WriteBarrierBase::setWithoutWriteBarrier): 2011-05-20 Brady Eidson <beidson@apple.com> Reviewed by Sam Weinig. <rdar://problem/9472883> and https://bugs.webkit.org/show_bug.cgi?id=61203 Horrendous bug in callOnMainThreadAndWait * wtf/MainThread.cpp: (WTF::dispatchFunctionsFromMainThread): Before signaling the background thread with the syncFlag condition, reacquire the mutex first. 2011-05-20 Oliver Hunt <oliver@apple.com> Reviewed by Sam Weinig. Remove unnecessary double->int conversion at the end of op_div https://bugs.webkit.org/show_bug.cgi?id=61198 We don't attempt this conversion on 64bit, removing it actually speeds up sunspider and v8 slightly, and it reduces code size. * jit/JITArithmetic32_64.cpp: (JSC::JIT::emit_op_div): 2011-05-19 Evan Martin <evan@chromium.org> Reviewed by Tony Chang. [chromium] remove <(library) variable https://bugs.webkit.org/show_bug.cgi?id=61158 This was for a build experiment; we can just use the correct value now. * JavaScriptCore.gyp/JavaScriptCore.gyp: 2011-05-20 Oliver Hunt <oliver@apple.com> Reviewed by Sam Weinig. Interpreter uses wrong bytecode offset for determining exception handler https://bugs.webkit.org/show_bug.cgi?id=61191 The bytecode offset given for the returnPC from the JIT is actually the offset for the start of the instruction triggering the call, whereas in the interpreter it is the actual return VPC. This means if the next instruction following a call was in an exception region we would incorrectly redirect to its handler. Long term we want to completely redo how exceptions are handled anyway so the simplest and lowest risk fix here is to simply subtract one from the return vPC so that we have an offset in the triggering instruction. It turns out this is caught by a couple of tests already. * interpreter/Interpreter.cpp: (JSC::Interpreter::unwindCallFrame): 2011-05-20 Xan Lopez <xlopez@igalia.com> Reviewed by Oliver Hunt. JIT requires VM overcommit (particularly on x86-64), Linux does not by default support this without swap? https://bugs.webkit.org/show_bug.cgi?id=42756 Use the MAP_NORESERVE flag for mmap on Linux to skip the kernel check of the available memory. This should give us an overcommit-like behavior in most systems, which is what we want. * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveAndCommit): pass MAP_NORSERVE to mmap. 2011-05-19 Gabor Loki <loki@webkit.org> Fix ARM build after r86919 * assembler/ARMAssembler.h: (JSC::ARMAssembler::nop): 2011-05-19 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Randomise code starting location a little https://bugs.webkit.org/show_bug.cgi?id=61161 Add a nop() function to the Assemblers so that we can randomise code offsets slightly at no real cost. * assembler/ARMAssembler.h: (JSC::ARMAssembler::nop): * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::nop): * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::nop): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::nop): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::nop): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::nop): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::nop): * assembler/X86Assembler.h: (JSC::X86Assembler::nop): * jit/JIT.cpp: (JSC::JIT::JIT): (JSC::JIT::privateCompile): * jit/JIT.h: * runtime/WeakRandom.h: (JSC::WeakRandom::getUint32): 2011-05-19 Oliver Hunt <oliver@apple.com> Fix windows build. * wtf/OSAllocatorWin.cpp: (WTF::OSAllocator::reserveUncommitted): (WTF::OSAllocator::reserveAndCommit): 2011-05-19 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Add guard pages to each end of the memory region used by the fixedvm allocator https://bugs.webkit.org/show_bug.cgi?id=61150 Add mechanism to notify the OSAllocator that pages at either end of an allocation should be considered guard pages. Update PageReservation, PageAllocation, etc to handle this. * JavaScriptCore.exp: * jit/ExecutableAllocatorFixedVMPool.cpp: (JSC::FixedVMPoolAllocator::FixedVMPoolAllocator): * wtf/OSAllocator.h: * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveUncommitted): (WTF::OSAllocator::reserveAndCommit): * wtf/PageAllocation.h: (WTF::PageAllocation::PageAllocation): * wtf/PageAllocationAligned.h: (WTF::PageAllocationAligned::PageAllocationAligned): * wtf/PageBlock.h: (WTF::PageBlock::PageBlock): * wtf/PageReservation.h: (WTF::PageReservation::reserve): (WTF::PageReservation::reserveWithGuardPages): Add a new function to make a reservation that will add guard pages to the ends of an allocation. (WTF::PageReservation::PageReservation): 2011-05-19 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make Executables release their JIT code as soon as they become dead https://bugs.webkit.org/show_bug.cgi?id=61134 Add an ability to clear an Executable's jit code without requiring it to be destroyed, and then call that from a finalizer. * heap/Weak.h: (JSC::Weak::Weak): (JSC::Weak::leak): * jit/JITCode.h: (JSC::JITCode::clear): * runtime/Executable.cpp: (JSC::ExecutableFinalizer::finalize): (JSC::ExecutableBase::executableFinalizer): * runtime/Executable.h: (JSC::ExecutableBase::ExecutableBase): (JSC::ExecutableBase::clearExecutableCode): 2011-05-19 Adam Roben <aroben@apple.com> Remove a redundant and broken data export Data can't be exported from JavaScriptCore.dll by listing it in the .def file. The JS_EXPORTDATA macro must be used instead. (In this case it was already being used, leading to a linker warning about multiple definitions.) * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Removed JSGlobalData::s_info. 2011-05-18 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Some tests crashing in JSC::MarkStack::validateValue beneath ScriptController::clearWindowShell on SnowLeopard Intel Release (WebKit2 Tests) https://bugs.webkit.org/show_bug.cgi?id=61064 Switch NonFinalObject to using WriteBarrier<> rather than WriteBarrierBase<> for its inline storage. This resolves the problem of GC occurring before a subclass has initialised its anonymous storage. * runtime/JSObject.h: 2011-05-18 Adam Barth <abarth@webkit.org> Reviewed by Sam Weinig. Delete WTFURL https://bugs.webkit.org/show_bug.cgi?id=61084 It's been a year and we've failed to complete this project. It's time to throw in the towel. * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/url: Removed. * wtf/url/api: Removed. * wtf/url/api/ParsedURL.cpp: Removed. * wtf/url/api/ParsedURL.h: Removed. * wtf/url/api/URLString.h: Removed. * wtf/url/src: Removed. * wtf/url/src/RawURLBuffer.h: Removed. * wtf/url/src/URLBuffer.h: Removed. * wtf/url/src/URLCharacterTypes.cpp: Removed. * wtf/url/src/URLCharacterTypes.h: Removed. * wtf/url/src/URLComponent.h: Removed. * wtf/url/src/URLEscape.cpp: Removed. * wtf/url/src/URLEscape.h: Removed. * wtf/url/src/URLParser.h: Removed. * wtf/url/src/URLQueryCanonicalizer.h: Removed. * wtf/url/src/URLSegments.cpp: Removed. * wtf/url/src/URLSegments.h: Removed. * wtf/url/wtfurl.gyp: Removed. 2011-05-18 Oliver Hunt <oliver@apple.com> Reviewed by Sam Weinig. JSGlobalObject and some others do GC allocation during initialization, which can cause heap corruption https://bugs.webkit.org/show_bug.cgi?id=61090 Remove the Structure-free JSGlobalObject constructor and instead always pass the structure into the JSGlobalObject constructor. Stop DebuggerActivation creating a new structure every time, and simply use a single shared structure held by the GlobalData. * API/JSContextRef.cpp: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::DebuggerActivation): * jsc.cpp: (GlobalObject::GlobalObject): (functionRun): (jscmain): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::clearBuiltinStructures): * runtime/JSGlobalData.h: * runtime/JSGlobalObject.h: 2011-05-18 Oliver Hunt <oliver@apple.com> Reviewed by Adam Roben. Disable gc validation in release builds https://bugs.webkit.org/show_bug.cgi?id=60680 Add back the NDEBUG check * wtf/Platform.h: 2011-05-17 Geoffrey Garen <ggaren@apple.com> Rolled out attempts to fix EFL build because they're not enough -- the build script needs to be fixed. * runtime/BooleanPrototype.cpp: * runtime/DateConstructor.cpp: * runtime/ErrorPrototype.cpp: 2011-05-17 Geoffrey Garen <ggaren@apple.com> More attempts to work around the EFL build system being borken. * runtime/DateConstructor.cpp: * runtime/ErrorPrototype.cpp: 2011-05-17 Geoffrey Garen <ggaren@apple.com> Try to fix the EFL build. * runtime/BooleanPrototype.cpp: 2011-05-16 Geoffrey Garen <ggaren@apple.com> Rolling back in r86653 with build fixed. Reviewed by Gavin Barraclough and Oliver Hunt. Global object initialization is expensive https://bugs.webkit.org/show_bug.cgi?id=60933 Changed a bunch of globals to allocate their properties lazily, and changed the global object to allocate a bunch of its globals lazily. This reduces the footprint of a global object from 287 objects with 58 functions for 24K to 173 objects with 20 functions for 15K. Large patch, but it's all mechanical. * DerivedSources.make: * JavaScriptCore.exp: Build! * create_hash_table: Added a special case for fromCharCode, since it uses a custom "thunk generator". * heap/Heap.cpp: (JSC::TypeCounter::operator()): Fixed a bug where the type counter would overcount objects that were owned through more than one mechanism because it was getting in the way of counting the results for this patch. * interpreter/CallFrame.h: (JSC::ExecState::arrayConstructorTable): (JSC::ExecState::arrayPrototypeTable): (JSC::ExecState::booleanPrototypeTable): (JSC::ExecState::dateConstructorTable): (JSC::ExecState::errorPrototypeTable): (JSC::ExecState::globalObjectTable): (JSC::ExecState::numberConstructorTable): (JSC::ExecState::numberPrototypeTable): (JSC::ExecState::objectPrototypeTable): (JSC::ExecState::regExpPrototypeTable): (JSC::ExecState::stringConstructorTable): Added new tables. * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::ArrayConstructor): (JSC::ArrayConstructor::getOwnPropertySlot): (JSC::ArrayConstructor::getOwnPropertyDescriptor): * runtime/ArrayConstructor.h: (JSC::ArrayConstructor::createStructure): * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::getOwnPropertySlot): (JSC::ArrayPrototype::getOwnPropertyDescriptor): * runtime/ArrayPrototype.h: * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::BooleanPrototype): (JSC::BooleanPrototype::getOwnPropertySlot): (JSC::BooleanPrototype::getOwnPropertyDescriptor): * runtime/BooleanPrototype.h: (JSC::BooleanPrototype::createStructure): * runtime/DateConstructor.cpp: (JSC::DateConstructor::DateConstructor): (JSC::DateConstructor::getOwnPropertySlot): (JSC::DateConstructor::getOwnPropertyDescriptor): * runtime/DateConstructor.h: (JSC::DateConstructor::createStructure): * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::ErrorPrototype): (JSC::ErrorPrototype::getOwnPropertySlot): (JSC::ErrorPrototype::getOwnPropertyDescriptor): * runtime/ErrorPrototype.h: (JSC::ErrorPrototype::createStructure): Standardized these objects to use static tables for function properties. * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::~JSGlobalData): * runtime/JSGlobalData.h: Added new tables. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): (JSC::JSGlobalObject::addStaticGlobals): (JSC::JSGlobalObject::getOwnPropertySlot): (JSC::JSGlobalObject::getOwnPropertyDescriptor): * runtime/JSGlobalObject.h: * runtime/JSGlobalObjectFunctions.cpp: * runtime/JSGlobalObjectFunctions.h: Changed JSGlobalObject to use a static table for its global functions. This required uninlining some things to avoid a circular header dependency. However, those things probably shouldn't have been inlined in the first place. Even more global object properties can be made lazy, but that requires more in-depth changes. * runtime/MathObject.cpp: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::getOwnPropertySlot): (JSC::NumberConstructor::getOwnPropertyDescriptor): * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::NumberPrototype): (JSC::NumberPrototype::getOwnPropertySlot): (JSC::NumberPrototype::getOwnPropertyDescriptor): * runtime/NumberPrototype.h: (JSC::NumberPrototype::createStructure): * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::ObjectPrototype): (JSC::ObjectPrototype::put): (JSC::ObjectPrototype::getOwnPropertySlot): (JSC::ObjectPrototype::getOwnPropertyDescriptor): * runtime/ObjectPrototype.h: (JSC::ObjectPrototype::createStructure): * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::RegExpPrototype): (JSC::RegExpPrototype::getOwnPropertySlot): (JSC::RegExpPrototype::getOwnPropertyDescriptor): * runtime/RegExpPrototype.h: (JSC::RegExpPrototype::createStructure): * runtime/StringConstructor.cpp: (JSC::StringConstructor::StringConstructor): (JSC::StringConstructor::getOwnPropertySlot): (JSC::StringConstructor::getOwnPropertyDescriptor): * runtime/StringConstructor.h: (JSC::StringConstructor::createStructure): Standardized these objects to use static tables for function properties. 2011-05-17 Sam Weinig <sam@webkit.org> Reviewed by Oliver Hunt. JSGlobalContextRelease should not trigger a synchronous garbage collection https://bugs.webkit.org/show_bug.cgi?id=60990 * API/JSContextRef.cpp: Change synchronous call to collectAllGarbage to a call to trigger the activityCallback. 2011-05-16 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Reduce code size for inline cache https://bugs.webkit.org/show_bug.cgi?id=60942 This patch introduces the concept of a "compact" address that allows individual architectures to control the maximum offset used for the inline path of get_by_id. This reduces the code size of get_by_id by 3 bytes on x86 and x86_64 and slightly improves performance on v8 tests. * assembler/ARMAssembler.h: (JSC::ARMAssembler::repatchCompact): * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::repatchCompact): * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::DataLabelCompact::DataLabelCompact): (JSC::AbstractMacroAssembler::differenceBetween): (JSC::AbstractMacroAssembler::repatchCompact): * assembler/CodeLocation.h: (JSC::CodeLocationDataLabelCompact::CodeLocationDataLabelCompact): (JSC::CodeLocationCommon::dataLabelCompactAtOffset): * assembler/LinkBuffer.h: (JSC::LinkBuffer::locationOf): * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::repatchCompact): * assembler/MacroAssembler.h: (JSC::MacroAssembler::loadPtrWithCompactAddressOffsetPatch): * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::load32WithCompactAddressOffsetPatch): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::load32WithCompactAddressOffsetPatch): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::load32WithCompactAddressOffsetPatch): * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::load32WithAddressOffsetPatch): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::repatchCompact): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::loadCompactWithAddressOffsetPatch): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::loadPtrWithCompactAddressOffsetPatch): * assembler/RepatchBuffer.h: (JSC::RepatchBuffer::repatch): * assembler/SH4Assembler.h: (JSC::SH4Assembler::repatchCompact): * assembler/X86Assembler.h: (JSC::X86Assembler::movl_mr_disp8): (JSC::X86Assembler::movq_mr_disp8): (JSC::X86Assembler::repatchCompact): (JSC::X86Assembler::setInt8): (JSC::X86Assembler::X86InstructionFormatter::oneByteOp_disp8): (JSC::X86Assembler::X86InstructionFormatter::oneByteOp64_disp8): (JSC::X86Assembler::X86InstructionFormatter::memoryModRM): * jit/JIT.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::patchGetByIdSelf): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::patchGetByIdSelf): * jit/JITStubs.cpp: (JSC::JITThunks::tryCacheGetByID): 2011-05-16 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r86653. http://trac.webkit.org/changeset/86653 https://bugs.webkit.org/show_bug.cgi?id=60944 "Caused regressions on Windows, OSX and EFL" (Requested by yutak on #webkit). * DerivedSources.make: * DerivedSources.pro: * GNUmakefile.am: * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * create_hash_table: * heap/Heap.cpp: (JSC::TypeCounter::operator()): * interpreter/CallFrame.h: (JSC::ExecState::arrayTable): (JSC::ExecState::numberTable): * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::ArrayConstructor): * runtime/ArrayConstructor.h: * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::getOwnPropertySlot): (JSC::ArrayPrototype::getOwnPropertyDescriptor): * runtime/ArrayPrototype.h: * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::BooleanPrototype): * runtime/BooleanPrototype.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::DateConstructor): * runtime/DateConstructor.h: * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::ErrorPrototype): * runtime/ErrorPrototype.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::~JSGlobalData): * runtime/JSGlobalData.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::addStaticGlobals): (JSC::JSGlobalObject::getOwnPropertySlot): (JSC::JSGlobalObject::getOwnPropertyDescriptor): * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncJSCPrint): * runtime/JSGlobalObjectFunctions.h: * runtime/MathObject.cpp: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::getOwnPropertySlot): (JSC::NumberConstructor::getOwnPropertyDescriptor): * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::NumberPrototype): * runtime/NumberPrototype.h: * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::ObjectPrototype): (JSC::ObjectPrototype::put): (JSC::ObjectPrototype::getOwnPropertySlot): * runtime/ObjectPrototype.h: * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::RegExpPrototype): * runtime/RegExpPrototype.h: * runtime/StringConstructor.cpp: (JSC::StringConstructor::StringConstructor): * runtime/StringConstructor.h: 2011-05-16 Geoffrey Garen <ggaren@apple.com> Reviewed by Geoffrey Garen. Global object initialization is expensive https://bugs.webkit.org/show_bug.cgi?id=60933 Changed a bunch of globals to allocate their properties lazily, and changed the global object to allocate a bunch of its globals lazily. This reduces the footprint of a global object from 287 objects with 58 functions for 24K to 173 objects with 20 functions for 15K. Large patch, but it's all mechanical. * DerivedSources.make: * JavaScriptCore.exp: Build! * create_hash_table: Added a special case for fromCharCode, since it uses a custom "thunk generator". * heap/Heap.cpp: (JSC::TypeCounter::operator()): Fixed a bug where the type counter would overcount objects that were owned through more than one mechanism because it was getting in the way of counting the results for this patch. * interpreter/CallFrame.h: (JSC::ExecState::arrayConstructorTable): (JSC::ExecState::arrayPrototypeTable): (JSC::ExecState::booleanPrototypeTable): (JSC::ExecState::dateConstructorTable): (JSC::ExecState::errorPrototypeTable): (JSC::ExecState::globalObjectTable): (JSC::ExecState::numberConstructorTable): (JSC::ExecState::numberPrototypeTable): (JSC::ExecState::objectPrototypeTable): (JSC::ExecState::regExpPrototypeTable): (JSC::ExecState::stringConstructorTable): Added new tables. * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::ArrayConstructor): (JSC::ArrayConstructor::getOwnPropertySlot): (JSC::ArrayConstructor::getOwnPropertyDescriptor): * runtime/ArrayConstructor.h: (JSC::ArrayConstructor::createStructure): * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::getOwnPropertySlot): (JSC::ArrayPrototype::getOwnPropertyDescriptor): * runtime/ArrayPrototype.h: * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::BooleanPrototype): (JSC::BooleanPrototype::getOwnPropertySlot): (JSC::BooleanPrototype::getOwnPropertyDescriptor): * runtime/BooleanPrototype.h: (JSC::BooleanPrototype::createStructure): * runtime/DateConstructor.cpp: (JSC::DateConstructor::DateConstructor): (JSC::DateConstructor::getOwnPropertySlot): (JSC::DateConstructor::getOwnPropertyDescriptor): * runtime/DateConstructor.h: (JSC::DateConstructor::createStructure): * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::ErrorPrototype): (JSC::ErrorPrototype::getOwnPropertySlot): (JSC::ErrorPrototype::getOwnPropertyDescriptor): * runtime/ErrorPrototype.h: (JSC::ErrorPrototype::createStructure): Standardized these objects to use static tables for function properties. * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::~JSGlobalData): * runtime/JSGlobalData.h: Added new tables. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): (JSC::JSGlobalObject::addStaticGlobals): (JSC::JSGlobalObject::getOwnPropertySlot): (JSC::JSGlobalObject::getOwnPropertyDescriptor): * runtime/JSGlobalObject.h: * runtime/JSGlobalObjectFunctions.cpp: * runtime/JSGlobalObjectFunctions.h: Changed JSGlobalObject to use a static table for its global functions. This required uninlining some things to avoid a circular header dependency. However, those things probably shouldn't have been inlined in the first place. Even more global object properties can be made lazy, but that requires more in-depth changes. * runtime/MathObject.cpp: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::getOwnPropertySlot): (JSC::NumberConstructor::getOwnPropertyDescriptor): * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::NumberPrototype): (JSC::NumberPrototype::getOwnPropertySlot): (JSC::NumberPrototype::getOwnPropertyDescriptor): * runtime/NumberPrototype.h: (JSC::NumberPrototype::createStructure): * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::ObjectPrototype): (JSC::ObjectPrototype::put): (JSC::ObjectPrototype::getOwnPropertySlot): (JSC::ObjectPrototype::getOwnPropertyDescriptor): * runtime/ObjectPrototype.h: (JSC::ObjectPrototype::createStructure): * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::RegExpPrototype): (JSC::RegExpPrototype::getOwnPropertySlot): (JSC::RegExpPrototype::getOwnPropertyDescriptor): * runtime/RegExpPrototype.h: (JSC::RegExpPrototype::createStructure): * runtime/StringConstructor.cpp: (JSC::StringConstructor::StringConstructor): (JSC::StringConstructor::getOwnPropertySlot): (JSC::StringConstructor::getOwnPropertyDescriptor): * runtime/StringConstructor.h: (JSC::StringConstructor::createStructure): Standardized these objects to use static tables for function properties. 2011-05-16 David Kilzer <ddkilzer@apple.com> <http://webkit.org/b/60913> C++ exceptions should not be enabled when building with llvm-gcc-4.2 <rdar://problem/9446430> Reviewed by Mark Rowe. * Configurations/Base.xcconfig: Fixed typo. 2011-05-16 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. JSWeakObjectMap finalisation may occur while gc is in inconsistent state https://bugs.webkit.org/show_bug.cgi?id=60908 <rdar://problem/9409491> We need to ensure that we have called all the weak map finalizers while the global object (and hence global context) is still in a consistent state. The best way to achieve this is to simply use a weak handle and finalizer on the global object. * JavaScriptCore.exp: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::WeakMapFinalizer::finalize): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::registerWeakMap): 2011-05-16 Siddharth Mathur <siddharth.mathur@nokia.com> Reviewed by Laszlo Gombos. [Qt][WK2][Symbian] Shared memory implementation for Symbian https://bugs.webkit.org/show_bug.cgi?id=55875 * wtf/Platform.h: Exclude Symbian OS from USE(UNIX_DOMAIN_SOCKETS) users 2011-05-16 Gavin Barraclough <barraclough@apple.com> Rubber stamped by Geoff Garen. https://bugs.webkit.org/show_bug.cgi?id=60866 Evaluation order broken for empty alternatives in subpatterns Reverting https://bugs.webkit.org/show_bug.cgi?id=51395 * yarr/YarrPattern.cpp: (JSC::Yarr::YarrPatternConstructor::atomParenthesesEnd): 2011-05-15 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen & Michael Saboff. https://bugs.webkit.org/show_bug.cgi?id=60860 Simplify backtracking in YARR JIT YARR JIT currently performs a single pass of code generation over the pattern, with special handling to allow the code generation for some backtracking code out of line. We can simplify things by moving to a common mechanism whereby all forwards matching code is generated in one pass, and all backtracking code is generated in another. Backtracking code can be generated in reverse order, to optimized the common fall-through case. To make it easier to walk over the pattern, we can first convert to a more byte-code like format before JIT generating. In time we should unify this with the YARR interpreter to more closely unify the two. * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::jumpIfNoAvailableInput): (JSC::Yarr::YarrGenerator::YarrOp::YarrOp): (JSC::Yarr::YarrGenerator::BacktrackingState::BacktrackingState): (JSC::Yarr::YarrGenerator::BacktrackingState::append): (JSC::Yarr::YarrGenerator::BacktrackingState::fallthrough): (JSC::Yarr::YarrGenerator::BacktrackingState::link): (JSC::Yarr::YarrGenerator::BacktrackingState::linkTo): (JSC::Yarr::YarrGenerator::BacktrackingState::takeBacktracksToJumpList): (JSC::Yarr::YarrGenerator::BacktrackingState::isEmpty): (JSC::Yarr::YarrGenerator::BacktrackingState::linkDataLabels): (JSC::Yarr::YarrGenerator::BacktrackingState::ReturnAddressRecord::ReturnAddressRecord): (JSC::Yarr::YarrGenerator::generateAssertionBOL): (JSC::Yarr::YarrGenerator::backtrackAssertionBOL): (JSC::Yarr::YarrGenerator::generateAssertionEOL): (JSC::Yarr::YarrGenerator::backtrackAssertionEOL): (JSC::Yarr::YarrGenerator::matchAssertionWordchar): (JSC::Yarr::YarrGenerator::generateAssertionWordBoundary): (JSC::Yarr::YarrGenerator::backtrackAssertionWordBoundary): (JSC::Yarr::YarrGenerator::generatePatternCharacterOnce): (JSC::Yarr::YarrGenerator::backtrackPatternCharacterOnce): (JSC::Yarr::YarrGenerator::generatePatternCharacterFixed): (JSC::Yarr::YarrGenerator::backtrackPatternCharacterFixed): (JSC::Yarr::YarrGenerator::generatePatternCharacterGreedy): (JSC::Yarr::YarrGenerator::backtrackPatternCharacterGreedy): (JSC::Yarr::YarrGenerator::generatePatternCharacterNonGreedy): (JSC::Yarr::YarrGenerator::backtrackPatternCharacterNonGreedy): (JSC::Yarr::YarrGenerator::generateCharacterClassOnce): (JSC::Yarr::YarrGenerator::backtrackCharacterClassOnce): (JSC::Yarr::YarrGenerator::generateCharacterClassFixed): (JSC::Yarr::YarrGenerator::backtrackCharacterClassFixed): (JSC::Yarr::YarrGenerator::generateCharacterClassGreedy): (JSC::Yarr::YarrGenerator::backtrackCharacterClassGreedy): (JSC::Yarr::YarrGenerator::generateCharacterClassNonGreedy): (JSC::Yarr::YarrGenerator::backtrackCharacterClassNonGreedy): (JSC::Yarr::YarrGenerator::generateTerm): (JSC::Yarr::YarrGenerator::backtrackTerm): (JSC::Yarr::YarrGenerator::generate): (JSC::Yarr::YarrGenerator::backtrack): (JSC::Yarr::YarrGenerator::opCompileParenthesesSubpattern): (JSC::Yarr::YarrGenerator::opCompileParentheticalAssertion): (JSC::Yarr::YarrGenerator::opCompileAlternative): (JSC::Yarr::YarrGenerator::opCompileBody): (JSC::Yarr::YarrGenerator::YarrGenerator): (JSC::Yarr::YarrGenerator::compile): 2011-05-15 Adam Barth <abarth@webkit.org> Enable strict PassOwnPtr on Qt. (Build fixes to follow.) * wtf/PassOwnPtr.h: 2011-05-15 Geoffrey Garen <ggaren@apple.com> Reviewed by Maciej Stachowiak. Partial fix for <rdar://problem/9417875> REGRESSION: SunSpider ~17% slower in browser than on command line This patch fixes a few issues in generated code that could unreasonably prolong object lifetimes. * heap/Heap.cpp: (JSC::Heap::collectAllGarbage): Throw away all function code before doing a major collection. We want to clear polymorphic caches, since they can keep alive large object graphs that have gone "stale". For the same reason, but to a lesser extent, we also want to clear linked functions and other one-off caches. This has the side-benefit of reducing memory footprint from run-once functions, and of allowing predictions and caches that have failed to re-specialize. Eventually, if compilation costs rise far enough, we may want a more limited strategy for de-specializing code without throwing it away completely, but this works for now, and it's the simplest solution. * jit/JITStubs.cpp: (JSC::JITThunks::hostFunctionStub): * jit/JITStubs.h: * runtime/JSFunction.cpp: Made the host function stub cache weak -- otherwise it's effectively a memory leak that can seriously fragment the GC and JIT heaps. (JSC::JSFunction::JSFunction): (JSC::JSFunction::visitChildren): Cleared up some comments that confused me when working with this code. 2011-05-13 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make GC validation more aggressive https://bugs.webkit.org/show_bug.cgi?id=60802 This patch makes the checks performed under GC_VALIDATION much more aggressive, and adds the checks to more places in order to allow us to catch GC bugs much closer to the point of failure. * JavaScriptCore.exp: * JavaScriptCore.xcodeproj/project.pbxproj: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::visitChildren): * heap/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): * heap/MarkedSpace.cpp: * runtime/Arguments.cpp: (JSC::Arguments::visitChildren): * runtime/Executable.cpp: (JSC::EvalExecutable::visitChildren): (JSC::ProgramExecutable::visitChildren): (JSC::FunctionExecutable::visitChildren): * runtime/Executable.h: * runtime/GetterSetter.cpp: (JSC::GetterSetter::visitChildren): * runtime/GetterSetter.h: * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): (JSC::JSAPIValueWrapper::JSAPIValueWrapper): * runtime/JSActivation.cpp: (JSC::JSActivation::visitChildren): * runtime/JSArray.cpp: (JSC::JSArray::visitChildren): * runtime/JSCell.cpp: (JSC::slowValidateCell): * runtime/JSCell.h: (JSC::JSCell::JSCell::unvalidatedStructure): (JSC::JSCell::JSCell::JSCell): * runtime/JSFunction.cpp: (JSC::JSFunction::visitChildren): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::visitChildren): (JSC::slowValidateCell): * runtime/JSONObject.h: * runtime/JSObject.cpp: (JSC::JSObject::visitChildren): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::visitChildren): * runtime/JSPropertyNameIterator.h: * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::visitChildren): * runtime/JSString.h: (JSC::RopeBuilder::JSString): * runtime/JSWrapperObject.cpp: (JSC::JSWrapperObject::visitChildren): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::visitChildren): * runtime/PropertyMapHashTable.h: (JSC::PropertyMapEntry::PropertyMapEntry): * runtime/RegExpObject.cpp: (JSC::RegExpObject::visitChildren): * runtime/ScopeChain.cpp: (JSC::ScopeChainNode::visitChildren): * runtime/ScopeChain.h: (JSC::ScopeChainNode::ScopeChainNode): * runtime/Structure.cpp: (JSC::Structure::Structure): (JSC::Structure::addPropertyTransition): (JSC::Structure::visitChildren): * runtime/Structure.h: (JSC::JSCell::classInfo): * runtime/StructureChain.cpp: (JSC::StructureChain::visitChildren): * runtime/StructureChain.h: * runtime/WriteBarrier.h: (JSC::validateCell): (JSC::JSCell): (JSC::JSGlobalObject): (JSC::WriteBarrierBase::set): (JSC::WriteBarrierBase::setMayBeNull): (JSC::WriteBarrierBase::setEarlyValue): (JSC::WriteBarrierBase::get): (JSC::WriteBarrierBase::operator*): (JSC::WriteBarrierBase::operator->): (JSC::WriteBarrierBase::unvalidatedGet): (JSC::WriteBarrier::WriteBarrier): * wtf/Assertions.h: 2011-05-13 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make GC validation more aggressive https://bugs.webkit.org/show_bug.cgi?id=60802 This patch makes the checks performed under GC_VALIDATION much more aggressive, and adds the checks to more places in order to allow us to catch GC bugs much closer to the point of failure. * JavaScriptCore.exp: * JavaScriptCore.xcodeproj/project.pbxproj: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::visitChildren): * heap/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): * heap/MarkedSpace.cpp: * runtime/Arguments.cpp: (JSC::Arguments::visitChildren): * runtime/Executable.cpp: (JSC::EvalExecutable::visitChildren): (JSC::ProgramExecutable::visitChildren): (JSC::FunctionExecutable::visitChildren): * runtime/Executable.h: * runtime/GetterSetter.cpp: (JSC::GetterSetter::visitChildren): * runtime/GetterSetter.h: * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): (JSC::JSAPIValueWrapper::JSAPIValueWrapper): * runtime/JSActivation.cpp: (JSC::JSActivation::visitChildren): * runtime/JSArray.cpp: (JSC::JSArray::visitChildren): * runtime/JSCell.cpp: (JSC::slowValidateCell): * runtime/JSCell.h: (JSC::JSCell::JSCell::unvalidatedStructure): (JSC::JSCell::JSCell::JSCell): * runtime/JSFunction.cpp: (JSC::JSFunction::visitChildren): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::visitChildren): (JSC::slowValidateCell): * runtime/JSONObject.h: * runtime/JSObject.cpp: (JSC::JSObject::visitChildren): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::visitChildren): * runtime/JSPropertyNameIterator.h: * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::visitChildren): * runtime/JSString.h: (JSC::RopeBuilder::JSString): * runtime/JSWrapperObject.cpp: (JSC::JSWrapperObject::visitChildren): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::visitChildren): * runtime/PropertyMapHashTable.h: (JSC::PropertyMapEntry::PropertyMapEntry): * runtime/RegExpObject.cpp: (JSC::RegExpObject::visitChildren): * runtime/ScopeChain.cpp: (JSC::ScopeChainNode::visitChildren): * runtime/ScopeChain.h: (JSC::ScopeChainNode::ScopeChainNode): * runtime/Structure.cpp: (JSC::Structure::Structure): (JSC::Structure::addPropertyTransition): (JSC::Structure::visitChildren): * runtime/Structure.h: (JSC::JSCell::classInfo): * runtime/StructureChain.cpp: (JSC::StructureChain::visitChildren): * runtime/StructureChain.h: * runtime/WriteBarrier.h: (JSC::validateCell): (JSC::JSCell): (JSC::JSGlobalObject): (JSC::WriteBarrierBase::set): (JSC::WriteBarrierBase::setMayBeNull): (JSC::WriteBarrierBase::setEarlyValue): (JSC::WriteBarrierBase::get): (JSC::WriteBarrierBase::operator*): (JSC::WriteBarrierBase::operator->): (JSC::WriteBarrierBase::unvalidatedGet): (JSC::WriteBarrier::WriteBarrier): * wtf/Assertions.h: 2011-05-14 Csaba Osztrogonác <ossy@webkit.org> Unreviewed, rolling out r86469 and r86471, because they made hundreds tests crash on Qt. Make GC validation more aggressive https://bugs.webkit.org/show_bug.cgi?id=60802 * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.xcodeproj/project.pbxproj: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::visitChildren): * heap/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): * heap/MarkedSpace.cpp: * runtime/Arguments.cpp: (JSC::Arguments::visitChildren): * runtime/Executable.cpp: (JSC::EvalExecutable::visitChildren): (JSC::ProgramExecutable::visitChildren): (JSC::FunctionExecutable::visitChildren): * runtime/Executable.h: (JSC::ProgramExecutable::createStructure): (JSC::FunctionExecutable::createStructure): * runtime/GetterSetter.cpp: (JSC::GetterSetter::visitChildren): * runtime/GetterSetter.h: (JSC::GetterSetter::createStructure): * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): * runtime/JSActivation.cpp: (JSC::JSActivation::visitChildren): * runtime/JSArray.cpp: (JSC::JSArray::visitChildren): * runtime/JSCell.cpp: * runtime/JSCell.h: (JSC::JSCell::JSCell::JSCell): * runtime/JSFunction.cpp: (JSC::JSFunction::visitChildren): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::visitChildren): * runtime/JSONObject.h: (JSC::JSONObject::createStructure): * runtime/JSObject.cpp: (JSC::JSObject::visitChildren): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::visitChildren): * runtime/JSPropertyNameIterator.h: * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::visitChildren): * runtime/JSString.h: (JSC::RopeBuilder::createStructure): * runtime/JSWrapperObject.cpp: (JSC::JSWrapperObject::visitChildren): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::visitChildren): * runtime/PropertyMapHashTable.h: (JSC::PropertyMapEntry::PropertyMapEntry): * runtime/RegExpObject.cpp: (JSC::RegExpObject::visitChildren): * runtime/ScopeChain.cpp: (JSC::ScopeChainNode::visitChildren): * runtime/ScopeChain.h: (JSC::ScopeChainNode::ScopeChainNode): * runtime/Structure.cpp: (JSC::Structure::Structure): (JSC::Structure::addPropertyTransition): (JSC::Structure::visitChildren): * runtime/Structure.h: (JSC::Structure::createStructure): (JSC::JSCell::classInfo): * runtime/StructureChain.cpp: (JSC::StructureChain::visitChildren): * runtime/StructureChain.h: * runtime/WriteBarrier.h: (JSC::WriteBarrierBase::set): (JSC::WriteBarrierBase::get): (JSC::WriteBarrierBase::operator*): (JSC::WriteBarrierBase::operator->): (JSC::WriteBarrier::WriteBarrier): * wtf/Assertions.h: 2011-05-13 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make GC validation more aggressive https://bugs.webkit.org/show_bug.cgi?id=60802 This patch makes the checks performed under GC_VALIDATION much more aggressive, and adds the checks to more places in order to allow us to catch GC bugs much closer to the point of failure. * JavaScriptCore.exp: * JavaScriptCore.xcodeproj/project.pbxproj: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::visitChildren): * heap/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): * heap/MarkedSpace.cpp: * runtime/Arguments.cpp: (JSC::Arguments::visitChildren): * runtime/Executable.cpp: (JSC::EvalExecutable::visitChildren): (JSC::ProgramExecutable::visitChildren): (JSC::FunctionExecutable::visitChildren): * runtime/Executable.h: * runtime/GetterSetter.cpp: (JSC::GetterSetter::visitChildren): * runtime/GetterSetter.h: * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): (JSC::JSAPIValueWrapper::JSAPIValueWrapper): * runtime/JSActivation.cpp: (JSC::JSActivation::visitChildren): * runtime/JSArray.cpp: (JSC::JSArray::visitChildren): * runtime/JSCell.cpp: (JSC::slowValidateCell): * runtime/JSCell.h: (JSC::JSCell::JSCell::unvalidatedStructure): (JSC::JSCell::JSCell::JSCell): * runtime/JSFunction.cpp: (JSC::JSFunction::visitChildren): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::visitChildren): (JSC::slowValidateCell): * runtime/JSONObject.h: * runtime/JSObject.cpp: (JSC::JSObject::visitChildren): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::visitChildren): * runtime/JSPropertyNameIterator.h: * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::visitChildren): * runtime/JSString.h: (JSC::RopeBuilder::JSString): * runtime/JSWrapperObject.cpp: (JSC::JSWrapperObject::visitChildren): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::visitChildren): * runtime/PropertyMapHashTable.h: (JSC::PropertyMapEntry::PropertyMapEntry): * runtime/RegExpObject.cpp: (JSC::RegExpObject::visitChildren): * runtime/ScopeChain.cpp: (JSC::ScopeChainNode::visitChildren): * runtime/ScopeChain.h: (JSC::ScopeChainNode::ScopeChainNode): * runtime/Structure.cpp: (JSC::Structure::Structure): (JSC::Structure::addPropertyTransition): (JSC::Structure::visitChildren): * runtime/Structure.h: (JSC::JSCell::classInfo): * runtime/StructureChain.cpp: (JSC::StructureChain::visitChildren): * runtime/StructureChain.h: * runtime/WriteBarrier.h: (JSC::validateCell): (JSC::JSCell): (JSC::JSGlobalObject): (JSC::WriteBarrierBase::set): (JSC::WriteBarrierBase::setMayBeNull): (JSC::WriteBarrierBase::setEarlyValue): (JSC::WriteBarrierBase::get): (JSC::WriteBarrierBase::operator*): (JSC::WriteBarrierBase::operator->): (JSC::WriteBarrierBase::unvalidatedGet): (JSC::WriteBarrier::WriteBarrier): * wtf/Assertions.h: 2011-05-01 Holger Hans Peter Freyther <holger@moiji-mobile.com> Reviewed by Steve Block. [android] OS(ANDROID) does not imply PLATFORM(ANDROID) https://bugs.webkit.org/show_bug.cgi?id=59888 It is possible to build QtWebKit and others for OS(ANDROID). Let the buildsystem decide which platform is to be build. * wtf/Platform.h: 2011-05-12 Maciej Stachowiak <mjs@apple.com> Reviewed by Darin Adler. XMLDocumentParserLibxml2 should play nice with strict OwnPtrs https://bugs.webkit.org/show_bug.cgi?id=59394 This portion of the change introduces a PassTraits template, which is used to enable takeFirst() to work for a Deque holding OwnPtrs, and optimize it for a Deque holding RefPtrs. In the future it can be deployed elsewhere to make our data structures work better with our smart pointers. * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/CMakeLists.txt: * wtf/Deque.h: (WTF::::takeFirst): * wtf/PassTraits.h: Added. (WTF::PassTraits::transfer): 2011-05-12 Nikolas Zimmermann <nzimmermann@rim.com> Not reviewed. Revert r86334, it broke the win build. WinCE build is fixed even without this patch. WinCairo remains broken atm, everything else works. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-05-12 Nikolas Zimmermann <nzimmermann@rim.com> Not reviewed. String operator+ reallocates unnecessarily when concatting > 2 strings https://bugs.webkit.org/show_bug.cgi?id=58420 Try to fix WinCE/WinCairo linking by exporting three symbols, not sure whether it's correct though. Win worked just fine before. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-05-12 Nikolas Zimmermann <nzimmermann@rim.com> Reviewed by Darin Adler. String operator+ reallocates unnecessarily when concatting > 2 strings https://bugs.webkit.org/show_bug.cgi?id=58420 Provide a faster String append operator. Up until now, "String operator+(const String& a, const String& b)" copied String a into a temporary object, and used a.append(b), which reallocates a new buffer of aLength+bLength. When concatting N strings using operator+, this leads to N-1 reallocations. Replace this with a flexible operator+ implementation, that avoids these reallocations. When concatting a 'String' with any string type (char*, UChar, Vector<char>, String, AtomicString, etc..) a StringAppend<String, T> object is created, which holds the intermediate string objects, and delays creation of the final string, until operator String() is invoked. template<typename T> StringAppend<String, T> operator+(const String& string1, T string2) { return StringAppend<String, T>(string1, string2); } template<typename U, typename V, typename W> StringAppend<U, StringAppend<V, W> > operator+(U string1, const StringAppend<V, W>& string2) { return StringAppend<U, StringAppend<V, W> >(string1, string2); } When concatting three strings - "String a, b, c; String result = a + b + c;" following happens: first a StringAppend<String, String> object is created by operator+(const String& string1, String string2). Then operator+(String string1, const StringAppend<String, String>& string2) is invoked, which returns a StringAppend<String, StringAppend<String, String> > object. Then operator String() is invoked, which allocates a StringImpl object, once, large enough to hold the final string - it uses tryMakeString provided by StringConcatenate.h under the hoods, which guards us against too big string allocations, etc. Note that the second template, defines a recursive way to concat an arbitary number of strings into a single String with just one allocation. * GNUmakefile.list.am: Add StringOperators.h to build. * JavaScriptCore.exp: Export WTF::emptyString(). Remove no longer needed symbols. * JavaScriptCore.gypi: Add StringOperators.h to build. * JavaScriptCore.vcproj/WTF/WTF.vcproj: Ditto. * JavaScriptCore.xcodeproj/project.pbxproj: Ditto. * wtf/text/AtomicString.h: Pull in StringConcatenate.h at the end of the file. * wtf/text/StringConcatenate.h: Conditionally include AtomicString.h to avoid a cyclic dependency. Pull in StringOperators.h at the end of the file. * wtf/text/StringOperators.h: Added. This is never meant to be included directly, including either WTFString.h or AtomicString.h automatically pulls in this file. (WTF::StringAppend::StringAppend): (WTF::StringAppend::operator String): (WTF::StringAppend::operator AtomicString): (WTF::StringAppend::writeTo): (WTF::StringAppend::length): (WTF::operator+): * wtf/text/WTFString.cpp: Remove operator+ implementations that use String::append(). (WTF::emptyString): Add new shared empty string free function. * wtf/text/WTFString.h: Replace operator+ implementations by StringAppend template solution. Pull in AtomicString.h at the end of the file. 2011-05-12 Philippe Normand <pnormand@igalia.com> Unreviewed, GTK build fix. * wtf/Platform.h: 2011-05-12 Keith Kyzivat <keith.kyzivat@nokia.com> Reviewed by Csaba Osztrogonác. [Qt] Arm debug build failing on ARMAssembler::debugOffset() https://bugs.webkit.org/show_bug.cgi?id=60688 Related to svn rev 85523 * assembler/ARMAssembler.h: (JSC::ARMAssembler::debugOffset): 2011-05-11 Igor Oliveira <igor.oliveira@openbossa.org> Reviewed by Eric Seidel. WebKit does not build with GCCE https://bugs.webkit.org/show_bug.cgi?id=60667 Allow compile WebKit with GCCE * wtf/Alignment.h: * wtf/Platform.h: 2011-05-11 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. Enable strict PassOwnPtr on Mac https://bugs.webkit.org/show_bug.cgi?id=60684 This should build cleanly now. * wtf/PassOwnPtr.h: 2011-05-11 Oliver Hunt <oliver@apple.com> Reviewed by Darin Adler. Protect JSC from WebCore executing JS during JS wrapper finalization https://bugs.webkit.org/show_bug.cgi?id=60672 <rdar://problem/9350997> Detect when we're trying to execute JS during GC and prevent the execution from happening. We also assert that this isn't happening as it implies incorrect behaviour of an object's destructor. * JavaScriptCore.exp: * heap/Heap.cpp: * heap/Heap.h: (JSC::Heap::isBusy): * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): (JSC::Interpreter::executeCall): (JSC::Interpreter::executeConstruct): * runtime/JSGlobalData.h: (JSC::JSGlobalData::isCollectorBusy): 2011-05-11 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Enable gc mark validation in temporarily in release builds https://bugs.webkit.org/show_bug.cgi?id=60678 Make it easier to turn the gc mark validation on and off, and temporarily turn it on for all builds. * heap/MarkStack.cpp: * heap/MarkStack.h: (JSC::MarkStack::append): (JSC::MarkStack::internalAppend): * runtime/WriteBarrier.h: (JSC::MarkStack::appendValues): * wtf/Platform.h: 2011-05-11 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. <rdar://problem/9331651> REGRESSION: RPRVT grows by 1MB / sec @ dvd2blu.com SunSpider reports no change. This bug was caused by changing Structure and Executable to being GC objects, and by a long-standing bug that would thrash the global object between dictionary and non-dictionary states. * runtime/BatchedTransitionOptimizer.h: (JSC::BatchedTransitionOptimizer::BatchedTransitionOptimizer): Don't eagerly transition to dictionary -- this can cause pathological dictionary churn, and it's not necessary, since objects know how to automatically transition to dictionary when necessary. * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): Be sure to report extra cost from compilation, because it can be quite high. This is especially important for program code, since DOM timers can repeatedly allocate program code without allocating any other objects. * runtime/JSObject.cpp: (JSC::JSObject::removeDirect): Don't transition to the uncacheable state if the thing we're trying to remove doesn't exist. This can happen during compilation, since the compiler needs to ensure that no pre-existing conflicting definitions exist for certain declarations. 2011-05-11 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Make mark stack validation functions do something useful in a release build https://bugs.webkit.org/show_bug.cgi?id=60645 Turn ASSERTs into actual if(...) CRASH(); statements. * heap/MarkStack.cpp: (JSC::MarkStack::validateValue): 2011-05-11 Xan Lopez <xlopez@igalia.com> Reviewed by Martin Robinson. Fix copy&paste error in comment. * jit/JITPropertyAccess.cpp: (JSC::JIT::stringGetByValStubGenerator): the value is stored in regT2, not regT1. 2011-05-11 Adam Roben <aroben@apple.com> WinCE build fixes for strict PassOwnPtr * wtf/unicode/CollatorDefault.cpp: (WTF::Collator::userDefault): Use adoptPtr. 2011-05-11 Holger Hans Peter Freyther <holger@moiji-mobile.com> Unreviewed build fix. [MIPS] Fix compilation of the MIPS JIT Include the MIPSAssembler.h first to indirectly include AssemblerBuffer.h before the AbstractMacroAssembler.h. This order is used for the ARM and X86 MacroAssembler*.h * assembler/MacroAssemblerMIPS.h: 2011-05-11 Adam Roben <aroben@apple.com> Turn on strict PassOwnPtr on Windows Fixes <http://webkit.org/b/60632> Windows should build with strict PassOwnPtr enabled Reviewed by Adam Barth. * wtf/PassOwnPtr.h: 2011-05-10 Stephanie Lewis <slewis@apple.com> Unreviewed. Revert accidental JavaScriptCore change in http://trac.webkit.org/changeset/86130 * Configurations/JavaScriptCore.xcconfig: 2011-05-10 Adam Barth <abarth@webkit.org> Reviewed by David Levin. Enable strict PassOwnPtr on Chromium https://bugs.webkit.org/show_bug.cgi?id=60502 Other platforms to follow. * wtf/PassOwnPtr.h: 2011-05-10 Geoffrey Garen <ggaren@apple.com> Reviewed by Darin Adler. Fixed up some #include dependencies so the WriteBarrier class can actually call Heap::writeBarrier https://bugs.webkit.org/show_bug.cgi?id=60532 * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.xcodeproj/project.pbxproj: Build! * heap/Handle.h: Moved HandleTypes to its own header because that's the WebKit style, and it was necessary to resolve a circular dependency between Handle.h and WriteBarrier.h. * heap/Heap.h: (JSC::Heap::writeBarrier): Added an inline no-op writeBarrier(), to verify that all the code is in the right place. * heap/MarkStack.h: Moved WriteBarrier operations to WriteBarrier.h to resolve a circular dependency. * runtime/ArgList.h: * runtime/JSCell.h: #include WriteBarrier.h since we don't get it for free anymore. * runtime/PropertyMapHashTable.h: (JSC::PropertyTable::PropertyTable): Call the real writeBarrier() function, now that it exists. * runtime/SmallStrings.h: Removed a stray #include to resolve a circular dependency. * runtime/WriteBarrier.h: (JSC::WriteBarrierBase::set): (JSC::MarkStack::append): (JSC::MarkStack::appendValues): Updated to match the changes above. 2011-05-10 Oliver Hunt <oliver@apple.com> Build fix. * heap/MarkStack.cpp: (JSC::MarkStack::validateValue): 2011-05-10 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Add some aggressive GC validation to debug builds. https://bugs.webkit.org/show_bug.cgi?id=60601 When assertions are enabled we now do some validity checking of objects being added to the mark stack. * bytecode/Instruction.h: (JSC::PolymorphicAccessStructureList::PolymorphicStubInfo::PolymorphicStubInfo): (JSC::PolymorphicAccessStructureList::visitAggregate): * heap/MarkStack.cpp: (JSC::MarkStack::validateSet): (JSC::MarkStack::validateValue): * heap/MarkStack.h: (JSC::MarkStack::appendValues): (JSC::MarkStack::append): (JSC::MarkStack::internalAppend): 2011-05-09 Darin Adler <darin@apple.com> Reviewed by Oliver Hunt. http://bugs.webkit.org/show_bug.cgi?id=60509 Wrong type used for return value from strlen * wtf/FastMalloc.cpp: (WTF::fastStrDup): Use size_t. Also don't bother checking for failure since fastMalloc won't return if it fails. 2011-05-09 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. CSP should block Function constructor https://bugs.webkit.org/show_bug.cgi?id=60240 When eval is disabled, we need to block the use of the function constructor. However, the WebCore JSC bindings call the function constructor directly to create inline event listeners. To support that use, this patch adds an entrypoint that bypasses the check for whether eval is enabled. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/FunctionConstructor.cpp: (JSC::constructFunction): (JSC::constructFunctionSkippingEvalEnabledCheck): * runtime/FunctionConstructor.h: 2011-05-09 Adam Roben <aroben@apple.com> Automatically touch WebKit.idl whenever any other WebKit1 IDL file changes Fixes <http://webkit.org/b/60468> WebKit.idl needs to be manually touched whenever any other WebKit1 IDL file changes to avoid build errors Reviewed by Tim Hatcher. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.make: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.vcproj: Updated for script rename. * JavaScriptCore.vcproj/JavaScriptCore/react-to-vsprops-changes.py: Removed. * JavaScriptCore.vcproj/JavaScriptCore/work-around-vs-dependency-tracking-bugs.py: Renamed from react-to-vsprops-changes.py. (top level): Moved a constant here from main. (main): Moved most code from here to react_to_vsprops_changes. Added a call to the new react_to_webkit1_interface_changes function. (react_to_vsprops_changes): Moved code here from main. Updated to use the TOP_LEVEL_DIRECTORY global. Moved some code from here to mtime_of_newest_file_matching_globa and touch_if_older_than. (react_to_webkit1_interface_changes): Added. Touches WebKit.idl if any other WebKit1 IDL file has changed. (mtime_of_newest_file_matching_glob): Added. Code came from main. (touch_if_older_than): Added. Code came from main. 2011-05-08 Jessie Berlin <jberlin@apple.com> Reviewed by Dan Bernstein. Make JSRetainPtr work with JSGlobalContextRefs. https://bugs.webkit.org/show_bug.cgi?id=60452 Add specialized functions for JSRetain and JSRelease when dealing with JSGlobalContextRefs. * API/JSRetainPtr.h: (JSRetain): (JSRelease): 2011-05-07 Dawit Alemayehu <adawit@kde.org> Reviewed by Daniel Bates. Fix compile with GCC 4.6.0 https://bugs.webkit.org/show_bug.cgi?id=60380 Remove unused local variable from code. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncMatch): 2011-05-06 Alexis Menard <alexis.menard@openbossa.org> Unreviewed build fix with gcc 4.6.0 on linux and c++0x support. std::tr1::has_trivial_constructor is in <tr1/memory>. * wtf/TypeTraits.h: 2011-05-05 Jay Civelli <jcivelli@chromium.org> Reviewed by Adam Barth. Added convenience methods to convert from a byte to hex ASCII digit characters and vice-versa. https://bugs.webkit.org/show_bug.cgi?id=59834 * wtf/ASCIICType.h: (WTF::toASCIIHexValue): (WTF::lowerNibbleToASCIIHexDigit): (WTF::upperNibbleToASCIIHexDigit): 2011-05-05 Alexis Menard <alexis.menard@openbossa.org> Reviewed by Benjamin Poulain. [Qt] Make QtWebKit build when using gcc 4.6.0 https://bugs.webkit.org/show_bug.cgi?id=60265 If QtWebKit is compiled with gcc 4.6.0 or later we don't want to deactivate the c++0x support because it works. * JavaScriptCore.pro: 2011-05-04 Fridrich Strba <fridrich.strba@bluewin.ch> Reviewed by Geoffrey Garen. Port MachineStackMarker.cpp to Windows x64 https://bugs.webkit.org/show_bug.cgi?id=60216 * heap/MachineStackMarker.cpp: (JSC::getPlatformThreadRegisters): the CONTEXT struct is usable also on 64-bit Windows. (JSC::otherThreadStackPointer): return the Rsp register on Windows x64. 2011-05-04 Fridrich Strba <fridrich.strba@bluewin.ch> Reviewed by Martin Robinson. Link libjavascriptcoregtk on Windows with winmm.dll https://bugs.webkit.org/show_bug.cgi?id=60215 * GNUmakefile.am: 2011-05-04 Tao Bai <michaelbai@chromium.org> Reviewed by David Kilzer. Populate touch-icon url to FrameLoaderClient https://bugs.webkit.org/show_bug.cgi?id=59143 * Configurations/FeatureDefines.xcconfig: 2011-05-03 Geoffrey Garen <ggaren@apple.com> Reviewed by Darin Adler. <rdar://problem/9366557> Various crashes due to bad DFG codegen at canalplus.fr * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::checkArgumentTypes): Removed a stray line of code that accidentally survived the conversion to a switch statement, causing a lot of important code not to run most of the time. Since this is not a trivial finger-picking mistake, I will not call it a typo. 2011-05-04 Adam Roben <aroben@apple.com> Another attempted build fix * wtf/OwnPtr.h: (WTF::OwnPtr::operator==): (WTF::OwnPtr::operator!=): * wtf/PassOwnPtr.h: (WTF::PassOwnPtr::operator==): (WTF::PassOwnPtr::operator!=): Added a return statement. And made a tweak based on a suggestion from Anders Carlsson. 2011-05-04 Adam Roben <aroben@apple.com> Try to fix Leopard, Qt, and probably others * wtf/OwnPtr.h: (WTF::OwnPtr::operator==): (WTF::OwnPtr::operator!=): * wtf/PassOwnPtr.h: (WTF::PassOwnPtr::operator==): (WTF::PassOwnPtr::operator!=): Try to get the compiler not to instantiate these function templates unnecessarily. 2011-05-03 Adam Roben <aroben@apple.com> Disallow equality comparisons between [Pass]OwnPtrs If you have two OwnPtrs that are equal, you've already lost. (Unless you're doing something really sneaky, in which case you should stop!) Fixes <http://webkit.org/b/60053> Testing OwnPtrs for equality should cause a compiler error Reviewed by Anders Carlsson and Antti Koivisto. * wtf/OwnPtr.h: (WTF::OwnPtr::operator==): (WTF::OwnPtr::operator!=): * wtf/PassOwnPtr.h: (WTF::PassOwnPtr::operator==): (WTF::PassOwnPtr::operator!=): Added private equality operators that fail to compile when used. (When not used, the compiler will skip over them because they are function templates.) 2011-05-04 Alexis Menard <alexis.menard@openbossa.org> Reviewed by Gavin Barraclough. JITArithmetic.cpp produces a warning on a unused variable. https://bugs.webkit.org/show_bug.cgi?id=60060 Just properly use what we already have converted. * jit/JITArithmetic.cpp: (JSC::JIT::emitSlow_op_add): (JSC::JIT::emitSlow_op_mul): 2011-05-04 Alexis Menard <alexis.menard@openbossa.org> Reviewed by Geoffrey Garen. JITPropertyAccess produces a unused but set variable warning in gcc 4.6.0. https://bugs.webkit.org/show_bug.cgi?id=60050 This patch fix a compilation warning. The new warning scenario -Wunused-but-set-variable in gcc 4.6.0 is included in -Wall and therefore stops the compilation when warnings are treated as errors. The patch introduces a new macro ASSERT_JIT_OFFSET_UNUSED and ASSERT_WITH_MESSAGE_UNUSED which copy the idea of ASSERT_UNUSED. * jit/JIT.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::emit_op_method_check): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::emit_op_put_by_id): * wtf/Assertions.h: (assertWithMessageUnused): 2011-04-29 Jer Noble <jer.noble@apple.com> Reviewed by Eric Seidel. Implement FULLSCREEN_API on Windows, Part 4: Enable it https://bugs.webkit.org/show_bug.cgi?id=59798 * wtf/Platform.h: Set ENABLE_FULLSCREEN_API on win. 2011-05-03 Alexis Menard <alexis.menard@openbossa.org> Reviewed by Eric Seidel. Unused but set variable warning in MacroAssemberX86_64 https://bugs.webkit.org/show_bug.cgi?id=59482 * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::call): (JSC::MacroAssemblerX86_64::tailRecursiveCall): (JSC::MacroAssemblerX86_64::makeTailRecursiveCall): 2011-05-03 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make malloc validation useful https://bugs.webkit.org/show_bug.cgi?id=57502 Reland this patch (rolled out in 82905) without turning it on by default. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/FastMalloc.cpp: (WTF::tryFastMalloc): (WTF::fastMalloc): (WTF::tryFastCalloc): (WTF::fastCalloc): (WTF::fastFree): (WTF::tryFastRealloc): (WTF::fastRealloc): (WTF::fastMallocSize): (WTF::TCMalloc_PageHeap::isScavengerSuspended): (WTF::TCMalloc_PageHeap::scheduleScavenger): (WTF::TCMalloc_PageHeap::suspendScavenger): (WTF::TCMalloc_PageHeap::signalScavenger): (WTF::TCMallocStats::malloc): (WTF::TCMallocStats::free): (WTF::TCMallocStats::fastCalloc): (WTF::TCMallocStats::tryFastCalloc): (WTF::TCMallocStats::calloc): (WTF::TCMallocStats::fastRealloc): (WTF::TCMallocStats::tryFastRealloc): (WTF::TCMallocStats::realloc): (WTF::TCMallocStats::fastMallocSize): * wtf/FastMalloc.h: (WTF::Internal::fastMallocValidationHeader): (WTF::Internal::fastMallocValidationSuffix): (WTF::Internal::fastMallocMatchValidationType): (WTF::Internal::setFastMallocMatchValidationType): (WTF::fastMallocMatchValidateFree): (WTF::fastMallocValidate): 2011-05-03 Xan Lopez <xlopez@igalia.com> Reviewed by Anders Carlsson. Compile error with GCC 4.6.0, tries to assign unsigned& to bitfield https://bugs.webkit.org/show_bug.cgi?id=59261 Use unary '+' to force proper type detection in template arguments with GCC 4.6.0. See bug report for more details. * runtime/Structure.cpp: (JSC::StructureTransitionTable::remove): Use '+' to force precise type detection. (JSC::StructureTransitionTable::add): ditto. * runtime/Structure.h: (JSC::StructureTransitionTable::keyForWeakGCMapFinalizer): ditto. 2011-05-03 Jessie Berlin <jberlin@apple.com> Rubber-stamped by Adam Roben. Revert r85550 and r85575. Variables cannot be exported via the .def file. Instead, they should be annotated with JS_EXPORTDATA. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/Structure.cpp: (JSC::Structure::materializePropertyMap): * runtime/Structure.h: (JSC::Structure::typeInfo): (JSC::Structure::previousID): (JSC::Structure::propertyStorageCapacity): (JSC::Structure::propertyStorageSize): (JSC::Structure::get): (JSC::Structure::materializePropertyMapIfNecessary): 2011-05-02 Adam Roben <aroben@apple.com> Allow implicit conversion from nullptr_t to PassOwnPtr This makes it a lot easier to write code that just wants a null PassOwnPtr, especially in strict PassOwnPtr mode. Fixes <http://webkit.org/b/59964> Implicit conversion from std::nullptr_t to PassOwnPtr doesn't work, but should Reviewed by Adam Barth. * wtf/PassOwnPtr.h: (WTF::PassOwnPtr::PassOwnPtr): Added a non-explicit constructor that takes a nullptr_t. * wtf/MessageQueue.h: (WTF::::waitForMessageFilteredWithTimeout): (WTF::::tryGetMessage): Use the new implicit conversion. 2011-05-02 Jessie Berlin <jberlin@apple.com> Rubber-stamped by Oliver Hunt. Remove an assertion that Windows was hitting on launch. * runtime/Structure.cpp: (JSC::Structure::materializePropertyMap): * runtime/Structure.h: (JSC::Structure::typeInfo): (JSC::Structure::previousID): (JSC::Structure::propertyStorageCapacity): (JSC::Structure::propertyStorageSize): (JSC::Structure::get): (JSC::Structure::materializePropertyMapIfNecessary): 2011-05-02 Mark Rowe <mrowe@apple.com> Reviewed by Geoff Garen. <rdar://problem/9371948> JavaScriptCore should build with GCC 4.2 * Configurations/CompilerVersion.xcconfig: 2011-05-02 Gavin Barraclough <barraclough@apple.com> ARMv7 build fix. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::Jump::link): (JSC::AbstractMacroAssembler::Jump::linkTo): 2011-05-02 Oliver Hunt <oliver@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-05-02 Michael Saboff <msaboff@apple.com> Reviewed by Geoffrey Garen. crash in JSC::RegExp::match https://bugs.webkit.org/show_bug.cgi?id=58922 Cleared chained backtrack data label when linking label even if that label doesn't chain itself. This is needed so that subsequent backtrack data labels point to the next outer paren and not within the current paren. * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::TermGenerationState::linkDataLabelToBacktrackIfExists): 2011-05-02 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Tiny bit of heap cleanup. * heap/MarkedBlock.h: (JSC::MarkedBlock::contains): Tightened up an assertion and a comment. * heap/MarkedSpace.h: (JSC::MarkedSpace::globalData): (JSC::MarkedSpace::highWaterMark): (JSC::MarkedSpace::setHighWaterMark): Moved inlines out of the class definition, for better clarity. 2011-05-02 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Correct marking of interpreter data in mixed mode builds https://bugs.webkit.org/show_bug.cgi?id=59962 We had a few places in mixed mode builds where we would not track data used by the interpreter for marking. This patch corrects the problem and adds a number of assertions to catch live Structures being collected. * JavaScriptCore.exp: * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::ARMInstructionFormatter::debugOffset): * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): * bytecode/CodeBlock.h: (JSC::CodeBlock::addPropertyAccessInstruction): (JSC::CodeBlock::addGlobalResolveInstruction): (JSC::CodeBlock::addStructureStubInfo): (JSC::CodeBlock::addGlobalResolveInfo): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitResolve): (JSC::BytecodeGenerator::emitResolveWithBase): (JSC::BytecodeGenerator::emitGetById): (JSC::BytecodeGenerator::emitPutById): (JSC::BytecodeGenerator::emitDirectPutById): * runtime/Structure.cpp: (JSC::Structure::materializePropertyMap): * runtime/Structure.h: (JSC::Structure::typeInfo): (JSC::Structure::previousID): (JSC::Structure::propertyStorageCapacity): (JSC::Structure::propertyStorageSize): (JSC::Structure::get): (JSC::Structure::materializePropertyMapIfNecessary): 2011-05-02 Xan Lopez <xlopez@igalia.com> Reviewed by Alexey Proskuryakov. Use native NullPtr when using GCC 4.6.0 and C++0x https://bugs.webkit.org/show_bug.cgi?id=59252 GCC 4.6.0 has nullptr support, use it when possible. * wtf/NullPtr.cpp: include config.h to pull in Platform.h before NullPtr.h, since we need the GCC_VERSION_AT_LEAST definition. * wtf/NullPtr.h: check for GCC >= 4.6.0 and C++0x in order to use native nullptr. 2011-05-02 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=59950 Clean up AssemblerBuffer to use a Vector internally. AssemblerBuffer handles reallocing a byte array itself - stop that. * assembler/ARMAssembler.cpp: (JSC::ARMAssembler::executableCopy): * assembler/AssemblerBuffer.h: (JSC::AssemblerLabel::AssemblerLabel): (JSC::AssemblerLabel::labelAtOffset): (JSC::AssemblerBuffer::AssemblerBuffer): (JSC::AssemblerBuffer::~AssemblerBuffer): (JSC::AssemblerBuffer::isAvailable): (JSC::AssemblerBuffer::ensureSpace): (JSC::AssemblerBuffer::isAligned): (JSC::AssemblerBuffer::putIntegral): (JSC::AssemblerBuffer::putIntegralUnchecked): (JSC::AssemblerBuffer::putByteUnchecked): (JSC::AssemblerBuffer::putByte): (JSC::AssemblerBuffer::putShortUnchecked): (JSC::AssemblerBuffer::putShort): (JSC::AssemblerBuffer::putIntUnchecked): (JSC::AssemblerBuffer::putInt): (JSC::AssemblerBuffer::putInt64Unchecked): (JSC::AssemblerBuffer::putInt64): (JSC::AssemblerBuffer::codeSize): (JSC::AssemblerBuffer::label): (JSC::AssemblerBuffer::executableCopy): (JSC::AssemblerBuffer::rewindToLabel): (JSC::AssemblerBuffer::debugOffset): (JSC::AssemblerBuffer::append): (JSC::AssemblerBuffer::grow): * assembler/AssemblerBufferWithConstantPool.h: * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::linkCall): * assembler/X86Assembler.h: (JSC::X86Assembler::X86InstructionFormatter::rewindToLabel): 2011-05-02 Jeff Miller <jeffm@apple.com> Reviewed by Alexy Proskuryakov. Avoid potential buffer overflow in WTFLog() and WTFLogVerbose() https://bugs.webkit.org/show_bug.cgi?id=59949 * wtf/Assertions.cpp: Check for 0 or empty format string in WTFLog() and WTFLogVerbose(). 2011-05-02 Adam Barth <abarth@webkit.org> Reviewed by Alexey Proskuryakov. StringImpl::endsWith has some insane code https://bugs.webkit.org/show_bug.cgi?id=59900 * wtf/text/StringImpl.cpp: (WTF::StringImpl::endsWith): - m_data shadows a member variable of the same name. 2011-05-02 Gabor Loki <loki@webkit.org> Buildfix for ARM after r85448 * assembler/ARMAssembler.h: (JSC::ARMAssembler::loadBranchTarget): 2011-05-01 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Strict-mode only reserved words not reserved https://bugs.webkit.org/show_bug.cgi?id=55342 Fix line number tracking when we rollback the lexer. * parser/JSParser.cpp: (JSC::JSParser::parseSourceElements): 2011-05-01 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. ES5 Strict mode does not allow getter and setter for same propId https://bugs.webkit.org/show_bug.cgi?id=57295 Simplify and correct the logic for strict mode object literals. * parser/JSParser.cpp: (JSC::JSParser::parseStrictObjectLiteral): 2011-05-01 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Assigning to function identifier under strict should throw https://bugs.webkit.org/show_bug.cgi?id=59289 Add logic to StaticScopeObject to ensure we don't silently consume writes to constant properties. * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::put): 2011-05-01 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=59903 Use AssemblerLabel throughout Assembler classes, AssemblerBuffer Creating a lable() into the AssemblerBuffer should return an AssemblerLabel, not an unsigned int. * assembler/ARMAssembler.cpp: (JSC::ARMAssembler::executableCopy): * assembler/ARMAssembler.h: (JSC::ARMAssembler::blx): (JSC::ARMAssembler::label): (JSC::ARMAssembler::loadBranchTarget): * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::b): (JSC::ARMv7Assembler::blx): (JSC::ARMv7Assembler::bx): (JSC::ARMv7Assembler::label): (JSC::ARMv7Assembler::ARMInstructionFormatter::label): * assembler/AssemblerBuffer.h: (JSC::AssemblerBuffer::label): * assembler/AssemblerBufferWithConstantPool.h: * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::label): (JSC::MIPSAssembler::relocateJumps): * assembler/SH4Assembler.h: (JSC::SH4Assembler::loadConstant): (JSC::SH4Assembler::loadConstantUnReusable): (JSC::SH4Assembler::call): (JSC::SH4Assembler::jmp): (JSC::SH4Assembler::jne): (JSC::SH4Assembler::je): (JSC::SH4Assembler::label): (JSC::SH4Assembler::oneShortOp): * assembler/X86Assembler.h: (JSC::X86Assembler::call): (JSC::X86Assembler::jmp_r): (JSC::X86Assembler::label): (JSC::X86Assembler::X86InstructionFormatter::immediateRel32): (JSC::X86Assembler::X86InstructionFormatter::label): 2011-05-01 Adam Barth <abarth@webkit.org> Reviewed by David Levin. Enable strict mode for OwnPtr and PassOwnPtr https://bugs.webkit.org/show_bug.cgi?id=59428 * wtf/OwnPtr.h: 2011-05-01 Patrick Gansterer <paroga@webkit.org> Reviewed by Adam Barth. Enable strict OwnPtr for PLATFORM(WIN) https://bugs.webkit.org/show_bug.cgi?id=59881 * wtf/OwnPtr.h: 2011-05-01 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. https://bugs.webkit.org/show_bug.cgi?id=59896 Remove JmpSrc/JmpDst types. The JmpSrc/JmpDst classes predate the MacroAssembler interface. Having these object be per-assembler in unhelpful, causes unnecessary code duplication, and prevents the AssemblerBuffer from providing a richer type for labels. The limited semantic meaning that they did convey is undermined by the manner in which their meanings have been overloaded (use of JmpSrc for Call, JmpDst for data labels). Jumps on ARMv7 have had additional information added to the object via the ARMv7 JmpSrc. This data should probably be in the instruction stream. This patch does not fix the problem, and moves the data (ifdefed) to AbstractMacroAssembler::Jump (which is effectively where it was before!). This at least closes the hole such that no further data may be added to JmpSrc, but this is unfortunate, and should be cleaned up. * assembler/ARMAssembler.h: (JSC::ARMAssembler::blx): (JSC::ARMAssembler::label): (JSC::ARMAssembler::align): (JSC::ARMAssembler::loadBranchTarget): (JSC::ARMAssembler::jmp): (JSC::ARMAssembler::linkPointer): (JSC::ARMAssembler::linkJump): (JSC::ARMAssembler::linkCall): (JSC::ARMAssembler::getRelocatedAddress): (JSC::ARMAssembler::getDifferenceBetweenLabels): (JSC::ARMAssembler::getCallReturnOffset): * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::b): (JSC::ARMv7Assembler::blx): (JSC::ARMv7Assembler::bx): (JSC::ARMv7Assembler::label): (JSC::ARMv7Assembler::align): (JSC::ARMv7Assembler::getRelocatedAddress): (JSC::ARMv7Assembler::getDifferenceBetweenLabels): (JSC::ARMv7Assembler::getCallReturnOffset): (JSC::ARMv7Assembler::linkJump): (JSC::ARMv7Assembler::linkCall): (JSC::ARMv7Assembler::linkPointer): * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::Label::isSet): (JSC::AbstractMacroAssembler::Call::Call): (JSC::AbstractMacroAssembler::Jump::Jump): (JSC::AbstractMacroAssembler::Jump::link): (JSC::AbstractMacroAssembler::Jump::linkTo): (JSC::AbstractMacroAssembler::linkPointer): (JSC::AbstractMacroAssembler::getLinkerAddress): * assembler/AssemblerBuffer.h: (JSC::AssemblerLabel::AssemblerLabel): (JSC::AssemblerLabel::isSet): * assembler/LinkBuffer.h: (JSC::LinkBuffer::patch): * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::label): (JSC::MIPSAssembler::align): (JSC::MIPSAssembler::getRelocatedAddress): (JSC::MIPSAssembler::getDifferenceBetweenLabels): (JSC::MIPSAssembler::getCallReturnOffset): (JSC::MIPSAssembler::linkJump): (JSC::MIPSAssembler::linkCall): (JSC::MIPSAssembler::linkPointer): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::branchDouble): (JSC::MacroAssemblerARMv7::branchDoubleZeroOrNaN): (JSC::MacroAssemblerARMv7::jump): (JSC::MacroAssemblerARMv7::nearCall): (JSC::MacroAssemblerARMv7::call): (JSC::MacroAssemblerARMv7::ret): (JSC::MacroAssemblerARMv7::tailRecursiveCall): (JSC::MacroAssemblerARMv7::makeBranch): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::nearCall): (JSC::MacroAssemblerMIPS::call): (JSC::MacroAssemblerMIPS::tailRecursiveCall): (JSC::MacroAssemblerMIPS::branchTrue): (JSC::MacroAssemblerMIPS::branchFalse): (JSC::MacroAssemblerMIPS::branchEqual): (JSC::MacroAssemblerMIPS::branchNotEqual): * assembler/SH4Assembler.h: (JSC::SH4Assembler::call): (JSC::SH4Assembler::jmp): (JSC::SH4Assembler::jne): (JSC::SH4Assembler::je): (JSC::SH4Assembler::label): (JSC::SH4Assembler::align): (JSC::SH4Assembler::linkJump): (JSC::SH4Assembler::linkCall): (JSC::SH4Assembler::linkPointer): (JSC::SH4Assembler::getCallReturnOffset): (JSC::SH4Assembler::getRelocatedAddress): (JSC::SH4Assembler::getDifferenceBetweenLabels): (JSC::SH4Assembler::patchPointer): * assembler/X86Assembler.h: (JSC::X86Assembler::call): (JSC::X86Assembler::jmp): (JSC::X86Assembler::jmp_r): (JSC::X86Assembler::jne): (JSC::X86Assembler::jnz): (JSC::X86Assembler::je): (JSC::X86Assembler::jz): (JSC::X86Assembler::jl): (JSC::X86Assembler::jb): (JSC::X86Assembler::jle): (JSC::X86Assembler::jbe): (JSC::X86Assembler::jge): (JSC::X86Assembler::jg): (JSC::X86Assembler::ja): (JSC::X86Assembler::jae): (JSC::X86Assembler::jo): (JSC::X86Assembler::jp): (JSC::X86Assembler::js): (JSC::X86Assembler::jCC): (JSC::X86Assembler::label): (JSC::X86Assembler::labelFor): (JSC::X86Assembler::align): (JSC::X86Assembler::linkJump): (JSC::X86Assembler::linkCall): (JSC::X86Assembler::linkPointer): (JSC::X86Assembler::getCallReturnOffset): (JSC::X86Assembler::getRelocatedAddress): (JSC::X86Assembler::getDifferenceBetweenLabels): (JSC::X86Assembler::rewindToLabel): (JSC::X86Assembler::X86InstructionFormatter::immediateRel32): (JSC::X86Assembler::X86InstructionFormatter::rewindToLabel): * jit/JIT.cpp: (JSC::JIT::privateCompileMainPass): * jit/JIT.h: * jit/JITInlineMethods.h: (JSC::JIT::atJumpTarget): (JSC::JIT::emitGetVirtualRegister): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_jmp): (JSC::JIT::emit_op_jfalse): (JSC::JIT::emit_op_jeq_null): (JSC::JIT::emit_op_jneq_null): (JSC::JIT::emit_op_jneq_ptr): (JSC::JIT::emit_op_jsr): (JSC::JIT::emit_op_jtrue): (JSC::JIT::emit_op_jmp_scopes): 2011-05-01 Chao-ying Fu <fu@mips.com> Reviewed by Eric Seidel. Fix MIPS build due to the split of "Condition" enum https://bugs.webkit.org/show_bug.cgi?id=59407 * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::debugOffset): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::branch32): (JSC::MacroAssemblerMIPS::compare32): 2011-04-30 Adam Barth <abarth@webkit.org> Reviewed by Adam Barth. Enable strict OwnPtr for GTK https://bugs.webkit.org/show_bug.cgi?id=59861 * wtf/OwnPtr.h: 2011-04-30 Gavin Barraclough <barraclough@apple.com> ARMv7 build fix. * assembler/AssemblerBufferWithConstantPool.h: 2011-04-30 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. Bug 59869 - AssemblerBuffer cleanup - disambiguate size() The method size() is called on the AssemblerBuffer both to acquire the complete size of the code, and to get a position to use as a label into the code. Instead, add an explicit 'label' method. * assembler/ARMAssembler.cpp: (JSC::ARMAssembler::executableCopy): * assembler/ARMAssembler.h: (JSC::ARMAssembler::blx): (JSC::ARMAssembler::codeSize): (JSC::ARMAssembler::label): (JSC::ARMAssembler::loadBranchTarget): * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::b): (JSC::ARMv7Assembler::blx): (JSC::ARMv7Assembler::bx): (JSC::ARMv7Assembler::label): (JSC::ARMv7Assembler::codeSize): (JSC::ARMv7Assembler::ARMInstructionFormatter::codeSize): (JSC::ARMv7Assembler::ARMInstructionFormatter::data): * assembler/AbstractMacroAssembler.h: * assembler/AssemblerBuffer.h: (JSC::AssemblerBuffer::codeSize): (JSC::AssemblerBuffer::label): * assembler/AssemblerBufferWithConstantPool.h: * assembler/LinkBuffer.h: (JSC::LinkBuffer::LinkBuffer): (JSC::LinkBuffer::linkCode): * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::newJmpSrc): (JSC::MIPSAssembler::appendJump): (JSC::MIPSAssembler::label): (JSC::MIPSAssembler::codeSize): (JSC::MIPSAssembler::relocateJumps): * assembler/SH4Assembler.h: (JSC::SH4Assembler::loadConstant): (JSC::SH4Assembler::loadConstantUnReusable): (JSC::SH4Assembler::call): (JSC::SH4Assembler::jmp): (JSC::SH4Assembler::jne): (JSC::SH4Assembler::je): (JSC::SH4Assembler::label): (JSC::SH4Assembler::executableCopy): (JSC::SH4Assembler::oneShortOp): (JSC::SH4Assembler::codeSize): * assembler/X86Assembler.h: (JSC::X86Assembler::call): (JSC::X86Assembler::jmp_r): (JSC::X86Assembler::codeSize): (JSC::X86Assembler::label): (JSC::X86Assembler::executableCopy): (JSC::X86Assembler::X86InstructionFormatter::immediateRel32): (JSC::X86Assembler::X86InstructionFormatter::codeSize): (JSC::X86Assembler::X86InstructionFormatter::label): (JSC::X86Assembler::X86InstructionFormatter::executableCopy): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * jit/JIT.cpp: (JSC::JIT::privateCompile): * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::compile): 2011-04-29 Adam Barth <abarth@webkit.org> Attempt to fix the Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-04-29 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. CSP script-src should block eval https://bugs.webkit.org/show_bug.cgi?id=59850 ggaren recommend a different approach to this patch, essentially installing a new function for function-eval and changing the AST representation of operator-eval to call function-eval. However, I'm not sure that approach is workable because the ASTBuilder doesn't know about global objects, and there is added complication due to the cache. This approach is more dynamic, adding a branch in EvalExecutable to detect whether eval is current disabled in the lexical scope. The spec is slightly unclear about whether we should return undefined or throw an exception. I've asked Brandon to clarify the spec, but throwing an exception seems natural. * JavaScriptCore.exp: * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::disableEval): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::JSGlobalObject): (JSC::JSGlobalObject::isEvalEnabled): 2011-04-29 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=59847 Remove linkOffset from LinkBuffer This is redundant since removal of recompilation for exception info. * assembler/LinkBuffer.h: (JSC::LinkBuffer::LinkBuffer): (JSC::LinkBuffer::linkCode): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * jit/JIT.cpp: (JSC::JIT::JIT): (JSC::JIT::privateCompile): * jit/JIT.h: (JSC::JIT::compile): (JSC::JIT::compileCTIMachineTrampolines): (JSC::JIT::compileCTINativeCall): * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::privateCompileCTINativeCall): * jit/JITPropertyAccess.cpp: (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::finalize): * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::compile): 2011-04-29 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt & Geoff Garen. https://bugs.webkit.org/show_bug.cgi?id=59221 [RegexFuzz] Regression blocking testing Okay, so the bug here is that when, in the case of a TypeParentheticalAssertion node, emitDisjunction recursively calls to itself to emit the nested disjunction the value of parenthesesInputCountAlreadyChecked is bogus (doesn't take into account the uncheck that has just taken place). Also, the special handling given to countToCheck in the case of parenthetical assertions is nonsense, delete it, along with the isParentheticalAssertion argument. * yarr/YarrInterpreter.cpp: (JSC::Yarr::ByteCompiler::emitDisjunction): 2011-04-29 Csaba Osztrogonác <ossy@webkit.org> Reviewed by Adam Barth. Enable strict OwnPtr for Qt https://bugs.webkit.org/show_bug.cgi?id=59667 * wtf/OwnPtr.h: 2011-04-29 Dean Jackson <dino@apple.com> Reviewed by Simon Fraser. Add ENABLE macro for WebKitAnimation https://bugs.webkit.org/show_bug.cgi?id=59729 Add new feature to toggle WebKit Animation API. * Configurations/FeatureDefines.xcconfig: 2011-04-28 Sam Weinig <sam@webkit.org> Reviewed by Mark Rowe. Install testapi.js along side testapi https://bugs.webkit.org/show_bug.cgi?id=59773 * JavaScriptCore.xcodeproj/project.pbxproj: Add new build phase to copy testapi.js to install path of testapi on install. 2011-04-28 David Levin <levin@chromium.org> Reviewed by Adam Barth. Remove IMAGE_RESIZER related code. https://bugs.webkit.org/show_bug.cgi?id=59735 * Configurations/FeatureDefines.xcconfig: 2011-04-28 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=59763 DFG JIT - Unify FPRReg & FPRegisterID (Following on from GPRReg/RegisterID unification). * dfg/DFGFPRInfo.h: (JSC::DFG::FPRInfo::toRegister): (JSC::DFG::FPRInfo::debugName): * dfg/DFGGPRInfo.h: * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::checkConsistency): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::boxDouble): (JSC::DFG::JITCodeGenerator::unboxDouble): (JSC::DFG::JITCodeGenerator::flushRegisters): (JSC::DFG::JITCodeGenerator::isFlushed): (JSC::DFG::JITCodeGenerator::setupTwoStubArgs): (JSC::DFG::JITCodeGenerator::setupStubArguments): (JSC::DFG::JITCodeGenerator::callOperation): (JSC::DFG::GPRResult::lockedResult): (JSC::DFG::FPRResult::lockedResult): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::fillNumericToDouble): (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGJITCompiler.h: * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::EntryLocation::EntryLocation): (JSC::DFG::NonSpeculativeJIT::valueToNumber): (JSC::DFG::NonSpeculativeJIT::valueToInt32): (JSC::DFG::NonSpeculativeJIT::numberToInt32): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGNonSpeculativeJIT.h: (JSC::DFG::NonSpeculativeJIT::silentSpillAllRegisters): (JSC::DFG::NonSpeculativeJIT::silentFillAllRegisters): * dfg/DFGRegisterBank.h: (JSC::DFG::RegisterBank::iterator::regID): (JSC::DFG::RegisterBank::iterator::debugName): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculationCheck::SpeculationCheck): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: 2011-04-28 David Kilzer <ddkilzer@apple.com> Revert "<http://webkit.org/b/59705> WTF::postTimer() leaks a CFRunLoopTimerRef every time it's called" This reverts commit r85195. It was crashing DumpRenderTree on Lion. * wtf/mac/MainThreadMac.mm: (WTF::postTimer): 2011-04-28 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. Remove WML https://bugs.webkit.org/show_bug.cgi?id=59678 Remove the WML configuration option from the Mac build system. * Configurations/FeatureDefines.xcconfig: 2011-04-28 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r85233 and r85235. http://trac.webkit.org/changeset/85233 http://trac.webkit.org/changeset/85235 https://bugs.webkit.org/show_bug.cgi?id=59754 Causes issues with jsc. (Requested by dave_levin on #webkit). * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * jit/ExecutableAllocator.h: (JSC::ExecutablePool::ExecutablePool): * parser/SourceProvider.h: (JSC::SourceProvider::SourceProvider): * runtime/RegExp.cpp: (JSC::RegExp::RegExp): * wtf/CMakeLists.txt: * wtf/RefCounted.h: (WTF::RefCountedBase::ref): (WTF::RefCountedBase::hasOneRef): (WTF::RefCountedBase::refCount): (WTF::RefCountedBase::derefBase): * wtf/SizeLimits.cpp: * wtf/ThreadRestrictionVerifier.h: Removed. * wtf/text/CString.h: (WTF::CStringBuffer::CStringBuffer): 2011-04-28 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. Bug 59740 - DFG JIT - Unify GPRReg & RegisterID Currently we use a mix of enum values throughout the DFG JIT to represent gpr registers - the RegisterID provided by the MacroAssembler, and the GPRReg enum giving the sequential register set over which the RegisterBank allocates. Unify the two. Patch to unify FPRReg in a similar fashion will follow. * JavaScriptCore.xcodeproj/project.pbxproj: * dfg/DFGFPRInfo.h: Added. (JSC::DFG::next): (JSC::DFG::FPRBankInfo::toRegister): (JSC::DFG::FPRBankInfo::toIndex): * dfg/DFGGPRInfo.h: Added. (JSC::DFG::GPRInfo::toRegister): (JSC::DFG::GPRInfo::toIndex): (JSC::DFG::GPRInfo::debugName): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillInteger): (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::dump): (JSC::DFG::JITCodeGenerator::checkConsistency): (JSC::DFG::GPRTemporary::GPRTemporary): (JSC::DFG::FPRTemporary::FPRTemporary): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::boxDouble): (JSC::DFG::JITCodeGenerator::unboxDouble): (JSC::DFG::JITCodeGenerator::spill): (JSC::DFG::JITCodeGenerator::flushRegisters): (JSC::DFG::JITCodeGenerator::isFlushed): (JSC::DFG::JITCodeGenerator::bitOp): (JSC::DFG::JITCodeGenerator::shiftOp): (JSC::DFG::JITCodeGenerator::setupTwoStubArgs): (JSC::DFG::JITCodeGenerator::setupStubArguments): (JSC::DFG::JITCodeGenerator::callOperation): (JSC::DFG::IntegerOperand::gpr): (JSC::DFG::DoubleOperand::gpr): (JSC::DFG::GPRTemporary::gpr): (JSC::DFG::FPRTemporary::gpr): (JSC::DFG::GPRResult::lockedResult): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::fillNumericToDouble): (JSC::DFG::JITCompiler::fillInt32ToInteger): (JSC::DFG::JITCompiler::fillToJS): (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): (JSC::DFG::JITCompiler::compileFunction): (JSC::DFG::JITCompiler::jitAssertIsInt32): (JSC::DFG::JITCompiler::jitAssertIsJSInt32): (JSC::DFG::JITCompiler::jitAssertIsJSNumber): (JSC::DFG::JITCompiler::jitAssertIsJSDouble): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::preserveReturnAddressAfterCall): (JSC::DFG::JITCompiler::restoreReturnAddressBeforeReturn): (JSC::DFG::JITCompiler::emitGetFromCallFrameHeaderPtr): (JSC::DFG::JITCompiler::emitPutToCallFrameHeader): (JSC::DFG::JITCompiler::emitPutImmediateToCallFrameHeader): (JSC::DFG::JITCompiler::addressForGlobalVar): (JSC::DFG::JITCompiler::addressFor): (JSC::DFG::JITCompiler::tagFor): (JSC::DFG::JITCompiler::payloadFor): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::EntryLocation::EntryLocation): (JSC::DFG::NonSpeculativeJIT::valueToNumber): (JSC::DFG::NonSpeculativeJIT::valueToInt32): (JSC::DFG::NonSpeculativeJIT::numberToInt32): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGNonSpeculativeJIT.h: (JSC::DFG::NonSpeculativeJIT::silentSpillGPR): (JSC::DFG::NonSpeculativeJIT::silentSpillFPR): (JSC::DFG::NonSpeculativeJIT::silentFillGPR): (JSC::DFG::NonSpeculativeJIT::silentFillFPR): (JSC::DFG::NonSpeculativeJIT::silentSpillAllRegisters): (JSC::DFG::NonSpeculativeJIT::silentFillAllRegisters): * dfg/DFGRegisterBank.h: (JSC::DFG::RegisterBank::allocate): (JSC::DFG::RegisterBank::retain): (JSC::DFG::RegisterBank::release): (JSC::DFG::RegisterBank::lock): (JSC::DFG::RegisterBank::unlock): (JSC::DFG::RegisterBank::isLocked): (JSC::DFG::RegisterBank::name): (JSC::DFG::RegisterBank::iterator::name): (JSC::DFG::RegisterBank::iterator::isLocked): (JSC::DFG::RegisterBank::iterator::release): (JSC::DFG::RegisterBank::iterator::gpr): (JSC::DFG::RegisterBank::iterator::debugName): (JSC::DFG::RegisterBank::iterator::operator++): (JSC::DFG::RegisterBank::iterator::operator!=): (JSC::DFG::RegisterBank::iterator::index): (JSC::DFG::RegisterBank::iterator::iterator): (JSC::DFG::RegisterBank::begin): (JSC::DFG::RegisterBank::end): (JSC::DFG::RegisterBank::isLockedAtIndex): (JSC::DFG::RegisterBank::nameAtIndex): (JSC::DFG::RegisterBank::releaseAtIndex): (JSC::DFG::RegisterBank::allocateInternal): (JSC::DFG::RegisterBank::MapEntry::MapEntry): * dfg/DFGScoreBoard.h: (JSC::DFG::ScoreBoard::~ScoreBoard): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculationCheck::SpeculationCheck): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::compilePeepHoleBranch): (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::initializeVariableTypes): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculateIntegerOperand::gpr): 2011-04-28 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Remove evil addressOfStructure() function https://bugs.webkit.org/show_bug.cgi?id=59739 Remove the addressOfStructure function from JSCell, and update callsites to use the same logic as testPrototype() * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdProtoList): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdProtoList): * runtime/JSCell.h: 2011-04-28 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Clean up testPrototype() https://bugs.webkit.org/show_bug.cgi?id=59734 Remove direct pointer to the inside of a GC object and just do the indirect load manually. Doesn't effect sunspider but does clean up the code quite a bit, and simplifies the handling of GC values. * jit/JITPropertyAccess.cpp: (JSC::JIT::testPrototype): 2011-04-28 David Levin <levin@chromium.org> Build fix. * wtf/RefCounted.h: Fix inverted ifdef. 2011-04-07 David Levin <levin@chromium.org> Reviewed by Darin Adler. Add asserts to RefCounted to make sure ref/deref happens on the right thread. https://bugs.webkit.org/show_bug.cgi?id=31639 * GNUmakefile.list.am: Added new files to the build. * JavaScriptCore.gypi: Ditto. * JavaScriptCore.vcproj/WTF/WTF.vcproj: Ditto. * JavaScriptCore.xcodeproj/project.pbxproj: Ditto. * jit/ExecutableAllocator.h: (JSC::ExecutablePool::ExecutablePool): Turned off checks for this due to not being able to figure out what was guarding it (bug 58091). * parser/SourceProvider.h: (JSC::SourceProvider::SourceProvider): Ditto. * runtime/RegExp.cpp: (JSC::RegExp::RegExp): Ditto. * wtf/CMakeLists.txt: Added new files to the build. * wtf/ThreadRestrictionVerifier.h: Added. Everything is done in the header to avoid the issue with exports that are only useful in debug but still needing to export them. * wtf/RefCounted.h: (WTF::RefCountedBase::ref): Added checks using the non thread safe verifier. and filed bug 58171 about making it stricter. (WTF::RefCountedBase::hasOneRef): Ditto. (WTF::RefCountedBase::refCount): Ditto. (WTF::RefCountedBase::setMutexForVerifier): Expose a way to change the checks to be based on a mutex. This is in the header to avoid adding more exports from JavaScriptCore. (WTF::RefCountedBase::deprecatedTurnOffVerifier): Temporary way to turn off verification. Filed bug 58174 to remove this method. (WTF::RefCountedBase::derefBase): * wtf/SizeLimits.cpp: Adjusted the debug size check for RefCounted. * wtf/text/CString.h: (WTF::CStringBuffer::CStringBuffer): Turned off checks for this while a fix is being done in Chromium's test_shell (bug 58093). 2011-04-28 Xan Lopez <xlopez@igalia.com> Unreviewed attempt to fix the build. * GNUmakefile.am: add -lpthread. 2011-04-28 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Only need a single implementation of testPrototype https://bugs.webkit.org/show_bug.cgi?id=59724 Remove excess copy of identical testPrototype() code * jit/JITPropertyAccess.cpp: (JSC::JIT::testPrototype): * jit/JITPropertyAccess32_64.cpp: 2011-04-28 Xan Lopez <xlopez@igalia.com> Reviewed by Martin Robinson. [Gtk] Split JSC and WebCore builds https://bugs.webkit.org/show_bug.cgi?id=19428 Build JavaScriptCore as a libtool shared library instead of a private convenience library. * GNUmakefile.am: define new jsc library and adapt to new name for javascriptcore target. * GNUmakefile.list.am: ditto. 2011-04-28 David Kilzer <ddkilzer@apple.com> <http://webkit.org/b/59705> WTF::postTimer() leaks a CFRunLoopTimerRef every time it's called Reviewed by Simon Fraser. * wtf/mac/MainThreadMac.mm: (WTF::postTimer): Use RetainPtr to plug the leak. 2011-04-27 Sam Weinig <sam@webkit.org> Reviewed by David Kilzer. Add way to install testapi in production builds https://bugs.webkit.org/show_bug.cgi?id=59674 * Configurations/TestAPI.xcconfig: Copied from Configurations/JavaScriptCore.xcconfig. Add configuration file for TestAPI. In addition to name, we now specify an install path and allow SKIP_INSTALL to be overridden by setting FORCE_TOOL_INSTALL. * JavaScriptCore.xcodeproj/project.pbxproj: Remove in-project build settings and add missing configuration files. Added missing CompilerVersion.xcconfig file. 2011-04-27 Adam Barth <abarth@webkit.org> Reviewed by David Levin. Enable strict OwnPtrs for Chromium https://bugs.webkit.org/show_bug.cgi?id=59666 * wtf/OwnPtr.h: 2011-04-27 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Add ability to remove keys from weakmap API https://bugs.webkit.org/show_bug.cgi?id=59645 Add JSWeakObjectMapRemove API * API/JSWeakObjectMapRefPrivate.cpp: * API/JSWeakObjectMapRefPrivate.h: * JavaScriptCore.exp: 2011-04-27 Adam Barth <abarth@webkit.org> Reviewed by David Levin. Enable strict mode for OwnPtr https://bugs.webkit.org/show_bug.cgi?id=59428 This patch enables strict mode for OwnPtr on PLATFORM(MAC) only. * wtf/OwnPtr.h: 2011-04-27 Steve Block <steveblock@google.com> Reviewed by David Levin. Remove Android build system https://bugs.webkit.org/show_bug.cgi?id=48111 This is to avoid the maintenance burden until the Android port is fully upstreamed. * Android.mk: Removed. * Android.v8.wtf.mk: Removed. 2011-04-27 Mark Rowe <mrowe@apple.com> Fix 32-bit build after r85036. * wtf/Platform.h: USE(PLUGIN_HOST_PROCESS) is only true for 64-bit. 2011-04-27 Csaba Osztrogonác <ossy@webkit.org> Unreviewed buildfix after r85036. Readd non-dead code. * wtf/OSAllocatorPosix.cpp: (WTF::OSAllocator::reserveAndCommit): 2011-04-27 Adam Barth <abarth@webkit.org> Reviewed by Kenneth Russell. OwnPtr assignment operator should be private https://bugs.webkit.org/show_bug.cgi?id=59487 Unfortunately we can't remove the copy constructor because of some detail about gcc. (The issue is documented in a comment already.) * wtf/OwnPtr.h: 2011-04-26 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r84977. http://trac.webkit.org/changeset/84977 https://bugs.webkit.org/show_bug.cgi?id=59568 caused crashes on the SL WK2 bots (Requested by jessieberlin on #webkit). * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::call): (JSC::MacroAssemblerX86_64::tailRecursiveCall): (JSC::MacroAssemblerX86_64::makeTailRecursiveCall): 2011-04-26 Kevin Ollivier <kevino@theolliviers.com> Rubberstamped by Eric Seidel. Enable waf to be used to build other ports https://bugs.webkit.org/show_bug.cgi?id=58213 * wscript: 2011-04-26 Sam Weinig <sam@webkit.org> Reviewed by David Hyatt. Remove Datagrid from the tree https://bugs.webkit.org/show_bug.cgi?id=59543 * Configurations/FeatureDefines.xcconfig: Remove feature. 2011-04-26 Adrienne Walker <enne@google.com> Reviewed by Geoffrey Garen. Fix incorrect use of OwnPtr<T*> in GCActivityCallback https://bugs.webkit.org/show_bug.cgi?id=59559 * runtime/GCActivityCallback.h: 2011-04-26 Xan Lopez <xlopez@igalia.com> Reviewed by Daniel Bates. Unused but set variable warning in MacroAssembelX86_64 https://bugs.webkit.org/show_bug.cgi?id=59482 * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::call): do not declare the label variable if we are not going to use it. (JSC::MacroAssemblerX86_64::tailRecursiveCall): ditto. (JSC::MacroAssemblerX86_64::makeTailRecursiveCall): ditto. 2011-04-26 Dan Bernstein <mitz@apple.com> Reviewed by Mark Rowe. Choose the compiler based on the Xcode version for Snow Leopard debug builds. * Configurations/Base.xcconfig: * Configurations/CompilerVersion.xcconfig: Added. 2011-04-25 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Nixed special finalizer handling for WebCore strings https://bugs.webkit.org/show_bug.cgi?id=59425 SunSpider reports no change. Not needed anymore, since weak handles have finalizers. * runtime/JSString.cpp: (JSC::JSString::resolveRope): (JSC::JSString::resolveRopeSlowCase): (JSC::JSString::outOfMemory): (JSC::JSString::substringFromRope): (JSC::JSString::replaceCharacter): Updated for removal of union. * runtime/JSString.h: (JSC::RopeBuilder::JSString): (JSC::RopeBuilder::~JSString): (JSC::RopeBuilder::appendStringInConstruct): (JSC::RopeBuilder::appendValueInConstructAndIncrementLength): No need for union or special constructor anymore. 2011-04-26 Gabor Loki <loki@webkit.org> Reviewed by Csaba Osztrogonác. Speeding up SVG filters with multicore (SMP) support https://bugs.webkit.org/show_bug.cgi?id=43903 Some SVG filters execute a huge number of pixel manipulations, which cannot be sped up by graphics accelerators, since their algorithm is too complex. Using the power of Symmetric Multi Processing (SMP) we can split up a task to smaller (data independent) tasks, which can be executed independently. The ParallelJobs framework provides a simple way for distributed programming. The framework is based on WebKit's threading infrastructure, Open Multi-Processing's (OpenMP) API, and libdispatch API. * GNUmakefile.list.am: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/CMakeLists.txt: * wtf/ParallelJobs.h: Added. (WTF::ParallelJobs::ParallelJobs): (WTF::ParallelJobs::numberOfJobs): (WTF::ParallelJobs::parameterForJob): (WTF::ParallelJobs::executeJobs): * wtf/ParallelJobsGeneric.cpp: Added. (WTF::ParallelEnvironment::ThreadPrivate::tryLockFor): (WTF::ParallelEnvironment::ThreadPrivate::executeJob): (WTF::ParallelEnvironment::ThreadPrivate::waitForFinish): (WTF::ParallelEnvironment::ThreadPrivate::workerThread): * wtf/ParallelJobsGeneric.h: Added. (WTF::ParallelEnvironment::ParallelEnvironment): (WTF::ParallelEnvironment::numberOfJobs): (WTF::ParallelEnvironment::parameterForJob): (WTF::ParallelEnvironment::executeJobs): (WTF::ParallelEnvironment::ThreadPrivate::ThreadPrivate): (WTF::ParallelEnvironment::ThreadPrivate::create): * wtf/ParallelJobsLibdispatch.h: Added. (WTF::ParallelEnvironment::ParallelEnvironment): (WTF::ParallelEnvironment::numberOfJobs): (WTF::ParallelEnvironment::parameterForJob): (WTF::ParallelEnvironment::executeJobs): * wtf/ParallelJobsOpenMP.h: Added. (WTF::ParallelEnvironment::ParallelEnvironment): (WTF::ParallelEnvironment::numberOfJobs): (WTF::ParallelEnvironment::parameterForJob): (WTF::ParallelEnvironment::executeJobs): * wtf/Platform.h: * wtf/wtf.pri: 2011-04-26 Mihai Parparita <mihaip@chromium.org> Reviewed by Adam Barth. Turn off make built-in implicit rules for derived sources makefile https://bugs.webkit.org/show_bug.cgi?id=59418 We don't use any of make's built-in implicit rules, turning them off speeds up parsing of the makefile. * JavaScriptCore.xcodeproj/project.pbxproj: * gyp/generate-derived-sources.sh: 2011-04-25 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Custom prototypes on DOM objects don't persist after garbage collection https://bugs.webkit.org/show_bug.cgi?id=59412 SunSpider reports no change. The hasCustomProperties() check didn't check for a custom prototype. * runtime/JSObject.h: (JSC::JSObject::hasCustomProperties): Changed to delegate to Structure because it is the "truth" about an object's pedigree. * runtime/Structure.cpp: (JSC::Structure::Structure): * runtime/Structure.h: (JSC::Structure::didTransition): Track whether a Structure has ever transitioned for any reason. If so, we have to assume that the object holding it is custom in some way. 2011-04-25 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen. https://bugs.webkit.org/show_bug.cgi?id=59405 DFG JIT - add type speculation for integer & array types, for vars & args. If a var or argument is used as the base for a GetByVal or PutByVal access we are speculating that it is of type Array (we only generate code on the speculative path to perform array accesses). By typing the var or args slot as Array, and checking on entry to the function (in the case of args), and each time the local is written to, we can avoid a type check at each point the array is accessed. This will typically hoist type checks out of loops. Similarly, any local that is incremented or decremented, or is the input or output or a bitwise operator, is likely to be an integer. By typing the local as int32 we can avoid speculation checks on access, and tagging when writing to the slot. All accesses can become 32bit instead of 64. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::set): (JSC::DFG::ByteCodeParser::predictArray): (JSC::DFG::ByteCodeParser::predictInt32): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.h: (JSC::DFG::PredictionSlot::PredictionSlot): (JSC::DFG::Graph::Graph): (JSC::DFG::Graph::predict): (JSC::DFG::Graph::getPrediction): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::tagFor): (JSC::DFG::JITCompiler::payloadFor): * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): (JSC::DFG::SpeculativeJIT::checkArgumentTypes): (JSC::DFG::SpeculativeJIT::initializeVariableTypes): * dfg/DFGSpeculativeJIT.h: * runtime/Executable.cpp: (JSC::tryDFGCompile): 2011-04-25 David Levin <levin@chromium.org> Reviewed by James Robinson. Fix OwnPtr strict mode violation in MessageQueue.h https://bugs.webkit.org/show_bug.cgi?id=59400 * wtf/MessageQueue.h: (WTF::::waitForMessage): (WTF::::waitForMessageFilteredWithTimeout): (WTF::::tryGetMessage): 2011-04-25 Adam Barth <abarth@webkit.org> Reviewed by Darin Adler. JavaScriptCore should play nice strict OwnPtrs https://bugs.webkit.org/show_bug.cgi?id=59401 * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parse): * heap/Heap.cpp: (JSC::TypeCounter::TypeCounter): * jit/JITStubs.cpp: (JSC::JITThunks::JITThunks): * parser/JSParser.cpp: (JSC::JSParser::Scope::Scope): * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::GenerationState::addParenthesesTail): 2011-04-25 Mark Rowe <mrowe@apple.com> Build fix. * wtf/ListHashSet.h: 2011-04-25 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. Bug 59370 - DFG JIT - fix leak of BlocksBlocks (put the blocks immediately into an OwnPtr). * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parse): 2011-04-25 James Robinson <jamesr@chromium.org> Reviewed by David Levin. Fix strict OwnPtr violations in ListHashSet and RenderLayerCompositor https://bugs.webkit.org/show_bug.cgi?id=59353 * wtf/ListHashSet.h: (WTF::::ListHashSet): 2011-04-25 David Levin <levin@chromium.org> Reviewed by Adam Barth. Fix PassOwnPtr issues in Structure and JSGlobalData.cpp https://bugs.webkit.org/show_bug.cgi?id=59347 * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/Structure.cpp: (JSC::Structure::copyPropertyTable): (JSC::Structure::createPropertyMap): * runtime/Structure.h: 2011-04-25 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make ClassInfo required when creating a Structure https://bugs.webkit.org/show_bug.cgi?id=59340 Add ClassInfo to all those types which currently don't have it, and add an assertion to Structure::create to ensure that the provided classInfo is not null. * runtime/Executable.h: (JSC::EvalExecutable::createStructure): (JSC::ProgramExecutable::createStructure): (JSC::FunctionExecutable::createStructure): * runtime/GetterSetter.cpp: * runtime/GetterSetter.h: (JSC::GetterSetter::createStructure): * runtime/JSAPIValueWrapper.cpp: * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): * runtime/JSCell.cpp: * runtime/JSCell.h: * runtime/JSString.cpp: * runtime/JSString.h: (JSC::RopeBuilder::createStructure): * runtime/Structure.h: (JSC::Structure::create): (JSC::JSCell::createDummyStructure): 2011-04-25 David Levin <levin@chromium.org> Reviewed by Adam Barth. PropertyMapHashTable.h should use adoptPtr instead of implicit conversions to PassRefPtr. https://bugs.webkit.org/show_bug.cgi?id=59342 This patch is to prepare for the strict OwnPtr hack-a-thon. * runtime/PropertyMapHashTable.h: (JSC::PropertyTable::copy): 2011-04-25 Thouraya ANDOLSI <thouraya.andolsi@st.com> Reviewed by Gavin Barraclough. Rationalize MacroAssembler branch methods https://bugs.webkit.org/show_bug.cgi?id=58950 split out the 'Condition' enum into 'RelationalCondition' and 'ResultCondition' and apply related changes (only for SH4 platforms). * assembler/MacroAssemblerSH4.cpp: * assembler/MacroAssemblerSH4.h: (JSC::MacroAssemblerSH4::compare32): (JSC::MacroAssemblerSH4::branch32WithUnalignedHalfWords): (JSC::MacroAssemblerSH4::branchDouble): (JSC::MacroAssemblerSH4::branch32): (JSC::MacroAssemblerSH4::branchTest8): (JSC::MacroAssemblerSH4::branch8): (JSC::MacroAssemblerSH4::branchTruncateDoubleToInt32): (JSC::MacroAssemblerSH4::test8): (JSC::MacroAssemblerSH4::branch16): (JSC::MacroAssemblerSH4::branchTest32): (JSC::MacroAssemblerSH4::branchAdd32): (JSC::MacroAssemblerSH4::branchMul32): (JSC::MacroAssemblerSH4::branchSub32): (JSC::MacroAssemblerSH4::branchOr32): (JSC::MacroAssemblerSH4::branchConvertDoubleToInt32): (JSC::MacroAssemblerSH4::branchPtrWithPatch): (JSC::MacroAssemblerSH4::SH4Condition): * assembler/SH4Assembler.h: (JSC::SH4Assembler::cmpEqImmR0): 2011-04-25 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. PropertyMapHashTable should work with strict OwnPtr https://bugs.webkit.org/show_bug.cgi?id=59337 This patch is in preparation for the strict OwnPtr hack-a-thon. * runtime/PropertyMapHashTable.h: (JSC::PropertyTable::PropertyTable): (JSC::PropertyTable::addDeletedOffset): 2011-04-25 Geoffrey Garen <ggaren@apple.com> Reviewed by Sam Weinig. Nixed MarkStack::deprecatedAppend, since it has no clients left. * heap/MarkStack.h: 2011-04-23 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. Bug 59287 - DFG JIT - Handle temporaries as vars, allowing support for ?: SetLocals to temporaries will only be generated if they are used within other blocks, due to the SSA based DCE. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::get): (JSC::DFG::ByteCodeParser::set): (JSC::DFG::ByteCodeParser::getLocal): (JSC::DFG::ByteCodeParser::setLocal): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::processPhiStack): (JSC::DFG::ByteCodeParser::allocateVirtualRegisters): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGraph.h: (JSC::DFG::BasicBlock::BasicBlock): 2011-04-22 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig & Geoff Garen. Bug 59266 - DFG JIT - Add SSA style DCE This works by making GetLocal nodes reference SetLocal nodes from prior blocks, via intermediate Phi nodes. Whenever we add a GetLocal to the graph, also add a matching child Phi, and add the Phi to a work queue to add references to prior definitions once we have the full CFG & can determine predecessors. This process is iterative, inserting new phis into predecessors as necessary. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getVariable): (JSC::DFG::ByteCodeParser::setVariable): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::setArgument): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::processWorkQueue): (JSC::DFG::ByteCodeParser::allocateVirtualRegisters): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): (JSC::DFG::Graph::refChildren): * dfg/DFGGraph.h: (JSC::DFG::Graph::ref): * dfg/DFGNode.h: (JSC::DFG::Node::ref): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGScoreBoard.h: (JSC::DFG::ScoreBoard::~ScoreBoard): (JSC::DFG::ScoreBoard::dump): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-04-22 Vitaly Repeshko <vitalyr@chromium.org> Reviewed by Adam Barth. Add missing default constructors for HashMap iterator specializations. https://bugs.webkit.org/show_bug.cgi?id=59250 * wtf/HashIterators.h: * wtf/HashTable.h: (WTF::HashTableConstIterator::HashTableConstIterator): Added cast to help compiler find the function template. 2011-04-22 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 59262 - DFG JIT - reduce size of VariableRecord We never need both the get & set node, only the most recent (which is always a set, if both exist). * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::getVariable): (JSC::DFG::ByteCodeParser::setVariable): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::setArgument): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.h: (JSC::DFG::VariableRecord::VariableRecord): 2011-04-22 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoffrey Garen. Bug 59254 - DFG JIT - retain VariableRecords for args/var in all basic blocks, such that this information is available for DCE. Also, since this enlarges the size of BasicBlock, make Graph hold a vector of pointers to basic blocks, not a vector of blocks. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::get): (JSC::DFG::ByteCodeParser::set): (JSC::DFG::ByteCodeParser::getVariable): (JSC::DFG::ByteCodeParser::setVariable): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::setArgument): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::setupPredecessors): (JSC::DFG::ByteCodeParser::allocateVirtualRegisters): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::VariableRecord::VariableRecord): (JSC::DFG::BasicBlock::BasicBlock): (JSC::DFG::BasicBlock::getBytecodeBegin): (JSC::DFG::Graph::blockIndexForBytecodeOffset): (JSC::DFG::Graph::blockForBytecodeOffset): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::detectPeepHoleBranch): 2011-04-22 Gavin Barraclough <barraclough@apple.com> Errk, build fix. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-04-22 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Quick cleanup to SpeculativeJIT/NonSpeculativeJIT compile loop, move out the call to checkConsistency(). * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::SpeculativeJIT): (JSC::DFG::SpeculativeJIT::terminateSpeculativeExecution): 2011-04-21 Vitaly Repeshko <vitalyr@chromium.org> Reviewed by Adam Barth. Provide default constructors for HashMap iterators. https://bugs.webkit.org/show_bug.cgi?id=59151 These will be used to implement an iterator over EventTarget's listeners. * wtf/HashTable.h: (WTF::HashTableConstIteratorAdapter::HashTableConstIteratorAdapter): (WTF::HashTableIteratorAdapter::HashTableIteratorAdapter): 2011-04-22 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen. Bug 59232 - DFG JIT - Add predecessor links to BasicBlocks These will be necessary for DCE support. Also factor allocateVirtualRegisters out into its own method. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::setupPredecessors): (JSC::DFG::ByteCodeParser::allocateVirtualRegisters): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGraph.h: (JSC::DFG::Graph::blockForBytecodeOffset): * dfg/DFGNode.h: (JSC::DFG::Node::isTerminal): 2011-04-22 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Object.create creates uncachable objects https://bugs.webkit.org/show_bug.cgi?id=59164 Use the prototype object's inheritorID, as we should always have done * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): (JSC::JSGlobalObject::visitChildren): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::nullPrototypeObjectStructure): * runtime/ObjectConstructor.cpp: (JSC::objectConstructorCreate): 2011-04-22 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 59222 - DFG JIT - don't allocate virtual registers to nodes with no result We currently allocate virtual registers to nodes which have no result - these are clearly unused, and may result in us allocating a larger than necessary stack frame. Encapsulate Node::virtualRegister such that we can ASSERT this is only called on nodes that have results, and improve the quality of output from the consistency check. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::Graph::ref): (JSC::DFG::Graph::deref): * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillInteger): (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): (JSC::DFG::JITCodeGenerator::dump): (JSC::DFG::JITCodeGenerator::checkConsistency): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::canReuse): (JSC::DFG::JITCodeGenerator::isFilled): (JSC::DFG::JITCodeGenerator::isFilledDouble): (JSC::DFG::JITCodeGenerator::use): (JSC::DFG::JITCodeGenerator::integerResult): (JSC::DFG::JITCodeGenerator::noResult): (JSC::DFG::JITCodeGenerator::cellResult): (JSC::DFG::JITCodeGenerator::jsValueResult): (JSC::DFG::JITCodeGenerator::doubleResult): (JSC::DFG::JITCodeGenerator::initConstantInfo): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::fillNumericToDouble): (JSC::DFG::JITCompiler::fillInt32ToInteger): (JSC::DFG::JITCompiler::fillToJS): (JSC::DFG::JITCompiler::jumpFromSpeculativeToNonSpeculative): * dfg/DFGNode.h: (JSC::DFG::Node::Node): (JSC::DFG::Node::hasResult): (JSC::DFG::Node::virtualRegister): (JSC::DFG::Node::setVirtualRegister): (JSC::DFG::Node::refCount): (JSC::DFG::Node::ref): (JSC::DFG::Node::deref): (JSC::DFG::Node::adjustedRefCount): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::isKnownInteger): (JSC::DFG::NonSpeculativeJIT::isKnownNumeric): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGScoreBoard.h: (JSC::DFG::ScoreBoard::use): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::compile): 2011-04-22 Sam Weinig <sam@webkit.org> Reviewed by Gavin Barraclough and Oliver Hunt. Arrays should participate in global object forwarding fun https://bugs.webkit.org/show_bug.cgi?id=59215 * runtime/JSGlobalObject.h: (JSC::constructEmptyArray): (JSC::constructArray): Add variants of constructArray that take a global object. 2011-04-22 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r84650 and r84654. http://trac.webkit.org/changeset/84650 http://trac.webkit.org/changeset/84654 https://bugs.webkit.org/show_bug.cgi?id=59218 Broke Windows build (Requested by bweinstein on #webkit). * API/JSCallbackObjectFunctions.h: (JSC::::init): * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * heap/Handle.h: (JSC::HandleBase::operator!): (JSC::HandleBase::operator UnspecifiedBoolType*): (JSC::HandleTypes::getFromSlot): * heap/HandleHeap.cpp: (JSC::HandleHeap::markStrongHandles): (JSC::HandleHeap::markWeakHandles): (JSC::HandleHeap::finalizeWeakHandles): (JSC::HandleHeap::writeBarrier): (JSC::HandleHeap::protectedGlobalObjectCount): (JSC::HandleHeap::isValidWeakNode): * heap/HandleHeap.h: (JSC::HandleHeap::copyWeak): (JSC::HandleHeap::makeWeak): (JSC::HandleHeap::Node::slot): * heap/HandleStack.cpp: (JSC::HandleStack::mark): (JSC::HandleStack::grow): * heap/HandleStack.h: (JSC::HandleStack::zapTo): (JSC::HandleStack::push): * heap/Heap.cpp: (JSC::HandleHeap::protectedObjectTypeCounts): * heap/Local.h: (JSC::::set): * heap/Strong.h: (JSC::Strong::set): * heap/Weak.h: (JSC::Weak::set): * runtime/StructureTransitionTable.h: (JSC::StructureTransitionTable::singleTransition): (JSC::StructureTransitionTable::setSingleTransition): * runtime/WeakGCMap.h: (JSC::WeakGCMap::add): (JSC::WeakGCMap::set): * runtime/WriteBarrier.h: 2011-04-22 Brian Weinstein <bweinstein@apple.com> Part of Windows build fix from r84650. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-04-22 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make it harder to use HandleSlot incorrectly https://bugs.webkit.org/show_bug.cgi?id=59205 Just add a little type fudging to make it harder to incorrectly assign through a HandleSlot. * API/JSCallbackObjectFunctions.h: (JSC::::init): * JavaScriptCore.exp: * heap/Handle.h: (JSC::HandleBase::operator!): (JSC::HandleBase::operator UnspecifiedBoolType*): (JSC::HandleTypes::getFromSlot): * heap/HandleHeap.cpp: (JSC::HandleHeap::markStrongHandles): (JSC::HandleHeap::markWeakHandles): (JSC::HandleHeap::finalizeWeakHandles): (JSC::HandleHeap::writeBarrier): (JSC::HandleHeap::protectedGlobalObjectCount): (JSC::HandleHeap::isValidWeakNode): * heap/HandleHeap.h: (JSC::HandleHeap::copyWeak): (JSC::HandleHeap::makeWeak): (JSC::HandleHeap::Node::slot): * heap/HandleStack.cpp: (JSC::HandleStack::mark): (JSC::HandleStack::grow): * heap/HandleStack.h: (JSC::HandleStack::zapTo): (JSC::HandleStack::push): * heap/Heap.cpp: (JSC::HandleHeap::protectedObjectTypeCounts): * heap/Local.h: (JSC::::set): * heap/Strong.h: (JSC::Strong::set): * heap/Weak.h: (JSC::Weak::set): * runtime/StructureTransitionTable.h: (JSC::StructureTransitionTable::singleTransition): (JSC::StructureTransitionTable::setSingleTransition): * runtime/WeakGCMap.h: (JSC::WeakGCMap::add): (JSC::WeakGCMap::set): * runtime/WriteBarrier.h: (JSC::OpaqueJSValue::toJSValue): (JSC::OpaqueJSValue::toJSValueRef): (JSC::OpaqueJSValue::fromJSValue): 2011-04-22 Patrick Gansterer <paroga@webkit.org> Unreviewed. Build fix for ENABLE(INTERPRETER) after r84556. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitAggregate): 2011-04-21 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r84583. http://trac.webkit.org/changeset/84583 https://bugs.webkit.org/show_bug.cgi?id=59173 "broke http://trac.webkit.org/export/84593/trunk/LayoutTests/fast/js /Object-create.html" (Requested by ggaren on #webkit). * runtime/ObjectConstructor.cpp: (JSC::objectConstructorCreate): 2011-04-21 Maciej Stachowiak <mjs@apple.com> Reviewed by Adam Roben. Add a feature define to allow <details> and <summary> to be disabled https://bugs.webkit.org/show_bug.cgi?id=59118 <rdar://problem/9257045> * Configurations/FeatureDefines.xcconfig: 2011-04-21 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Object.create creates uncachable objects https://bugs.webkit.org/show_bug.cgi?id=59164 Use the prototype object's inheritorID, as we should always have done * runtime/ObjectConstructor.cpp: (JSC::objectConstructorCreate): 2011-04-21 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Start moving to a general visitor pattern for GC traversal https://bugs.webkit.org/show_bug.cgi?id=59141 This is just a rename: markChildren -> visitChildren markAggregate -> visitAggregate markStack -> visitor MarkStack -> typedef'd to SlotVisitor * API/JSCallbackObject.h: (JSC::JSCallbackObjectData::visitChildren): (JSC::JSCallbackObjectData::JSPrivatePropertyMap::visitChildren): (JSC::JSCallbackObject::visitChildren): * JavaScriptCore.exp: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::visitStructures): (JSC::EvalCodeCache::visitAggregate): (JSC::CodeBlock::visitAggregate): * bytecode/CodeBlock.h: * bytecode/EvalCodeCache.h: * bytecode/Instruction.h: (JSC::PolymorphicAccessStructureList::visitAggregate): * bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::visitAggregate): * bytecode/StructureStubInfo.h: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::visitChildren): * debugger/DebuggerActivation.h: * heap/HandleHeap.cpp: (JSC::WeakHandleOwner::isReachableFromOpaqueRoots): (JSC::HandleHeap::markStrongHandles): (JSC::HandleHeap::markWeakHandles): * heap/HandleHeap.h: * heap/HandleStack.cpp: (JSC::HandleStack::mark): * heap/HandleStack.h: * heap/Heap.cpp: (JSC::Heap::markProtectedObjects): (JSC::Heap::markTempSortVectors): (JSC::Heap::markRoots): * heap/Heap.h: * heap/MarkStack.cpp: (JSC::MarkStack::visitChildren): (JSC::MarkStack::drain): * heap/MarkStack.h: (JSC::HeapRootVisitor::HeapRootVisitor): (JSC::HeapRootVisitor::mark): (JSC::HeapRootVisitor::visitor): * heap/MarkedSpace.h: * runtime/ArgList.cpp: (JSC::MarkedArgumentBuffer::markLists): * runtime/ArgList.h: * runtime/Arguments.cpp: (JSC::Arguments::visitChildren): * runtime/Arguments.h: * runtime/Executable.cpp: (JSC::EvalExecutable::visitChildren): (JSC::ProgramExecutable::visitChildren): (JSC::FunctionExecutable::visitChildren): * runtime/Executable.h: * runtime/GetterSetter.cpp: (JSC::GetterSetter::visitChildren): * runtime/GetterSetter.h: (JSC::GetterSetter::createStructure): * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): * runtime/JSActivation.cpp: (JSC::JSActivation::visitChildren): * runtime/JSActivation.h: * runtime/JSArray.cpp: (JSC::JSArray::visitChildren): * runtime/JSArray.h: (JSC::JSArray::visitDirect): * runtime/JSCell.h: (JSC::JSCell::JSCell::visitChildren): * runtime/JSFunction.cpp: (JSC::JSFunction::visitChildren): * runtime/JSFunction.h: * runtime/JSGlobalObject.cpp: (JSC::visitIfNeeded): (JSC::JSGlobalObject::visitChildren): * runtime/JSGlobalObject.h: * runtime/JSONObject.cpp: * runtime/JSObject.cpp: (JSC::JSObject::visitChildren): * runtime/JSObject.h: (JSC::JSObject::visitDirect): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::visitChildren): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::createStructure): * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::visitChildren): * runtime/JSStaticScopeObject.h: * runtime/JSTypeInfo.h: (JSC::TypeInfo::TypeInfo): (JSC::TypeInfo::overridesVisitChildren): * runtime/JSWrapperObject.cpp: (JSC::JSWrapperObject::visitChildren): * runtime/JSWrapperObject.h: * runtime/JSZombie.h: (JSC::JSZombie::visitChildren): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::visitChildren): * runtime/NativeErrorConstructor.h: * runtime/RegExpObject.cpp: (JSC::RegExpObject::visitChildren): * runtime/RegExpObject.h: * runtime/ScopeChain.cpp: (JSC::ScopeChainNode::visitChildren): * runtime/ScopeChain.h: * runtime/SmallStrings.cpp: (JSC::SmallStrings::visitChildren): * runtime/SmallStrings.h: * runtime/Structure.cpp: (JSC::Structure::Structure): (JSC::Structure::visitChildren): * runtime/Structure.h: * runtime/StructureChain.cpp: (JSC::StructureChain::visitChildren): * runtime/StructureChain.h: (JSC::StructureChain::createStructure): 2011-04-21 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r84548. http://trac.webkit.org/changeset/84548 https://bugs.webkit.org/show_bug.cgi?id=59144 Broke chromium-win build (Requested by aklein on #webkit). * wtf/Platform.h: 2011-04-21 Adam Klein <adamk@chromium.org> Reviewed by David Levin. [fileapi] Worker File API calls that create Blobs fail in debug builds due to random number generator thread assertion https://bugs.webkit.org/show_bug.cgi?id=55728 Enable WTF_MULTIPLE_THREADS for Chromium. * wtf/Platform.h: 2011-04-20 Michael Saboff <msaboff@apple.com> Reviewed by Geoff Garen. JSString::resolveRope inefficient for common 2 fiber case https://bugs.webkit.org/show_bug.cgi?id=58994 Split JSString::resolveRope into three routines. resolveRope allocates the new buffer and handles the 1 or 2 fiber case with single level fibers. resolveRopeSlowCase handles the general case. outOfMemory handles the rare out of memory exception case. * runtime/JSString.cpp: (JSC::JSString::resolveRope): (JSC::JSString::resolveRopeSlowCase): (JSC::JSString::outOfMemory): * runtime/JSString.h: 2011-04-20 Adam Klein <adamk@chromium.org> Reviewed by David Levin. Rename all uses of JSC_MULTIPLE_THREADS under wtf/... to WTF_MULTIPLE_THREADS https://bugs.webkit.org/show_bug.cgi?id=59040 This will be used to fix https://bugs.webkit.org/show_bug.cgi?id=55728 by enabling WTF_MULTIPLE_THREADS for Chromium. * wtf/CryptographicallyRandomNumber.cpp: (WTF::ARC4Stream::ARC4RandomNumberGenerator::randomNumber): (WTF::ARC4Stream::ARC4RandomNumberGenerator::randomValues): * wtf/FastMalloc.cpp: * wtf/Platform.h: Enable WTF_MULTIPLE_THREADS whenever JSC_MULTIPLE_THREADS is enabled. * wtf/RandomNumber.cpp: (WTF::randomNumber): * wtf/RefCountedLeakCounter.cpp: (WTF::RefCountedLeakCounter::increment): (WTF::RefCountedLeakCounter::decrement): * wtf/dtoa.cpp: (WTF::pow5mult): 2011-04-20 Gavin Barraclough <barraclough@apple.com> Rubber stamped by Geoff Garen Bug 59069 - DFG JIT - register allocate r8, r9, r10 * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::gprToRegisterID): 2011-04-20 Gavin Barraclough <barraclough@apple.com> Build fix - revert accidental change. * wtf/Platform.h: 2011-04-20 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Add SAMPLING_FLAGS tool to DFG JIT. * bytecode/SamplingTool.h: (JSC::SamplingFlags::addressOfFlags): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::setSamplingFlag): (JSC::DFG::JITCompiler::clearSamplingFlag): * dfg/DFGJITCompiler.h: * jit/JITInlineMethods.h: (JSC::JIT::setSamplingFlag): (JSC::JIT::clearSamplingFlag): * wtf/Platform.h: 2011-04-20 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. Bug 59022 - DFG JIT - Optimize branch-on-relational-compare If a relational compare (< or <=) is immediately followed by a branch, we can combine the two, avoiding generation of a boolean into a register. * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::branch32): (JSC::MacroAssemblerX86Common::invert): (JSC::MacroAssemblerX86Common::commute): * dfg/DFGNode.h: (JSC::DFG::Node::adjustedRefCount): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compilePeepHoleBranch): (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isJSConstantWithInt32Value): (JSC::DFG::SpeculativeJIT::detectPeepHoleBranch): 2011-04-20 Gavin Barraclough <barraclough@apple.com> ARMv7 build fix II. * jit/JITOpcodes32_64.cpp: (JSC::JIT::softModulo): 2011-04-20 Gavin Barraclough <barraclough@apple.com> ARMv7 build fix. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::test8): 2011-04-19 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. Rationalize MacroAssembler branch methods https://bugs.webkit.org/show_bug.cgi?id=58950 The MacroAssembler currently exposes x86's weird behaviour that the 'setcc' instruction only sets the low 8 bits of a register. Stop that. Having done so, to clarify remove the 'set32' prefix from test & compare instructions - these methods all now set a full 32/64 bit register (Ptr size). The size in the function name should indicate the amount of data being compared. Also split out the 'Condition' enum into 'RelationalCondition' and 'ResultCondition'. The former is used in binary comparison, the latter is a unary condition check on the result of an operation. * JavaScriptCore.xcodeproj/project.pbxproj: * assembler/MacroAssembler.h: (JSC::MacroAssembler::branchPtr): (JSC::MacroAssembler::branch32): (JSC::MacroAssembler::branch16): (JSC::MacroAssembler::branchTestPtr): (JSC::MacroAssembler::comparePtr): (JSC::MacroAssembler::branchAddPtr): (JSC::MacroAssembler::branchSubPtr): (JSC::MacroAssembler::branchTest8): * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::branch8): (JSC::MacroAssemblerARM::branch32): (JSC::MacroAssemblerARM::branch32WithUnalignedHalfWords): (JSC::MacroAssemblerARM::branch16): (JSC::MacroAssemblerARM::branchTest8): (JSC::MacroAssemblerARM::branchTest32): (JSC::MacroAssemblerARM::branchAdd32): (JSC::MacroAssemblerARM::branchMul32): (JSC::MacroAssemblerARM::branchSub32): (JSC::MacroAssemblerARM::branchNeg32): (JSC::MacroAssemblerARM::branchOr32): (JSC::MacroAssemblerARM::compare32): (JSC::MacroAssemblerARM::test32): (JSC::MacroAssemblerARM::test8): (JSC::MacroAssemblerARM::branchPtrWithPatch): (JSC::MacroAssemblerARM::ARMCondition): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::branch32): (JSC::MacroAssemblerARMv7::branch32WithUnalignedHalfWords): (JSC::MacroAssemblerARMv7::branch16): (JSC::MacroAssemblerARMv7::branch8): (JSC::MacroAssemblerARMv7::branchTest32): (JSC::MacroAssemblerARMv7::branchTest8): (JSC::MacroAssemblerARMv7::branchAdd32): (JSC::MacroAssemblerARMv7::branchMul32): (JSC::MacroAssemblerARMv7::branchOr32): (JSC::MacroAssemblerARMv7::branchSub32): (JSC::MacroAssemblerARMv7::compare32): (JSC::MacroAssemblerARMv7::test32): (JSC::MacroAssemblerARMv7::test8): (JSC::MacroAssemblerARMv7::branchPtrWithPatch): (JSC::MacroAssemblerARMv7::makeBranch): (JSC::MacroAssemblerARMv7::armV7Condition): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::branch8): (JSC::MacroAssemblerMIPS::branch32): (JSC::MacroAssemblerMIPS::branch32WithUnalignedHalfWords): (JSC::MacroAssemblerMIPS::branch16): (JSC::MacroAssemblerMIPS::branchTest32): (JSC::MacroAssemblerMIPS::branchTest8): (JSC::MacroAssemblerMIPS::branchAdd32): (JSC::MacroAssemblerMIPS::branchMul32): (JSC::MacroAssemblerMIPS::branchSub32): (JSC::MacroAssemblerMIPS::branchOr32): (JSC::MacroAssemblerMIPS::compare32): (JSC::MacroAssemblerMIPS::test8): (JSC::MacroAssemblerMIPS::test32): (JSC::MacroAssemblerMIPS::branchPtrWithPatch): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::branch32): (JSC::MacroAssemblerX86::branchPtrWithPatch): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::branch8): (JSC::MacroAssemblerX86Common::branch32): (JSC::MacroAssemblerX86Common::branch32WithUnalignedHalfWords): (JSC::MacroAssemblerX86Common::branch16): (JSC::MacroAssemblerX86Common::branchTest32): (JSC::MacroAssemblerX86Common::branchTest8): (JSC::MacroAssemblerX86Common::branchAdd32): (JSC::MacroAssemblerX86Common::branchMul32): (JSC::MacroAssemblerX86Common::branchSub32): (JSC::MacroAssemblerX86Common::branchNeg32): (JSC::MacroAssemblerX86Common::branchOr32): (JSC::MacroAssemblerX86Common::compare32): (JSC::MacroAssemblerX86Common::test8): (JSC::MacroAssemblerX86Common::test32): (JSC::MacroAssemblerX86Common::x86Condition): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::comparePtr): (JSC::MacroAssemblerX86_64::branchPtr): (JSC::MacroAssemblerX86_64::branchTestPtr): (JSC::MacroAssemblerX86_64::branchAddPtr): (JSC::MacroAssemblerX86_64::branchSubPtr): (JSC::MacroAssemblerX86_64::branchPtrWithPatch): (JSC::MacroAssemblerX86_64::branchTest8): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_eq): (JSC::JIT::emit_op_neq): (JSC::JIT::compileOpStrictEq): (JSC::JIT::emit_op_eq_null): (JSC::JIT::emit_op_neq_null): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_eq): (JSC::JIT::emit_op_neq): (JSC::JIT::compileOpStrictEq): (JSC::JIT::emit_op_eq_null): (JSC::JIT::emit_op_neq_null): 2011-04-20 Balazs Kelemen <kbalazs@webkit.org> Reviewed by Csaba Osztrogonác. [Qt] Cleanup includepath adjustment for generated files https://bugs.webkit.org/show_bug.cgi?id=58869 * JavaScriptCore.pri: Add the directory of generated files to the include path with absolute path to make it valid in the final build step. 2011-04-19 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Remove unneeded deprecated methods from MarkStack https://bugs.webkit.org/show_bug.cgi?id=58853 Remove deprecated methods * heap/MarkStack.h: 2011-04-19 Mark Rowe <mrowe@apple.com> Things work best when the Xcode project refers to the file at a path that exists. * JavaScriptCore.xcodeproj/project.pbxproj: 2011-04-19 Renata Hodovan <reni@webkit.org> Reviewed by Eric Seidel. Move the alignment related macros in Vector.h to new Alignment.h. https://bugs.webkit.org/show_bug.cgi?id=56000 * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/Alignment.h: Added. * wtf/CMakeLists.txt: * wtf/Vector.h: 2011-04-19 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Remove DeprecatedPtr https://bugs.webkit.org/show_bug.cgi?id=58718 Remove the class an all functions that still exist to support it. * heap/MarkStack.h: (JSC::MarkStack::append): * runtime/JSValue.h: * runtime/WriteBarrier.h: 2011-04-19 Jungshik Shin <jshin@chromium.org> Reviewed by David Levin Add U+FEFF (Zero width no-break space) to CharacterNames.h. It's added to the list of characters to treat as zero-width in WebCore. https://bugs.webkit.org/show_bug.cgi?id=48860 * wtf/unicode/CharacterNames.h: 2011-04-19 Csaba Osztrogonác <ossy@webkit.org> [Qt] REGRESSION(84176): http/tests/xmlhttprequest/event-listener-gc.html fails https://bugs.webkit.org/show_bug.cgi?id=58871 Unreviewed, rolling out r84176, r84178, r84186, r84212 and r84231. http://trac.webkit.org/changeset/84176 (original patch) http://trac.webkit.org/changeset/84178 (original patch - part 2) http://trac.webkit.org/changeset/84186 (build fix) http://trac.webkit.org/changeset/84212 http://trac.webkit.org/changeset/84231 (skip failing test) original bugs: - https://bugs.webkit.org/show_bug.cgi?id=58718 - https://bugs.webkit.org/show_bug.cgi?id=58853 * heap/MarkStack.h: (JSC::MarkStack::deprecatedAppendValues): (JSC::MarkStack::append): (JSC::MarkStack::deprecatedAppend): * runtime/JSValue.h: * 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::operator==): 2011-04-18 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Remove unneeded deprecated methods from MarkStack https://bugs.webkit.org/show_bug.cgi?id=58853 Remove deprecated methods * heap/MarkStack.h: 2011-04-18 Oliver Hunt <oliver@apple.com> Reviewed by Adam Roben. Off by one initialising repeat callframe https://bugs.webkit.org/show_bug.cgi?id=58838 <rdar://problem/8756810> If the end of a callframe made for a repeat call landed on a page boundary the following page may not have been committed which means that the off by one could lead to a crash. However it could only happen in this case and only on windows which is why it was so hard to repro. Alas given the steps needed to reproduce are such that it's not really possible to make a testcase. This fix makes the code a little less squirrely by not trying to avoid the unnecessary initialisation of |this|. * interpreter/Interpreter.cpp: (JSC::Interpreter::prepareForRepeatCall): 2011-04-18 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen. Bug 58829 - DFG JIT - Optimize add/sub immediate, multiply. Add code generation for add/subtract instruction with immediate operands (where a child is a constant), and don't bail to non-speculative if an integer multiple results in a +0 result (only if it should be generating -0). * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculativeJIT::isDoubleConstantWithInt32Value): 2011-04-18 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen. Bug 58817 - DFG JIT - if speculative compilation fails, throw away code. If we detect a logical conflict, throw away generated code, and only compile through the NonSpeculativeJIT. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::rewindToLabel): * assembler/AssemblerBuffer.h: (JSC::AssemblerBuffer::rewindToOffset): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::branchAdd32): (JSC::MacroAssemblerX86Common::branchSub32): * assembler/X86Assembler.h: (JSC::X86Assembler::rewindToLabel): (JSC::X86Assembler::X86InstructionFormatter::rewindToLabel): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::linkSpeculationChecks): (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: (JSC::DFG::SpeculationCheckIndexIterator::SpeculationCheckIndexIterator): 2011-04-18 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Remove DeprecatedPtr https://bugs.webkit.org/show_bug.cgi?id=58718 As simple as it sounds. * runtime/JSValue.h: * runtime/WriteBarrier.h: 2011-04-17 Cameron Zwarich <zwarich@apple.com> Reviewed by Dan Bernstein. JSC no longer builds with Clang due to -Woverloaded-virtual warning https://bugs.webkit.org/show_bug.cgi?id=58760 Rename Structure's specificValue overload of put to putSpecificValue to avoid Clang's warning for overloading a virtual function. * runtime/Structure.cpp: (JSC::Structure::addPropertyTransition): (JSC::Structure::addPropertyWithoutTransition): (JSC::Structure::putSpecificValue): * runtime/Structure.h: 2011-04-17 Patrick Gansterer <paroga@webkit.org> Reviewed by Adam Barth. Remove WTF_PLATFORM_SGL https://bugs.webkit.org/show_bug.cgi?id=58743 WTF_PLATFORM_SGL and PLATFORM(SGL) are not used in the code anywhere. * wtf/Platform.h: 2011-04-17 Patrick Gansterer <paroga@webkit.org> Reviewed by Adam Barth. Rename PLATFORM(CA) to USE(CA) https://bugs.webkit.org/show_bug.cgi?id=58742 * wtf/Platform.h: 2011-04-17 Patrick Gansterer <paroga@webkit.org> Reviewed by Adam Barth. Rename PLATFORM(CG) to USE(CG) https://bugs.webkit.org/show_bug.cgi?id=58729 * wtf/Platform.h: 2011-04-16 Patrick Gansterer <paroga@webkit.org> Reviewed by Eric Seidel. Rename PLATFORM(CAIRO) to USE(CAIRO) https://bugs.webkit.org/show_bug.cgi?id=55192 * wtf/Platform.h: * wtf/gobject/GTypedefs.h: 2011-04-15 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r84067. http://trac.webkit.org/changeset/84067 https://bugs.webkit.org/show_bug.cgi?id=58724 qt build are failing. (Requested by loislo2 on #webkit). * heap/MarkStack.h: (JSC::MarkStack::append): * runtime/JSValue.h: * 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::operator==): 2011-04-15 Shishir Agrawal <shishir@chromium.org> Reviewed by James Robinson. Add a flag to guard Page Visibility API changes. https://bugs.webkit.org/show_bug.cgi?id=58464 * Configurations/FeatureDefines.xcconfig: 2011-04-15 Gavin Barraclough <barraclough@apple.com> Errrk! - build fix from !x86-64. * dfg/DFGNode.h: 2011-04-15 David Levin <levin@chromium.org> Revert of r83974. JavaScriptCore shouldn't depend on ../ThirdParty/gtest/xcode/gtest.xcodeproj https://bugs.webkit.org/show_bug.cgi?id=58716 * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/tests/RunAllWtfTests.cpp: Removed. * wtf/tests/StringTests.cpp: Removed. 2011-04-15 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Remove DeprecatedPtr https://bugs.webkit.org/show_bug.cgi?id=58718 As simple as it sounds. * heap/MarkStack.h: (JSC::MarkStack::append): * runtime/JSValue.h: * runtime/WriteBarrier.h: 2011-04-15 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. Add a simple tool to gather statistics on whether functions are completed through the new or old JIT. * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JIT.cpp: (JSC::JIT::privateCompile): 2011-04-15 Oliver Hunt <oliver@apple.com> GC allocate Structure https://bugs.webkit.org/show_bug.cgi?id=58483 Rolling r83894 r83827 r83810 r83809 r83808 back in with a workaround for the gcc bug seen by the gtk bots * API/JSCallbackConstructor.cpp: (JSC::JSCallbackConstructor::JSCallbackConstructor): * API/JSCallbackConstructor.h: (JSC::JSCallbackConstructor::createStructure): * API/JSCallbackFunction.h: (JSC::JSCallbackFunction::createStructure): * API/JSCallbackObject.h: (JSC::JSCallbackObject::createStructure): * API/JSCallbackObjectFunctions.h: (JSC::::JSCallbackObject): * API/JSContextRef.cpp: * JavaScriptCore.JSVALUE32_64only.exp: * JavaScriptCore.JSVALUE64only.exp: * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::~CodeBlock): (JSC::CodeBlock::markStructures): (JSC::CodeBlock::markAggregate): * bytecode/CodeBlock.h: (JSC::MethodCallLinkInfo::setSeen): (JSC::GlobalResolveInfo::GlobalResolveInfo): * bytecode/Instruction.h: (JSC::PolymorphicAccessStructureList::PolymorphicStubInfo::set): (JSC::PolymorphicAccessStructureList::PolymorphicAccessStructureList): (JSC::PolymorphicAccessStructureList::markAggregate): (JSC::Instruction::Instruction): * bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::deref): (JSC::StructureStubInfo::markAggregate): * bytecode/StructureStubInfo.h: (JSC::StructureStubInfo::initGetByIdSelf): (JSC::StructureStubInfo::initGetByIdProto): (JSC::StructureStubInfo::initGetByIdChain): (JSC::StructureStubInfo::initPutByIdTransition): (JSC::StructureStubInfo::initPutByIdReplace): * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::DebuggerActivation): * debugger/DebuggerActivation.h: (JSC::DebuggerActivation::createStructure): * heap/Handle.h: * heap/MarkStack.cpp: (JSC::MarkStack::markChildren): (JSC::MarkStack::drain): * heap/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): (JSC::MarkedBlock::sweep): * heap/Strong.h: (JSC::Strong::Strong): (JSC::Strong::set): * interpreter/Interpreter.cpp: (JSC::Interpreter::resolveGlobal): (JSC::Interpreter::resolveGlobalDynamic): (JSC::Interpreter::tryCachePutByID): (JSC::Interpreter::uncachePutByID): (JSC::Interpreter::tryCacheGetByID): (JSC::Interpreter::uncacheGetByID): (JSC::Interpreter::privateExecute): * jit/JIT.h: * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::patchMethodCallProto): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::patchMethodCallProto): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): * jit/JITStubs.cpp: (JSC::JITThunks::tryCachePutByID): (JSC::JITThunks::tryCacheGetByID): (JSC::DEFINE_STUB_FUNCTION): (JSC::getPolymorphicAccessStructureListSlot): * jit/JSInterfaceJIT.h: (JSC::JSInterfaceJIT::storePtrWithWriteBarrier): * jsc.cpp: (cleanupGlobalData): * runtime/Arguments.h: (JSC::Arguments::createStructure): (JSC::Arguments::Arguments): (JSC::JSActivation::copyRegisters): * runtime/ArrayConstructor.cpp: (JSC::ArrayConstructor::ArrayConstructor): (JSC::constructArrayWithSizeQuirk): * runtime/ArrayConstructor.h: * runtime/ArrayPrototype.cpp: (JSC::ArrayPrototype::ArrayPrototype): (JSC::arrayProtoFuncSplice): * runtime/ArrayPrototype.h: (JSC::ArrayPrototype::createStructure): * runtime/BatchedTransitionOptimizer.h: (JSC::BatchedTransitionOptimizer::BatchedTransitionOptimizer): * runtime/BooleanConstructor.cpp: (JSC::BooleanConstructor::BooleanConstructor): * runtime/BooleanConstructor.h: * runtime/BooleanObject.cpp: (JSC::BooleanObject::BooleanObject): * runtime/BooleanObject.h: (JSC::BooleanObject::createStructure): * runtime/BooleanPrototype.cpp: (JSC::BooleanPrototype::BooleanPrototype): * runtime/BooleanPrototype.h: * runtime/DateConstructor.cpp: (JSC::DateConstructor::DateConstructor): * runtime/DateConstructor.h: * runtime/DateInstance.cpp: (JSC::DateInstance::DateInstance): * runtime/DateInstance.h: (JSC::DateInstance::createStructure): * runtime/DatePrototype.cpp: (JSC::DatePrototype::DatePrototype): * runtime/DatePrototype.h: (JSC::DatePrototype::createStructure): * runtime/Error.cpp: (JSC::StrictModeTypeErrorFunction::StrictModeTypeErrorFunction): * runtime/ErrorConstructor.cpp: (JSC::ErrorConstructor::ErrorConstructor): * runtime/ErrorConstructor.h: * runtime/ErrorInstance.cpp: (JSC::ErrorInstance::ErrorInstance): (JSC::ErrorInstance::create): * runtime/ErrorInstance.h: (JSC::ErrorInstance::createStructure): * runtime/ErrorPrototype.cpp: (JSC::ErrorPrototype::ErrorPrototype): * runtime/ErrorPrototype.h: * runtime/ExceptionHelpers.cpp: (JSC::InterruptedExecutionError::InterruptedExecutionError): (JSC::TerminatedExecutionError::TerminatedExecutionError): * runtime/Executable.cpp: * runtime/Executable.h: (JSC::ExecutableBase::ExecutableBase): (JSC::ExecutableBase::createStructure): (JSC::NativeExecutable::createStructure): (JSC::NativeExecutable::NativeExecutable): (JSC::ScriptExecutable::ScriptExecutable): (JSC::EvalExecutable::createStructure): (JSC::ProgramExecutable::createStructure): (JSC::FunctionExecutable::createStructure): * runtime/FunctionConstructor.cpp: (JSC::FunctionConstructor::FunctionConstructor): * runtime/FunctionConstructor.h: * runtime/FunctionPrototype.cpp: (JSC::FunctionPrototype::FunctionPrototype): * runtime/FunctionPrototype.h: (JSC::FunctionPrototype::createStructure): * runtime/GetterSetter.h: (JSC::GetterSetter::GetterSetter): (JSC::GetterSetter::createStructure): * runtime/InitializeThreading.cpp: (JSC::initializeThreadingOnce): * runtime/InternalFunction.cpp: (JSC::InternalFunction::InternalFunction): * runtime/InternalFunction.h: (JSC::InternalFunction::createStructure): * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): (JSC::JSAPIValueWrapper::JSAPIValueWrapper): * runtime/JSActivation.cpp: (JSC::JSActivation::JSActivation): * runtime/JSActivation.h: (JSC::JSActivation::createStructure): * runtime/JSArray.cpp: (JSC::JSArray::JSArray): * runtime/JSArray.h: (JSC::JSArray::createStructure): * runtime/JSByteArray.cpp: (JSC::JSByteArray::JSByteArray): (JSC::JSByteArray::createStructure): * runtime/JSByteArray.h: (JSC::JSByteArray::JSByteArray): * runtime/JSCell.cpp: (JSC::isZombie): * runtime/JSCell.h: (JSC::JSCell::JSCell::JSCell): (JSC::JSCell::JSCell::addressOfStructure): (JSC::JSCell::JSCell::structure): (JSC::JSCell::JSCell::markChildren): (JSC::JSCell::JSValue::isZombie): * runtime/JSFunction.cpp: (JSC::JSFunction::JSFunction): * runtime/JSFunction.h: (JSC::JSFunction::createStructure): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::storeVPtrs): (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::clearBuiltinStructures): (JSC::JSGlobalData::createLeaked): * runtime/JSGlobalData.h: (JSC::allocateGlobalHandle): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): (JSC::JSGlobalObject::markChildren): (JSC::JSGlobalObject::copyGlobalsFrom): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::JSGlobalObject): (JSC::JSGlobalObject::createStructure): (JSC::Structure::prototypeChain): (JSC::Structure::isValid): (JSC::constructEmptyArray): * runtime/JSNotAnObject.h: (JSC::JSNotAnObject::JSNotAnObject): (JSC::JSNotAnObject::createStructure): * runtime/JSONObject.cpp: (JSC::JSONObject::JSONObject): * runtime/JSONObject.h: (JSC::JSONObject::createStructure): * runtime/JSObject.cpp: (JSC::JSObject::defineGetter): (JSC::JSObject::defineSetter): (JSC::JSObject::seal): (JSC::JSObject::freeze): (JSC::JSObject::preventExtensions): (JSC::JSObject::removeDirect): (JSC::JSObject::createInheritorID): * runtime/JSObject.h: (JSC::JSObject::createStructure): (JSC::JSObject::JSObject): (JSC::JSNonFinalObject::createStructure): (JSC::JSNonFinalObject::JSNonFinalObject): (JSC::JSFinalObject::create): (JSC::JSFinalObject::createStructure): (JSC::JSFinalObject::JSFinalObject): (JSC::constructEmptyObject): (JSC::createEmptyObjectStructure): (JSC::JSObject::~JSObject): (JSC::JSObject::setPrototype): (JSC::JSObject::setStructure): (JSC::JSObject::inheritorID): (JSC::JSObject::putDirectInternal): (JSC::JSObject::transitionTo): (JSC::JSObject::markChildrenDirect): * runtime/JSObjectWithGlobalObject.cpp: (JSC::JSObjectWithGlobalObject::JSObjectWithGlobalObject): * runtime/JSObjectWithGlobalObject.h: (JSC::JSObjectWithGlobalObject::createStructure): (JSC::JSObjectWithGlobalObject::JSObjectWithGlobalObject): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::JSPropertyNameIterator): (JSC::JSPropertyNameIterator::create): (JSC::JSPropertyNameIterator::get): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::createStructure): (JSC::JSPropertyNameIterator::setCachedStructure): (JSC::Structure::setEnumerationCache): * runtime/JSStaticScopeObject.h: (JSC::JSStaticScopeObject::JSStaticScopeObject): (JSC::JSStaticScopeObject::createStructure): * runtime/JSString.h: (JSC::RopeBuilder::JSString): (JSC::RopeBuilder::createStructure): * runtime/JSType.h: * runtime/JSTypeInfo.h: (JSC::TypeInfo::TypeInfo): * runtime/JSValue.h: * runtime/JSVariableObject.h: (JSC::JSVariableObject::createStructure): (JSC::JSVariableObject::JSVariableObject): (JSC::JSVariableObject::copyRegisterArray): * runtime/JSWrapperObject.h: (JSC::JSWrapperObject::createStructure): (JSC::JSWrapperObject::JSWrapperObject): * runtime/JSZombie.cpp: * runtime/JSZombie.h: (JSC::JSZombie::JSZombie): (JSC::JSZombie::createStructure): * runtime/MathObject.cpp: (JSC::MathObject::MathObject): * runtime/MathObject.h: (JSC::MathObject::createStructure): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::NativeErrorConstructor): (JSC::NativeErrorConstructor::markChildren): (JSC::constructWithNativeErrorConstructor): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::createStructure): * runtime/NativeErrorPrototype.cpp: (JSC::NativeErrorPrototype::NativeErrorPrototype): * runtime/NativeErrorPrototype.h: * runtime/NumberConstructor.cpp: (JSC::NumberConstructor::NumberConstructor): * runtime/NumberConstructor.h: (JSC::NumberConstructor::createStructure): * runtime/NumberObject.cpp: (JSC::NumberObject::NumberObject): * runtime/NumberObject.h: (JSC::NumberObject::createStructure): * runtime/NumberPrototype.cpp: (JSC::NumberPrototype::NumberPrototype): * runtime/NumberPrototype.h: * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::ObjectConstructor): * runtime/ObjectConstructor.h: (JSC::ObjectConstructor::createStructure): * runtime/ObjectPrototype.cpp: (JSC::ObjectPrototype::ObjectPrototype): * runtime/ObjectPrototype.h: * runtime/PropertyMapHashTable.h: (JSC::PropertyTable::PropertyTable): * runtime/RegExpConstructor.cpp: (JSC::RegExpConstructor::RegExpConstructor): (JSC::RegExpMatchesArray::RegExpMatchesArray): * runtime/RegExpConstructor.h: (JSC::RegExpConstructor::createStructure): * runtime/RegExpObject.cpp: (JSC::RegExpObject::RegExpObject): * runtime/RegExpObject.h: (JSC::RegExpObject::createStructure): * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::RegExpPrototype): * runtime/RegExpPrototype.h: * runtime/ScopeChain.h: (JSC::ScopeChainNode::ScopeChainNode): (JSC::ScopeChainNode::createStructure): * runtime/StrictEvalActivation.cpp: (JSC::StrictEvalActivation::StrictEvalActivation): * runtime/StringConstructor.cpp: (JSC::StringConstructor::StringConstructor): * runtime/StringConstructor.h: * runtime/StringObject.cpp: (JSC::StringObject::StringObject): * runtime/StringObject.h: (JSC::StringObject::createStructure): * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::StringObjectThatMasqueradesAsUndefined): (JSC::StringObjectThatMasqueradesAsUndefined::createStructure): * runtime/StringPrototype.cpp: (JSC::StringPrototype::StringPrototype): * runtime/StringPrototype.h: (JSC::StringPrototype::createStructure): * runtime/Structure.cpp: (JSC::StructureTransitionTable::remove): (JSC::StructureTransitionTable::add): (JSC::Structure::Structure): (JSC::Structure::~Structure): (JSC::Structure::materializePropertyMap): (JSC::Structure::addPropertyTransitionToExistingStructure): (JSC::Structure::addPropertyTransition): (JSC::Structure::removePropertyTransition): (JSC::Structure::changePrototypeTransition): (JSC::Structure::despecifyFunctionTransition): (JSC::Structure::getterSetterTransition): (JSC::Structure::toDictionaryTransition): (JSC::Structure::toCacheableDictionaryTransition): (JSC::Structure::toUncacheableDictionaryTransition): (JSC::Structure::sealTransition): (JSC::Structure::freezeTransition): (JSC::Structure::preventExtensionsTransition): (JSC::Structure::flattenDictionaryStructure): (JSC::Structure::copyPropertyTable): (JSC::Structure::put): (JSC::Structure::markChildren): * runtime/Structure.h: (JSC::Structure::create): (JSC::Structure::setPrototypeWithoutTransition): (JSC::Structure::createStructure): (JSC::JSCell::createDummyStructure): (JSC::StructureTransitionTable::keyForWeakGCMapFinalizer): * runtime/StructureChain.cpp: (JSC::StructureChain::StructureChain): (JSC::StructureChain::markChildren): * runtime/StructureChain.h: (JSC::StructureChain::create): (JSC::StructureChain::head): (JSC::StructureChain::createStructure): * runtime/StructureTransitionTable.h: (JSC::StructureTransitionTable::WeakGCMapFinalizerCallback::finalizerContextFor): (JSC::StructureTransitionTable::WeakGCMapFinalizerCallback::keyForFinalizer): (JSC::StructureTransitionTable::~StructureTransitionTable): (JSC::StructureTransitionTable::slot): (JSC::StructureTransitionTable::setMap): (JSC::StructureTransitionTable::singleTransition): (JSC::StructureTransitionTable::clearSingleTransition): (JSC::StructureTransitionTable::setSingleTransition): * runtime/WeakGCMap.h: (JSC::DefaultWeakGCMapFinalizerCallback::finalizerContextFor): (JSC::DefaultWeakGCMapFinalizerCallback::keyForFinalizer): (JSC::WeakGCMap::contains): (JSC::WeakGCMap::find): (JSC::WeakGCMap::remove): (JSC::WeakGCMap::add): (JSC::WeakGCMap::set): (JSC::WeakGCMap::finalize): * runtime/WriteBarrier.h: (JSC::writeBarrier): (JSC::WriteBarrierBase::set): (JSC::WriteBarrierBase::operator*): (JSC::WriteBarrierBase::operator->): (JSC::WriteBarrierBase::setWithoutWriteBarrier): 2011-04-15 Fridrich Strba <fridrich.strba@bluewin.ch> Reviewed by Gavin Barraclough. Correctly prefix symbols. Since gcc 4.5.0, Windows x64 symbols are not prefixed by underscore anymore. This is consistent with what MSVC does. https://bugs.webkit.org/show_bug.cgi?id=58573 * jit/JITStubs.cpp: 2011-04-15 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen. Bug 58705 - DFG JIT Add support for flow control (branch, jump). Add support for control flow by breaking the CodeBlock up into multiple basic blocks, generating code for each basic block in turn through the speculative JIT & then the non-speculative JIT. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::setTemporary): (JSC::DFG::ByteCodeParser::addToGraph): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: (JSC::DFG::BasicBlock::BasicBlock): (JSC::DFG::BasicBlock::getBytecodeOffset): (JSC::DFG::Graph::blockIndexForBytecodeOffset): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::JITCodeGenerator): (JSC::DFG::JITCodeGenerator::addBranch): (JSC::DFG::JITCodeGenerator::linkBranches): (JSC::DFG::JITCodeGenerator::BranchRecord::BranchRecord): * dfg/DFGNode.h: (JSC::DFG::Node::Node): (JSC::DFG::Node::isJump): (JSC::DFG::Node::isBranch): (JSC::DFG::Node::takenBytecodeOffset): (JSC::DFG::Node::notTakenBytecodeOffset): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGNonSpeculativeJIT.h: * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: 2011-04-15 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen. Bug 58701 - DFG JIT - add GetLocal/SetLocal nodes Use these for both access to arguments & local variables, adds ability to set locals, such that values will persist between basic blocks. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::get): (JSC::DFG::ByteCodeParser::set): (JSC::DFG::ByteCodeParser::getVariable): (JSC::DFG::ByteCodeParser::setVariable): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::setArgument): (JSC::DFG::ByteCodeParser::getThis): (JSC::DFG::ByteCodeParser::setThis): (JSC::DFG::ByteCodeParser::VariableRecord::VariableRecord): (JSC::DFG::ByteCodeParser::parseBlock): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): (JSC::DFG::Graph::derefChildren): * dfg/DFGGraph.h: (JSC::DFG::Graph::ref): (JSC::DFG::Graph::deref): * dfg/DFGNode.h: (JSC::DFG::Node::hasLocal): (JSC::DFG::Node::local): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-04-15 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 58696 - DFG JIT split handling of vars/temporaries Presently all callee registers are treated as having single block scope, since the DFG JIT can only compile single block functions. In order to expand the JIT to support control flow we will need to change to retaining locals (but not temporaries) across basic block boundaries. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::get): (JSC::DFG::ByteCodeParser::set): (JSC::DFG::ByteCodeParser::getVariable): (JSC::DFG::ByteCodeParser::setVariable): (JSC::DFG::ByteCodeParser::getTemporary): (JSC::DFG::ByteCodeParser::setTemporary): (JSC::DFG::ByteCodeParser::getArgument): (JSC::DFG::ByteCodeParser::getInt32Constant): (JSC::DFG::ByteCodeParser::getDoubleConstant): (JSC::DFG::ByteCodeParser::getJSConstant): (JSC::DFG::ByteCodeParser::constantUndefined): (JSC::DFG::ByteCodeParser::constantNull): (JSC::DFG::ByteCodeParser::one): (JSC::DFG::ByteCodeParser::parseBlock): (JSC::DFG::ByteCodeParser::parse): (JSC::DFG::parse): * dfg/DFGNode.h: * dfg/DFGScoreBoard.h: (JSC::DFG::ScoreBoard::ScoreBoard): (JSC::DFG::ScoreBoard::~ScoreBoard): (JSC::DFG::ScoreBoard::allocate): (JSC::DFG::ScoreBoard::use): 2011-04-15 Michael Saboff <msaboff@apple.com> Reviewed by Oliver Hunt. globalObject moved to JSObjectWithGlobalObject.cpp inhibits inlining https://bugs.webkit.org/show_bug.cgi?id=58677 Moved JSObjectWithGlobalObject::globalObject() to runtime/JSObjectWithGlobalObject.h to allow the compiler to inline it for a performance benefit. An equivalent instance had been in a header file before r60057. * JavaScriptCore.exp: * runtime/JSObjectWithGlobalObject.cpp: * runtime/JSObjectWithGlobalObject.h: (JSC::JSObjectWithGlobalObject::globalObject): 2011-04-14 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make JSNodeFilterCondition handle its lifetime correctly https://bugs.webkit.org/show_bug.cgi?id=58622 Add export * JavaScriptCore.exp: 2011-04-14 Alexey Proskuryakov <ap@apple.com> Reviewed by Dan Bernstein. WebKit2: Password field input does not switch to ASCII-compatible source https://bugs.webkit.org/show_bug.cgi?id=58583 <rdar://problem/9059651> * wtf/Platform.h: Removed WTF_USE_CARBON_SECURE_INPUT_MODE. It's now only used by Chromium, and shouldn't be enabled on any other platforms, so there is no reason to make it configurable via Platform.h. 2011-04-15 Dmitry Lomov <dslomov@google.com> Reviewed by David Levin. Add a sample test case for GTest framework https://bugs.webkit.org/show_bug.cgi?id=58509 Add an example of GTest testcase, complete with a runner, to JavaScriptCore. * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/tests/RunAllWtfTests.cpp: Added. (main): * wtf/tests/StringTests.cpp: Added. 2011-04-15 Anna Cavender <annacc@chromium.org> Reviewed by Eric Carlson. Renaming TRACK feature define to VIDEO_TRACK https://bugs.webkit.org/show_bug.cgi?id=53556 * Configurations/FeatureDefines.xcconfig: 2011-04-14 Gavin Barraclough <barraclough@apple.com> Rubber stamped by Geoffrey Garen. Hide DFG_JIT_RESTRICTIONS behind ARITHMETIC_OP() macro, and rename m_regressionGuard to m_parseFailed, such that it can be reused for other failure cases. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::parse): 2011-04-14 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoffrey Garen. Bug 58620 - DFG JIT - loading of arguments should not be lazy This optimization is overly simplistic. It only works because we never write out definitions to arguments (since we currently only compile single block functions). Revert this for now, we may want to reintroduce something like this again in the future, but it will need to be aware how to schedule definitions to arguments versus lazy loads that have not yet been performed. * dfg/DFGGenerationInfo.h: (JSC::DFG::GenerationInfo::needsSpill): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillInteger): (JSC::DFG::JITCodeGenerator::fillDouble): (JSC::DFG::JITCodeGenerator::fillJSValue): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::initConstantInfo): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::fillNumericToDouble): (JSC::DFG::JITCompiler::fillInt32ToInteger): (JSC::DFG::JITCompiler::fillToJS): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::isKnownInteger): (JSC::DFG::NonSpeculativeJIT::isKnownNumeric): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateIntInternal): (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::compile): 2011-04-14 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoffrey Garen. Bug 58600 - DFG JIT bugs in ValueToInt, PutByVal The bug in PutByVal is that an operand is in JSValueOperand - when this locks an integer into a register it will always retag the value without checking if the register is already locked. This is a problem where the value being stored by a PutByVal is the same as the subscript. The subscript is locked into a register first, as a strict integer. Locking the value results in the subscript being modified. The bug in ValueToInt related to the function of sillentFillAllRegisters. The problem is that this method will restore all register values from prior to the call, overwriting the result of the call out. Allow a register to be passed to specifically be excluded from being preserved. * assembler/ARMAssembler.h: (JSC::ARMAssembler::debugOffset): * assembler/ARMv7Assembler.h: (JSC::ARMv7Assembler::ARMInstructionFormatter::debugOffset): * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::debugOffset): * assembler/AssemblerBuffer.h: (JSC::AssemblerBuffer::debugOffset): * assembler/LinkBuffer.h: (JSC::LinkBuffer::debugAddress): * assembler/MIPSAssembler.h: (JSC::MIPSAssembler::debugOffset): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::orPtr): * assembler/X86Assembler.h: (JSC::X86Assembler::debugOffset): (JSC::X86Assembler::X86InstructionFormatter::debugOffset): * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parse): * dfg/DFGGenerationInfo.h: * dfg/DFGJITCodeGenerator.cpp: (JSC::DFG::JITCodeGenerator::fillJSValue): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::isConstant): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::isConstant): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::valueToNumber): (JSC::DFG::NonSpeculativeJIT::valueToInt32): (JSC::DFG::NonSpeculativeJIT::numberToInt32): (JSC::DFG::NonSpeculativeJIT::isKnownInteger): (JSC::DFG::NonSpeculativeJIT::isKnownNumeric): (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGNonSpeculativeJIT.h: (JSC::DFG::NonSpeculativeJIT::silentSpillGPR): (JSC::DFG::NonSpeculativeJIT::silentSpillFPR): (JSC::DFG::NonSpeculativeJIT::silentFillGPR): (JSC::DFG::NonSpeculativeJIT::silentFillFPR): (JSC::DFG::NonSpeculativeJIT::silentSpillAllRegisters): (JSC::DFG::NonSpeculativeJIT::silentFillAllRegisters): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): 2011-04-14 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Drain the mark stack while marking weak handles, not after. https://bugs.webkit.org/show_bug.cgi?id=58574 Otherwise, items that would have caused more weak handle marking are processed after all weak handle marking has finished, and referenced weak handles get recycled. * heap/HandleHeap.cpp: (JSC::HandleHeap::markWeakHandles): Removed looping from here, since we want Heap::markRoots to be responsible for draining the mark stack. * heap/Heap.cpp: (JSC::Heap::markRoots): Moved looping to here, as explained above. For efficiency's sake, drain the mark stack before starting to mark weak handles. Otherwise, items drained while marking weak handles may force an extra trip through the weak handle list. For correctness's sake, drain the mark stack each time through the weak handle list. Otherwise, opaque roots that would make weak handles reachable are not discovered until after weak handle marking is over. 2011-04-14 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make protected object list in caches window reflect reality https://bugs.webkit.org/show_bug.cgi?id=58565 Make sure the heap includes objects protected by Strong handles in its list of protected objects. * heap/HandleHeap.h: * heap/Heap.cpp: (JSC::HandleHeap::protectedObjectTypeCounts): 2011-04-14 Satish Sampath <satish@chromium.org> Reviewed by Anders Carlsson. Don't emit RegExp tables for chromium where they are not used https://bugs.webkit.org/show_bug.cgi?id=58544 * JavaScriptCore.gyp/JavaScriptCore.gyp: * create_regex_tables: Added the "--notables" command line argument. 2011-04-13 Geoffrey Garen <ggaren@apple.com> Try to fix ASSERTs seen on Windows bots. * wtf/HashTable.h: (WTF::hashTableSwap): Force MSVC to use the right version of swap. 2011-04-13 Ryuan Choi <ryuan.choi@samsung.com> Reviewed by Kenneth Rohde Christiansen. [CMAKE] Separate DerivedSources. https://bugs.webkit.org/show_bug.cgi?id=58427 * CMakeLists.txt: Change DERIVED_SOURCES_DIR to DERIVED_SOURCES_JAVASCRIPTCORE_DIR. 2011-04-13 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Switched DOM wrappers to use HashMap of Weak<T> instead of WeakGCMap<T> https://bugs.webkit.org/show_bug.cgi?id=58482 This will allow wrappers to make individual decisions about their lifetimes. * heap/HandleHeap.h: (JSC::HandleHeap::copyWeak): New function for copying a weak handle. It's wasn't previously possible to perform this operation using HandleHeap API because the HandleHeap doesn't expose its underlying Node structure. * heap/Local.h: (JSC::::set): * heap/Strong.h: (JSC::Strong::set): Added ASSERTs to verify that dead objects are not resurrected by placement into handles. (JSC::swap): Added a swap helper, so use of Strong<T> inside a hash table is efficient. * heap/Weak.h: (JSC::Weak::Weak): Fixed a bug where copying a weak pointer would not copy its weak callback and context. (JSC::Weak::operator=): Added an assignment operator, since the default C++ assignment operator did the wrong thing. (JSC::Weak::set): Added ASSERTs to verify that dead objects are not resurrected by placement into handles. (JSC::swap): Added a swap helper, so use of Strong<T> inside a hash table is efficient, and can be done without copying, which is illegal during the handle finalization phase. 2011-04-13 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Make PropertyMapEntry use a WriteBarrier for specificValue https://bugs.webkit.org/show_bug.cgi?id=58407 Make PropertyMapEntry use a WriteBarrier for specificValue, and then propagate the required JSGlobalData through all the methods it ends up being needed. * API/JSClassRef.cpp: (OpaqueJSClass::prototype): * API/JSContextRef.cpp: * API/JSObjectRef.cpp: (JSObjectMake): (JSObjectSetPrototype): * JavaScriptCore.exp: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): * interpreter/Interpreter.cpp: (JSC::appendSourceToError): (JSC::Interpreter::tryCacheGetByID): (JSC::Interpreter::privateExecute): * jit/JITStubs.cpp: (JSC::JITThunks::tryCacheGetByID): (JSC::DEFINE_STUB_FUNCTION): * runtime/BatchedTransitionOptimizer.h: (JSC::BatchedTransitionOptimizer::BatchedTransitionOptimizer): * runtime/InternalFunction.cpp: (JSC::InternalFunction::name): (JSC::InternalFunction::displayName): * runtime/JSActivation.cpp: (JSC::JSActivation::getOwnPropertySlot): * runtime/JSFunction.cpp: (JSC::JSFunction::name): (JSC::JSFunction::displayName): (JSC::JSFunction::getOwnPropertySlot): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::putWithAttributes): (JSC::JSGlobalObject::reset): (JSC::JSGlobalObject::resetPrototype): * runtime/JSGlobalObject.h: * runtime/JSObject.cpp: (JSC::JSObject::put): (JSC::JSObject::deleteProperty): (JSC::JSObject::defineGetter): (JSC::JSObject::defineSetter): (JSC::JSObject::lookupGetter): (JSC::JSObject::lookupSetter): (JSC::JSObject::getPropertySpecificValue): (JSC::JSObject::getOwnPropertyNames): (JSC::JSObject::seal): (JSC::JSObject::freeze): (JSC::JSObject::preventExtensions): (JSC::JSObject::removeDirect): (JSC::JSObject::getOwnPropertyDescriptor): (JSC::JSObject::defineOwnProperty): * runtime/JSObject.h: (JSC::JSObject::getDirect): (JSC::JSObject::getDirectLocation): (JSC::JSObject::isSealed): (JSC::JSObject::isFrozen): (JSC::JSObject::setPrototypeWithCycleCheck): (JSC::JSObject::setPrototype): (JSC::JSObject::inlineGetOwnPropertySlot): (JSC::JSObject::putDirectInternal): (JSC::JSObject::putDirectWithoutTransition): (JSC::JSObject::putDirectFunctionWithoutTransition): * runtime/Lookup.cpp: (JSC::setUpStaticFunctionSlot): * runtime/ObjectConstructor.cpp: (JSC::objectConstructorCreate): (JSC::objectConstructorSeal): (JSC::objectConstructorFreeze): (JSC::objectConstructorPreventExtensions): (JSC::objectConstructorIsSealed): (JSC::objectConstructorIsFrozen): * runtime/Operations.h: (JSC::normalizePrototypeChain): * runtime/PropertyMapHashTable.h: (JSC::PropertyMapEntry::PropertyMapEntry): (JSC::PropertyTable::PropertyTable): (JSC::PropertyTable::copy): * runtime/Structure.cpp: (JSC::Structure::materializePropertyMap): (JSC::Structure::despecifyDictionaryFunction): (JSC::Structure::addPropertyTransition): (JSC::Structure::removePropertyTransition): (JSC::Structure::changePrototypeTransition): (JSC::Structure::despecifyFunctionTransition): (JSC::Structure::getterSetterTransition): (JSC::Structure::toDictionaryTransition): (JSC::Structure::toCacheableDictionaryTransition): (JSC::Structure::toUncacheableDictionaryTransition): (JSC::Structure::sealTransition): (JSC::Structure::freezeTransition): (JSC::Structure::preventExtensionsTransition): (JSC::Structure::isSealed): (JSC::Structure::isFrozen): (JSC::Structure::addPropertyWithoutTransition): (JSC::Structure::removePropertyWithoutTransition): (JSC::Structure::copyPropertyTable): (JSC::Structure::get): (JSC::Structure::despecifyFunction): (JSC::Structure::despecifyAllFunctions): (JSC::Structure::put): (JSC::Structure::getPropertyNames): * runtime/Structure.h: (JSC::Structure::get): (JSC::Structure::materializePropertyMapIfNecessary): 2011-04-13 Paul Knight <pknight@apple.com> Reviewed by Gavin Barraclough. BACKTRACE() macro should check for Debug configuration in macro, not WTFReportBacktrace definition https://bugs.webkit.org/show_bug.cgi?id=58405 The BACKTRACE() macro requires JavaScriptCore be built with a Debug configuration in order for it to be enabled. Move the NDEBUG check to the header so it will be enabled when the calling framework or application is built with a Debug configuration, similar to how ASSERT() and friends work. * wtf/Assertions.cpp: * wtf/Assertions.h: 2011-04-12 Ben Taylor <bentaylor.solx86@gmail.com> Reviewed by Alexey Proskuryakov. https://bugs.webkit.org/show_bug.cgi?id=58131 Provide a workaround for an obscure Studio 12 compiler bug, which couldn't call src->~T() on a const T *src. * wtf/Vector.h: 2011-04-12 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=58395 Exceptions thrown from property getters called from Array prototype functions can be missed This is caught by an ASSERT in the top of Interpreter::executeCall. Check for exceptions after accessing properties that could be getters. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncSort): (JSC::arrayProtoFuncFilter): (JSC::arrayProtoFuncMap): (JSC::arrayProtoFuncEvery): (JSC::arrayProtoFuncForEach): (JSC::arrayProtoFuncSome): (JSC::arrayProtoFuncReduce): (JSC::arrayProtoFuncReduceRight): - Add exception checks. 2011-04-12 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make API callback objects use weak handles to run their finalizers https://bugs.webkit.org/show_bug.cgi?id=58389 Make the API object's private data struct act as a finalizer for an api object if the callback object has a API defined finalizer. * API/JSCallbackObject.cpp: (JSC::JSCallbackObjectData::finalize): * API/JSCallbackObject.h: * API/JSCallbackObjectFunctions.h: (JSC::::init): * heap/Handle.h: 2011-04-12 Geoffrey Garen <ggaren@apple.com> Reviewed by Geoffrey Garen. Cleaned up hash traits, and added hash traits for handles https://bugs.webkit.org/show_bug.cgi?id=58381 * heap/Handle.h: (JSC::HandleBase::swap): (JSC::Handle::Handle): (JSC::Handle::swap): Implemented swap, so we can rehash efficiently, and without creating new handles (which is not allowed during handle finalization). * heap/Strong.h: (JSC::Strong::swap): Use new SimpleClassHashTraits to avoid duplication. * heap/Weak.h: (JSC::Weak::isHashTableDeletedValue): (JSC::Weak::Weak): (JSC::Weak::swap): (JSC::Weak::hashTableDeletedValue): Ditto. * wtf/HashTraits.h: (WTF::SimpleClassHashTraits::constructDeletedValue): (WTF::SimpleClassHashTraits::isDeletedValue): Added SimpleClassHashTraits, which are analogous to SimpleClassVectorTraits, since they are used in a bunch of places. * wtf/RetainPtr.h: Use new SimpleClassHashTraits to avoid duplication. * wtf/text/StringHash.h: Use new SimpleClassHashTraits to avoid duplication. 2011-04-12 Geoffrey Garen <ggaren@apple.com> Reviewed by Sam Weinig. Cleaned up some Vector traits, and added missing Vector traits for handles https://bugs.webkit.org/show_bug.cgi?id=58372 * heap/Local.h: Inherit from SimpleClassVectorTraits to avoid duplication. * heap/Strong.h: Ditto. * heap/Weak.h: Ditto. * parser/JSParser.cpp: Fixed a traits error. No test case because this particular trait is not currently exercised by the parser. * runtime/UString.h: No need to override canInitializeWithMemset, since our base class sets it to true. * wtf/VectorTraits.h: Inherit from VectorTraitsBase to avoid duplication. * wtf/text/WTFString.h: No need to override canInitializeWithMemset, since our base class sets it to true. 2011-04-12 Thouraya ANDOLSI <thouraya.andolsi@st.com> Reviewed by Eric Seidel. [Qt] Enable JIT build for SH4 platforms. https://bugs.webkit.org/show_bug.cgi?id=58317 enable JIT build for QT backend for SH4 platforms. * JavaScriptCore.pro: * wtf/Platform.h: 2011-04-11 Ben Taylor <bentaylor.solx86@gmail.com> Reviewed by Alexey Proskuryakov. https://bugs.webkit.org/show_bug.cgi?id=58289 Fix compilation on Solaris/Studio 12 C++ in wtf/FastMalloc.cpp, WTF::TCMalloc_PageHeap::runScavengerThread(void*) expected to return a value. * wtf/FastMalloc.cpp: (WTF::TCMalloc_PageHeap::runScavengerThread): 2011-04-11 Mark Rowe <mrowe@apple.com> Fix the build. * JavaScriptCore.xcodeproj/project.pbxproj: Headers used outside of JavaScriptCore need to be marked as private. 2011-04-11 Anna Cavender <annacc@chromium.org> Reviewed by Eric Carlson. Setup ENABLE(TRACK) feature define https://bugs.webkit.org/show_bug.cgi?id=53556 * Configurations/FeatureDefines.xcconfig: 2011-04-11 Geoffrey Garen <ggaren@apple.com> Try to fix a few builds. Updated a few more build configurations for file moves. * CMakeListsWinCE.txt: 2011-04-11 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 58263 - Use EncodedValueDescriptor on both JSVALUE32_64, JSVALUE64 The JSJITInterface already uses EncodedValueDescriptor to access the tag/payload separately on JSVALUE64, even though EncodedValueDescriptor is not used in JSVALUE64's implementation of JSValue. Remove the separate definition for m_ptr on X86_64. Using the union allows us to remove a layer of makeImmediate()/ immedaiteValue() methods. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * jit/JITInlineMethods.h: (JSC::JIT::emitTagAsBoolImmediate): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_not): (JSC::JIT::emit_op_jeq_null): (JSC::JIT::emit_op_jneq_null): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_eq_null): (JSC::JIT::emit_op_neq_null): (JSC::JIT::emitSlow_op_not): * runtime/JSCell.h: * runtime/JSValue.h: * runtime/JSValueInlineMethods.h: (JSC::JSValue::encode): (JSC::JSValue::decode): (JSC::JSValue::operator==): (JSC::JSValue::operator!=): (JSC::JSValue::JSValue): (JSC::JSValue::operator bool): (JSC::JSValue::asInt32): (JSC::JSValue::isUndefinedOrNull): (JSC::JSValue::isBoolean): (JSC::JSValue::isCell): (JSC::JSValue::isInt32): (JSC::JSValue::asDouble): (JSC::JSValue::isNumber): (JSC::JSValue::asCell): 2011-04-11 Geoffrey Garen <ggaren@apple.com> Try to fix a few builds. Updated a few more build configurations for file moves. * CMakeListsEfl.txt: * wscript: 2011-04-11 Geoffrey Garen <ggaren@apple.com> Build fix: Updated a file name. * CMakeLists.txt: 2011-04-11 Geoffrey Garen <ggaren@apple.com> Rubber-stamped by Sam Weinig. Moved remaining heap implementation files to the heap folder. * Android.mk: * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * heap/ConservativeRoots.cpp: Copied from runtime/ConservativeSet.cpp. * heap/ConservativeRoots.h: Copied from runtime/ConservativeSet.h. * heap/Handle.h: * heap/Heap.cpp: * heap/MachineStackMarker.cpp: Copied from runtime/MachineStackMarker.cpp. * heap/MachineStackMarker.h: Copied from runtime/MachineStackMarker.h. * heap/MarkStack.cpp: Copied from runtime/MarkStack.cpp. * heap/MarkStack.h: Copied from runtime/MarkStack.h. * heap/MarkStackPosix.cpp: Copied from runtime/MarkStackPosix.cpp. * heap/MarkStackSymbian.cpp: Copied from runtime/MarkStackSymbian.cpp. * heap/MarkStackWin.cpp: Copied from runtime/MarkStackWin.cpp. * heap/MarkedBlock.cpp: Copied from runtime/MarkedBlock.cpp. * heap/MarkedBlock.h: Copied from runtime/MarkedBlock.h. * heap/MarkedSpace.cpp: Copied from runtime/MarkedSpace.cpp. * heap/MarkedSpace.h: Copied from runtime/MarkedSpace.h. * interpreter/RegisterFile.cpp: * runtime/ConservativeSet.cpp: Removed. * runtime/ConservativeSet.h: Removed. * runtime/MachineStackMarker.cpp: Removed. * runtime/MachineStackMarker.h: Removed. * runtime/MarkStack.cpp: Removed. * runtime/MarkStack.h: Removed. * runtime/MarkStackPosix.cpp: Removed. * runtime/MarkStackSymbian.cpp: Removed. * runtime/MarkStackWin.cpp: Removed. * runtime/MarkedBlock.cpp: Removed. * runtime/MarkedBlock.h: Removed. * runtime/MarkedSpace.cpp: Removed. * runtime/MarkedSpace.h: Removed. 2011-04-11 Gavin Barraclough <barraclough@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-04-09 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 58198 - Clean up JSValue implementation for JSVALUE64 Remove JSNumberCell, JSImmediate, unify some methods between JSVALUE32_64/JSVALUE64 JSNumberCell.h largely just contained the constructors for JSValue on JSVALUE64, which should not have been here. JSImmediate mostly contained uncalled methods, along with the internal implementation of the JSValue constructors split unnecessarily across a number of layers of function calls. These could largely be merged back together. Many methods and constructors from JSVALUE32_64 and JSVALUE64 can by unified. The .cpp files were empty. Moving all these methods into JSValue.h seems to be a repro measurable regression, so I have kept these methods in a separate JSValueInlineMethods.h. Adding the 64-bit tag values as static const members of JSValue also measures as a repro regression, so I have made these #defines. * Android.mk: * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: - Removed JSImmediate.h, JSNumberCell.h. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitLoad): - Removed class JSImmediate. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): - Removed class JSImmediate. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): - Removed class JSImmediate. * jit/JITArithmetic.cpp: (JSC::JIT::compileBinaryArithOpSlowCase): - Removed class JSImmediate. * jit/JITInlineMethods.h: (JSC::JIT::emitJumpIfJSCell): (JSC::JIT::emitJumpIfNotJSCell): (JSC::JIT::emitJumpIfImmediateInteger): (JSC::JIT::emitJumpIfNotImmediateInteger): (JSC::JIT::emitFastArithDeTagImmediate): (JSC::JIT::emitFastArithDeTagImmediateJumpIfZero): (JSC::JIT::emitFastArithReTagImmediate): (JSC::JIT::emitTagAsBoolImmediate): - Removed class JSImmediate. * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_not): (JSC::JIT::emit_op_jeq_null): (JSC::JIT::emit_op_jneq_null): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_eq_null): (JSC::JIT::emit_op_neq_null): (JSC::JIT::emitSlow_op_not): - Removed class JSImmediate. * jit/JSInterfaceJIT.h: - Removed class JSImmediate. * runtime/JSCell.h: - Removed JSImmediate.h, JSNumberCell.h. * runtime/JSImmediate.cpp: Removed. * runtime/JSImmediate.h: Removed. * runtime/JSNumberCell.cpp: Removed. * runtime/JSNumberCell.h: Removed. - Removed. * runtime/JSObject.h: - Removed JSImmediate.h, JSNumberCell.h. * runtime/JSString.h: - Removed JSImmediate.h, JSNumberCell.h. * runtime/JSValue.h: - Added tags for JSVALUE64, moved out some JSVALUE32_64 methods, unified with JSVALUE64. * runtime/JSValueInlineMethods.h: Added. (JSC::JSValue::toInt32): (JSC::JSValue::toUInt32): (JSC::JSValue::isUInt32): (JSC::JSValue::asUInt32): (JSC::JSValue::uncheckedGetNumber): (JSC::JSValue::toJSNumber): (JSC::jsNaN): (JSC::JSValue::getNumber): (JSC::JSValue::getBoolean): (JSC::JSValue::JSValue): (JSC::JSValue::encode): (JSC::JSValue::decode): (JSC::JSValue::operator bool): (JSC::JSValue::operator==): (JSC::JSValue::operator!=): (JSC::JSValue::isUndefined): (JSC::JSValue::isNull): (JSC::JSValue::isUndefinedOrNull): (JSC::JSValue::isCell): (JSC::JSValue::isInt32): (JSC::JSValue::isDouble): (JSC::JSValue::isTrue): (JSC::JSValue::isFalse): (JSC::JSValue::tag): (JSC::JSValue::payload): (JSC::JSValue::asInt32): (JSC::JSValue::asDouble): (JSC::JSValue::asCell): (JSC::JSValue::isNumber): (JSC::JSValue::isBoolean): (JSC::JSValue::makeImmediate): (JSC::JSValue::immediateValue): (JSC::reinterpretDoubleToIntptr): (JSC::reinterpretIntptrToDouble): - Methods moved here from JSImmediate.h/JSNumberCell.h/JSValue.h. * runtime/Operations.h: - Removed JSImmediate.h, JSNumberCell.h. * wtf/StdLibExtras.h: - Export bitwise_cast. 2011-04-11 Thouraya ANDOLSI <thouraya.andolsi@st.com> Reviewed by Oliver Hunt. SH4 JIT SUPPORT. https://bugs.webkit.org/show_bug.cgi?id=44329 Add JIT remaining part for SH4 platforms. * assembler/MacroAssemblerSH4.h: * jit/JIT.h: * jit/JITInlineMethods.h: * jit/JITOpcodes32_64.cpp: * jit/JITPropertyAccess32_64.cpp: * jit/JITStubs.cpp: * jit/JITStubs.h: * jit/JSInterfaceJIT.h: 2011-04-10 Geoffrey Garen <ggaren@apple.com> Rubber-stamped by Beth Dakin. Moved Heap.h and Heap.cpp to the heap folder, because anything less would be uncivilized. * Android.mk: * CMakeLists.txt: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * heap/Heap.cpp: Copied from JavaScriptCore/runtime/Heap.cpp. * heap/Heap.h: Copied from JavaScriptCore/runtime/Heap.h. * runtime/Heap.cpp: Removed. * runtime/Heap.h: Removed. 2011-04-10 Patrick Gansterer <paroga@webkit.org> Reviewed by Darin Adler. Remove duplicated code from AtomicString::fromUTF8() https://bugs.webkit.org/show_bug.cgi?id=53711 * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/text/AtomicString.cpp: (WTF::AtomicString::fromUTF8Internal): * wtf/text/AtomicString.h: (WTF::AtomicString::fromUTF8): * wtf/unicode/UTF8.cpp: (WTF::Unicode::calculateStringHashAndLengthFromUTF8): * wtf/unicode/UTF8.h: 2011-04-10 Maciej Stachowiak <mjs@apple.com> Not reviewed. Fix build (at least on Lion) by adding some newish header files to PrivateHeaders. * JavaScriptCore.xcodeproj/project.pbxproj: 2011-04-09 Geoffrey Garen <ggaren@apple.com> Not reviewed. Try recommitting some things svn left out of its last commit. * heap/Handle.h: (JSC::HandleBase::operator!): (JSC::HandleBase::HandleBase): (JSC::HandleBase::slot): (JSC::HandleBase::setSlot): (JSC::Handle::Handle): * heap/HandleHeap.cpp: (JSC::HandleHeap::markWeakHandles): (JSC::HandleHeap::finalizeWeakHandles): (JSC::HandleHeap::isValidWeakNode): * heap/HandleHeap.h: (JSC::HandleHeap::globalData): 2011-04-08 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. A few heap-related renames and file moves. WeakGCPtr<T> => Weak<T> Global<T> => Strong<T> collector/ => heap/ collector/* => heap/* runtime/WeakGCPtr.h => heap/Weak.h (Eventually, even more files should move into the heap directory. Like Heap.h and Heap.cpp, for example.) * API/JSClassRef.h: * CMakeLists.txt: * GNUmakefile.am: * GNUmakefile.list.am: * JavaScriptCore.gypi: * JavaScriptCore.pri: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreCommon.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: * JavaScriptCore.vcproj/jsc/jscCommon.vsprops: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/SamplingTool.h: * bytecompiler/BytecodeGenerator.h: * collector: Removed. * collector/handles: Removed. * collector/handles/Global.h: Removed. * collector/handles/Handle.h: Removed. * collector/handles/HandleHeap.cpp: Removed. * collector/handles/HandleHeap.h: Removed. * collector/handles/HandleStack.cpp: Removed. * collector/handles/HandleStack.h: Removed. * collector/handles/Local.h: Removed. * collector/handles/LocalScope.h: Removed. * heap: Copied from collector. * heap/Handle.h: Copied from collector/handles/Handle.h. * heap/HandleHeap.cpp: Copied from collector/handles/HandleHeap.cpp. * heap/HandleHeap.h: Copied from collector/handles/HandleHeap.h. * heap/HandleStack.cpp: Copied from collector/handles/HandleStack.cpp. * heap/HandleStack.h: Copied from collector/handles/HandleStack.h. * heap/Local.h: Copied from collector/handles/Local.h. * heap/LocalScope.h: Copied from collector/handles/LocalScope.h. * heap/Strong.h: Copied from collector/handles/Global.h. (JSC::Strong::Strong): (JSC::Strong::~Strong): (JSC::Strong::operator=): * heap/Weak.h: Copied from runtime/WeakGCPtr.h. (JSC::Weak::Weak): (JSC::Weak::~Weak): * heap/handles: Removed. * interpreter/RegisterFile.h: * jit/JITStubs.cpp: (JSC::JITThunks::hostFunctionStub): * jit/JITStubs.h: * runtime/Structure.h: * runtime/WeakGCPtr.h: Removed. 2011-04-08 Alpha Lam <hclam@chromium.org> Unreviewed, rolling out r83335. http://trac.webkit.org/changeset/83335 https://bugs.webkit.org/show_bug.cgi?id=53556 GTK and QT bots are broken * Configurations/FeatureDefines.xcconfig: 2011-04-08 Gavin Barraclough <barraclough@apple.com> Ooops, typo, build fix. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parse): 2011-04-08 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. Bug 58154 - Add support for comparison operators to the DFG JIT. Add support for <, <=, ==, ===, and also !. Add support for all corresponding bytecode ops, including the not- and -null forms. Initially add functionally correct support, we'll revisit the performance. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::constantNull): (JSC::DFG::ByteCodeParser::parse): - Add support for parsing of bytecode opcodes, * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::callOperation): - Add new operation call types, return bool values. * dfg/DFGNode.h: - Add new node types. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): - Add code generation for new nodes. * dfg/DFGOperations.cpp: (JSC::DFG::operationCompareLess): (JSC::DFG::operationCompareLessEq): (JSC::DFG::operationCompareEq): (JSC::DFG::operationCompareStrictEq): (JSC::DFG::dfgConvertJSValueToBoolean): * dfg/DFGOperations.h: - Add operation callbacks to implement new ops. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): - Add code generation for new nodes. * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): - Switched to a simpler <0 check, rather than relying on an internal value in JSImmediate. * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompilePatchGetArrayLength): - Switched to a simpler <0 check, rather than relying on an internal value in JSImmediate. * runtime/JSImmediate.h: - Make tag values public, rather than relying on a friend - this matches JSVALUE32_64. 2011-04-07 Anna Cavender <annacc@chromium.org> Reviewed by Eric Carlson. Setup ENABLE(TRACK) feature define https://bugs.webkit.org/show_bug.cgi?id=53556 * Configurations/FeatureDefines.xcconfig: 2011-04-07 Balazs Kelemen <kbalazs@webkit.org> Reviewed by Kenneth Rohde Christiansen. [WK2][Qt][GTK] Introduce common use flag for the shared UNIX domain socket IPC implementation https://bugs.webkit.org/show_bug.cgi?id=58030 * wtf/Platform.h: Introduce USE(UNIX_DOMAIN_SOCKETS) for WebKit2. 2011-04-08 Adam Roben <aroben@apple.com> Clean build fix * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: Don't try to delete directories that don't exist. Also switched from del /s to rmdir /s, which has the benefit of deleting the directory itself in addition to the files it contains. 2011-04-07 Geoffrey Garen <ggaren@apple.com> Reviewed by Maciej Stachowiak. Some Handle<T> cleanup https://bugs.webkit.org/show_bug.cgi?id=58109 * bytecode/SamplingTool.h: Sorted alphabetically because that's the WebKit style. Added a Global.h #include that was previously missing but harmless. * collector/handles/Global.h: (JSC::Global::Global): Added a null constructor. No need for a special tag, and the tag is incompatible with some data structures. (JSC::Global::isHashTableDeletedValue): (JSC::Global::~Global): (JSC::Global::set): (JSC::Global::operator=): (JSC::Global::clear): (JSC::Global::hashTableDeletedValue): Reordered constructors to be near each other. (JSC::Global::setWithWriteBarrier): Renamed internalSet to setWithWriteBarrier for clarity, and funneled more code into using set and setWithWriteBarrier to reduce duplication. * collector/handles/Handle.h: (JSC::HandleBase::operator!): (JSC::HandleBase::HandleBase): Removed isEmpty(), since we already have boolean and ! operators. (JSC::HandleBase::slot): (JSC::HandleBase::setSlot): (JSC::Handle::Handle): Added general support for null Handles. This was previously outlawed by ASSERTs, but our code has grown to support and rely on null Handles. * collector/handles/HandleHeap.cpp: (JSC::HandleHeap::markWeakHandles): (JSC::HandleHeap::finalizeWeakHandles): (JSC::HandleHeap::isValidWeakNode): Migrated from isValidWeakHandle, and beefed this up a bit. * collector/handles/HandleHeap.h: (JSC::HandleHeap::globalData): Added accessor, used by some new set functions. * collector/handles/Local.h: Moved hash traits to the bottom of the file, since this file is about the Local class, not the traits. (JSC::::Local): Updated for removal of invalidate(). (JSC::::operator): Deployed "using" to avoid a lot of this-> template funny business. (JSC::::setWithSlotCheck): Renamed from internalSet, more specific now. * interpreter/RegisterFile.h: (JSC::RegisterFile::RegisterFile): Updated to use null constructor. * jit/JITStubs.cpp: (JSC::JITThunks::hostFunctionStub): * runtime/JSPropertyNameIterator.h: (JSC::Structure::setEnumerationCache): * runtime/Structure.h: Removed clearEnumerationCache because it was an unused holdover from when the enumeration cache was not a handle. * runtime/WeakGCMap.h: (JSC::WeakGCMap::set): Finish initializing our handle before putting it in the table. This seemed more logical, and at one point was required to avoid triggering an ASSERT. * runtime/WeakGCPtr.h: Inherit from Handle instead of rolling our own handle-like behavior, to avoid duplication. (JSC::WeakGCPtr::WeakGCPtr): (JSC::WeakGCPtr::~WeakGCPtr): (JSC::WeakGCPtr::get): (JSC::WeakGCPtr::clear): (JSC::WeakGCPtr::set): (JSC::WeakGCPtr::setWithWriteBarrier): Removed duplicate code and standardized on Handle idioms. 2011-04-07 Adam Barth <abarth@webkit.org> Reviewed by Martin Robinson. Refactor Gtk build system to separate list of files https://bugs.webkit.org/show_bug.cgi?id=58090 This is the first step towards generating part of the GTK build system using GYP. In the first iteration, our plan is to just generate the list of files. This patch is the first step, which is to separate out the part of JavaScriptCore build system that we intend to generate from the rest of the build system. * GNUmakefile.am: * GNUmakefile.list.am: Added. 2011-04-07 Zoltan Herczeg <zherczeg@webkit.org> Reviewed by Gavin Barraclough. Mapping booleans the same way as integers https://bugs.webkit.org/show_bug.cgi?id=56913 Instead of having a seperate tag field for booleans, the logical values are stored in the payload field (for JSValue32_64 representation). 1.007x speedup on SunSpider. * jit/JIT.h: * jit/JITInlineMethods.h: (JSC::JIT::emitStoreBool): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_instanceof): (JSC::JIT::emit_op_not): (JSC::JIT::emit_op_jfalse): (JSC::JIT::emitSlow_op_jfalse): (JSC::JIT::emit_op_jtrue): (JSC::JIT::emitSlow_op_jtrue): (JSC::JIT::emit_op_jeq_null): (JSC::JIT::emit_op_jneq_null): (JSC::JIT::emit_op_eq): (JSC::JIT::emitSlow_op_eq): (JSC::JIT::emit_op_neq): (JSC::JIT::emitSlow_op_neq): (JSC::JIT::compileOpStrictEq): (JSC::JIT::emit_op_eq_null): (JSC::JIT::emit_op_neq_null): * jit/JSInterfaceJIT.h: * runtime/JSValue.h: (JSC::JSValue::JSValue): (JSC::JSValue::isTrue): (JSC::JSValue::isFalse): (JSC::JSValue::getBoolean): 2011-04-07 Eric Seidel <eric@webkit.org> Reviewed by Adam Barth. Add stub support for generating Gtk build system from gyp https://bugs.webkit.org/show_bug.cgi?id=58086 This does not produce a buildable JavaScriptCore, but it does allow running gyp/configure --port=gtk and having it generate a gtk.Makefile which we can use for testing the rest of the plumbing. * gyp/gtk.gyp: Added. 2011-04-07 Andrew Scherkus <scherkus@chromium.org> Revert ENABLE_TRACK patch due to compile failures. * Configurations/FeatureDefines.xcconfig: 2011-04-07 Adam Barth <abarth@webkit.org> Fix whitespace in GNUmakefile.am. * GNUmakefile.am: 2011-04-07 Gavin Barraclough <barraclough@apple.com> Fix a couple of typos in comments that Darin spotted. * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_not): * runtime/JSImmediate.h: 2011-04-06 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoff Garen. Bug 58057 - Store boolean payload in low bit of JSImmediate And remove some uncalled functions from JSImmediate.h * jit/JITInlineMethods.h: (JSC::JIT::emitTagAsBoolImmediate): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_not): * runtime/JSImmediate.h: (JSC::JSImmediate::makeInt): (JSC::JSImmediate::makeBool): (JSC::JSImmediate::intValue): (JSC::JSImmediate::boolValue): (JSC::JSImmediate::asInt32): (JSC::JSImmediate::toDouble): (JSC::JSValue::asInt32): (JSC::JSValue::isUInt32): (JSC::JSValue::asUInt32): 2011-04-07 Liang Qi <liang.qi@nokia.com> Reviewed by Laszlo Gombos. [Qt][Symbian] Enable webkit build with GCCE on Symbian. https://bugs.webkit.org/show_bug.cgi?id=57841 * wtf/MathExtras.h: GCCE compiler doesn't support those std static functions. 2011-04-06 Dai Mikurube <dmikurube@chromium.org> Reviewed by David Levin. Add QUOTA build flag for unified quota API https://bugs.webkit.org/show_bug.cgi?id=57918 * Configurations/FeatureDefines.xcconfig: Added QUOTA build flag 2011-04-06 Kevin Ollivier <kevino@theolliviers.com> Reviewed by Darin Adler. Make sure JS_EXPORT_PRIVATE is an empty define when we aren't using the export macros. https://bugs.webkit.org/show_bug.cgi?id=27551 * config.h: 2011-04-06 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Stop JSObject::isUsingInlineStorage() from using the structure https://bugs.webkit.org/show_bug.cgi?id=57986 Make the isUsingInlineStorage() implementation just look at whether the property storage is inside the object. * runtime/JSObject.h: (JSC::JSObject::isUsingInlineStorage): (JSC::JSObject::JSObject): 2011-04-06 Gavin Barraclough <barraclough@apple.com> Rubber stamped by Geoff Garen. Update comments documenting JSVALUE64/JSVALUE32_64 JSValue representations. * runtime/JSImmediate.h: * runtime/JSValue.h: 2011-04-06 Lucas De Marchi <lucas.demarchi@profusion.mobi> cmake: Fix build for ARMv7 * CMakeLists.txt: add missing file. 2011-04-06 Liang Qi <liang.qi@nokia.com> Reviewed by Benjamin Poulain. Correct a include file name. https://bugs.webkit.org/show_bug.cgi?id=57839 * wtf/PageAllocatorSymbian.h: It should be case sensitive. This fix builds on Unix hosts. 2011-04-06 Adam Roben <aroben@apple.com> Build fix after r83056 * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: Added property svn:executable. 2011-04-06 Adam Roben <aroben@apple.com> Move JavaScriptCoreGenerated's file-copying logic out to a new script Hopefully this will make it easier to modify this logic in the future. I also made the script much quieter than the old logic, since it didn't seem helpful to see long lists of filenames during the copying phase. If we like this new style, we could copy it for our other projects. Fixes <http://webkit.org/b/57950> JavaScriptCoreGenerated's file-copying logic is hard to modify and noisy Reviewed by Steve Falkenburg. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.make: Moved logic to copy files from here... * JavaScriptCore.vcproj/JavaScriptCore/copy-files.cmd: ...to here. (Added.) * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.vcproj: Added copy-files.cmd for convenience. 2011-04-05 Geoffrey Garen <ggaren@apple.com> Try to fix the Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Who likes export files? I do! 2011-04-05 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Introduced the concept of opaque roots, in preparation for marking the DOM with them https://bugs.webkit.org/show_bug.cgi?id=57903 * JavaScriptCore.exp: Who likes export files? I do! * collector/handles/HandleHeap.cpp: (JSC::isValidWeakHandle): Factored out a helper function for ASSERTs. (JSC::WeakHandleOwner::~WeakHandleOwner): Moved from header to avoid weak linkage problems. (JSC::WeakHandleOwner::isReachableFromOpaqueRoots): New callback. Currently unused. (JSC::WeakHandleOwner::finalize): Switched from pure virtual to a default empty implementation, since not all clients necessarily want or need non-trivial finalizers. (JSC::HandleHeap::markWeakHandles): Split updateWeakHandles into two passes. The first pass marks all reachable weak handles. The second pass finalizes all unreachable weak handles. This must be two passes because we don't know the set of finalizable weak handles until we're done marking all weak handles. (JSC::HandleHeap::finalizeWeakHandles): Use new helper function. * collector/handles/HandleHeap.h: Ditto. * runtime/Heap.cpp: (JSC::Heap::destroy): (JSC::Heap::markRoots): (JSC::Heap::reset): Split out handle marking from handle finalization. * runtime/MarkStack.cpp: (JSC::MarkStack::reset): * runtime/MarkStack.h: (JSC::MarkStack::addOpaqueRoot): (JSC::MarkStack::containsOpaqueRoot): (JSC::MarkStack::opaqueRootCount): (JSC::HeapRootMarker::markStack): New helper functions for managing the set of opaque roots. * runtime/WeakGCMap.h: (JSC::WeakGCMap::finalize): Renamed to match parent class declaration. 2011-04-05 Balazs Kelemen <kbalazs@webkit.org> Reviewed by Darin Adler. Build fix for YarrParser.h https://bugs.webkit.org/show_bug.cgi?id=57822 * yarr/YarrParser.h: (JSC::Yarr::Parser::CharacterClassParserDelegate::CharacterClassParserDelegate): 2011-04-05 Steve Falkenburg <sfalken@apple.com> Follow-up Windows build fix. Don't skip react-to-vsprops-changes.py for all production builds, only those initiated via JavaScriptCore.make. * JavaScriptCore.vcproj/JavaScriptCore.make: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.make: 2011-04-05 Oliver Hunt <oliver@apple.com> Reviewed by Darin Adler. REGRESSION (r82849): 85,000+ JSC-related leaks seen on SnowLeopard Intel Leaks https://bugs.webkit.org/show_bug.cgi?id=57857 Whoops, accidentally removed a deref(). * bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::deref): 2011-04-05 Steve Falkenburg <sfalken@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.vcproj: Add per-configuration vsprops files. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedCommon.vsprops: Removed inheritance from common.vsprops. Set production environment variable before calling make. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedDebug.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedDebugAll.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedDebugCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedProduction.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedRelease.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedReleaseCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGeneratedReleasePGO.vsprops: Added. 2011-04-05 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Make caches window show more info about non-jsobject GC values https://bugs.webkit.org/show_bug.cgi?id=57874 Add ClassInfo to the various internal JS types that currently don't have any, and make the text for caches window show the classname for non-JSObject instances. * runtime/Executable.cpp: * runtime/Executable.h: (JSC::ExecutableBase::createStructure): (JSC::NativeExecutable::createStructure): (JSC::NativeExecutable::NativeExecutable): (JSC::EvalExecutable::createStructure): (JSC::ProgramExecutable::createStructure): (JSC::FunctionExecutable::createStructure): * runtime/Heap.cpp: (JSC::TypeCounter::typeName): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: * runtime/ScopeChain.cpp: * runtime/ScopeChain.h: (JSC::ScopeChainNode::createStructure): * runtime/StructureChain.cpp: * runtime/StructureChain.h: (JSC::StructureChain::createStructure): 2011-04-05 Nikolas Zimmermann <nzimmermann@rim.com> Reviewed by Andreas Kling. Cleanup StringConcatenate https://bugs.webkit.org/show_bug.cgi?id=57836 Don't use PassRefPtr in local variables, properly store in RefPtrs and release on return. Add a makeString() variant taking 9 arguments, needed by a follow-up patch. * wtf/text/StringConcatenate.h: (WTF::tryMakeString): (WTF::makeString): 2011-04-04 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r82876. http://trac.webkit.org/changeset/82876 https://bugs.webkit.org/show_bug.cgi?id=57816 Caused a lot of test crashes (Requested by tkent on #webkit). * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * wtf/FastMalloc.cpp: (WTF::tryFastMalloc): (WTF::fastMalloc): (WTF::tryFastCalloc): (WTF::fastCalloc): (WTF::fastFree): (WTF::tryFastRealloc): (WTF::fastRealloc): (WTF::fastMallocSize): (WTF::TCMalloc_PageHeap::isScavengerSuspended): (WTF::TCMalloc_PageHeap::scheduleScavenger): (WTF::TCMalloc_PageHeap::suspendScavenger): (WTF::TCMalloc_PageHeap::signalScavenger): (WTF::TCMallocStats::malloc): (WTF::TCMallocStats::free): (WTF::TCMallocStats::fastCalloc): (WTF::TCMallocStats::tryFastCalloc): (WTF::TCMallocStats::calloc): (WTF::TCMallocStats::fastRealloc): (WTF::TCMallocStats::tryFastRealloc): (WTF::TCMallocStats::realloc): (WTF::TCMallocStats::fastMallocSize): * wtf/FastMalloc.h: (WTF::Internal::fastMallocMatchValidationType): (WTF::Internal::fastMallocMatchValidationValue): (WTF::Internal::setFastMallocMatchValidationType): (WTF::fastMallocMatchValidateFree): * wtf/Platform.h: 2011-04-04 Oliver Hunt <oliver@apple.com> Reviewed by Antti Koivisto. Stop JSCell.h from including Structure.h https://bugs.webkit.org/show_bug.cgi?id=57809 * runtime/GetterSetter.h: * runtime/JSAPIValueWrapper.h: * runtime/JSCell.h: (JSC::JSCell::JSValue::toThisObject): * runtime/JSString.h: * runtime/ScopeChain.h: * runtime/Structure.h: (JSC::JSCell::isObject): (JSC::JSCell::isString): (JSC::JSCell::classInfo): (JSC::JSCell::createDummyStructure): (JSC::JSValue::needsThisConversion): (JSC::MarkStack::internalAppend): * runtime/StructureChain.h: 2011-04-04 Oliver Hunt <oliver@apple.com> Fix clang build. * wtf/FastMalloc.cpp: (WTF::fastMalloc): (WTF::fastCalloc): (WTF::fastRealloc): 2011-04-04 Oliver Hunt <oliver@apple.com> Remove accidental change to Platform.h * wtf/Platform.h: 2011-04-04 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Fixed a weak-handle-related leak in RegisterFile https://bugs.webkit.org/show_bug.cgi?id=57793 * interpreter/RegisterFile.cpp: Nixed leaky GlobalObjectNotifier. * interpreter/RegisterFile.h: (JSC::RegisterFile::GlobalObjectOwner::finalize): (JSC::RegisterFile::RegisterFile): Replaced GlobalObjectNotifier with a per-RegisterFile weak handle owner, which does not leak. * runtime/WeakGCPtr.h: (JSC::WeakGCPtr::set): Allow set() to take a context argument, since RegisterFile now needs this. (Seems like it was an accidental omission all along.) 2011-04-04 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make malloc validation useful https://bugs.webkit.org/show_bug.cgi?id=57502 This patch changes FAST_MALLOC_MATCH_VALIDATION with a general corruption check that tags the beginning and end of all allocations to check for write overflows and overwrites the contents of memory on free in order to (hopefully) show up use-after-free issues sooner. We also turn it on by default for debug builds. * JavaScriptCore.exp: * wtf/FastMalloc.cpp: (WTF::tryFastMalloc): (WTF::fastMalloc): (WTF::tryFastCalloc): (WTF::fastCalloc): (WTF::fastFree): (WTF::tryFastRealloc): (WTF::fastRealloc): (WTF::TCMalloc_PageHeap::isScavengerSuspended): (WTF::TCMalloc_PageHeap::scheduleScavenger): (WTF::TCMalloc_PageHeap::suspendScavenger): (WTF::TCMalloc_PageHeap::signalScavenger): (WTF::TCMallocStats::malloc): (WTF::TCMallocStats::free): (WTF::TCMallocStats::fastCalloc): (WTF::TCMallocStats::tryFastCalloc): (WTF::TCMallocStats::calloc): (WTF::TCMallocStats::fastRealloc): (WTF::TCMallocStats::tryFastRealloc): (WTF::TCMallocStats::realloc): * wtf/FastMalloc.h: (WTF::Internal::fastMallocValidationHeader): (WTF::Internal::fastMallocValidationSuffix): (WTF::Internal::fastMallocMatchValidationType): (WTF::Internal::setFastMallocMatchValidationType): (WTF::fastMallocMatchValidateFree): (WTF::fastMallocValidate): * wtf/Platform.h: 2011-04-04 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Renamed clearWeakPointers => updateWeakHandles and removed misleading comment https://bugs.webkit.org/show_bug.cgi?id=57790 * collector/handles/HandleHeap.cpp: (JSC::HandleHeap::updateWeakHandles): Updated for rename. * collector/handles/HandleHeap.h: Removed comment claiming that this function should only be called during teardown, because it's actually called after every GC pass. * runtime/Heap.cpp: (JSC::Heap::destroy): (JSC::Heap::markRoots): Updated for rename. 2011-04-04 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Standardized handling of handles for immediate values https://bugs.webkit.org/show_bug.cgi?id=57788 * collector/handles/HandleHeap.cpp: (JSC::HandleHeap::clearWeakPointers): Don't check for null or non-cell values here, because our write barrier guarantees that such values are not in the weak list. (JSC::HandleHeap::writeBarrier): Standardized on checking for null before checking for cell, and on using early return instead of if/else. * collector/handles/HandleHeap.h: (JSC::HandleHeap::deallocate): (JSC::HandleHeap::makeWeak): Ditto. 2011-04-04 Geoffrey Garen <ggaren@apple.com> Not reviewed. Removed a redundant variable from HandleHeap https://bugs.webkit.org/show_bug.cgi?id=57786 Forgot to commit the file that actually removes the data member! * collector/handles/HandleHeap.h: 2011-04-04 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed a redundant variable from HandleHeap https://bugs.webkit.org/show_bug.cgi?id=57786 We don't need a specific variable to indicate that we're in the middle of the finalization phase, since m_nextToFinalize already does this. * collector/handles/HandleHeap.cpp: (JSC::HandleHeap::HandleHeap): (JSC::HandleHeap::clearWeakPointers): (JSC::HandleHeap::writeBarrier): 2011-04-04 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Renamed Finalizer => WeakHandleOwner (in preparation for adding a reachability callback) https://bugs.webkit.org/show_bug.cgi?id=57775 Also renamed noFinalizer => emptyWeakOwner, since this is really an optimization for a weak owner with empty callbacks. * collector/handles/HandleHeap.cpp: (JSC::HandleHeap::clearWeakPointers): Updated for renames. Removed redundant initialization of m_nextToFinalize. Moved deletion check inside weak owner check, since the weak owner can't delete the node if there is no weak owner! * collector/handles/HandleHeap.h: (JSC::WeakHandleOwner::~WeakHandleOwner): (JSC::HandleHeap::makeWeak): Updated for renames. (JSC::HandleHeap::hasWeakOwner): Changed getFinalizer to hasWeakOwner, to clarify this function's role in assertions. (JSC::HandleHeap::Node::Node): (JSC::HandleHeap::Node::makeWeak): (JSC::HandleHeap::Node::isWeak): (JSC::HandleHeap::Node::weakOwner): (JSC::HandleHeap::Node::weakOwnerContext): (JSC::HandleHeap::Node::emptyWeakOwner): * interpreter/RegisterFile.cpp: (JSC::RegisterFile::globalObjectCollectedNotifier): * interpreter/RegisterFile.h: * runtime/WeakGCMap.h: * runtime/WeakGCPtr.h: (JSC::WeakGCPtr::WeakGCPtr): (JSC::WeakGCPtr::set): Updated for renames. 2011-04-04 Oliver Hunt <oliver@apple.com> Fix WinCE build. * bytecode/Instruction.h: * interpreter/Interpreter.cpp: (JSC::Interpreter::tryCachePutByID): (JSC::Interpreter::tryCacheGetByID): 2011-04-04 Adam Roben <aroben@apple.com> Delete mt.dep files when doing a clean build due to .vsprops file changes Apparently this is yet another file that Visual Studio can't figure out it needs to rebuild. Fixes <http://webkit.org/b/57777> r82850 failed to build on Windows Debug (Build) Reviewed by Brian Weinstein. * JavaScriptCore.vcproj/JavaScriptCore/react-to-vsprops-changes.py: (main): Added dep to the list of extensions we look for when choosing files to delete. 2011-04-01 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make StructureChain GC allocated https://bugs.webkit.org/show_bug.cgi?id=56695 Make StructureChain GC allocated, and make the various owners mark it correctly. * JavaScriptCore.exp: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): (JSC::CodeBlock::derefStructures): (JSC::CodeBlock::refStructures): (JSC::CodeBlock::markAggregate): * bytecode/Instruction.h: (JSC::PolymorphicAccessStructureList::PolymorphicStubInfo::set): (JSC::PolymorphicAccessStructureList::PolymorphicAccessStructureList): (JSC::PolymorphicAccessStructureList::derefStructures): (JSC::PolymorphicAccessStructureList::markAggregate): (JSC::Instruction::Instruction): * bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::deref): (JSC::StructureStubInfo::markAggregate): * bytecode/StructureStubInfo.h: (JSC::StructureStubInfo::initGetByIdChain): (JSC::StructureStubInfo::initPutByIdTransition): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitJumpIfNotFunctionCall): (JSC::BytecodeGenerator::emitJumpIfNotFunctionApply): * collector/handles/Handle.h: (JSC::HandleConverter::operator->): (JSC::HandleConverter::operator*): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_jneq_ptr): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_jneq_ptr): * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompileGetByIdChainList): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::privateCompileGetByIdChainList): * jit/JITStubs.cpp: (JSC::JITThunks::tryCachePutByID): (JSC::JITThunks::tryCacheGetByID): (JSC::getPolymorphicAccessStructureListSlot): (JSC::DEFINE_STUB_FUNCTION): * runtime/JSCell.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: * runtime/JSGlobalObject.cpp: (JSC::markIfNeeded): * runtime/JSGlobalObject.h: (JSC::Structure::prototypeChain): * runtime/JSObject.h: (JSC::JSObject::putDirectInternal): (JSC::JSObject::markChildrenDirect): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::create): (JSC::JSPropertyNameIterator::get): (JSC::JSPropertyNameIterator::markChildren): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::setCachedPrototypeChain): * runtime/JSZombie.cpp: (JSC::JSZombie::leakedZombieStructure): * runtime/JSZombie.h: * runtime/MarkStack.h: (JSC::MarkStack::append): * runtime/MarkedBlock.cpp: (JSC::MarkedBlock::sweep): * runtime/Structure.cpp: (JSC::Structure::addPropertyTransition): * runtime/Structure.h: (JSC::Structure::markAggregate): * runtime/StructureChain.cpp: (JSC::StructureChain::StructureChain): (JSC::StructureChain::~StructureChain): (JSC::StructureChain::markChildren): * runtime/StructureChain.h: (JSC::StructureChain::create): (JSC::StructureChain::createStructure): * runtime/WriteBarrier.h: (JSC::WriteBarrierBase::get): (JSC::WriteBarrierBase::operator*): (JSC::WriteBarrierBase::operator->): 2011-04-01 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed some complexity from HandleHeap https://bugs.webkit.org/show_bug.cgi?id=57650 Eliminated pointer-tagging flags. Tied being weak to having a finalizer (or at least a finalizer sentinel). * collector/handles/HandleHeap.cpp: (JSC::HandleHeap::clearWeakPointers): Removed the special self-destroying flag. It was unused. If we bring it back, we'll probably use a shared autodeallocating finalizer instead. * collector/handles/HandleHeap.h: (JSC::HandleHeap::makeWeak): makeWeak and adding a finalizer are now a single, atomic operation -- this makes the relationship between finalizers and weak pointers clearer, and impossible to get wrong. (JSC::HandleHeap::Node::Node): (JSC::HandleHeap::Node::handleHeap): No more flags. (JSC::HandleHeap::Node::makeWeak): (JSC::HandleHeap::Node::isWeak): Ditto above. We use a special sentienl value in the finalizer slot to indicate that a handle is weak but doesn't require an external function call for finalization. 2011-04-01 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed WeakGCMap::deprecatedRemove because it was deprecated and unused https://bugs.webkit.org/show_bug.cgi?id=57648 * runtime/WeakGCMap.h: 2011-04-01 Adam Roben <aroben@apple.com> Maintain the invariant that Lexer::m_current is set to -1 when at the end of the code buffer Covered by existing tests. Fixes <http://webkit.org/b/56699>. Reviewed by Oliver Hunt. * parser/Lexer.h: (JSC::Lexer::setOffset): Copied code from Lexer::shift to update m_current, because supposedly the idiom that function uses is fast. 2011-03-31 Thouraya ANDOLSI <thouraya.andolsi@st.com> Reviewed by Oliver Hunt. SH4 JIT SUPPORT. https://bugs.webkit.org/show_bug.cgi?id=44329 Add YARR support for SH4 platforms (disabled by default). * GNUmakefile.am: * assembler/MacroAssembler.h: * assembler/MacroAssemblerSH4.cpp: Added. * assembler/MacroAssemblerSH4.h: Added. * assembler/SH4Assembler.h: Added. * yarr/YarrJIT.cpp: 2011-03-30 Adam Roben <aroben@apple.com> Clean build fix * JavaScriptCore.vcproj/JavaScriptCore.sln: Serialized project dependencies so projects don't try to build in parallel (which doesn't mesh with our buildfailed mechanism). 2011-03-30 Oliver Hunt <oliver@apple.com> Rollout r82500 * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): (JSC::CodeBlock::derefStructures): (JSC::CodeBlock::refStructures): (JSC::CodeBlock::markAggregate): * bytecode/Instruction.h: (JSC::PolymorphicAccessStructureList::PolymorphicStubInfo::set): (JSC::PolymorphicAccessStructureList::PolymorphicAccessStructureList): (JSC::PolymorphicAccessStructureList::derefStructures): (JSC::Instruction::Instruction): * bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::deref): * bytecode/StructureStubInfo.h: (JSC::StructureStubInfo::initGetByIdChain): (JSC::StructureStubInfo::initPutByIdTransition): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitJumpIfNotFunctionCall): (JSC::BytecodeGenerator::emitJumpIfNotFunctionApply): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_jneq_ptr): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_jneq_ptr): * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompileGetByIdChainList): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::privateCompileGetByIdChainList): * jit/JITStubs.cpp: (JSC::getPolymorphicAccessStructureListSlot): (JSC::DEFINE_STUB_FUNCTION): * runtime/JSCell.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: * runtime/JSGlobalObject.cpp: (JSC::markIfNeeded): * runtime/JSGlobalObject.h: (JSC::Structure::prototypeChain): * runtime/JSObject.h: (JSC::JSObject::markChildrenDirect): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::create): (JSC::JSPropertyNameIterator::get): (JSC::JSPropertyNameIterator::markChildren): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::setCachedPrototypeChain): * runtime/MarkStack.h: (JSC::MarkStack::append): * runtime/Structure.h: * runtime/StructureChain.cpp: (JSC::StructureChain::StructureChain): * runtime/StructureChain.h: (JSC::StructureChain::create): 2011-03-29 Matthew Delaney <mdelaney@apple.com> Reviewed by Simon Fraser. Use the Accelerate vImage vectorized (un)premultiplyImageData functions for ImageBufferCG https://bugs.webkit.org/show_bug.cgi?id=53134 * wtf/Platform.h: Added in WTF flag for using the Accelerate framework 2011-03-30 Steve Falkenburg <sfalken@apple.com> Reviewed by Adam Roben. Share most vsprops between Release and Production builds in releaseproduction.vsprops https://bugs.webkit.org/show_bug.cgi?id=57508 * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreProduction.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreRelease.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleaseCairoCFLite.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleasePGO.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleasePGOOptimize.vsprops: * JavaScriptCore.vcproj/WTF/WTFProduction.vsprops: * JavaScriptCore.vcproj/WTF/WTFRelease.vsprops: * JavaScriptCore.vcproj/WTF/WTFReleaseCairoCFLite.vsprops: * JavaScriptCore.vcproj/WTF/WTFReleasePGO.vsprops: * JavaScriptCore.vcproj/jsc/jscProduction.vsprops: * JavaScriptCore.vcproj/jsc/jscRelease.vsprops: * JavaScriptCore.vcproj/jsc/jscReleaseCairoCFLite.vsprops: * JavaScriptCore.vcproj/jsc/jscReleasePGO.vsprops: * JavaScriptCore.vcproj/testapi/testapiProduction.vsprops: * JavaScriptCore.vcproj/testapi/testapiRelease.vsprops: * JavaScriptCore.vcproj/testapi/testapiReleaseCairoCFLite.vsprops: 2011-03-30 Mark Rowe <mrowe@apple.com> Reviewed by Adam Roben. Explicitly prevent testapi and minidom from being installed rather than relying on Xcode's current behavior of not installing if INSTALL_PATH is not explicitly set at the target level. <rdar://problem/9206357> * JavaScriptCore.xcodeproj/project.pbxproj: 2011-03-30 Timur Iskhodzhanov <timurrrr@google.com> Reviewed by Alexey Proskuryakov. Add some dynamic annotations to JavaScriptCore/wtf https://bugs.webkit.org/show_bug.cgi?id=53747 By using these annotations we can improve the precision of finding WebKit errors using dynamic analysis tools like ThreadSanitizer and Valgrind. These annotations don't affect the compiled binaries unless USE(DYNAMIC_ANNOTATIONS) is "1". These files don't add new functionality, so don't need extra tests. * GNUmakefile.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/CMakeLists.txt: * wtf/DynamicAnnotations.cpp: Added. (WTFAnnotateBenignRaceSized): (WTFAnnotateHappensBefore): (WTFAnnotateHappensAfter): * wtf/DynamicAnnotations.h: Added. * wtf/ThreadSafeRefCounted.h: (WTF::ThreadSafeRefCountedBase::derefBase): * wtf/text/StringStatics.cpp: (WTF::StringImpl::empty): 2011-03-30 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make StructureChain GC allocated https://bugs.webkit.org/show_bug.cgi?id=56695 Make StructureChain GC allocated, and make the various owners mark it correctly. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): (JSC::CodeBlock::derefStructures): (JSC::CodeBlock::refStructures): (JSC::CodeBlock::markAggregate): * bytecode/Instruction.h: (JSC::PolymorphicAccessStructureList::PolymorphicStubInfo::set): (JSC::PolymorphicAccessStructureList::PolymorphicAccessStructureList): (JSC::PolymorphicAccessStructureList::derefStructures): (JSC::PolymorphicAccessStructureList::markAggregate): (JSC::Instruction::Instruction): * bytecode/StructureStubInfo.cpp: (JSC::StructureStubInfo::deref): (JSC::StructureStubInfo::markAggregate): * bytecode/StructureStubInfo.h: (JSC::StructureStubInfo::initGetByIdChain): (JSC::StructureStubInfo::initPutByIdTransition): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitJumpIfNotFunctionCall): (JSC::BytecodeGenerator::emitJumpIfNotFunctionApply): * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_jneq_ptr): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_jneq_ptr): * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompileGetByIdChainList): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::privateCompileGetByIdChainList): * jit/JITStubs.cpp: (JSC::getPolymorphicAccessStructureListSlot): (JSC::DEFINE_STUB_FUNCTION): * runtime/JSCell.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: * runtime/JSGlobalObject.cpp: (JSC::markIfNeeded): * runtime/JSGlobalObject.h: (JSC::Structure::prototypeChain): * runtime/JSObject.h: (JSC::JSObject::markChildrenDirect): * runtime/JSPropertyNameIterator.cpp: (JSC::JSPropertyNameIterator::create): (JSC::JSPropertyNameIterator::get): (JSC::JSPropertyNameIterator::markChildren): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::setCachedPrototypeChain): * runtime/MarkStack.h: (JSC::MarkStack::append): * runtime/Structure.h: (JSC::Structure::cachedPrototypeChainSlot): * runtime/StructureChain.cpp: (JSC::StructureChain::StructureChain): * runtime/StructureChain.h: (JSC::StructureChain::create): (JSC::StructureChain::createStructure): 2011-03-30 Steve Falkenburg <sfalken@apple.com> Reviewed by Adam Roben. Update Windows production build logic for new production configurations https://bugs.webkit.org/show_bug.cgi?id=57494 * JavaScriptCore.vcproj/JavaScriptCore.make: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreProduction.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleasePGO.vsprops: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleasePGOOptimize.vsprops: * JavaScriptCore.vcproj/WTF/WTFProduction.vsprops: * JavaScriptCore.vcproj/WTF/WTFReleasePGO.vsprops: * JavaScriptCore.vcproj/jsc/jscProduction.vsprops: * JavaScriptCore.vcproj/jsc/jscReleasePGO.vsprops: * JavaScriptCore.vcproj/testapi/testapiProduction.vsprops: 2011-03-30 Steve Falkenburg <sfalken@apple.com> Reviewed by Adam Roben. Rename Windows configuration Release_LTCG to Production for clarity https://bugs.webkit.org/show_bug.cgi?id=57465 * JavaScriptCore.vcproj/JavaScriptCore.sln: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.vcproj: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreProduction.vsprops: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleaseLTCG.vsprops. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleaseLTCG.vsprops: Removed. * JavaScriptCore.vcproj/JavaScriptCoreSubmit.sln: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.vcproj/WTF/WTFProduction.vsprops: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/WTF/WTFReleaseLTCG.vsprops. * JavaScriptCore.vcproj/WTF/WTFReleaseLTCG.vsprops: Removed. * JavaScriptCore.vcproj/jsc/jsc.vcproj: * JavaScriptCore.vcproj/jsc/jscProduction.vsprops: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/jsc/jscReleaseLTCG.vsprops. * JavaScriptCore.vcproj/jsc/jscReleaseLTCG.vsprops: Removed. * JavaScriptCore.vcproj/testapi/testapi.vcproj: * JavaScriptCore.vcproj/testapi/testapiProduction.vsprops: Copied from Source/JavaScriptCore/JavaScriptCore.vcproj/testapi/testapiReleaseLTCG.vsprops. * JavaScriptCore.vcproj/testapi/testapiReleaseLTCG.vsprops: Removed. 2011-03-30 Zoltan Herczeg <zherczeg@inf.u-szeged.hu> Reviewed by Maciej Stachowiak. Add the NEXT_OPCODE() macro to the DFG-JIT parser https://bugs.webkit.org/show_bug.cgi?id=57322 In JavaScriptCore we use macros to jump to the next opcode (both in interpreter and JIT). This macro is added to the DFG-JIT parser as well. * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::parse): 2011-03-29 Geoffrey Garen <ggaren@apple.com> Reviewed by Darin Adler. ~25% regression on v8-splay in the SunSpider harness https://bugs.webkit.org/show_bug.cgi?id=56128 I'm not sure if this is the root cause of the regression Stephanie measured, but it seems to get us back to previous v8-splay times. SunSpider reports no change. v8-splay says 41% faster. * runtime/Heap.cpp: (JSC::Heap::reset): Make marking proportional to 1X the size of the heap, not .5X the size of the heap. When the heap is large, this makes a big difference. (Our old heap growth policy matched this. You can see by looking at resizeBlocks in revisions prior to r77699.) 2011-03-29 Steve Falkenburg <sfalken@apple.com> Reviewed by Darin Adler. Use per-configuration vsprops in JavaScriptCore to avoid WebKitVSPropsRedirectionDir removal by MSVC IDE https://bugs.webkit.org/show_bug.cgi?id=57350 Visual Studio's IDE was removing instances of $(WebKitVSPropsRedirectionDir) from InheritedPropertySheet rules in our vcproj files when the vcproj was edited from within the IDE. To avoid this, add a separate vsprops file for each project configuration that contains the required inherited property sheets. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreDebug.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreDebugAll.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreDebugCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreRelease.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleaseCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleaseLTCG.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleasePGO.vsprops: Added. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreReleasePGOOptimize.vsprops: Added. * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.vcproj/WTF/WTFDebug.vsprops: Added. * JavaScriptCore.vcproj/WTF/WTFDebugAll.vsprops: Added. * JavaScriptCore.vcproj/WTF/WTFDebugCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/WTF/WTFRelease.vsprops: Added. * JavaScriptCore.vcproj/WTF/WTFReleaseCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/WTF/WTFReleaseLTCG.vsprops: Added. * JavaScriptCore.vcproj/WTF/WTFReleasePGO.vsprops: Added. * JavaScriptCore.vcproj/jsc/jsc.vcproj: * JavaScriptCore.vcproj/jsc/jscDebug.vsprops: Added. * JavaScriptCore.vcproj/jsc/jscDebugAll.vsprops: Added. * JavaScriptCore.vcproj/jsc/jscDebugCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/jsc/jscRelease.vsprops: Added. * JavaScriptCore.vcproj/jsc/jscReleaseCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/jsc/jscReleaseLTCG.vsprops: Added. * JavaScriptCore.vcproj/jsc/jscReleasePGO.vsprops: Added. * JavaScriptCore.vcproj/testapi/testapi.vcproj: * JavaScriptCore.vcproj/testapi/testapiDebug.vsprops: Added. * JavaScriptCore.vcproj/testapi/testapiDebugAll.vsprops: Added. * JavaScriptCore.vcproj/testapi/testapiDebugCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/testapi/testapiRelease.vsprops: Added. * JavaScriptCore.vcproj/testapi/testapiReleaseCairoCFLite.vsprops: Added. * JavaScriptCore.vcproj/testapi/testapiReleaseLTCG.vsprops: Added. 2011-03-29 Oliver Hunt <oliver@apple.com> Reviewed by Eric Seidel. REGRESSION(r82173): Causes assertion and test failures in run-javascriptcore-tests on Windows (Requested by aroben on #webkit). https://bugs.webkit.org/show_bug.cgi?id=57333 constructDate now takes the global object explicitly as it may be called by functions other than the constructor itself. * API/JSObjectRef.cpp: (JSObjectMakeDate): * runtime/DateConstructor.cpp: (JSC::constructDate): (JSC::constructWithDateConstructor): * runtime/DateConstructor.h: 2011-03-29 Ben Taylor <bentaylor.solx86@gmail.com> Reviewed by Benjamin Poulain. https://bugs.webkit.org/show_bug.cgi?id=41953 Fix compile error on Solaris 10/Sun Studio 12 CC emanating from MathExtras.h * wtf/MathExtras.h: 2011-03-29 Ben Taylor <bentaylor.solx86@gmail.com> Reviewed by Darin Adler. https://bugs.webkit.org/show_bug.cgi?id=57231 Add conditional for SUNCC supporting alignment macros Compile fix for Solaris 10/Sun Studio 12 CC * wtf/Vector.h: 2011-03-29 Ben Taylor <bentaylor.solx86@gmail.com> Reviewed by Darin Adler. https://bugs.webkit.org/show_bug.cgi?id=57256 Fix crash on misaligned reads on Solaris 10/Sparc * wtf/text/AtomicString.cpp: (WTF::equal): 2011-03-28 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. instanceof Array test fails when using iframes https://bugs.webkit.org/show_bug.cgi?id=17250 This is a problem with all built in constructors, the use of lexicalGlobalObject rather than the constructors own global object reference means that a builtin will always use the prototype from the lexical global object rather than that of the constructors origin. * API/JSObjectRef.cpp: (JSObjectMakeFunction): (JSObjectMakeRegExp): * JavaScriptCore.exp: * runtime/ArrayConstructor.cpp: (JSC::constructArrayWithSizeQuirk): * runtime/BooleanConstructor.cpp: (JSC::constructBoolean): (JSC::constructBooleanFromImmediateBoolean): * runtime/BooleanConstructor.h: * runtime/DateConstructor.cpp: (JSC::constructDate): * runtime/DateInstance.cpp: * runtime/DateInstance.h: * runtime/ErrorConstructor.cpp: (JSC::constructWithErrorConstructor): (JSC::callErrorConstructor): * runtime/FunctionConstructor.cpp: (JSC::constructWithFunctionConstructor): (JSC::callFunctionConstructor): (JSC::constructFunction): * runtime/FunctionConstructor.h: * runtime/JSCell.cpp: (JSC::JSCell::getOwnPropertySlot): (JSC::JSCell::put): (JSC::JSCell::deleteProperty): (JSC::JSCell::toThisObject): (JSC::JSCell::toObject): * runtime/JSCell.h: (JSC::JSCell::JSValue::toObject): * runtime/JSNotAnObject.cpp: (JSC::JSNotAnObject::toObject): * runtime/JSNotAnObject.h: * runtime/JSObject.cpp: (JSC::JSObject::toObject): * runtime/JSObject.h: * runtime/JSString.cpp: (JSC::StringObject::create): (JSC::JSString::toObject): (JSC::JSString::toThisObject): * runtime/JSString.h: * runtime/JSValue.cpp: (JSC::JSValue::toObjectSlowCase): (JSC::JSValue::toThisObjectSlowCase): (JSC::JSValue::synthesizeObject): * runtime/JSValue.h: * runtime/NumberConstructor.cpp: (JSC::constructWithNumberConstructor): * runtime/NumberObject.cpp: (JSC::constructNumber): * runtime/NumberObject.h: * runtime/ObjectConstructor.cpp: (JSC::constructObject): (JSC::constructWithObjectConstructor): (JSC::callObjectConstructor): * runtime/RegExpConstructor.cpp: (JSC::constructRegExp): (JSC::constructWithRegExpConstructor): (JSC::callRegExpConstructor): * runtime/RegExpConstructor.h: * runtime/StringConstructor.cpp: (JSC::constructWithStringConstructor): * runtime/StringObject.h: 2011-03-28 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. REGRESSION [r78794-r79249] Allocation of memory is slow when number of active objects is large https://bugs.webkit.org/show_bug.cgi?id=56823 Partial fix for most of the problem. (TOT still shows a regression, though.) * runtime/Heap.cpp: (JSC::Heap::reportExtraMemoryCostSlowCase): Use highWaterMark(), instead of capacity(), since capacity() is O(n) relative to the size of the heap. In limited circumstances, capacity() is also worse than highWaterMark() for measuring extra cost relative to heap size, since capacity() only measures the *current* capacity of the heap, but the heap will grow if necessary to attain highWaterMark(). 2011-03-28 Oliver Hunt <oliver@apple.com> REGRESSION(r82130): It made all tests crash (Requested by Ossy on #webkit). https://bugs.webkit.org/show_bug.cgi?id=57251 Build fix, had remnant of another patch in r82130 * jit/JITPropertyAccess.cpp: (JSC::JIT::privateCompileGetByIdChainList): 2011-03-27 Oliver Hunt <oliver@apple.com> Reviewed by Maciej Stachowiak. Add additional immediate types to allow us to distinguish the source of a JIT immediate https://bugs.webkit.org/show_bug.cgi?id=57190 Allow us to distinguish whether a JIT immediate is a value that we control (TrustedImm32 and TrustedImmPtr) vs. ones that can be controlled or influenced by code we are compiling. Currently we do nothing with this information -- this change is large and mechanical but would obscure any logic changes that we would have made. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::TrustedImmPtr::TrustedImmPtr): (JSC::AbstractMacroAssembler::ImmPtr::ImmPtr): (JSC::AbstractMacroAssembler::TrustedImm32::TrustedImm32): (JSC::AbstractMacroAssembler::Imm32::Imm32): * assembler/MacroAssembler.h: (JSC::MacroAssembler::pop): (JSC::MacroAssembler::poke): (JSC::MacroAssembler::branchPtr): (JSC::MacroAssembler::branch32): (JSC::MacroAssembler::addPtr): (JSC::MacroAssembler::andPtr): (JSC::MacroAssembler::orPtr): (JSC::MacroAssembler::subPtr): (JSC::MacroAssembler::xorPtr): (JSC::MacroAssembler::setPtr): (JSC::MacroAssembler::storePtr): (JSC::MacroAssembler::branchTestPtr): (JSC::MacroAssembler::branchSubPtr): (JSC::MacroAssembler::branchTest8): * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::add32): (JSC::MacroAssemblerARM::and32): (JSC::MacroAssemblerARM::lshift32): (JSC::MacroAssemblerARM::mul32): (JSC::MacroAssemblerARM::or32): (JSC::MacroAssemblerARM::rshift32): (JSC::MacroAssemblerARM::urshift32): (JSC::MacroAssemblerARM::sub32): (JSC::MacroAssemblerARM::xor32): (JSC::MacroAssemblerARM::store32): (JSC::MacroAssemblerARM::push): (JSC::MacroAssemblerARM::move): (JSC::MacroAssemblerARM::branch8): (JSC::MacroAssemblerARM::branch32): (JSC::MacroAssemblerARM::branch32WithUnalignedHalfWords): (JSC::MacroAssemblerARM::branch16): (JSC::MacroAssemblerARM::branchTest8): (JSC::MacroAssemblerARM::branchTest32): (JSC::MacroAssemblerARM::branchAdd32): (JSC::MacroAssemblerARM::branchMul32): (JSC::MacroAssemblerARM::branchSub32): (JSC::MacroAssemblerARM::set32Compare32): (JSC::MacroAssemblerARM::set8Compare32): (JSC::MacroAssemblerARM::set32Test32): (JSC::MacroAssemblerARM::set32Test8): (JSC::MacroAssemblerARM::moveWithPatch): (JSC::MacroAssemblerARM::branchPtrWithPatch): (JSC::MacroAssemblerARM::storePtrWithPatch): * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::add32): (JSC::MacroAssemblerARMv7::and32): (JSC::MacroAssemblerARMv7::lshift32): (JSC::MacroAssemblerARMv7::mul32): (JSC::MacroAssemblerARMv7::or32): (JSC::MacroAssemblerARMv7::rshift32): (JSC::MacroAssemblerARMv7::urshift32): (JSC::MacroAssemblerARMv7::sub32): (JSC::MacroAssemblerARMv7::xor32): (JSC::MacroAssemblerARMv7::load32): (JSC::MacroAssemblerARMv7::load32WithAddressOffsetPatch): (JSC::MacroAssemblerARMv7::load16): (JSC::MacroAssemblerARMv7::store32WithAddressOffsetPatch): (JSC::MacroAssemblerARMv7::store32): (JSC::MacroAssemblerARMv7::loadDouble): (JSC::MacroAssemblerARMv7::storeDouble): (JSC::MacroAssemblerARMv7::push): (JSC::MacroAssemblerARMv7::move): (JSC::MacroAssemblerARMv7::compare32): (JSC::MacroAssemblerARMv7::test32): (JSC::MacroAssemblerARMv7::branch32): (JSC::MacroAssemblerARMv7::branch32WithUnalignedHalfWords): (JSC::MacroAssemblerARMv7::branch16): (JSC::MacroAssemblerARMv7::branch8): (JSC::MacroAssemblerARMv7::branchTest32): (JSC::MacroAssemblerARMv7::branchTest8): (JSC::MacroAssemblerARMv7::branchAdd32): (JSC::MacroAssemblerARMv7::branchMul32): (JSC::MacroAssemblerARMv7::branchSub32): (JSC::MacroAssemblerARMv7::nearCall): (JSC::MacroAssemblerARMv7::call): (JSC::MacroAssemblerARMv7::set32Compare32): (JSC::MacroAssemblerARMv7::set8Compare32): (JSC::MacroAssemblerARMv7::set32Test32): (JSC::MacroAssemblerARMv7::set32Test8): (JSC::MacroAssemblerARMv7::moveWithPatch): (JSC::MacroAssemblerARMv7::branchPtrWithPatch): (JSC::MacroAssemblerARMv7::storePtrWithPatch): (JSC::MacroAssemblerARMv7::tailRecursiveCall): (JSC::MacroAssemblerARMv7::makeJump): (JSC::MacroAssemblerARMv7::makeBranch): (JSC::MacroAssemblerARMv7::setupArmAddress): (JSC::MacroAssemblerARMv7::makeBaseIndexBase): (JSC::MacroAssemblerARMv7::moveFixedWidthEncoding): * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::add32): (JSC::MacroAssemblerMIPS::and32): (JSC::MacroAssemblerMIPS::lshift32): (JSC::MacroAssemblerMIPS::mul32): (JSC::MacroAssemblerMIPS::or32): (JSC::MacroAssemblerMIPS::rshift32): (JSC::MacroAssemblerMIPS::urshift32): (JSC::MacroAssemblerMIPS::sub32): (JSC::MacroAssemblerMIPS::xor32): (JSC::MacroAssemblerMIPS::load32): (JSC::MacroAssemblerMIPS::load32WithAddressOffsetPatch): (JSC::MacroAssemblerMIPS::store32WithAddressOffsetPatch): (JSC::MacroAssemblerMIPS::store32): (JSC::MacroAssemblerMIPS::push): (JSC::MacroAssemblerMIPS::move): (JSC::MacroAssemblerMIPS::branch8): (JSC::MacroAssemblerMIPS::branch32): (JSC::MacroAssemblerMIPS::branch32WithUnalignedHalfWords): (JSC::MacroAssemblerMIPS::branch16): (JSC::MacroAssemblerMIPS::branchTest32): (JSC::MacroAssemblerMIPS::branchTest8): (JSC::MacroAssemblerMIPS::branchAdd32): (JSC::MacroAssemblerMIPS::branchMul32): (JSC::MacroAssemblerMIPS::branchSub32): (JSC::MacroAssemblerMIPS::set8Compare32): (JSC::MacroAssemblerMIPS::set32Compare32): (JSC::MacroAssemblerMIPS::set32Test8): (JSC::MacroAssemblerMIPS::set32Test32): (JSC::MacroAssemblerMIPS::moveWithPatch): (JSC::MacroAssemblerMIPS::branchPtrWithPatch): (JSC::MacroAssemblerMIPS::storePtrWithPatch): (JSC::MacroAssemblerMIPS::tailRecursiveCall): (JSC::MacroAssemblerMIPS::loadDouble): (JSC::MacroAssemblerMIPS::storeDouble): (JSC::MacroAssemblerMIPS::branchTruncateDoubleToInt32): * assembler/MacroAssemblerX86.h: (JSC::MacroAssemblerX86::add32): (JSC::MacroAssemblerX86::addWithCarry32): (JSC::MacroAssemblerX86::and32): (JSC::MacroAssemblerX86::or32): (JSC::MacroAssemblerX86::sub32): (JSC::MacroAssemblerX86::store32): (JSC::MacroAssemblerX86::branch32): (JSC::MacroAssemblerX86::moveWithPatch): (JSC::MacroAssemblerX86::branchPtrWithPatch): (JSC::MacroAssemblerX86::storePtrWithPatch): * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::add32): (JSC::MacroAssemblerX86Common::and32): (JSC::MacroAssemblerX86Common::lshift32): (JSC::MacroAssemblerX86Common::mul32): (JSC::MacroAssemblerX86Common::or32): (JSC::MacroAssemblerX86Common::rshift32): (JSC::MacroAssemblerX86Common::urshift32): (JSC::MacroAssemblerX86Common::sub32): (JSC::MacroAssemblerX86Common::xor32): (JSC::MacroAssemblerX86Common::store32): (JSC::MacroAssemblerX86Common::branchTruncateDoubleToInt32): (JSC::MacroAssemblerX86Common::push): (JSC::MacroAssemblerX86Common::move): (JSC::MacroAssemblerX86Common::branch8): (JSC::MacroAssemblerX86Common::branch32): (JSC::MacroAssemblerX86Common::branch32WithUnalignedHalfWords): (JSC::MacroAssemblerX86Common::branch16): (JSC::MacroAssemblerX86Common::branchTest32): (JSC::MacroAssemblerX86Common::branchTest8): (JSC::MacroAssemblerX86Common::branchAdd32): (JSC::MacroAssemblerX86Common::branchMul32): (JSC::MacroAssemblerX86Common::branchSub32): (JSC::MacroAssemblerX86Common::set8Compare32): (JSC::MacroAssemblerX86Common::set32Compare32): (JSC::MacroAssemblerX86Common::set32Test8): (JSC::MacroAssemblerX86Common::set32Test32): * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::add32): (JSC::MacroAssemblerX86_64::and32): (JSC::MacroAssemblerX86_64::or32): (JSC::MacroAssemblerX86_64::sub32): (JSC::MacroAssemblerX86_64::loadDouble): (JSC::MacroAssemblerX86_64::addDouble): (JSC::MacroAssemblerX86_64::convertInt32ToDouble): (JSC::MacroAssemblerX86_64::store32): (JSC::MacroAssemblerX86_64::call): (JSC::MacroAssemblerX86_64::tailRecursiveCall): (JSC::MacroAssemblerX86_64::makeTailRecursiveCall): (JSC::MacroAssemblerX86_64::addPtr): (JSC::MacroAssemblerX86_64::andPtr): (JSC::MacroAssemblerX86_64::orPtr): (JSC::MacroAssemblerX86_64::subPtr): (JSC::MacroAssemblerX86_64::xorPtr): (JSC::MacroAssemblerX86_64::storePtr): (JSC::MacroAssemblerX86_64::setPtr): (JSC::MacroAssemblerX86_64::branchPtr): (JSC::MacroAssemblerX86_64::branchTestPtr): (JSC::MacroAssemblerX86_64::branchSubPtr): (JSC::MacroAssemblerX86_64::moveWithPatch): (JSC::MacroAssemblerX86_64::branchPtrWithPatch): (JSC::MacroAssemblerX86_64::storePtrWithPatch): (JSC::MacroAssemblerX86_64::branchTest8): * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::callOperation): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::jitAssertIsInt32): (JSC::DFG::JITCompiler::emitCount): * dfg/DFGJITCompiler.h: (JSC::DFG::JITCompiler::emitPutImmediateToCallFrameHeader): * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::fillSpeculateCell): (JSC::DFG::SpeculativeJIT::compile): * jit/JIT.cpp: (JSC::JIT::emitTimeoutCheck): (JSC::JIT::privateCompile): * jit/JIT.h: * jit/JITArithmetic.cpp: (JSC::JIT::emit_op_urshift): (JSC::JIT::emitSlow_op_urshift): (JSC::JIT::emit_op_post_inc): (JSC::JIT::emit_op_post_dec): (JSC::JIT::emit_op_pre_inc): (JSC::JIT::emit_op_pre_dec): (JSC::JIT::emit_op_mod): * jit/JITArithmetic32_64.cpp: (JSC::JIT::emit_op_negate): (JSC::JIT::emit_op_jnless): (JSC::JIT::emit_op_jless): (JSC::JIT::emit_op_jlesseq): (JSC::JIT::emit_op_lshift): (JSC::JIT::emitRightShift): (JSC::JIT::emitRightShiftSlowCase): (JSC::JIT::emit_op_bitand): (JSC::JIT::emit_op_bitor): (JSC::JIT::emit_op_bitxor): (JSC::JIT::emit_op_bitnot): (JSC::JIT::emit_op_post_inc): (JSC::JIT::emit_op_post_dec): (JSC::JIT::emitSlow_op_post_dec): (JSC::JIT::emit_op_pre_inc): (JSC::JIT::emit_op_pre_dec): (JSC::JIT::emit_op_add): (JSC::JIT::emitAdd32Constant): (JSC::JIT::emit_op_sub): (JSC::JIT::emitSub32Constant): (JSC::JIT::emitBinaryDoubleOp): (JSC::JIT::emit_op_mul): (JSC::JIT::emitSlow_op_mul): (JSC::JIT::emit_op_div): (JSC::JIT::emit_op_mod): * jit/JITCall.cpp: (JSC::JIT::compileOpCallVarargs): (JSC::JIT::compileOpCall): (JSC::JIT::compileOpCallSlowCase): * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCallVarargs): (JSC::JIT::emit_op_ret_object_or_this): (JSC::JIT::compileOpCall): (JSC::JIT::compileOpCallSlowCase): * jit/JITInlineMethods.h: (JSC::JIT::emitPutCellToCallFrameHeader): (JSC::JIT::emitPutIntToCallFrameHeader): (JSC::JIT::emitPutImmediateToCallFrameHeader): (JSC::JIT::emitLoadCharacterString): (JSC::JIT::restoreArgumentReferenceForTrampoline): (JSC::JIT::checkStructure): (JSC::JIT::setSamplingFlag): (JSC::JIT::clearSamplingFlag): (JSC::JIT::emitCount): (JSC::JIT::sampleInstruction): (JSC::JIT::sampleCodeBlock): (JSC::JIT::emitStoreInt32): (JSC::JIT::emitStoreCell): (JSC::JIT::emitStoreBool): (JSC::JIT::emitJumpSlowCaseIfNotJSCell): (JSC::JIT::emitInitRegister): (JSC::JIT::emitJumpIfJSCell): (JSC::JIT::emitJumpIfNotJSCell): (JSC::JIT::emitJumpIfImmediateInteger): (JSC::JIT::emitJumpIfNotImmediateInteger): (JSC::JIT::emitFastArithDeTagImmediate): (JSC::JIT::emitFastArithDeTagImmediateJumpIfZero): (JSC::JIT::emitFastArithReTagImmediate): (JSC::JIT::emitTagAsBoolImmediate): * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::privateCompileCTINativeCall): (JSC::JIT::emit_op_check_has_instance): (JSC::JIT::emit_op_instanceof): (JSC::JIT::emit_op_ret_object_or_this): (JSC::JIT::emit_op_resolve): (JSC::JIT::emit_op_to_primitive): (JSC::JIT::emit_op_resolve_base): (JSC::JIT::emit_op_ensure_property_exists): (JSC::JIT::emit_op_resolve_skip): (JSC::JIT::emit_op_resolve_global): (JSC::JIT::emitSlow_op_resolve_global): (JSC::JIT::emit_op_not): (JSC::JIT::emit_op_jfalse): (JSC::JIT::emit_op_jeq_null): (JSC::JIT::emit_op_jneq_null): (JSC::JIT::emit_op_jneq_ptr): (JSC::JIT::emit_op_jsr): (JSC::JIT::emit_op_resolve_with_base): (JSC::JIT::emit_op_new_func_exp): (JSC::JIT::emit_op_jtrue): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_next_pname): (JSC::JIT::emit_op_to_jsnumber): (JSC::JIT::emit_op_push_new_scope): (JSC::JIT::emit_op_catch): (JSC::JIT::emit_op_eq_null): (JSC::JIT::emit_op_neq_null): (JSC::JIT::emit_op_init_lazy_reg): (JSC::JIT::emit_op_convert_this): (JSC::JIT::emit_op_convert_this_strict): (JSC::JIT::emitSlow_op_not): (JSC::JIT::emitSlow_op_neq): (JSC::JIT::emit_op_get_arguments_length): (JSC::JIT::emitSlow_op_get_arguments_length): (JSC::JIT::emit_op_get_argument_by_val): (JSC::JIT::emitSlow_op_resolve_global_dynamic): (JSC::JIT::emit_op_new_regexp): (JSC::JIT::emit_op_load_varargs): (JSC::JIT::emitSlow_op_load_varargs): (JSC::JIT::emit_op_new_func): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTIMachineTrampolines): (JSC::JIT::privateCompileCTINativeCall): (JSC::JIT::emit_op_loop_if_lesseq): (JSC::JIT::emit_op_check_has_instance): (JSC::JIT::emit_op_instanceof): (JSC::JIT::emit_op_get_scoped_var): (JSC::JIT::emit_op_put_scoped_var): (JSC::JIT::emit_op_tear_off_activation): (JSC::JIT::emit_op_tear_off_arguments): (JSC::JIT::emit_op_resolve): (JSC::JIT::emit_op_to_primitive): (JSC::JIT::emit_op_resolve_base): (JSC::JIT::emit_op_ensure_property_exists): (JSC::JIT::emit_op_resolve_skip): (JSC::JIT::emit_op_resolve_global): (JSC::JIT::emitSlow_op_resolve_global): (JSC::JIT::emit_op_not): (JSC::JIT::emit_op_jfalse): (JSC::JIT::emit_op_jtrue): (JSC::JIT::emit_op_jeq_null): (JSC::JIT::emit_op_jneq_null): (JSC::JIT::emit_op_jneq_ptr): (JSC::JIT::emit_op_jsr): (JSC::JIT::emit_op_eq): (JSC::JIT::emitSlow_op_eq): (JSC::JIT::emit_op_neq): (JSC::JIT::emitSlow_op_neq): (JSC::JIT::compileOpStrictEq): (JSC::JIT::emit_op_eq_null): (JSC::JIT::emit_op_neq_null): (JSC::JIT::emit_op_resolve_with_base): (JSC::JIT::emit_op_new_func_exp): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_next_pname): (JSC::JIT::emit_op_to_jsnumber): (JSC::JIT::emit_op_push_new_scope): (JSC::JIT::emit_op_catch): (JSC::JIT::emit_op_create_activation): (JSC::JIT::emit_op_create_arguments): (JSC::JIT::emit_op_convert_this): (JSC::JIT::emit_op_convert_this_strict): (JSC::JIT::emit_op_get_arguments_length): (JSC::JIT::emitSlow_op_get_arguments_length): (JSC::JIT::emit_op_get_argument_by_val): (JSC::JIT::softModulo): * jit/JITPropertyAccess.cpp: (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::emit_op_get_by_val): (JSC::JIT::emitSlow_op_get_by_val): (JSC::JIT::emit_op_get_by_pname): (JSC::JIT::emit_op_put_by_val): (JSC::JIT::emit_op_put_by_index): (JSC::JIT::emit_op_put_getter): (JSC::JIT::emit_op_put_setter): (JSC::JIT::emit_op_del_by_id): (JSC::JIT::emit_op_get_by_id): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::emit_op_method_check): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::compileGetByIdSlowCase): (JSC::JIT::emitSlow_op_put_by_id): (JSC::JIT::testPrototype): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_put_getter): (JSC::JIT::emit_op_put_setter): (JSC::JIT::emit_op_del_by_id): (JSC::JIT::emit_op_get_by_id): (JSC::JIT::emit_op_put_by_id): (JSC::JIT::emit_op_method_check): (JSC::JIT::stringGetByValStubGenerator): (JSC::JIT::emit_op_get_by_val): (JSC::JIT::emitSlow_op_get_by_val): (JSC::JIT::emit_op_put_by_val): (JSC::JIT::compileGetByIdHotPath): (JSC::JIT::compileGetByIdSlowCase): (JSC::JIT::emitSlow_op_put_by_id): (JSC::JIT::testPrototype): (JSC::JIT::privateCompilePutByIdTransition): (JSC::JIT::privateCompilePatchGetArrayLength): (JSC::JIT::privateCompileGetByIdProto): (JSC::JIT::privateCompileGetByIdSelfList): (JSC::JIT::privateCompileGetByIdProtoList): (JSC::JIT::privateCompileGetByIdChainList): (JSC::JIT::privateCompileGetByIdChain): (JSC::JIT::emit_op_get_by_pname): * jit/JITStubCall.h: (JSC::JITStubCall::addArgument): * jit/JITStubs.cpp: (JSC::getPolymorphicAccessStructureListSlot): (JSC::DEFINE_STUB_FUNCTION): * jit/JSInterfaceJIT.h: (JSC::JSInterfaceJIT::emitJumpIfNotJSCell): (JSC::JSInterfaceJIT::emitLoadInt32): (JSC::JSInterfaceJIT::emitLoadDouble): * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::SpecializedThunkJIT): (JSC::SpecializedThunkJIT::loadJSStringArgument): (JSC::SpecializedThunkJIT::tagReturnAsInt32): (JSC::SpecializedThunkJIT::tagReturnAsJSCell): * jit/ThunkGenerators.cpp: (JSC::charToString): (JSC::powThunkGenerator): * yarr/YarrJIT.cpp: (JSC::Yarr::YarrGenerator::matchCharacterClass): (JSC::Yarr::YarrGenerator::storeToFrame): (JSC::Yarr::YarrGenerator::storeToFrameWithPatch): (JSC::Yarr::YarrGenerator::ParenthesesTail::generateCode): (JSC::Yarr::YarrGenerator::generatePatternCharacterSingle): (JSC::Yarr::YarrGenerator::generatePatternCharacterFixed): (JSC::Yarr::YarrGenerator::generatePatternCharacterGreedy): (JSC::Yarr::YarrGenerator::generatePatternCharacterNonGreedy): (JSC::Yarr::YarrGenerator::generateCharacterClassFixed): (JSC::Yarr::YarrGenerator::generateCharacterClassGreedy): (JSC::Yarr::YarrGenerator::generateCharacterClassNonGreedy): (JSC::Yarr::YarrGenerator::generateParenthesesSingle): (JSC::Yarr::YarrGenerator::generateDisjunction): 2011-03-28 Andras Becsi <abecsi@webkit.org> Reviewed by Csaba Osztrogonác. [Qt] Fix the linking of jsc with MinGW after r81963. * jsc.pro: add -l and remove the lib suffix. 2011-03-27 Ben Taylor <bentaylor.solx86@gmail.com> Reviewed by Alexey Proskuryakov. https://bugs.webkit.org/show_bug.cgi?id=57170 Fix last elements in an enum to remove a trailing comma. Sun Studio 12 CC errors out. Compile fix only, no actual code change. * wtf/MessageQueue.h: 2011-03-25 Oliver Hunt <oliver@apple.com> Reviewed by Darin Adler. Allow defineOwnProperty to work on DOMObjects https://bugs.webkit.org/show_bug.cgi?id=57129 Fix a couple of places where we uses getter()/setter() rather than [gs]etterPresent(). * runtime/JSObject.cpp: (JSC::JSObject::defineOwnProperty): 2011-03-25 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Crash when paused at a breakpoint caused by inaccurate Activation records. https://bugs.webkit.org/show_bug.cgi?id=57120 * runtime/JSActivation.cpp: (JSC::JSActivation::symbolTableGet): (JSC::JSActivation::symbolTablePut): (JSC::JSActivation::getOwnPropertyNames): (JSC::JSActivation::symbolTablePutWithAttributes): 2011-03-24 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Crash in debugger beneath MarkStack::drain @ me.com, ibm.com https://bugs.webkit.org/show_bug.cgi?id=57080 <rdar://problem/8525907> The crash was caused by changes in the executable after recompilation. The fix is for the activation to copy the data it needs instead of relying on the data in the executable. SunSpider and v8 report no change. * runtime/Arguments.h: (JSC::JSActivation::copyRegisters): Use our own data members instead of reading data out of the executable. * runtime/JSActivation.cpp: (JSC::JSActivation::JSActivation): Initialize our data members. (JSC::JSActivation::markChildren): (JSC::JSActivation::symbolTableGet): (JSC::JSActivation::symbolTablePut): (JSC::JSActivation::getOwnPropertyNames): (JSC::JSActivation::symbolTablePutWithAttributes): (JSC::JSActivation::isDynamicScope): (JSC::JSActivation::argumentsGetter): Use our own data members instead of reading data out of the executable. * runtime/JSActivation.h: Added new data members to track data previously tracked by the executable. Since I've removed the executable pointer, on a 64bit system, I've only made activations bigger by an int. 2011-03-25 David Kilzer <ddkilzer@apple.com> Remove duplicate entry from JavaScriptCore.exp JSC::createStackOverflowError(JSC::ExecState*) was originally exported in r60057, then duplicated in r60392. * JavaScriptCore.exp: Removed duplicate entry. 2011-03-25 Jarred Nicholls <jarred@sencha.com> Reviewed by Ariya Hidayat. [Qt] MSVC Build Error - need to link advapi32.lib for jsc.exe https://bugs.webkit.org/show_bug.cgi?id=56098 Need to link advapi32.lib for jsc.exe since wtf/OSRandomSource.cpp uses the Win32 Crypto API * jsc.pro: 2011-03-24 Nikolas Zimmermann <nzimmermann@rim.com> Reviewed by Darin Adler. Introduce WTF HexNumber.h https://bugs.webkit.org/show_bug.cgi?id=56099 Introduce a set of functions that ease converting from a bye or a number to a hex string, replacing several of these conversions and String::format("%x") usages all over WebCore. * GNUmakefile.am: Add HexNumber.h to build. * JavaScriptCore.exp: Export StringBuilder::reserveCapacity. * JavaScriptCore.gypi: Add HexNumber.h to build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Export StringBuilder::reserveCapacity. * JavaScriptCore.vcproj/WTF/WTF.vcproj: Add HexNumber.h to build. * JavaScriptCore.xcodeproj/project.pbxproj: Ditto. * wtf/CMakeLists.txt: Ditto. * wtf/HexNumber.h: Added. (WTF::Internal::hexDigitsForMode): Internal helper. (WTF::appendByteAsHex): Free function, that appends a byte as hex string into a destination. (WTF::placeByteAsHex): Ditto, but places the result using *foo++ = '..' or foo[index++] = '..' (WTF::appendUnsignedAsHex): Free function, that appends a number as hex string into a destination. 2011-03-24 Geoffrey Garen <ggaren@apple.com> Windows build fix take 2: Add new symobl. (I should have used the EWS bots for this!) * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-03-24 Geoffrey Garen <ggaren@apple.com> Windows build fix take 1: Removed old symobl. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-03-24 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Ensure that all compilation takes place within a dynamic global object scope https://bugs.webkit.org/show_bug.cgi?id=57054 <rdar://problem/9083011> Otherwise, entry to the global object scope might throw away the code we just compiled, causing a crash. * JavaScriptCore.exp: Updated for signature change. * debugger/Debugger.cpp: (JSC::evaluateInGlobalCallFrame): * debugger/DebuggerCallFrame.cpp: (JSC::DebuggerCallFrame::evaluate): Removed explicit compilation calls here because (a) they took place outside a dynamic global object scope and (b) they were redundant. * interpreter/CachedCall.h: (JSC::CachedCall::CachedCall): Updated for signature change. * interpreter/Interpreter.cpp: (JSC::Interpreter::execute): (JSC::Interpreter::executeCall): (JSC::Interpreter::executeConstruct): Declare our dynamic global object scope earlier, to ensure that compilation takes place within it. * runtime/Completion.cpp: (JSC::evaluate): Removed explicit compilation calls here because (a) they took place outside a dynamic global object scope and (b) they were redundant. * runtime/Executable.h: (JSC::EvalExecutable::compile): (JSC::ProgramExecutable::compile): (JSC::FunctionExecutable::compileForCall): (JSC::FunctionExecutable::compileForConstruct): Added an ASSERT to verify our new invariant that all compilation takes place within a dynamic global object scope. * runtime/JSGlobalObject.cpp: (JSC::DynamicGlobalObjectScope::DynamicGlobalObjectScope): * runtime/JSGlobalObject.h: Changed the signature of DynamicGlobalObjectScope to require a JSGlobalData instead of an ExecState* since it is often easier to provide the former, and the latter was not necessary. 2011-03-24 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. REGRESSION (r79987-r80210): Crash in JSWeakObjectMapClear https://bugs.webkit.org/show_bug.cgi?id=55671 This is no longer necessary, and it seems that with the new weakmap model it's simply unsafe, so this reduces it to a no-op. * API/JSWeakObjectMapRefPrivate.cpp: 2011-03-24 Ben Taylor <bentaylor.solx86@gmail.com> Reviewed by Darin Adler. https://bugs.webkit.org/show_bug.cgi?id=20302 Correct implementation of signbit on Solaris * wtf/MathExtras.h: (signbit): 2011-03-23 Mark Rowe <mrowe@apple.com> Reviewed by Darin Adler. <rdar://problem/7959320> Threads that use APIs above the BSD layer must be registered with the Obj-C GC. * wtf/ThreadingPthreads.cpp: (WTF::initializeCurrentThreadInternal): 2011-03-23 Mark Rowe <mrowe@apple.com> Stop setting OTHER_OPTIONS in JavaScriptCore's Makefile. It's not necessary to pass "-target All" as xcodebuild always builds the first target in the project unless otherwise specified. The presence of that option also breaks "make clean" since that results in both the -target and -alltargets options being passed to xcodebuild. * Makefile: 2011-03-23 Pavel Feldman <pfeldman@chromium.org> Not reviewed: bring back Vector::contains that was removed as a part of roll back. * wtf/Vector.h: (WTF::::contains): 2011-03-23 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r81686. http://trac.webkit.org/changeset/81686 https://bugs.webkit.org/show_bug.cgi?id=56914 Breaks webkit_tests in Chromium again. (Requested by pfeldman on #webkit). * wtf/Vector.h: 2011-03-23 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. JavaScriptCore GYP build should work on a case-sensitive file system https://bugs.webkit.org/show_bug.cgi?id=56911 The issue is that there are two UString.h headers, one named UString.h and one named ustring.h. This patch excludes ustring.h from the header map to avoid confusion. While I was editing this part of the GYP file, I cleaned up the exclude rules to be more modern. * gyp/JavaScriptCore.gyp: 2011-03-22 Geoffrey Garen <ggaren@apple.com> Reviewed by Maciej Stachowiak. REGRESSION (r78382): No scripts appear in the Web Inspector's Scripts panel on Windows, and many inspector regression tests are failing https://bugs.webkit.org/show_bug.cgi?id=54490 The bug was caused by two different classes using the same name (Recompiler). * debugger/Debugger.cpp: * runtime/JSGlobalData.cpp: (WTF::Recompiler::operator()): Put Recompiler in an anonymous namespace, so our two recompilers' inline functions don't stomp each other at link time. 2011-03-22 Sam Weinig <sam@webkit.org> Reviewed by Mark Rowe. Remove USE_WK_SCROLLBAR_PAINTER_AND_CONTROLLER. <rdar://problem/8944718> * DerivedSources.make: Remove generation of USE_WK_SCROLLBAR_PAINTER_AND_CONTROLLER. 2011-03-22 Gabor Loki <loki@webkit.org> Reviewed by Csaba Osztrogonác. [Qt] Add DFG module to build system (disabled by default). https://bugs.webkit.org/show_bug.cgi?id=56845 * JavaScriptCore.pri: * JavaScriptCore.pro: 2011-03-22 Eric Seidel <eric@webkit.org> Reviewed by Adam Barth. Add support to build-webkit for building with gyp-generated project files https://bugs.webkit.org/show_bug.cgi?id=56877 Found a couple missing Private headers while trying to make WebCore build. * JavaScriptCore.gypi: 2011-03-22 Eric Seidel <eric@webkit.org> Reviewed by Adam Barth. Make it possible to build JavaScriptCore and WebCore gyp builds outside of Source https://bugs.webkit.org/show_bug.cgi?id=56867 This should make it possible to build the gyp-generated JavaScriptCore.xcodeproj from a JavaScriptCore directory outside of Source. * gyp/JavaScriptCore.gyp: * gyp/run-if-exists.sh: Added. * gyp/update-info-plist.sh: Added. 2011-03-22 Eric Seidel <eric@webkit.org> Reviewed by Adam Barth. Add Profiling Configuration to JavaScriptCore gyp build https://bugs.webkit.org/show_bug.cgi?id=56862 It appears this is identical to Release, but I suspect there is someone/thing who uses the Profiling target so we're adding it for completeness. * gyp/JavaScriptCore.gyp: 2011-03-22 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. Remove os_win32_files variable from the GYP build https://bugs.webkit.org/show_bug.cgi?id=56804 Now that our understanding of GYP is sufficiently advanced, we don't need os_win32_files any more. (Turns out Eric was right, as he always is.) * JavaScriptCore.gypi: 2011-03-22 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. GYP build of JavaScriptCore should be able to link from an empty WebKitBuild directory https://bugs.webkit.org/show_bug.cgi?id=56803 Previously, we thought we should generate the derived source files in the shared intermediate build products directory, but there are assumptions built into other parts of the Mac build system that the derived source files will be generated in a particular subdirectory of the build products directory. This patch is a partial revert of the change that moved the derived source files to the shared intermediate directory. After this patch, the GYP build can build JavaScriptCore without help from the main normal build system. * JavaScriptCore.gypi: * gyp/JavaScriptCore.gyp: * gyp/generate-derived-sources.sh: * gyp/generate-dtrace-header.sh: 2011-03-22 Jay Civelli <jcivelli@chromium.org> Reviewed by David Levin. Adding a contains method to Vector. https://bugs.webkit.org/show_bug.cgi?id=55859 * wtf/Vector.h: (WTF::Vector::contains): 2011-03-22 Gabor Loki <loki@webkit.org> Reviewed by Alexey Proskuryakov. Fix a bunch of typos in DFG. https://bugs.webkit.org/show_bug.cgi?id=56813 * dfg/DFGByteCodeParser.cpp: (JSC::DFG::parse): * dfg/DFGGenerationInfo.h: (JSC::DFG::GenerationInfo::setSpilled): * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): * dfg/DFGGraph.h: * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::setupStubArguments): * dfg/DFGJITCompiler.cpp: (JSC::DFG::JITCompiler::compileFunction): * dfg/DFGJITCompiler.h: * dfg/DFGNode.h: * dfg/DFGNonSpeculativeJIT.h: * dfg/DFGOperations.h: * dfg/DFGRegisterBank.h: (JSC::DFG::RegisterBank::allocate): * dfg/DFGScoreBoard.h: (JSC::DFG::ScoreBoard::~ScoreBoard): (JSC::DFG::ScoreBoard::allocate): * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: 2011-03-22 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. Production configuration in GYP isn't set up correctly https://bugs.webkit.org/show_bug.cgi?id=56786 Update JavaScriptCore.gyp with information mined from JavaScriptCore.xcodeproj. * JavaScriptCore.gypi: * gyp/JavaScriptCore.gyp: 2011-03-22 Kent Tamura <tkent@chromium.org> Reviewed by Eric Seidel. REGRESSION(r80096): Number type input unexpectedly rounds fractional values https://bugs.webkit.org/show_bug.cgi?id=56367 Introduce clampToInteger(unsigned). * wtf/MathExtras.h: (clampToInteger): Added. 2011-03-21 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. GYP build should not have include paths that point within the source tree https://bugs.webkit.org/show_bug.cgi?id=56788 Turns out we don't need these include paths anymore now that we have header maps working properly. * gyp/JavaScriptCore.gyp: - Also, remove jsc.cpp from the excluded list because it's not part of the jsc_files variable instead of the javascriptcore_files variable. 2011-03-21 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. Solve the Assertions.cpp / -Wno-missing-format-attribute mystery https://bugs.webkit.org/show_bug.cgi?id=56780 The reason we couldn't resolve this warning in the GYP build was that the normal build disables this warning specifically for this file. This patch takes the same approach as the previous patch to WebCoreObjCExtras.mm in that it uses a pragma to suppress the warning (rather than a build system configuration). * JavaScriptCore.xcodeproj/project.pbxproj: - Remove the special-case for this file. * gyp/JavaScriptCore.gyp: - Remove the work-around for this issue. * wtf/Assertions.cpp: - Add a pragma disabling this warning for this file. 2011-03-21 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. WebCore GYP build shouldn't crash on startup https://bugs.webkit.org/show_bug.cgi?id=56776 Debug builds shouldn't define NDEBUG. This same logic exists in the project.pbxproj file. * gyp/JavaScriptCore.gyp: 2011-03-21 Robert Kroeger <rjkroege@chromium.org> Reviewed by Antonio Gomes. Flag to enable/disable a GestureReocognizer framework https://bugs.webkit.org/show_bug.cgi?id=49345 * wtf/Platform.h: 2011-03-21 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. Add new files to JavaScriptCore.gypi https://bugs.webkit.org/show_bug.cgi?id=56766 * JavaScriptCore.gypi: 2011-03-21 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r81377. http://trac.webkit.org/changeset/81377 https://bugs.webkit.org/show_bug.cgi?id=56765 WebPageSerializerTest.MultipleFrames is broken (Requested by simonjam on #webkit). * wtf/Vector.h: 2011-03-21 Gabor Loki <loki@webkit.org> Reviewed by Csaba Osztrogonác. Extend constant pool to be able to store 16 bit instructions with a constant https://bugs.webkit.org/show_bug.cgi?id=46796 The putShortWithConstantInt function inserts a 16 bit instruction which refers a 32 bits constant or literal. This is a vital function for those target which loads a PC relative value with a 16 bit instruction (like Thumb-2 instruction set and SH4 architecture). * assembler/AssemblerBuffer.h: (JSC::AssemblerBuffer::putIntegral): (JSC::AssemblerBuffer::putIntegralUnchecked): * assembler/AssemblerBufferWithConstantPool.h: 2011-03-21 Philippe Normand <pnormand@igalia.com> Unreviewed, GTK distcheck build fix. * GNUmakefile.am: 2011-03-20 Bill Budge <bbudge@chromium.org> Reviewed by Adam Barth. Rename ThreadSafeShared to ThreadSafeRefCounted https://bugs.webkit.org/show_bug.cgi?id=56714 No new tests. Exposes no new functionality. * API/JSClassRef.h: * API/OpaqueJSString.h: * GNUmakefile.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/Atomics.h: * wtf/CMakeLists.txt: * wtf/CrossThreadRefCounted.h: (WTF::CrossThreadRefCounted::CrossThreadRefCounted): (WTF::::crossThreadCopy): * wtf/ThreadSafeRefCounted.h: Copied from wtf/ThreadSafeShared.h. (WTF::ThreadSafeRefCountedBase::ThreadSafeRefCountedBase): (WTF::ThreadSafeRefCountedBase::ref): (WTF::ThreadSafeRefCountedBase::refCount): (WTF::ThreadSafeRefCountedBase::derefBase): (WTF::ThreadSafeRefCounted::ThreadSafeRefCounted): * wtf/ThreadSafeShared.h: Removed. * wtf/Threading.h: 2011-03-19 Patrick Gansterer <paroga@webkit.org> Reviewed by Darin Adler. Remove StringImpl::computeHash() https://bugs.webkit.org/show_bug.cgi?id=49894 Replace remainig StringImpl::computeHash with StringImpl::computeHashStringHasher. * wtf/text/AtomicString.cpp: (WTF::CStringTranslator::hash): (WTF::UCharBufferTranslator::hash): (WTF::HashAndCharactersTranslator::hash): * wtf/text/StringImpl.h: (WTF::StringImpl::setHash): (WTF::StringImpl::hash): 2011-03-19 Patrick Gansterer <paroga@webkit.org> Reviewed by Darin Adler. Rename WTF::StringHasher methods https://bugs.webkit.org/show_bug.cgi?id=53532 Rename createHash to computeHash and createBlobHash to hashMemory. Also add a using WTF::StringHasher in the header file. * profiler/CallIdentifier.h: (JSC::CallIdentifier::Hash::hash): * runtime/Identifier.cpp: (JSC::IdentifierCStringTranslator::hash): (JSC::IdentifierUCharBufferTranslator::hash): * wtf/StringHasher.h: (WTF::StringHasher::computeHash): (WTF::StringHasher::hashMemory): * wtf/text/StringHash.h: (WTF::CaseFoldingHash::hash): * wtf/text/StringImpl.h: (WTF::StringImpl::computeHash): * wtf/unicode/UTF8.cpp: (WTF::Unicode::calculateStringHashAndLengthFromUTF8Internal): 2011-03-18 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. [GTK] JSC crashes in 32bit Release bots after r80743 https://bugs.webkit.org/show_bug.cgi?id=56180 The crash was caused by referencing GC memory from a GC destructor. This is not safe because destruction time / order is not guaranteed. * profiler/ProfileGenerator.cpp: (JSC::ProfileGenerator::create): (JSC::ProfileGenerator::ProfileGenerator): (JSC::ProfileGenerator::willExecute): (JSC::ProfileGenerator::didExecute): * profiler/ProfileGenerator.h: (JSC::ProfileGenerator::origin): Made ExecState* the first argument, to match the rest of this class and JSC. Use a JSGlobalObject* instead of an ExecState* with an indirect reference to a JSGlobalObject* to track our origin. This is simpler and more efficient, and it removes the destruction order dependency that was causing our crash. * profiler/Profiler.cpp: (JSC::Profiler::startProfiling): Updated for change to JSGlobalObject*. (JSC::Profiler::stopProfiling): New function for stopping all profiles for a given global object. This is more straight-forward than multiplexing through the old function. (JSC::dispatchFunctionToProfiles): Updated for change to JSGlobalObject*. * profiler/Profiler.h: Ditto. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::~JSGlobalObject): Ditto. 2011-03-17 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. 1 Structure leaked beneath JSGlobalData::storeVPtrs() https://bugs.webkit.org/show_bug.cgi?id=56595 * runtime/Executable.cpp: (JSC::EvalExecutable::EvalExecutable): (JSC::ProgramExecutable::ProgramExecutable): (JSC::FunctionExecutable::FunctionExecutable): * runtime/Executable.h: (JSC::ExecutableBase::ExecutableBase): (JSC::NativeExecutable::NativeExecutable): (JSC::VPtrHackExecutable::VPtrHackExecutable): (JSC::ScriptExecutable::ScriptExecutable): Use a raw pointer instead of PassRefPtr, like JSString does, since JSGlobalData owns the singleton exectuable structure. 2011-03-17 Geoffrey Garen <ggaren@apple.com> Reviewed by Mark Rowe. Fixed some string leaks seen on the buildbot https://bugs.webkit.org/show_bug.cgi?id=56619 * runtime/PropertyMapHashTable.h: (JSC::PropertyTable::~PropertyTable): DEref! 2011-03-17 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Crash in JSC::MarkStack::drain Under Stress https://bugs.webkit.org/show_bug.cgi?id=56470 We perform a number of gc allocations while when we are setting up new globals in a piece of global code. We do this by adding new properties to the symbol table, and then expanding the storage to fit at the end. If a GC happens during this time we will report an incorrect size for the global object's symbol table storage. This patch corrects this by growing the storage size before we starting adding entries to the symbol table. * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::resizeRegisters): 2011-03-17 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. 1 Structure leaked beneath JSGlobalData::storeVPtrs() https://bugs.webkit.org/show_bug.cgi?id=56595 * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::storeVPtrs): Take local ownership of the Structure we're using, since the Executable is not designed to own the Structure. 2011-03-17 Gavin Barraclough <barraclough@apple.com> Rubber Stamped by Sam Weinig. Add missing register-register branchTest8 to MacroAssemblerX86Common/X86Assembler. * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::branchTest8): * assembler/X86Assembler.h: (JSC::X86Assembler::testb_rr): 2011-03-17 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 56603 - DFG JIT related cleanup Move node generation out to separate function, move binarySearch algorithm out to StdLibExtras, fix Graph::dump() to print comma between non-node children, even if there are no node children. * bytecode/CodeBlock.h: (JSC::getCallReturnOffset): (JSC::CodeBlock::getStubInfo): (JSC::CodeBlock::getCallLinkInfo): (JSC::CodeBlock::getMethodCallLinkInfo): (JSC::CodeBlock::bytecodeOffset): - Move binaryChop to binarySearch in StdLibExtras * dfg/DFGByteCodeParser.cpp: (JSC::DFG::ByteCodeParser::ByteCodeParser): (JSC::DFG::ByteCodeParser::parse): (JSC::DFG::parse): - Make m_noArithmetic a member, initialize m_currentIndex in the constructor. * dfg/DFGByteCodeParser.h: - Change parse() to not take a start index (always 0). * dfg/DFGGraph.cpp: (JSC::DFG::Graph::dump): - Fix Graph::dump() to print comma between non-node children, even if there are no node children. * dfg/DFGJITCodeGenerator.h: (JSC::DFG::JITCodeGenerator::JITCodeGenerator): - Initialize m_compileIndex in constructor. * dfg/DFGNonSpeculativeJIT.cpp: (JSC::DFG::NonSpeculativeJIT::compile): * dfg/DFGNonSpeculativeJIT.h: - Spilt out compilation of individual node. * dfg/DFGOperations.cpp: (JSC::DFG::operationConvertThis): * dfg/DFGOperations.h: - Cleanup parameter name. * dfg/DFGSpeculativeJIT.cpp: (JSC::DFG::SpeculativeJIT::compile): * dfg/DFGSpeculativeJIT.h: - Spilt out compilation of individual node. * runtime/Executable.cpp: (JSC::tryDFGCompile): - Change parse() to not take a start index (always 0). * wtf/StdLibExtras.h: (WTF::binarySearch): - Move binaryChop to binarySearch in StdLibExtras 2011-03-17 Anders Carlsson <andersca@apple.com> Reviewed by Geoffrey Garen. Fix clang build. * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::storeVPtrs): 2011-03-17 Geoffrey Garen <ggaren@apple.com> Reviewed by Darin Adler. 1 Structure leaked beneath JSGlobalData::storeVPtrs() https://bugs.webkit.org/show_bug.cgi?id=56595 * JavaScriptCore.xcodeproj/project.pbxproj: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::storeVPtrs): Now that we have an executable, we need to explicitly run its destructor. 2011-03-17 Jeff Miller <jeffm@apple.com> Use a consistent set of file patterns in the svn:ignore property for all .xcodeproj directories, specifically: *.mode* *.pbxuser *.perspective* project.xcworkspace xcuserdata * JavaScriptCore.xcodeproj: Modified property svn:ignore. 2011-03-17 Gavin Barraclough <barraclough@apple.com> Reverting r81197, breaks JIT + INTERPRETER build. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::hasGlobalResolveInstructionAtBytecodeOffset): (JSC::CodeBlock::hasGlobalResolveInfoAtBytecodeOffset): * bytecode/CodeBlock.h: (JSC::CodeBlock::addPropertyAccessInstruction): (JSC::CodeBlock::addGlobalResolveInstruction): (JSC::CodeBlock::addStructureStubInfo): * bytecode/Opcode.h: * bytecode/StructureStubInfo.h: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitResolve): (JSC::BytecodeGenerator::emitResolveWithBase): (JSC::BytecodeGenerator::emitGetById): (JSC::BytecodeGenerator::emitPutById): (JSC::BytecodeGenerator::emitDirectPutById): (JSC::BytecodeGenerator::emitCall): (JSC::BytecodeGenerator::emitConstruct): (JSC::BytecodeGenerator::emitCatch): 2011-03-17 Ben Taylor <bentaylor.solx86@gmail.com> Reviewed by Alexey Proskuryakov. Add a COMPILER(SUNCC) define for Sun Studio 12. https://bugs.webkit.org/show_bug.cgi?56444 derived from patch 1 of 16 originally from https://bugs.webkit.org/show_bug.cgi?id=24932 * wtf/Platform.h: 2011-03-17 Jay Civelli <jcivelli@chromium.org> Reviewed by David Levin. Adding a contains method to Vector. https://bugs.webkit.org/show_bug.cgi?id=55859 * wtf/Vector.h: (WTF::::operator): (WTF::::contains): 2011-03-17 Patrick Gansterer <paroga@webkit.org> Fix the interpreter build. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): Added globalData to inheritorID(). 2011-03-16 Sam Weinig <sam@webkit.org> Fix the interpreter build. * interpreter/Interpreter.cpp: (JSC::Interpreter::resolve): (JSC::Interpreter::resolveSkip): (JSC::Interpreter::resolveGlobal): (JSC::Interpreter::resolveGlobalDynamic): (JSC::Interpreter::resolveBaseAndProperty): (JSC::Interpreter::privateExecute): Remove .get()s. 2011-03-16 Adam Barth <abarth@webkit.org> Reviewed by James Robinson. Remove USE(BUILTIN_UTF8_CODEC) https://bugs.webkit.org/show_bug.cgi?id=56508 We added this recently when we were unsure about the stability of the built-in UTF-8 codec. However, the codec seems to be stable, so we don't need the macro. * wtf/Platform.h: 2011-03-16 Daniel Bates <dbates@rim.com> Reviewed by Darin Adler. Make JIT build for ARM Thumb-2 with RVCT https://bugs.webkit.org/show_bug.cgi?id=56440 Derived from a patch by Dave Tapuska. Also, modify the RVCT stub template to indicate that it preserves 8 byte stack alignment. * jit/JITStubs.cpp: 2011-03-16 Chao-ying Fu <fu@mips.com> Reviewed by Darin Adler. Fix MIPS build with const *void https://bugs.webkit.org/show_bug.cgi?id=56513 * assembler/MacroAssemblerMIPS.h: (JSC::MacroAssemblerMIPS::load32): (JSC::MacroAssemblerMIPS::store32): 2011-03-16 Oliver Hunt <oliver@apple.com> Reviewed by Darin Adler. Remove unnecessary caller tracking shenanigans from CodeBlock https://bugs.webkit.org/show_bug.cgi?id=56483 This removes some leftover cruft from when we made CodeBlock mark its callees. Removing it gives us a 0.7% progression, reducing the overall regression to ~1.3%. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::shrinkToFit): * bytecode/CodeBlock.h: (JSC::CallLinkInfo::CallLinkInfo): * jit/JIT.cpp: (JSC::JIT::linkCall): (JSC::JIT::linkConstruct): 2011-03-15 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make Structure creation require a JSGlobalData https://bugs.webkit.org/show_bug.cgi?id=56438 Mechanical change to make Structure::create require JSGlobalData&, and require all users to provide the globalData. * API/JSCallbackConstructor.h: (JSC::JSCallbackConstructor::createStructure): * API/JSCallbackFunction.h: (JSC::JSCallbackFunction::createStructure): * API/JSCallbackObject.h: (JSC::JSCallbackObject::createStructure): * API/JSContextRef.cpp: * JavaScriptCore.exp: * debugger/DebuggerActivation.cpp: (JSC::DebuggerActivation::DebuggerActivation): * debugger/DebuggerActivation.h: (JSC::DebuggerActivation::createStructure): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jsc.cpp: (GlobalObject::GlobalObject): (functionRun): (jscmain): * runtime/Arguments.h: (JSC::Arguments::createStructure): * runtime/ArrayPrototype.h: (JSC::ArrayPrototype::createStructure): * runtime/BooleanObject.h: (JSC::BooleanObject::createStructure): * runtime/DateInstance.h: (JSC::DateInstance::createStructure): * runtime/DatePrototype.h: (JSC::DatePrototype::createStructure): * runtime/ErrorInstance.h: (JSC::ErrorInstance::createStructure): * runtime/Executable.h: (JSC::ExecutableBase::createStructure): (JSC::EvalExecutable::createStructure): (JSC::ProgramExecutable::createStructure): (JSC::FunctionExecutable::createStructure): * runtime/FunctionPrototype.h: (JSC::FunctionPrototype::createStructure): * runtime/GetterSetter.h: (JSC::GetterSetter::createStructure): * runtime/InternalFunction.h: (JSC::InternalFunction::createStructure): * runtime/JSAPIValueWrapper.h: (JSC::JSAPIValueWrapper::createStructure): * runtime/JSActivation.h: (JSC::JSActivation::createStructure): * runtime/JSArray.cpp: (JSC::JSArray::JSArray): * runtime/JSArray.h: (JSC::JSArray::createStructure): * runtime/JSByteArray.cpp: (JSC::JSByteArray::createStructure): * runtime/JSByteArray.h: (JSC::JSByteArray::JSByteArray): * runtime/JSCell.h: (JSC::JSCell::JSCell::createDummyStructure): * runtime/JSFunction.h: (JSC::JSFunction::createStructure): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::storeVPtrs): (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::JSGlobalObject): (JSC::JSGlobalObject::createStructure): * runtime/JSNotAnObject.h: (JSC::JSNotAnObject::createStructure): * runtime/JSONObject.h: (JSC::JSONObject::createStructure): * runtime/JSObject.cpp: (JSC::JSObject::createInheritorID): * runtime/JSObject.h: (JSC::JSObject::createStructure): (JSC::JSNonFinalObject::createStructure): (JSC::JSFinalObject::createStructure): (JSC::createEmptyObjectStructure): (JSC::JSObject::inheritorID): * runtime/JSObjectWithGlobalObject.h: (JSC::JSObjectWithGlobalObject::createStructure): * runtime/JSPropertyNameIterator.h: (JSC::JSPropertyNameIterator::createStructure): * runtime/JSStaticScopeObject.h: (JSC::JSStaticScopeObject::createStructure): * runtime/JSString.h: (JSC::RopeBuilder::createStructure): * runtime/JSVariableObject.h: (JSC::JSVariableObject::createStructure): * runtime/JSWrapperObject.h: (JSC::JSWrapperObject::createStructure): * runtime/JSZombie.h: (JSC::JSZombie::createStructure): * runtime/MathObject.h: (JSC::MathObject::createStructure): * runtime/NativeErrorConstructor.cpp: (JSC::NativeErrorConstructor::NativeErrorConstructor): * runtime/NativeErrorConstructor.h: (JSC::NativeErrorConstructor::createStructure): * runtime/NumberConstructor.h: (JSC::NumberConstructor::createStructure): * runtime/NumberObject.h: (JSC::NumberObject::createStructure): * runtime/ObjectConstructor.h: (JSC::ObjectConstructor::createStructure): * runtime/RegExpConstructor.h: (JSC::RegExpConstructor::createStructure): * runtime/RegExpObject.h: (JSC::RegExpObject::createStructure): * runtime/ScopeChain.h: (JSC::ScopeChainNode::createStructure): * runtime/StringObject.h: (JSC::StringObject::createStructure): * runtime/StringObjectThatMasqueradesAsUndefined.h: (JSC::StringObjectThatMasqueradesAsUndefined::createStructure): * runtime/StringPrototype.h: (JSC::StringPrototype::createStructure): * runtime/Structure.h: (JSC::Structure::create): 2011-03-16 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Some conservative root gathering cleanup https://bugs.webkit.org/show_bug.cgi?id=56447 SunSpider says 0.5% - 1.8% faster. * interpreter/RegisterFile.cpp: (JSC::RegisterFile::gatherConservativeRoots): * interpreter/RegisterFile.h: New helper function for doing the conservative gathering of the register file. It's still conservative, since the register file may contain uninitialized values, but it's moving-safe, because it only visits values tagged as pointers, so there's no risk of mistaking an integer for a pointer and accidentally changing it. * runtime/ConservativeSet.cpp: (JSC::ConservativeRoots::add): * runtime/ConservativeSet.h: Added a single-value add function, used above. * runtime/Heap.cpp: (JSC::Heap::markRoots): Separated machine stack conservative roots from register file conservative roots because machine stack roots must be pinned, but register file roots need not be pinned. Adopted new interface for passing the current stack extent to the machine stack root gathering routine. This allows us to exclude marking-related data structures on the stack, and thus avoid double-marking the set of machine roots. * runtime/MachineStackMarker.cpp: (JSC::MachineThreads::gatherFromCurrentThread): (JSC::MachineThreads::gatherConservativeRoots): * runtime/MachineStackMarker.h: Added new interface, described above. * runtime/MarkedBlock.h: (JSC::MarkedBlock::firstAtom): * wtf/StdLibExtras.h: (WTF::roundUpToMultipleOf): Moved roundUpToMultipleOf so it could be used by MachineStacks. 2011-03-16 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. A little bit of MarkStack cleanup https://bugs.webkit.org/show_bug.cgi?id=56443 Moved MarkStack functions into MarkStack.h/.cpp. SunSpider reports no change. * runtime/JSArray.h: * runtime/JSCell.h: Moved from here... * runtime/MarkStack.cpp: (JSC::MarkStack::markChildren): (JSC::MarkStack::drain): ...to here. Also, no need to inline drain. It's a huge function, and not called many times. * runtime/MarkStack.h: (JSC::MarkStack::~MarkStack): Moved near constructor, per style guide. (JSC::MarkStack::append): (JSC::MarkStack::deprecatedAppend): (JSC::MarkStack::internalAppend): Moved to here. 2011-03-15 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed another deprecatedAppend https://bugs.webkit.org/show_bug.cgi?id=56429 * collector/handles/HandleHeap.cpp: (JSC::HandleHeap::markStrongHandles): * collector/handles/HandleHeap.h: Use HeapRootMarker, since handles are marked directly by the Heap. * runtime/Heap.cpp: (JSC::Heap::markRoots): Ditto. 2011-03-15 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed some more deprecated / unsafe append https://bugs.webkit.org/show_bug.cgi?id=56428 * collector/handles/HandleStack.cpp: (JSC::HandleStack::mark): * collector/handles/HandleStack.h: Mark the handle stack using a HeapRoot marker, since it's a heap root. * runtime/ArgList.cpp: (JSC::MarkedArgumentBuffer::markLists): (JSC::MarkedArgumentBuffer::slowAppend): * runtime/ArgList.h: Ditto. * runtime/Heap.cpp: (JSC::Heap::markRoots): Added a mark call for marking the handle stack. It seems like Oliver forgot this in his last patch. (!) * runtime/MarkStack.h: Removed appendSlots, since it would allow an object to embed JSValues directly instead of using WriteBarrier. (JSC::MarkStack::append): Added a private append for a list of values. (JSC::HeapRootMarker::mark): Access to the above. 2011-03-15 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed a few more deprecatedAppends, and removed HeapRoot<T> https://bugs.webkit.org/show_bug.cgi?id=56422 Added HeapRootMarker, a privileged class for marking direct heap roots that are iterated during each garbage collection. This is easier to use and more reliable than HeapRoot<T>, so I've removed HeapRoot<T>. * debugger/Debugger.cpp: (JSC::evaluateInGlobalCallFrame): * debugger/DebuggerCallFrame.cpp: (JSC::DebuggerCallFrame::evaluate): * interpreter/CallFrame.h: (JSC::ExecState::exception): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/Completion.cpp: (JSC::evaluate): exception is no longer a HeapRoot<T>, so no need to call .get() on it. * runtime/Heap.cpp: (JSC::Heap::markProtectedObjects): (JSC::Heap::markTempSortVectors): (JSC::Heap::markRoots): * runtime/Heap.h: Updated to use HeapRootMarker. * runtime/JSCell.h: (JSC::JSCell::MarkStack::append): Added private functions for HeapRootMarker to use. * runtime/JSGlobalData.h: exception is no longer a HeapRoot<T>. * runtime/MarkStack.h: (JSC::HeapRootMarker::HeapRootMarker): (JSC::HeapRootMarker::mark): Added private functions for HeapRootMarker to use. * runtime/SmallStrings.cpp: (JSC::SmallStrings::markChildren): Updated to use HeapRootMarker. * runtime/SmallStrings.h: (JSC::SmallStrings::emptyString): (JSC::SmallStrings::singleCharacterString): (JSC::SmallStrings::singleCharacterStrings): Updated to use HeapRootMarker. * runtime/WriteBarrier.h: Removed HeapRoot<T>. 2011-03-14 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Made the global object moving-GC-safe https://bugs.webkit.org/show_bug.cgi?id=56348 SunSpider reports no change. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::markChildren): Removed a dubious comment that suggested we do not need to visit all our references during GC, since that is not true in a moving GC. Re-sorted data members by type, removed one duplicate, and added back the one missing mark I found. * runtime/JSGlobalObject.h: Re-sorted data members by type. 2011-03-15 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Introduce Local<T> to allow us to start moving to precise marking of locals https://bugs.webkit.org/show_bug.cgi?id=56394 Introduce a new handle type, Local<T> and a scoping mechanism LocalScope to allow us to start moving towards precise marking of temporaries and local variables. We also start to use the new Local<> type in the JSON stringifier so that we can have some coverage of their behaviour in the initial checkin. * GNUmakefile.am: * JavaScriptCore.gypi: * JavaScriptCore.pro: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * collector/handles/Handle.h: (JSC::::asObject): * collector/handles/HandleStack.cpp: Added. (JSC::HandleStack::HandleStack): (JSC::HandleStack::mark): (JSC::HandleStack::grow): * collector/handles/HandleStack.h: Added. (JSC::HandleStack::enterScope): (JSC::HandleStack::zapTo): (JSC::HandleStack::leaveScope): (JSC::HandleStack::push): * collector/handles/Local.h: Added. (JSC::Local::internalSet): (JSC::::Local): (JSC::::operator): (JSC::LocalStack::LocalStack): (JSC::LocalStack::peek): (JSC::LocalStack::pop): (JSC::LocalStack::push): (JSC::LocalStack::isEmpty): (JSC::LocalStack::size): * collector/handles/LocalScope.h: Added. (JSC::LocalScope::LocalScope): (JSC::LocalScope::~LocalScope): (JSC::LocalScope::release): * runtime/Heap.cpp: (JSC::Heap::markRoots): * runtime/Heap.h: (JSC::Heap::allocateLocalHandle): (JSC::Heap::handleStack): * runtime/JSCell.h: (JSC::JSCell::::getString): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: (JSC::JSGlobalData::allocateLocalHandle): * runtime/JSONObject.cpp: (JSC::Stringifier::Stringifier): (JSC::Stringifier::stringify): (JSC::Stringifier::appendStringifiedValue): (JSC::Stringifier::Holder::Holder): (JSC::Walker::Walker): (JSC::Walker::walk): (JSC::JSONProtoFuncParse): (JSC::JSONProtoFuncStringify): (JSC::JSONStringify): * runtime/JSONObject.h: * runtime/MarkStack.h: (JSC::MarkStack::appendValues): (JSC::MarkStack::appendSlots): 2011-03-15 Gavin Barraclough <barraclough@apple.com> Rubber Stamped by Sam Weinig. Bug 56420 - Remove ENABLE(JIT) code from ByteCompiler Some methods have unnecessary differences in name/arguments for interpreter/JIT. * bytecode/CodeBlock.cpp: * bytecode/CodeBlock.h: (JSC::HandlerInfo::HandlerInfo): (JSC::CodeBlock::addPropertyAccessInfo): (JSC::CodeBlock::addGlobalResolveInfo): (JSC::CodeBlock::addCallLinkInfo): (JSC::CodeBlock::globalResolveInfo): * bytecode/Opcode.h: * bytecode/StructureStubInfo.h: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::emitResolve): (JSC::BytecodeGenerator::emitResolveWithBase): (JSC::BytecodeGenerator::emitGetById): (JSC::BytecodeGenerator::emitPutById): (JSC::BytecodeGenerator::emitDirectPutById): (JSC::BytecodeGenerator::emitCall): (JSC::BytecodeGenerator::emitConstruct): (JSC::BytecodeGenerator::emitCatch): 2011-03-15 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Fix broken assert in new code. * dfg/DFGAliasTracker.h: (JSC::DFG::AliasTracker::recordPutByVal): - recordPutByVal is called for both PutByVal & PutByValAlias. 2011-03-15 Gavin Barraclough <barraclough@apple.com> Rubber stamped by Sam Weinig. Removed redundant code from BytecodeGenerator. * bytecompiler/BytecodeGenerator.cpp: * bytecompiler/BytecodeGenerator.h: - delete uncalled code missed when reparsing was removed. 2011-03-15 Kevin Ollivier <kevino@theolliviers.com> Reviewed by Darin Adler. Introduce WTF_USE_EXPORT_MACROS, which will allow us to put shared library import/export info into the headers rather than in export symbol definition files, but disable it on all platforms initially so we can deal with port build issues one port at a time. https://bugs.webkit.org/show_bug.cgi?id=27551 * API/JSBase.h: * config.h: * wtf/Assertions.h: * wtf/ExportMacros.h: Added. * wtf/Platform.h: 2011-03-14 Laszlo Gombos <laszlo.1.gombos@nokia.com> Unreviewed build fix. Buildfix when JIT is not enabled after r81079 https://bugs.webkit.org/show_bug.cgi?id=56361 * runtime/Executable.cpp: 2011-03-14 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Made the global object moving-GC-safe https://bugs.webkit.org/show_bug.cgi?id=56348 SunSpider reports no change. * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::markChildren): Removed a dubious comment that suggested we do not need to visit all our references during GC, since that is not true in a moving GC. Re-sorted data members by type, removed one duplicate, and added back the one missing mark I found. * runtime/JSGlobalObject.h: Re-sorted data members by type. 2011-03-14 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Made JSWrapperObject and subclasses moving-GC-safe https://bugs.webkit.org/show_bug.cgi?id=56346 SunSpider reports no change. * runtime/BooleanObject.cpp: (JSC::BooleanObject::BooleanObject): * runtime/DateInstance.cpp: (JSC::DateInstance::DateInstance): No more need for JSGlobalData, since we don't initialize the wrapped value in our constructor. * runtime/DateInstance.h: Don't set the OverridesMarkChildren flag because we do not in fact override markChildren. * runtime/DatePrototype.h: Declare an anonymous slot, since wrapper object no longer does so for us. Also added an ASSERT to catch a latent bug, where DatePrototype stomped on its base class's anonymous slot. Hard-coded anonymous slots are a plague on our code. This doesn't cause any problems in our existing code since the base class never reads the anonymous slot it declares, but it caused crashes when I tried to start using the slot in an initial version of this patch. * runtime/JSWrapperObject.h: (JSC::JSWrapperObject::JSWrapperObject): (JSC::JSWrapperObject::internalValue): (JSC::JSWrapperObject::setInternalValue): Resolved a problem where our internal value was stored in two places: an anonymous slot, and a data member which was not always visited during GC. Now, we only use the data member, and we always visit it. (Instead of relying on certain subclasses to set the OverridesMarkChildren bit, we set it ourselves.) * runtime/NumberObject.cpp: (JSC::NumberObject::NumberObject): No more need for JSGlobalData, since we don't initialize the wrapped value in our constructor. * runtime/NumberObject.h: Removed meaningless declaration. * runtime/StringObject.cpp: (JSC::StringObject::StringObject): No more need for JSGlobalData, since we don't initialize the wrapped value in our constructor. * runtime/StringObject.h: Don't set the OverridesMarkChildren flag because we do not in fact override markChildren. * runtime/StringPrototype.h: Declare an anonymous slot, since wrapper object no longer does so for us. Also added an ASSERT to catch a latent bug, where DatePrototype stomped on its base class's anonymous slot. Hard-coded anonymous slots are a plague on our code. 2011-03-14 Michael Saboff <msaboff@apple.com> Reviewed by Gavin Barraclough. Look-ahead assertions with back references don’t work as expected https://bugs.webkit.org/show_bug.cgi?id=56082 Changed parentheses assertion processing to temporarily back out the number of known characters after the assertion while processing the assertion. This was done so that assertions don't fail due to checking the number of required characters as additional to the rest of the express since assertions don't "consume" input. Added a byte code to uncheck characters to support the change. * yarr/YarrInterpreter.cpp: (JSC::Yarr::Interpreter::matchDisjunction): (JSC::Yarr::ByteCompiler::uncheckInput): (JSC::Yarr::ByteCompiler::emitDisjunction): * yarr/YarrInterpreter.h: (JSC::Yarr::ByteTerm::UncheckInput): 2011-03-14 Viatcheslav Ostapenko <ostapenko.viatcheslav@nokia.com> Reviewed by Laszlo Gombos. [Qt] Warning that round/roundf functions are already defined when compiled with RVCT 4 on symbian. https://bugs.webkit.org/show_bug.cgi?id=56133 Add condition to not compile webkit internal math round functions on RVCT compiler versions from 3.0.0 because they are already defined in compiler math library. * wtf/MathExtras.h: 2011-03-14 Gavin Barraclough <barraclough@apple.com> Reviewed by Geoffrey Garen & Oliver Hunt. Bug 56284 - Add a dataflow intermediate representation for use in JIT generation. The JSC JIT presently generates code directly from the bytecode used by the interpreter. This is not an optimal intermediate representation for JIT code generation, since it does not capture liveness information of values, and provides little opportunity to perform any static analysis for even primitive types. The JIT currently generates two code paths, a fast path handling common cases, and a slower path handling less common operand types. However the slow path jumps back into the fast path, meaning that information arising from the earlier type checks cannot be propagated to later operations. This patch adds: * a dataflow intermediate representation capable of describing a single basic block of operations, * a mechanism to convert a simple, single-block bytecode functions to the new IR, * and a JIT code generator capable of generating code from this representation. The JIT generates two code paths, with the slower path not reentering the fast path mid-block, allowing speculative optimizations to be made on the hot path, with type information arising from these speculative decisions able to be propagated through the dataflow. Code generation of both speculative and non-speculative paths exploits the type and liveness information represented in the dataflow graph to attempt to avoid redundant boxing and type-checking of values, and to remove unnecessary spills of temporary values to the RegisterFile. The dataflow JIT currently can only support a subset of bytecode operations, limited to arithmetic, bit-ops, and basic property access. Functions that cannot be compiled by the dataflow JIT will be run using the existing JIT. The coverage of the dataflow JIT will be expanded to include, control-flow, function calls, and then the long-tail of remaining bytecode instructions. The JIT presently only support JSVALUE64, and as a consequence of this only supports x86-64. The status of the dataflow JIT is currently work-in-progress. Limitations of the present JIT code generation may cause performance regressions, particularly: * the policy to only generate arithmetic code on the speculative path using integer instructions, never using floating point. * the policy to only generate arithmetic code on the non-speculative path using floating point instructions, never using integer. * always generating JSValue adds on the non-speculative path as a call out to a C-function, never handling this in JIT code. * always assuming by-Value property accesses on the speculative path to be array accesses. * generating all by-Value property accesses from the non-speculative path as a call out to a C-function. * generating all by-Indentifer property accesses as a call out to a C-function. Due to these regressions, the code is landed in a state where it is disabled in most cases by the ENABLE_DFG_JIT_RESTRICTIONS guard in Platform.h. As these regressions are addressed, the JIT will be allowed to trigger in more cases. * JavaScriptCore.xcodeproj/project.pbxproj: - Added new files to Xcode project. * dfg: Added. - Added directory for new code. * dfg/DFGByteCodeParser.cpp: Added. * dfg/DFGByteCodeParser.h: Added. - Contruct a DFG::Graph representation from a bytecode CodeBlock. * dfg/DFGGenerationInfo.h: Added. - Track type & register information for VirtualRegisters during JIT code generation. * dfg/DFGGraph.cpp: Added. * dfg/DFGGraph.h: Added. - Dataflow graph intermediate representation for code generation. * dfg/DFGJITCodeGenerator.cpp: Added. * dfg/DFGJITCodeGenerator.h: Added. - Base class for SpeculativeJIT & NonSpeculativeJIT to share common functionality. * dfg/DFGJITCompiler.cpp: Added. * dfg/DFGJITCompiler.h: Added. - Class responsible for driving code generation of speculativeJIT & non-speculative code paths from the dataflow graph. * dfg/DFGNonSpeculativeJIT.cpp: Added. * dfg/DFGNonSpeculativeJIT.h: Added. - Used to generate the non-speculative code path, this make no assumptions about operand types. * dfg/DFGOperations.cpp: Added. * dfg/DFGOperations.h: Added. - Helper functions called from the JIT generated code. * dfg/DFGRegisterBank.h: Added. - Used to track contents of physical registers during JIT code generation. * dfg/DFGSpeculativeJIT.cpp: Added. * dfg/DFGSpeculativeJIT.h: Added. - Used to generate the speculative code path, this make assumptions about operand types to enable optimization. * runtime/Executable.cpp: - Add code to attempt to use the DFG JIT to compile a function, with fallback to the existing JIT. * wtf/Platform.h: - Added compile guards to enable the DFG JIT. 2011-03-14 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed more cases of DeprecatedPtr (exception, SmallStrings) https://bugs.webkit.org/show_bug.cgi?id=56332 * runtime/Identifier.cpp: (JSC::Identifier::add): (JSC::Identifier::addSlowCase): Use a variable instead of a hard-coded constant, to make this code less brittle. * runtime/JSGlobalData.h: Use HeapRoot instead of DeprecatedPtr because this reference is owned and managed directly by the heap. * runtime/JSString.cpp: (JSC::JSString::substringFromRope): * runtime/JSString.h: (JSC::jsSingleCharacterString): (JSC::jsSingleCharacterSubstring): (JSC::jsString): (JSC::jsStringWithFinalizer): (JSC::jsSubstring): (JSC::jsOwnedString): Use a variable instead of a hard-coded constant, to make this code less brittle. * runtime/SmallStrings.cpp: (JSC::SmallStringsStorage::rep): (JSC::SmallStringsStorage::SmallStringsStorage): (JSC::SmallStrings::SmallStrings): (JSC::SmallStrings::markChildren): (JSC::SmallStrings::clear): (JSC::SmallStrings::count): Use a variable instead of a hard-coded constant, to make this code less brittle. * runtime/SmallStrings.h: (JSC::SmallStrings::singleCharacterString): Use HeapRoot instead of DeprecatedPtr because these references are owned and managed directly by the heap. Stop using FixedArray because we only want a very limited set of classes to be able to use HeapRoot. (Replaced with manual ASSERTs.) * runtime/WriteBarrier.h: (JSC::operator==): (JSC::WriteBarrier::WriteBarrier): (JSC::HeapRoot::HeapRoot): (JSC::HeapRoot::operator=): Added HeapRoot, which is allowed to set without write barrier because we assume all HeapRoots are scanned during all GC passes. 2011-03-14 Brian Weinstein <bweinstein@apple.com> Reviewed by Adam Roben and Gavin Barraclough. FileSystemWin.cpp needs listDirectory() implementation https://bugs.webkit.org/show_bug.cgi?id=56331 <rdar://problem/9126635> Give StringConcatenate the ability to deal with const UChar*'s as a String type to append. * wtf/text/StringConcatenate.h: 2011-03-14 Mark Rowe <mrowe@apple.com> Reviewed by Oliver Hunt. <http://webkit.org/b/56304> REGRESSION(r80892): 100,000+ leaks seen on the build bot * API/JSClassRef.cpp: (OpaqueJSClass::OpaqueJSClass): Don't leak any existing entry for the given name if the class definition contains duplicates. This also removes what look to be leaks of the StringImpl instances that are used as keys: the HashMap key type is a RefPtr which retains / releases the instances at the appropriate time, so explicitly calling ref is not necessary. 2011-03-14 Oliver Hunt <oliver@apple.com> Fix windows build * jit/JSInterfaceJIT.h: (JSC::JSInterfaceJIT::emitLoadInt32): (JSC::JSInterfaceJIT::tagFor): (JSC::JSInterfaceJIT::payloadFor): (JSC::JSInterfaceJIT::intPayloadFor): (JSC::JSInterfaceJIT::intTagFor): (JSC::JSInterfaceJIT::addressFor): 2011-03-11 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Ensure all values are correctly tagged in the registerfile https://bugs.webkit.org/show_bug.cgi?id=56214 This patch makes sure that all JSCell pointers written to the registerfile are correctly tagged as JSCells, and replaces raw int usage with the immediate representation. For performance, register pressure, and general saneness reasons I've added abstractions for reading and writing the tag and payload of integer registers directly for the JSVALUE64 encoding. * interpreter/Register.h: (JSC::Register::withInt): (JSC::Register::withCallee): (JSC::Register::operator=): (JSC::Register::i): (JSC::Register::activation): (JSC::Register::function): (JSC::Register::propertyNameIterator): (JSC::Register::scopeChain): * jit/JIT.h: * jit/JITCall.cpp: (JSC::JIT::compileOpCallInitializeCallFrame): (JSC::JIT::compileOpCallVarargs): (JSC::JIT::compileOpCall): * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCallInitializeCallFrame): (JSC::JIT::compileOpCallVarargs): (JSC::JIT::compileOpCall): (JSC::JIT::compileOpCallSlowCase): * jit/JITInlineMethods.h: (JSC::JIT::emitPutToCallFrameHeader): (JSC::JIT::emitPutCellToCallFrameHeader): (JSC::JIT::emitPutIntToCallFrameHeader): * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTINativeCall): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_next_pname): (JSC::JIT::emit_op_load_varargs): (JSC::JIT::emitSlow_op_load_varargs): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTINativeCall): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_next_pname): * jit/JSInterfaceJIT.h: (JSC::JSInterfaceJIT::intPayloadFor): (JSC::JSInterfaceJIT::intTagFor): * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::returnJSValue): (JSC::SpecializedThunkJIT::returnDouble): (JSC::SpecializedThunkJIT::returnInt32): (JSC::SpecializedThunkJIT::returnJSCell): 2011-03-13 Geoffrey Garen <ggaren@apple.com> Reviewed by Sam Weinig. A few Heap-related renames (sans file moves, which should come next) https://bugs.webkit.org/show_bug.cgi?id=56283 ConservativeSet => ConservativeRoots. "Set" was misleading, since items are not uniqued. Also, "Roots" is more specific about what's in the set. MachineStackMarker => MachineThreads. "Threads" is more descriptive of the fact that this class maintains a set of all threads using JSC. "Stack" was misleading, since this class traverses stacks and registers. "Mark" was misleading, since this class doesn't mark anything anymore. registerThread => addCurrentThread. "Current" is more specific. unregisterThread => removeCurrentThread. "Current" is more specific. "currentThreadRegistrar" => threadSpecific. The only point of this data structure is to register a thread-specific destructor with a pointer to this. "mark...Conservatively" => "gather". "Mark" is not true, since these functions don't mark anything. "Conservatively" is redundant, since they take "ConservativeRoots" as an argument. * API/APIShims.h: (JSC::APIEntryShimWithoutLock::APIEntryShimWithoutLock): * JavaScriptCore.exp: * runtime/ConservativeSet.cpp: (JSC::ConservativeRoots::grow): (JSC::ConservativeRoots::add): * runtime/ConservativeSet.h: (JSC::ConservativeRoots::ConservativeRoots): (JSC::ConservativeRoots::~ConservativeRoots): (JSC::ConservativeRoots::size): (JSC::ConservativeRoots::roots): * runtime/Heap.cpp: (JSC::Heap::Heap): (JSC::Heap::markRoots): * runtime/Heap.h: (JSC::Heap::machineThreads): * runtime/JSGlobalData.h: (JSC::JSGlobalData::makeUsableFromMultipleThreads): * runtime/MachineStackMarker.cpp: (JSC::MachineThreads::MachineThreads): (JSC::MachineThreads::~MachineThreads): (JSC::MachineThreads::makeUsableFromMultipleThreads): (JSC::MachineThreads::addCurrentThread): (JSC::MachineThreads::removeThread): (JSC::MachineThreads::removeCurrentThread): (JSC::MachineThreads::gatherFromCurrentThreadInternal): (JSC::MachineThreads::gatherFromCurrentThread): (JSC::MachineThreads::gatherFromOtherThread): (JSC::MachineThreads::gatherConservativeRoots): * runtime/MachineStackMarker.h: * runtime/MarkStack.h: (JSC::MarkStack::append): 2011-03-13 David Kilzer <ddkilzer@apple.com> BUILD FIX for armv7 after r80969 Bug 56270 - The JIT 'friend's many classes in JSC; start unwinding this. <https://bugs.webkit.org/show_bug.cgi?id=56270> * assembler/MacroAssemblerARMv7.h: (JSC::MacroAssemblerARMv7::load32): Made void* address argument const. (JSC::MacroAssemblerARMv7::store32): Ditto. 2011-03-13 Geoffrey Garen <ggaren@apple.com> Not reviewed. Try to fix the Mac build. * JavaScriptCore.xcodeproj/project.pbxproj: Make sure to forward ConervativeSet.h, since it's now visible when compiling other projects. 2011-03-13 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed another case of DeprecatedPtr (ConservativeSet) https://bugs.webkit.org/show_bug.cgi?id=56281 The ConservativeSet is an internal data structure used during marking, so direct pointers are fine. * runtime/ConservativeSet.cpp: (JSC::ConservativeSet::grow): * runtime/ConservativeSet.h: Added some accessors, for use by MarkStack::append. (JSC::ConservativeSet::~ConservativeSet): Fixed a typo where we calculated the size of the set based on sizeof(DeprecatedPtr<T>*) instead of sizeof(DeprecatedPtr<T>). I'm not sure if this had real-world implications or not. (JSC::ConservativeSet::size): (JSC::ConservativeSet::set): Use direct pointers, as stated above. * runtime/Heap.cpp: (JSC::Heap::markRoots): * runtime/MarkStack.h: (JSC::MarkStack::append): Created a special case of append for ConservativeSet. I didn't want to add back a generic "append JSCell*" function, since other class might start using that wrong. (In the end, this function might go away, since the Heap will want to do something slightly more interesting with the conservative set, but this is OK for now.) 2011-03-13 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed another case of DeprecatedPtr (PutPropertySlot) https://bugs.webkit.org/show_bug.cgi?id=56278 * runtime/PutPropertySlot.h: (JSC::PutPropertySlot::setExistingProperty): (JSC::PutPropertySlot::setNewProperty): (JSC::PutPropertySlot::base): Direct pointer is fine for PutPropertySlot, since it's a stack-allocated temporary. 2011-03-13 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Removed one case of DeprecatedPtr (ScopeChainIterator) https://bugs.webkit.org/show_bug.cgi?id=56277 * runtime/ScopeChain.h: Direct pointer is fine for ScopeChainIterator, since it's a stack-allocated temporary. 2011-03-13 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 56273 - Add three operand forms to MacroAssember operations. Adding for X86(_64) for now, should be rolled out to other backends as necessary. These may allow more efficient code generation in some cases, avoiding the need for unnecessary register-register move instructions. * assembler/AbstractMacroAssembler.h: (JSC::AbstractMacroAssembler::Jump::link): (JSC::AbstractMacroAssembler::Jump::linkTo): - marked these methods const. (JSC::AbstractMacroAssembler::Jump::isSet): - add a method to check whether a Jump object has been set to reference an instruction, or is in a null, unset state. * assembler/MacroAssemblerCodeRef.h: (JSC::FunctionPtr::FunctionPtr): - add non-explicit constructor, for FunctionPtr's to C/C++ functions. * assembler/MacroAssemblerX86Common.h: (JSC::MacroAssemblerX86Common::and32): (JSC::MacroAssemblerX86Common::lshift32): (JSC::MacroAssemblerX86Common::or32): (JSC::MacroAssemblerX86Common::rshift32): (JSC::MacroAssemblerX86Common::urshift32): (JSC::MacroAssemblerX86Common::xor32): (JSC::MacroAssemblerX86Common::moveDouble): (JSC::MacroAssemblerX86Common::addDouble): (JSC::MacroAssemblerX86Common::divDouble): (JSC::MacroAssemblerX86Common::subDouble): (JSC::MacroAssemblerX86Common::mulDouble): (JSC::MacroAssemblerX86Common::branchTruncateDoubleToInt32): (JSC::MacroAssemblerX86Common::branchTest32): (JSC::MacroAssemblerX86Common::branchTest8): (JSC::MacroAssemblerX86Common::branchAdd32): (JSC::MacroAssemblerX86Common::branchMul32): (JSC::MacroAssemblerX86Common::branchSub32): - add three operand forms of these instructions. * assembler/MacroAssemblerX86_64.h: (JSC::MacroAssemblerX86_64::addDouble): (JSC::MacroAssemblerX86_64::convertInt32ToDouble): (JSC::MacroAssemblerX86_64::loadPtr): (JSC::MacroAssemblerX86_64::branchTestPtr): * assembler/X86Assembler.h: (JSC::X86Assembler::JmpSrc::isSet): - add a method to check whether a JmpSrc object has been set to reference an instruction, or is in a null, unset state. (JSC::X86Assembler::movsd_rr): - added FP register-register move. (JSC::X86Assembler::linkJump): - Add an assert to check jumps aren't linked more than once. * jit/JITInlineMethods.h: (JSC::JIT::emitLoadInt32ToDouble): - load integers to the FPU via regsiters on x86-64. 2011-03-13 Gavin Barraclough <barraclough@apple.com> ARM build fix. * assembler/MacroAssemblerARM.h: (JSC::MacroAssemblerARM::load32): 2011-03-13 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 56270 - The JIT 'friend's many classes in JSC; start unwinding this. The JIT need to 'friend' other classes in order to be able to calculate offsets of various properties, or the absolute addresses of members within specific objects, in order to JIT generate code that will access members within the class when run. Instead of using friends in these cases, switch to providing specific accessor methods to provide this information. In the case of offsets, these can be static functions, and in the case of pointers to members within a specific object these can be const methods returning pointers to const values, to prevent clients from modifying values otherwise encapsulated within classes. * bytecode/SamplingTool.h: * interpreter/Register.h: * interpreter/RegisterFile.h: * runtime/JSArray.h: * runtime/JSCell.h: * runtime/JSTypeInfo.h: * runtime/JSVariableObject.h: * runtime/Structure.h: * wtf/RefCounted.h: - Change these classes to no longer friend the JIT, add accessors for member offsets. * jit/JIT.cpp: * jit/JITCall32_64.cpp: * jit/JITInlineMethods.h: * jit/JITOpcodes.cpp: * jit/JITOpcodes32_64.cpp: * jit/JITPropertyAccess.cpp: * jit/JITPropertyAccess32_64.cpp: - Change the JIT to use class accessors, rather than taking object ofsets directly. * assembler/AbstractMacroAssembler.h: * assembler/MacroAssemblerX86_64.h: * assembler/X86Assembler.h: - Since the accessors for objects members return const pointers to retain encapsulation, methods generating code with absolute addresses must be able to handle const pointers (the JIT doesn't write to these values, do dies treat the pointer to value as const from within the C++ code of the JIT, if not at runtime!). 2011-03-12 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r80919. http://trac.webkit.org/changeset/80919 https://bugs.webkit.org/show_bug.cgi?id=56251 all windows bots failed to compile this change (Requested by loislo on #webkit). * JavaScriptCore.xcodeproj/project.pbxproj: * bytecode/StructureStubInfo.cpp: * interpreter/Register.h: (JSC::Register::withInt): (JSC::Register::withCallee): (JSC::Register::operator=): (JSC::Register::i): (JSC::Register::activation): (JSC::Register::function): (JSC::Register::propertyNameIterator): (JSC::Register::scopeChain): * jit/JIT.h: * jit/JITCall.cpp: (JSC::JIT::compileOpCallInitializeCallFrame): (JSC::JIT::compileOpCallVarargs): (JSC::JIT::compileOpCall): * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCallInitializeCallFrame): (JSC::JIT::compileOpCallVarargs): (JSC::JIT::compileOpCall): (JSC::JIT::compileOpCallSlowCase): * jit/JITInlineMethods.h: (JSC::JIT::emitPutToCallFrameHeader): * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTINativeCall): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_next_pname): (JSC::JIT::emit_op_load_varargs): (JSC::JIT::emitSlow_op_load_varargs): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTINativeCall): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_next_pname): * jit/JSInterfaceJIT.h: (JSC::JSInterfaceJIT::payloadFor): * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::returnJSValue): (JSC::SpecializedThunkJIT::returnDouble): (JSC::SpecializedThunkJIT::returnInt32): (JSC::SpecializedThunkJIT::returnJSCell): * runtime/ArgList.cpp: * runtime/DateConversion.cpp: * runtime/GCActivityCallbackCF.cpp: * runtime/Identifier.cpp: * runtime/JSActivation.h: (JSC::asActivation): * runtime/JSLock.cpp: * runtime/JSNumberCell.cpp: * runtime/JSObject.h: * runtime/JSPropertyNameIterator.h: * runtime/JSValue.h: * runtime/JSZombie.cpp: * runtime/MarkedBlock.cpp: * runtime/MarkedSpace.cpp: * runtime/PropertyNameArray.cpp: * runtime/ScopeChain.h: (JSC::ExecState::globalThisValue): * wtf/DateMath.cpp: 2011-03-11 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Ensure all values are correctly tagged in the registerfile https://bugs.webkit.org/show_bug.cgi?id=56214 This patch makes sure that all JSCell pointers written to the registerfile are correctly tagged as JSCells, and replaces raw int usage with the immediate representation. For performance, register pressure, and general saneness reasons I've added abstractions for reading and writing the tag and payload of integer registers directly for the JSVALUE64 encoding. * interpreter/Register.h: (JSC::Register::withInt): (JSC::Register::withCallee): (JSC::Register::operator=): (JSC::Register::i): (JSC::Register::activation): (JSC::Register::function): (JSC::Register::propertyNameIterator): (JSC::Register::scopeChain): * jit/JIT.h: * jit/JITCall.cpp: (JSC::JIT::compileOpCallInitializeCallFrame): (JSC::JIT::compileOpCallVarargs): (JSC::JIT::compileOpCall): * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCallInitializeCallFrame): (JSC::JIT::compileOpCallVarargs): (JSC::JIT::compileOpCall): (JSC::JIT::compileOpCallSlowCase): * jit/JITInlineMethods.h: (JSC::JIT::emitPutToCallFrameHeader): (JSC::JIT::emitPutCellToCallFrameHeader): (JSC::JIT::emitPutIntToCallFrameHeader): * jit/JITOpcodes.cpp: (JSC::JIT::privateCompileCTINativeCall): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_next_pname): (JSC::JIT::emit_op_load_varargs): (JSC::JIT::emitSlow_op_load_varargs): * jit/JITOpcodes32_64.cpp: (JSC::JIT::privateCompileCTINativeCall): (JSC::JIT::emit_op_get_pnames): (JSC::JIT::emit_op_next_pname): * jit/JSInterfaceJIT.h: (JSC::JSInterfaceJIT::intPayloadFor): (JSC::JSInterfaceJIT::intTagFor): * jit/SpecializedThunkJIT.h: (JSC::SpecializedThunkJIT::returnJSValue): (JSC::SpecializedThunkJIT::returnDouble): (JSC::SpecializedThunkJIT::returnInt32): (JSC::SpecializedThunkJIT::returnJSCell): 2011-03-11 Dimitri Glazkov <dglazkov@chromium.org> Reviewed by Eric Seidel. Introduce project_dir variable and make paths a whole lot saner. Ok, a little bit saner. https://bugs.webkit.org/show_bug.cgi?id=56231 * JavaScriptCore.gypi: Added project_dir variable. * gyp/JavaScriptCore.gyp: Changed to use project_dir, rather than DEPTH/JavaScriptCore. * gyp/generate-dtrace-header.sh: Changed to use project_dir. 2011-03-11 Dimitri Glazkov <dglazkov@chromium.org> Reviewed by Adam Barth. Start using derived sources correctly and link minidom with JavaScriptCore gyp project. https://bugs.webkit.org/show_bug.cgi?id=56217 * gyp/JavaScriptCore.gyp: Added derived source files and passing of shared directory to the scripts. * gyp/generate-derived-sources.sh: Changed to use passed directory. * gyp/generate-dtrace-header.sh: Ditto. 2011-03-11 Eric Carlson <eric.carlson@apple.com> Reviewed by Sam Weinig. <rdar://problem/8955589> Adopt AVFoundation media back end on Lion. No new tests, existing media tests cover this. * JavaScriptCore.exp: Export cancelCallOnMainThread * wtf/Platform.h: Define WTF_USE_AVFOUNDATION. 2011-03-11 Dimitri Glazkov <dglazkov@chromium.org> Reviewed by Adam Barth. Tweak dylib paths and add dtrace header generation action to JavaScriptCore gyp project. https://bugs.webkit.org/show_bug.cgi?id=56207 * JavaScriptCore.gypi: Added Tracing.d to the sources. * gyp/generate-dtrace-header.sh: Added. * gyp/JavaScriptCore.gyp: Updated dylib paths (now the project can see them), and added DTrace header generating step. 2011-03-10 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Fix allocation of native function with a cached thunk https://bugs.webkit.org/show_bug.cgi?id=56127 Fix this race condition found while fixing zombies. * collector/handles/HandleHeap.cpp: (JSC::HandleHeap::clearWeakPointers): * runtime/Heap.cpp: (JSC::Heap::reset): * runtime/JSFunction.cpp: (JSC::JSFunction::JSFunction): (JSC::JSFunction::markChildren): * runtime/JSValue.h: (JSC::JSValue::decode): * runtime/JSZombie.cpp: (JSC::JSZombie::leakedZombieStructure): * runtime/JSZombie.h: (JSC::JSZombie::createStructure): * runtime/MarkedBlock.cpp: 2011-03-10 Luiz Agostini <luiz.agostini@openbossa.org> Reviewed by Andreas Kling. [Qt] fast/workers/stress-js-execution.html is crashing on Qt bot (intermittently) https://bugs.webkit.org/show_bug.cgi?id=33008 Defining WTF_USE_PTHREAD_BASED_QT=1 for platforms where QThread uses pthread internally. Symbian is excluded because pthread_kill does not work on it. Mac is excluded because it has its own ways to do JSC threading. Defining WTF_USE_PTHREADS inside MachineStackMarker.cpp if USE(PTHREAD_BASED_QT) is true. * runtime/MachineStackMarker.cpp: * wtf/Platform.h: 2011-03-10 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. Bug 56077 - ES5 conformance issues with RegExp.prototype There are three issues causing test failures in sputnik. (1) lastIndex should be converted at the point it is used, not the point it is set (this is visible if valueOf is overridden). (2) The 'length' property of the test/exec functions should be 1. (3) If no input is specified, the input to test()/exec() is "undefined" (i.e. ToString(undefined)) - not RegExp.input. * runtime/RegExpObject.cpp: (JSC::RegExpObject::markChildren): - Added to mark lastIndex (JSC::regExpObjectLastIndex): (JSC::setRegExpObjectLastIndex): - lastIndex is now stored as a JSValue. (JSC::RegExpObject::match): - Use accessor methods to get/set lastIndex, add fast case for isUInt32 (don't convert to double). * runtime/RegExpObject.h: (JSC::RegExpObject::setLastIndex): (JSC::RegExpObject::setLastIndex): - Set lastIndex, either from a size_t or a JSValue. (JSC::RegExpObject::getLastIndex): - Get lastIndex. (JSC::RegExpObject::RegExpObjectData::RegExpObjectData): - Initialize as a JSValue. * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::RegExpPrototype): - Add test/exec properties with length 1. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncMatch): (JSC::stringProtoFuncSearch): - Do not read RegExp.input if none is provided. * tests/mozilla/js1_2/regexp/RegExp_input.js: * tests/mozilla/js1_2/regexp/RegExp_input_as_array.js: - Update these tests (they relied on non-ES5 behaviour). 2011-03-10 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Rolled back in 80277 and 80280 with event handler layout test failures fixed. https://bugs.webkit.org/show_bug.cgi?id=55653 The failures were caused by a last minute typo: assigning to currentEvent instead of m_currentEvent. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecompiler/BytecodeGenerator.cpp: * jit/JITOpcodes.cpp: * jit/JITOpcodes32_64.cpp: * runtime/Arguments.h: * runtime/JSActivation.cpp: * runtime/JSActivation.h: * runtime/JSCell.h: * runtime/JSGlobalObject.cpp: * runtime/JSGlobalObject.h: * runtime/JSObject.cpp: * runtime/JSStaticScopeObject.cpp: * runtime/JSStaticScopeObject.h: * runtime/JSVariableObject.h: * runtime/MarkedSpace.cpp: * runtime/MarkedSpace.h: 2011-03-09 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. jquery/manipulation.html fails after r80598 https://bugs.webkit.org/show_bug.cgi?id=56019 When linking a call, codeblock now takes ownership of the linked function This removes the need for unlinking, and thus the incorrectness that was showing up in these tests. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::~CodeBlock): (JSC::CodeBlock::markAggregate): * bytecode/CodeBlock.h: (JSC::CallLinkInfo::CallLinkInfo): (JSC::CallLinkInfo::setUnlinked): (JSC::CodeBlock::addCaller): * jit/JIT.cpp: (JSC::JIT::privateCompile): (JSC::JIT::linkCall): (JSC::JIT::linkConstruct): * jit/JIT.h: * runtime/Executable.cpp: * runtime/Executable.h: 2011-03-09 Daniel Bates <dbates@rim.com> Attempt to fix the WinCE build after changeset 80684 <http://trac.webkit.org/changeset/80684> (Bug #56041<https://bugs.webkit.org/show_bug.cgi?id=56041>). * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): Substitute variable callFrame for exec in call to createSyntaxError(). 2011-03-09 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 56041 - RexExp constructor should only accept flags "gim" Fix for issues introduced in r80667. Invalid flags to a RegExp literal are a late syntax error! * bytecode/CodeBlock.h: (JSC::CodeBlock::addRegExp): - Pass a PassRefPtr<RegExp> * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::addRegExp): (JSC::BytecodeGenerator::emitNewRegExp): * bytecompiler/BytecodeGenerator.h: - Pass a PassRefPtr<RegExp> * bytecompiler/NodesCodegen.cpp: (JSC::RegExpNode::emitBytecode): - Should not be ASSERTing that the flags are valid - this is a late(er) error. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): - Need to check for error from RegExp constructor. * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): - Need to check for error from RegExp constructor. * runtime/RegExp.h: (JSC::RegExp::isValid): - Make isValid check that the regexp was created with valid flags. * runtime/RegExpKey.h: - Since we'll not create RegExp objects with invalid flags, separate out the deleted value. 2011-03-09 Gavin Barraclough <barraclough@apple.com> Windows build fix part 2. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-03-09 Gavin Barraclough <barraclough@apple.com> Windows build fix part 1. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-03-09 Gavin Barraclough <barraclough@apple.com> Reviewed by Darin Adler. Bug 56041 - RexExp constructor should only accept flags "gim" We also should be passing the flags around as a bitfield rather than a string, and should not have redundant, incompatible code for converting the string to a bitfield! * JavaScriptCore.exp: * bytecompiler/NodesCodegen.cpp: (JSC::RegExpNode::emitBytecode): - Need to parse flags string to enum. * runtime/RegExp.cpp: (JSC::regExpFlags): (JSC::RegExp::RegExp): (JSC::RegExp::create): - Add method to parse flags string to enum, change constructor/create args to take enum. * runtime/RegExp.h: (JSC::RegExp::global): (JSC::RegExp::ignoreCase): (JSC::RegExp::multiline): - Change to use new enum values. * runtime/RegExpCache.cpp: (JSC::RegExpCache::lookupOrCreate): (JSC::RegExpCache::create): * runtime/RegExpCache.h: - Changed to use regExpFlags enum instead of int/const UString&. * runtime/RegExpConstructor.cpp: (JSC::constructRegExp): - Add use new enum parsing, check for error. * runtime/RegExpKey.h: (JSC::RegExpKey::RegExpKey): * runtime/RegExpPrototype.cpp: (JSC::RegExpPrototype::RegExpPrototype): - Pass NoFlags value instead of empty string. (JSC::regExpProtoFuncCompile): - Add use new enum parsing, check for error. * runtime/StringPrototype.cpp: (JSC::stringProtoFuncMatch): (JSC::stringProtoFuncSearch): - Pass NoFlags value instead of empty string. 2011-03-08 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig Bug 55994 - Functions on Array.prototype should check length first. These methods are designed to work on generic objects too, and if 'length' is a getter that throws an exception, ensure this is correctly thrown (even if other exceptions would be thrown, too). Make the length check the first thing we do. This change shows a progression on SunSpider on my machine, but this is likely bogus. * runtime/ArrayPrototype.cpp: (JSC::arrayProtoFuncToString): (JSC::arrayProtoFuncToLocaleString): (JSC::arrayProtoFuncJoin): (JSC::arrayProtoFuncPop): (JSC::arrayProtoFuncPush): (JSC::arrayProtoFuncReverse): (JSC::arrayProtoFuncShift): (JSC::arrayProtoFuncSlice): (JSC::arrayProtoFuncSort): (JSC::arrayProtoFuncSplice): (JSC::arrayProtoFuncUnShift): (JSC::arrayProtoFuncFilter): (JSC::arrayProtoFuncMap): (JSC::arrayProtoFuncEvery): (JSC::arrayProtoFuncForEach): (JSC::arrayProtoFuncSome): (JSC::arrayProtoFuncReduce): (JSC::arrayProtoFuncReduceRight): (JSC::arrayProtoFuncIndexOf): (JSC::arrayProtoFuncLastIndexOf): 2011-03-07 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Make CodeBlock GC write barrier safe https://bugs.webkit.org/show_bug.cgi?id=55910 In order to make CodeBlock WriteBarrier safe it was necessary to make it have a single GC owner, and for that reason I have made ExecutableBase a GC allocated object. This required updating their creation routines as well as all sites that hold a reference to them. GC objects that held Executable's have been converted to WriteBarriers, and all other sites now use Global<>. As an added benefit this gets rid of JSGlobalData's list of GlobalCodeBlocks. Perf testing shows a 0.5% progression on v8, vs. a 0.3% regression on SunSpider. Given none of the tests that show regressions demonstrate a regression on their own, and sampling shows up nothing. I suspect we're just getting one or two additional gc passes at the end of the run. * bytecode/CodeBlock.cpp: (JSC::CodeBlock::dump): (JSC::CodeBlock::CodeBlock): (JSC::EvalCodeCache::markAggregate): (JSC::CodeBlock::markAggregate): * bytecode/CodeBlock.h: (JSC::CodeBlock::ownerExecutable): (JSC::CodeBlock::addConstant): (JSC::CodeBlock::constantRegister): (JSC::CodeBlock::getConstant): (JSC::CodeBlock::addFunctionDecl): (JSC::CodeBlock::addFunctionExpr): (JSC::GlobalCodeBlock::GlobalCodeBlock): (JSC::ExecState::r): * bytecode/EvalCodeCache.h: (JSC::EvalCodeCache::get): * bytecode/SamplingTool.h: (JSC::ScriptSampleRecord::ScriptSampleRecord): * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::addConstantValue): (JSC::BytecodeGenerator::emitEqualityOp): * bytecompiler/BytecodeGenerator.h: (JSC::BytecodeGenerator::makeFunction): * debugger/Debugger.cpp: (JSC::evaluateInGlobalCallFrame): * debugger/DebuggerCallFrame.cpp: (JSC::DebuggerCallFrame::evaluate): * interpreter/Interpreter.cpp: (JSC::Interpreter::callEval): * jit/JITInlineMethods.h: (JSC::JIT::emitLoadDouble): (JSC::JIT::emitLoadInt32ToDouble): * jit/JITStubs.cpp: (JSC::JITThunks::JITThunks): (JSC::JITThunks::hostFunctionStub): (JSC::JITThunks::clearHostFunctionStubs): * jit/JITStubs.h: * runtime/Completion.cpp: (JSC::checkSyntax): (JSC::evaluate): * runtime/Executable.cpp: (JSC::EvalExecutable::EvalExecutable): (JSC::ProgramExecutable::ProgramExecutable): (JSC::FunctionExecutable::FunctionExecutable): (JSC::FunctionExecutable::~FunctionExecutable): (JSC::EvalExecutable::markChildren): (JSC::ProgramExecutable::markChildren): (JSC::FunctionExecutable::markChildren): (JSC::FunctionExecutable::fromGlobalCode): * runtime/Executable.h: (JSC::ExecutableBase::ExecutableBase): (JSC::ExecutableBase::createStructure): (JSC::NativeExecutable::create): (JSC::NativeExecutable::NativeExecutable): (JSC::VPtrHackExecutable::VPtrHackExecutable): (JSC::ScriptExecutable::ScriptExecutable): (JSC::EvalExecutable::create): (JSC::EvalExecutable::createStructure): (JSC::ProgramExecutable::create): (JSC::ProgramExecutable::createStructure): (JSC::FunctionExecutable::create): (JSC::FunctionExecutable::createStructure): * runtime/FunctionConstructor.cpp: (JSC::constructFunction): * runtime/Heap.cpp: (JSC::Heap::destroy): (JSC::Heap::markRoots): * runtime/Heap.h: * runtime/JSActivation.cpp: (JSC::JSActivation::JSActivation): (JSC::JSActivation::markChildren): * runtime/JSActivation.h: (JSC::JSActivation::JSActivationData::JSActivationData): * runtime/JSCell.h: * runtime/JSFunction.cpp: (JSC::JSFunction::JSFunction): (JSC::JSFunction::~JSFunction): (JSC::JSFunction::markChildren): * runtime/JSFunction.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::storeVPtrs): (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::getHostFunction): * runtime/JSGlobalData.h: * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncEval): * runtime/JSObject.cpp: * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::markChildren): * runtime/JSStaticScopeObject.h: (JSC::JSStaticScopeObject::JSStaticScopeObjectData::JSStaticScopeObjectData): (JSC::JSStaticScopeObject::JSStaticScopeObject): * runtime/JSZombie.cpp: (JSC::JSZombie::leakedZombieStructure): * runtime/JSZombie.h: (JSC::JSZombie::createStructure): * runtime/MarkedSpace.h: 2011-03-07 Andy Estes <aestes@apple.com> Reviewed by Dan Bernstein. REGRESSION (r79060): Timestamp is missing from tweets in twitter. https://bugs.webkit.org/show_bug.cgi?id=55228 A change to the date parser to handle the case where the year is specified before the time zone inadvertently started accepting strings such as '+0000' as valid years. Those strings actually represent time zones in an offset of hours and minutes from UTC, not years. * wtf/DateMath.cpp: (WTF::parseDateFromNullTerminatedCharacters): If the current character in dateString is '+' or '-', do not try to parse the next token as a year. 2011-03-06 Yuta Kitamura <yutak@chromium.org> Reviewed by Kent Tamura. Add SHA-1 for new WebSocket protocol https://bugs.webkit.org/show_bug.cgi?id=55039 The code is based on Chromium's portable SHA-1 implementation (src/base/sha1_portable.cc). Modifications were made in order to make the code comply with WebKit coding style. * GNUmakefile.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * wtf/CMakeLists.txt: * wtf/MD5.cpp: (WTF::MD5::MD5): * wtf/SHA1.cpp: Added. (WTF::testSHA1): This function will be run the first time SHA1 constructor is called. This function computes a few hash values and checks the results in debug builds. However, constructor is probably not a good place to run these tests, so we need to find a good place for it (bug 55853). (WTF::expectSHA1): (WTF::f): (WTF::k): (WTF::rotateLeft): (WTF::SHA1::SHA1): (WTF::SHA1::addBytes): (WTF::SHA1::computeHash): (WTF::SHA1::finalize): (WTF::SHA1::processBlock): (WTF::SHA1::reset): * wtf/SHA1.h: Added. (WTF::SHA1::addBytes): * wtf/wtf.pri: 2011-03-05 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. Add Derived Sources to WebCore GYP build https://bugs.webkit.org/show_bug.cgi?id=55813 Rename the action to be friendlier. * gyp/JavaScriptCore.gyp: 2011-03-04 Viatcheslav Ostapenko <ostapenko.viatcheslav@nokia.com> Reviewed by Laszlo Gombos. [Qt] Need symbian version of cryptographicallyRandomValuesFromOS https://bugs.webkit.org/show_bug.cgi?id=55782 Implement Symbian version of cryptographicallyRandomValuesFromOS * wtf/OSRandomSource.cpp: (WTF::cryptographicallyRandomValuesFromOS): 2011-03-04 Gavin Barraclough <barraclough@apple.com> Reviewed by Cameron Zwarich. Bug 55815 - Should throw an exception from JSObject::defineOwnProperty if !isExtensible(). * runtime/JSObject.cpp: (JSC::JSObject::defineOwnProperty): Add missing check. 2011-03-04 Gavin Barraclough <barraclough@apple.com> Rubber stamped by olliej. Bug 54945 - The web page hangs towards the end of page load in Interpreter enabled javascript code in the latest webkit trunk. * interpreter/Interpreter.cpp: (JSC::Interpreter::privateExecute): (1) don't infinite loop. (2) goto 1. 2011-03-04 Gavin Barraclough <barraclough@apple.com> cmake build fix. * CMakeLists.txt: 2011-03-04 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. Add Copy Files step to JavaScriptCore GYP build for apitest and minidom https://bugs.webkit.org/show_bug.cgi?id=55798 * JavaScriptCore.gypi: * gyp/JavaScriptCore.gyp: 2011-03-04 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. Remove unneeded round-trips through ../Source in the Chromium GYP build https://bugs.webkit.org/show_bug.cgi?id=55795 * JavaScriptCore.gyp/JavaScriptCore.gyp: 2011-03-04 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. Use target_defaults to reduce boilerplate in GYP build system https://bugs.webkit.org/show_bug.cgi?id=55790 Instead of setting up the configuration in each target, just defer to target_defaults. Also, removed a define that was redundant with the xcconfig. * gyp/JavaScriptCore.gyp: 2011-03-03 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 55736 - Implement seal/freeze/preventExtensions for normal object types. Provide basic functionallity from section 15.2.4 of ECMA-262. This support will need expanding to cover arrays, too. Shows a 0.5% progression on SunSpidey, this seems to be due to changing ObjectConstructor to use a static table. * DerivedSources.make: * JavaScriptCore.exp: * interpreter/CallFrame.h: (JSC::ExecState::objectConstructorTable): Add a static table for ObjectConstructor. * runtime/CommonIdentifiers.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): (JSC::JSGlobalData::~JSGlobalData): Add a static table for ObjectConstructor. * runtime/JSGlobalData.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::reset): Add a static table for ObjectConstructor. * runtime/JSObject.cpp: (JSC::JSObject::seal): (JSC::JSObject::freeze): (JSC::JSObject::preventExtensions): Transition the object's structure. (JSC::JSObject::defineOwnProperty): Check isExtensible. * runtime/JSObject.h: (JSC::JSObject::isSealed): (JSC::JSObject::isFrozen): (JSC::JSObject::isExtensible): These wrap method on structure. (JSC::JSObject::putDirectInternal): Check isExtensible. * runtime/ObjectConstructor.cpp: (JSC::ObjectConstructor::ObjectConstructor): (JSC::ObjectConstructor::getOwnPropertySlot): (JSC::ObjectConstructor::getOwnPropertyDescriptor): Change ObjectConstructor to use a static table. (JSC::objectConstructorSeal): (JSC::objectConstructorFreeze): (JSC::objectConstructorPreventExtensions): (JSC::objectConstructorIsSealed): (JSC::objectConstructorIsFrozen): (JSC::objectConstructorIsExtensible): Add new methods on Object. * runtime/ObjectConstructor.h: (JSC::ObjectConstructor::createStructure): * runtime/Structure.cpp: (JSC::Structure::Structure): init/propagate m_preventExtensions (JSC::Structure::sealTransition): (JSC::Structure::freezeTransition): (JSC::Structure::preventExtensionsTransition): transition the structure, materializing the property map, setting m_preventExtensions & changing attributes. (JSC::Structure::isSealed): (JSC::Structure::isFrozen): check attributes to detect if object is sealed/frozen. * runtime/Structure.h: (JSC::Structure::isExtensible): checks the m_preventExtensions flag. 2011-03-04 Steve Falkenburg <sfalken@apple.com> Reviewed by Jon Honeycutt. Adopt VersionStamper tool for Windows WebKit DLLs https://bugs.webkit.org/show_bug.cgi?id=55784 <rdar://problem/9021273> We now use a tool to stamp the version number onto the Apple WebKit DLLs during the post-build step. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.rc: Removed. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.vcproj: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCorePostBuild.cmd: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCorePreBuild.cmd: 2011-03-04 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. JavaScriptCore GYP build should use a header map https://bugs.webkit.org/show_bug.cgi?id=55712 This patch moves the os-win32 files into their own variable so that we can use a header map in the Apple Mac Xcode build. The problem is that the header map searches the whole project rather than just the files included in a given target. Another solution to this problem is to make GYP smarter about filtering out what files are added to the project file. * JavaScriptCore.gypi: * gyp/JavaScriptCore.gyp: 2011-03-03 Ryosuke Niwa <rniwa@webkit.org> Reviewed by Darin Adler. Remove LOOSE_PASS_OWN_ARRAY_PTR from PassOwnArrayPtr.h https://bugs.webkit.org/show_bug.cgi?id=55554 * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::copyGlobalsTo): Pass nullptr instead of 0. (JSC::JSGlobalObject::resizeRegisters): Ditto; also use OwnArrayPtr instead of a raw pointer. * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::addStaticGlobals): Ditto. * wtf/PassOwnArrayPtr.h: Removed #define LOOSE_PASS_OWN_ARRAY_PTR (WTF::PassOwnArrayPtr::PassOwnArrayPtr): Added a constructor that takes nullptr_t. 2011-03-03 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. Add jsc to JavaScriptCore GYP build https://bugs.webkit.org/show_bug.cgi?id=55711 * JavaScriptCore.gypi: - Move jsc.cpp into jsc_files because it's really part of the jsc target. * JavaScriptCore.xcodeproj/project.pbxproj: - Remove extraneous files from the normal jsc build. I probably added these by mistake at some point. * gyp/JavaScriptCore.gyp: - Add the jsc target to the GYP file. 2011-03-03 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. Add testapi to JavaScriptCore GYP build https://bugs.webkit.org/show_bug.cgi?id=55707 The new testapi target is slightly incomplete. There's a resource copying step that we don't quite have yet. This patch also cleans up some of the configuration issues in JavaScriptCore.xcodeproj. It seems kind of wordy to repeat these for each target. I suspect there's a more compact way of defining the configurations, but this removes the "Default" configuration, which is progress. * JavaScriptCore.gypi: * gyp/JavaScriptCore.gyp: 2011-03-03 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. Teach JavaScriptCore GYP build about private headers https://bugs.webkit.org/show_bug.cgi?id=55532 This patch distinguishes between public and private framework headers so that public headers are copied into the Headers directory and private headers are copied into the PrivateHeaders directory. * gyp/JavaScriptCore.gyp: 2011-03-03 Geoffrey Garen <ggaren@apple.com> Rolled out 80277 and 80280 because they caused event handler layout test failures. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.xcodeproj/project.pbxproj: * bytecompiler/BytecodeGenerator.cpp: * jit/JITOpcodes.cpp: * jit/JITOpcodes32_64.cpp: * runtime/Arguments.h: * runtime/JSActivation.cpp: * runtime/JSActivation.h: * runtime/JSCell.h: * runtime/JSGlobalObject.cpp: * runtime/JSGlobalObject.h: * runtime/JSObject.cpp: * runtime/JSStaticScopeObject.cpp: * runtime/JSStaticScopeObject.h: * runtime/JSVariableObject.h: * runtime/MarkedSpace.cpp: * runtime/MarkedSpace.h: 2011-03-03 Kevin Ollivier <kevino@theolliviers.com> [wx] Build fix. Alter order of headers included to make sure windows.h is configured by wx, and skip Posix implementation file we don't use on Win. * wscript: * wtf/wx/StringWx.cpp: 2011-03-03 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. JSVariableObject needs to use WriteBarrier for symboltable property storage https://bugs.webkit.org/show_bug.cgi?id=55698 Replace the direct usage of Register in JSVariableObject (and descendents) with WriteBarrier. This requires updating the Arguments object to use WriteBarrier as well. * interpreter/Interpreter.cpp: (JSC::Interpreter::unwindCallFrame): (JSC::Interpreter::privateExecute): (JSC::Interpreter::retrieveArguments): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * runtime/ArgList.h: (JSC::MarkedArgumentBuffer::initialize): * runtime/Arguments.cpp: (JSC::Arguments::markChildren): (JSC::Arguments::copyToRegisters): (JSC::Arguments::fillArgList): (JSC::Arguments::getOwnPropertySlot): (JSC::Arguments::getOwnPropertyDescriptor): (JSC::Arguments::put): * runtime/Arguments.h: (JSC::Arguments::setActivation): (JSC::Arguments::Arguments): (JSC::Arguments::copyRegisters): (JSC::JSActivation::copyRegisters): * runtime/JSActivation.cpp: (JSC::JSActivation::markChildren): (JSC::JSActivation::symbolTableGet): (JSC::JSActivation::symbolTablePut): (JSC::JSActivation::symbolTablePutWithAttributes): (JSC::JSActivation::put): (JSC::JSActivation::putWithAttributes): (JSC::JSActivation::argumentsGetter): * runtime/JSActivation.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::put): (JSC::JSGlobalObject::putWithAttributes): (JSC::JSGlobalObject::markChildren): (JSC::JSGlobalObject::copyGlobalsFrom): (JSC::JSGlobalObject::copyGlobalsTo): (JSC::JSGlobalObject::resizeRegisters): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::setRegisters): (JSC::JSGlobalObject::addStaticGlobals): * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::put): (JSC::JSStaticScopeObject::putWithAttributes): * runtime/JSVariableObject.cpp: (JSC::JSVariableObject::symbolTableGet): * runtime/JSVariableObject.h: (JSC::JSVariableObject::registerAt): (JSC::JSVariableObject::JSVariableObjectData::JSVariableObjectData): (JSC::JSVariableObject::symbolTableGet): (JSC::JSVariableObject::symbolTablePut): (JSC::JSVariableObject::symbolTablePutWithAttributes): (JSC::JSVariableObject::copyRegisterArray): (JSC::JSVariableObject::setRegisters): 2011-03-03 Geoffrey Garen <ggaren@apple.com> Try to fix Windows build. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Removed obsolete symbol. * runtime/JSStaticScopeObject.cpp: (JSC::JSStaticScopeObject::getOwnPropertySlot): Don't mark this function inline -- it's virtual. 2011-03-02 Geoffrey Garen <ggaren@apple.com> Reviewed by Darin Adler. Moved all variable object storage inline -- upping the object size limit to 1K https://bugs.webkit.org/show_bug.cgi?id=55653 * JavaScriptCore.exp: * bytecompiler/BytecodeGenerator.cpp: * jit/JITOpcodes.cpp: * runtime/Arguments.h: * runtime/JSActivation.h: Removed out-of-line storage. Changed d-> to m_. * runtime/JSCell.h: (JSC::JSCell::MarkedSpace::sizeClassFor): Added an imprecise size class to accomodate objects up to 1K. * runtime/JSGlobalObject.cpp: * runtime/JSGlobalObject.h: Removed out-of-line storage. Changed d-> to m_. * runtime/JSObject.cpp: Don't ASSERT that JSFinalObject fills the maximum object size, since it doesn't anymore. * runtime/JSStaticScopeObject.cpp: * runtime/JSStaticScopeObject.h: * runtime/JSVariableObject.h: Removed out-of-line storage. Changed d-> to m_. * runtime/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::reset): * runtime/MarkedSpace.h: Added an imprecise size class to accomodate objects up to 1K. 2011-03-03 Timothy Hatcher <timothy@apple.com> Make APIShims usable from WebCore. Reviewed by Oliver Hunt. * ForwardingHeaders/JavaScriptCore/APIShims.h: Added. * GNUmakefile.am: * JavaScriptCore.exp: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCoreGenerated.make: * JavaScriptCore.xcodeproj/project.pbxproj: 2011-03-03 Peter Varga <pvarga@webkit.org> Reviewed by Oliver Hunt. Begin Characters Optimization Causes YARR Interpreter Errors https://bugs.webkit.org/show_bug.cgi?id=55479 The addBeginTerm function is removed because it doesn't correctly handle those cases when an "invalid" term has been collected (e.g. CharacterClass). Move the removed function to the setupAlternativeBeginTerms method's switch-case where the non-allowed cases are correctly handled. Reenable the Beginning Character Optimization in the YARR Interpreter again. * yarr/YarrPattern.cpp: (JSC::Yarr::YarrPatternConstructor::setupAlternativeBeginTerms): (JSC::Yarr::YarrPattern::compile): 2011-03-02 Jessie Berlin <jberlin@apple.com> Reviewed by Adam Roben. WebKit2: Use CFNetwork Sessions API. https://bugs.webkit.org/show_bug.cgi?id=55435 Add the ability to create a Private Browsing storage session. * wtf/Platform.h: Add a new #define for using CF Storage Sessions. 2011-03-02 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Remove "register slot" concept from PropertySlot https://bugs.webkit.org/show_bug.cgi?id=55621 PropertySlot had already stopped storing Register "slots" so this patch is simply removing that api entirely. This exposed a problem in the ProgramNode constructor for BytecodeGenerator where it reads from the registerfile before it has initialised it. This bug wasn't a problem before as we were merely testing for property existence rather than the actual value, and used to work because setRegisterSlot didn't check that the provided slot contained an initialised value. To get around this issue we now use symbolTableHasProperty to do the symbol table check without trying to read the RegisterFile. * JavaScriptCore.xcodeproj/project.pbxproj: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::BytecodeGenerator): * runtime/Arguments.cpp: (JSC::Arguments::getOwnPropertySlot): * runtime/JSActivation.cpp: (JSC::JSActivation::symbolTableGet): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::symbolTableHasProperty): * runtime/JSVariableObject.h: (JSC::JSVariableObject::symbolTableGet): * runtime/PropertySlot.h: 2011-03-02 Daniel Cheng <dcheng@chromium.org> Reviewed by David Levin. Add feature define for data transfer items https://bugs.webkit.org/show_bug.cgi?id=55510 * Configurations/FeatureDefines.xcconfig: * wtf/Platform.h: 2011-03-02 Adam Roben <aroben@apple.com> Delete old .res files whenever any .vsprops file changes Prospective fix for <http://webkit.org/b/55599> r80079 caused incremental Windows builds to fail Reviewed by Tony Chang. * JavaScriptCore.vcproj/JavaScriptCore/react-to-vsprops-changes.py: (main): Restructured code to loop over a set of file extensions, deleting any old files that have that extension. Now deletes .res files, too. (We previously deleted any file matching *.manifest*, but that turned out to just be the union of *.manifest and *.res.) 2011-03-02 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. Teach JavaScriptCore GYP build how to build minidom https://bugs.webkit.org/show_bug.cgi?id=55536 * JavaScriptCore.gypi: * gyp/JavaScriptCore.gyp: 2011-03-01 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. JavaScriptCore GYP build should copy some headers into the target framework https://bugs.webkit.org/show_bug.cgi?id=55524 After this patch, all the framework headers are exported as public headers. We need to teach GYP how to handle private headers. I struggled to determine how to store the information about whether a header was public, private, or project (i.e., not exported). Generally, the GYPI should just list the files, but it seemed siliy to have an almost duplicated list of files in the GYP file itself. If this design doesn't scale, we might have to revisit it in the future. * JavaScriptCore.gyp/JavaScriptCore.gyp: * JavaScriptCore.gypi: * gyp/JavaScriptCore.gyp: 2011-03-01 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r80079. http://trac.webkit.org/changeset/80079 https://bugs.webkit.org/show_bug.cgi?id=55547 "Broke the Win debug build?" (Requested by dcheng on #webkit). * wtf/Platform.h: 2011-03-01 Daniel Cheng <dcheng@chromium.org> Reviewed by David Levin. Add feature define for data transfer items https://bugs.webkit.org/show_bug.cgi?id=55510 * wtf/Platform.h: 2011-03-01 Oliver Hunt <oliver@apple.com> Reviewed by Joseph Pecoraro. Misaligned memory access in CloneDeserializer on all ARM arch. https://bugs.webkit.org/show_bug.cgi?id=48742 Add a CPU class for architectures that need aligned addresses for memory access. * wtf/Platform.h: 2011-03-01 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. Add pre- and post-build actions for JavaScriptCore GYP build https://bugs.webkit.org/show_bug.cgi?id=55507 After this patch, we have all the steps for building the main JavaScriptCore framework except the "copy headers" step, which I'll do next. * gyp/JavaScriptCore.gyp: 2011-03-01 Geoffrey Garen <ggaren@apple.com> Reviewed by Sam Weinig. Rolled back in r79627 now that the underlying cause for it crashing is fixed. https://bugs.webkit.org/show_bug.cgi?id=55159 * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * JavaScriptCore.xcodeproj/project.pbxproj: * runtime/Heap.cpp: (JSC::Heap::allocateSlowCase): * runtime/Heap.h: * runtime/JSCell.h: (JSC::JSCell::MarkedSpace::sizeClassFor): (JSC::JSCell::Heap::allocate): (JSC::JSCell::JSCell::operator new): * runtime/MarkedBlock.h: * runtime/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::allocateBlock): (JSC::MarkedSpace::reset): * runtime/MarkedSpace.h: (JSC::MarkedSpace::SizeClass::SizeClass): 2011-03-01 Mark Rowe <mrowe@apple.com> Reviewed by Sam Weinig. Replace two script phases that do nothing but copy files with copy files build phases. This speeds up the build by a few seconds on high-end Mac Pros. * JavaScriptCore.xcodeproj/project.pbxproj: 2011-03-01 David Kilzer <ddkilzer@apple.com> Spring cleaning! Rubber-stamped by Mark Rowe. * JavaScriptCore.xcodeproj/project.pbxproj: (Copy Into Framework): Remove "set -x" and its comment. 2011-03-01 Michael Saboff <msaboff@apple.com> Reviewed by Darin Adler. TinyMCE not working in nightlies https://bugs.webkit.org/show_bug.cgi?id=54978 Disabling setupBeginChars() to temporarily work arround the test failure. Filed https://bugs.webkit.org/show_bug.cgi?id=55479 to track fixing the issue. * yarr/YarrPattern.cpp: (JSC::Yarr::YarrPattern::compile): 2011-02-23 Joseph Pecoraro <joepeck@webkit.org> Reviewed by Kenneth Rohde Christiansen. Viewport parsing no longer accepts "1.0;" value as valid. https://bugs.webkit.org/show_bug.cgi?id=53705 Include a didReadNumber parameter to String -> float / double conversion functions. This way, if the "ok" boolean out parameter is false, you can check to see if there in fact was a valid number parsed with garbage at the end. Examples of that would be parsing "123x456" would have ok = false, but didReadNumber = true. * JavaScriptCore.exp: * wtf/text/StringImpl.cpp: (WTF::StringImpl::toDouble): (WTF::StringImpl::toFloat): * wtf/text/StringImpl.h: * wtf/text/WTFString.cpp: (WTF::String::toDouble): (WTF::String::toFloat): (WTF::charactersToDouble): (WTF::charactersToFloat): * wtf/text/WTFString.h: 2011-02-28 Geoffrey Garen <ggaren@apple.com> Reviewed by Gavin Barraclough. Past-the-end writes in VM exceptions (caused crashes in r79627) https://bugs.webkit.org/show_bug.cgi?id=55448 Some exceptions had the wrong structures, so they misoverestimated their inline storage sizes. * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): Use the right structure. * runtime/JSObject.h: (JSC::JSNonFinalObject::JSNonFinalObject): (JSC::JSFinalObject::JSFinalObject): ASSERT that our structure capacity is correct to verify this doesn't happen again. 2011-03-01 Andras Becsi <abecsi@webkit.org> Reviewed by Csaba Osztrogonác. [Qt] Clean up the project files and move common options to WebKit.pri. * JavaScriptCore.pri: Move options also needed in WebCore into WebKit.pri. * JavaScriptCore.pro: Deduplicate options. * jsc.pro: Ditto. 2011-03-01 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. Teach JavaScriptCore GYP build about DEPTH https://bugs.webkit.org/show_bug.cgi?id=55425 In addition to teaching the JavaScriptCore GYP build about DEPTH, this change overrides the GCC warning configuration to disable a warning that's causing probems in Assertions.cpp. With that warning disabled, JavaScriptCore builds again. * gyp/JavaScriptCore.gyp: 2011-02-28 Gavin Barraclough <barraclough@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-02-28 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r79948. http://trac.webkit.org/changeset/79948 https://bugs.webkit.org/show_bug.cgi?id=55439 "caused crashes on the SL release bot" (Requested by ggaren on #webkit). * runtime/JSGlobalData.h: * runtime/WriteBarrier.h: 2011-02-28 Gavin Barraclough <barraclough@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-02-28 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig & Darin Adler. Bug 55423 - Clean up property tables in Structure Encapsulate, reduce duplication of table search code, and reduce the size of the tables (remove the index, just maintain the tables in the correct order). Shows a 0.5% - 1% progression on sunspider. * JavaScriptCore.exp: * runtime/PropertyMapHashTable.h: (JSC::isPowerOf2): (JSC::nextPowerOf2): bit ops used to calculate table size. (JSC::PropertyMapEntry::PropertyMapEntry): (JSC::PropertyTable::ordered_iterator::operator++): (JSC::PropertyTable::ordered_iterator::operator==): (JSC::PropertyTable::ordered_iterator::operator!=): (JSC::PropertyTable::ordered_iterator::operator*): (JSC::PropertyTable::ordered_iterator::operator->): (JSC::PropertyTable::ordered_iterator::ordered_iterator): implementation of the iterator types (JSC::PropertyTable::PropertyTable): (JSC::PropertyTable::~PropertyTable): constructors take an initial capacity for the table, a table to copy, or both. (JSC::PropertyTable::begin): (JSC::PropertyTable::end): create in-order iterators. (JSC::PropertyTable::find): search the hash table (JSC::PropertyTable::add): add a value to the hash table (JSC::PropertyTable::remove): remove a value from the hash table (JSC::PropertyTable::size): (JSC::PropertyTable::isEmpty): accessors. (JSC::PropertyTable::propertyStorageSize): (JSC::PropertyTable::clearDeletedOffsets): (JSC::PropertyTable::hasDeletedOffset): (JSC::PropertyTable::getDeletedOffset): (JSC::PropertyTable::addDeletedOffset): cache deleted (available) offsets in the property storage array. (JSC::PropertyTable::copy): take a copy of the PropertyTable, potentially expanding the capacity. (JSC::PropertyTable::sizeInMemory): used for DEBUG build statistics (JSC::PropertyTable::reinsert): (JSC::PropertyTable::rehash): (JSC::PropertyTable::tableCapacity): (JSC::PropertyTable::deletedEntryIndex): (JSC::PropertyTable::skipDeletedEntries): (JSC::PropertyTable::table): (JSC::PropertyTable::usedCount): (JSC::PropertyTable::dataSize): (JSC::PropertyTable::sizeForCapacity): (JSC::PropertyTable::canInsert): these methods provide internal implementation. * runtime/Structure.cpp: (JSC::Structure::dumpStatistics): (JSC::Structure::~Structure): (JSC::Structure::materializePropertyMap): (JSC::Structure::despecifyDictionaryFunction): (JSC::Structure::addPropertyTransition): (JSC::Structure::flattenDictionaryStructure): (JSC::Structure::copyPropertyTable): (JSC::Structure::get): (JSC::Structure::despecifyFunction): (JSC::Structure::despecifyAllFunctions): (JSC::Structure::put): (JSC::Structure::remove): (JSC::Structure::createPropertyMap): (JSC::Structure::getPropertyNames): (JSC::PropertyTable::checkConsistency): (JSC::Structure::checkConsistency): factored out code to PropertyMapHashTable.h * runtime/Structure.h: (JSC::Structure::propertyStorageSize): (JSC::Structure::isEmpty): (JSC::Structure::get): factored out code to PropertyMapHashTable.h 2011-02-28 Xan Lopez <xlopez@igalia.com> Another fix build :( Fix typo. * runtime/MachineStackMarker.cpp: (JSC::freePlatformThreadRegisters): 2011-02-28 Xan Lopez <xlopez@igalia.com> Unreviewed build fix for Snow Leopard. * runtime/MachineStackMarker.cpp: (JSC::freePlatformThreadRegisters): 2011-02-28 Alejandro G. Castro <alex@igalia.com> Unreviewed, fix SnowLeopard compilation after r79952. * runtime/MachineStackMarker.cpp: (JSC::freePlatformThreadRegisters): 2011-02-28 Mark Rowe <mrowe@apple.com> Reviewed by Darin Adler. <http://webkit.org/b/55430> OwnArrayPtr.h's LOOSE_OWN_ARRAY_PTR results in link errors. * wtf/OwnArrayPtr.h: (WTF::::set): Implement OwnArrayPtr::set. 2011-02-28 Martin Zoubek <martin.zoubek@acision.com> and Alejandro G. Castro <alex@igalia.com> Reviewed by Martin Robinson. Multithread support for JSC on UNIX https://bugs.webkit.org/show_bug.cgi?id=26838 Implement suspendThread() and resumeThread() for systems with pthread.h using thread signal handler. * runtime/MachineStackMarker.cpp: (JSC::pthreadSignalHandlerSuspendResume): (JSC::MachineStackMarker::Thread::Thread): (JSC::getCurrentPlatformThread): (JSC::suspendThread): (JSC::resumeThread): (JSC::getPlatformThreadRegisters): (JSC::otherThreadStackPointer): (JSC::freePlatformThreadRegisters): (JSC::MachineStackMarker::markOtherThreadConservatively): * wtf/Platform.h: Added Gtk port to use ENABLE_JSC_MULTIPLE_THREADS. 2011-02-28 Oliver Hunt <oliver@apple.com> Reviewed by Darin Adler. Stop using DeprecatedPtr for the global exception slot https://bugs.webkit.org/show_bug.cgi?id=55424 Create GCRootPtr to signify that the exception slot is a gcroot, and so is exempt from the usual writebarrier restrictions. * runtime/JSGlobalData.h: * runtime/WriteBarrier.h: (JSC::GCRootPtr::GCRootPtr): (JSC::GCRootPtr::operator=): 2011-02-28 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. Use more xcconfig files in JavaScriptCore gyp build https://bugs.webkit.org/show_bug.cgi?id=55391 The GYP experts tell me that we have have a total of two xcconfig files: one for the xcodeproj as a whole and one for each target. This patch uses that technique to re-use the existing xcconfig files and eliminate the duplication. Technically, this patch introduces some build errors because the xcconfig files assume that the xcodeproj file is one level higher in the directory hierarchy. Specifically, the xcodeproj file can no longer find the Info.plist or the prefix header. I plan to fix that in a subsequent patch. Also, this patch introduces the Release and Production configurations, which should work correctly now. * gyp/JavaScriptCore.gyp: 2011-02-28 Jon Honeycutt <jhoneycutt@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Add symbol to export. 2011-02-28 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Make ScopeChainNode GC allocated https://bugs.webkit.org/show_bug.cgi?id=55283 Simplify lifetime and other issues with the scopechain by making it gc allocated. This allows us to simplify function exit and unwinding, as well as making the current iterative refcounting go away. * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * bytecode/CodeBlock.cpp: (JSC::CodeBlock::createActivation): * bytecode/StructureStubInfo.cpp: * bytecompiler/BytecodeGenerator.cpp: (JSC::BytecodeGenerator::generate): (JSC::BytecodeGenerator::BytecodeGenerator): (JSC::BytecodeGenerator::emitJumpIfNotFunctionCall): (JSC::BytecodeGenerator::emitJumpIfNotFunctionApply): * bytecompiler/BytecodeGenerator.h: * debugger/Debugger.cpp: (JSC::Recompiler::operator()): * debugger/DebuggerCallFrame.h: (JSC::DebuggerCallFrame::scopeChain): * interpreter/CachedCall.h: (JSC::CachedCall::CachedCall): * interpreter/CallFrame.h: * interpreter/Interpreter.cpp: (JSC::depth): (JSC::Interpreter::unwindCallFrame): (JSC::Interpreter::throwException): (JSC::Interpreter::execute): (JSC::Interpreter::executeCall): (JSC::Interpreter::executeConstruct): (JSC::Interpreter::privateExecute): * jit/JITCall.cpp: (JSC::JIT::compileOpCallInitializeCallFrame): (JSC::JIT::compileOpCall): * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCallInitializeCallFrame): (JSC::JIT::emit_op_ret): (JSC::JIT::emit_op_ret_object_or_this): (JSC::JIT::compileOpCall): * jit/JITOpcodes.cpp: (JSC::JIT::emit_op_end): (JSC::JIT::emit_op_ret): (JSC::JIT::emit_op_ret_object_or_this): * jit/JITOpcodes32_64.cpp: (JSC::JIT::emit_op_end): * jit/JITStubs.cpp: (JSC::DEFINE_STUB_FUNCTION): * jit/JITStubs.h: * runtime/ArgList.cpp: * runtime/Completion.cpp: (JSC::evaluate): * runtime/Completion.h: * runtime/DateConversion.cpp: * runtime/Executable.cpp: (JSC::EvalExecutable::compileInternal): (JSC::ProgramExecutable::compileInternal): (JSC::FunctionExecutable::compileForCallInternal): (JSC::FunctionExecutable::compileForConstructInternal): * runtime/FunctionConstructor.cpp: (JSC::constructFunction): * runtime/GCActivityCallbackCF.cpp: * runtime/Identifier.cpp: * runtime/JSCell.h: * runtime/JSChunk.cpp: Added. * runtime/JSChunk.h: Added. * runtime/JSFunction.cpp: (JSC::JSFunction::JSFunction): (JSC::JSFunction::markChildren): (JSC::JSFunction::getCallData): (JSC::JSFunction::getOwnPropertySlot): (JSC::JSFunction::getConstructData): * runtime/JSFunction.h: (JSC::JSFunction::scope): (JSC::JSFunction::setScope): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::init): (JSC::JSGlobalObject::markChildren): * runtime/JSGlobalObject.h: (JSC::JSGlobalObject::JSGlobalObjectData::JSGlobalObjectData): (JSC::JSGlobalObject::globalScopeChain): * runtime/JSGlobalObjectFunctions.cpp: (JSC::globalFuncEval): * runtime/JSLock.cpp: * runtime/JSNumberCell.cpp: * runtime/JSZombie.cpp: * runtime/MarkedBlock.cpp: * runtime/MarkedSpace.cpp: * runtime/PropertyNameArray.cpp: * runtime/ScopeChain.cpp: (JSC::ScopeChainNode::print): (JSC::ScopeChainNode::localDepth): (JSC::ScopeChainNode::markChildren): * runtime/ScopeChain.h: (JSC::ScopeChainNode::ScopeChainNode): (JSC::ScopeChainNode::createStructure): (JSC::ScopeChainNode::push): (JSC::ScopeChainNode::pop): (JSC::ScopeChainIterator::ScopeChainIterator): (JSC::ScopeChainIterator::operator*): (JSC::ScopeChainIterator::operator->): (JSC::ScopeChainIterator::operator++): (JSC::ScopeChainNode::begin): (JSC::ScopeChainNode::end): (JSC::ExecState::globalData): (JSC::ExecState::lexicalGlobalObject): (JSC::ExecState::globalThisValue): * runtime/ScopeChainMark.h: * wtf/DateMath.cpp: 2011-02-27 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. Implement WTF::randomNumber in terms of WTF::cryptographicallyRandomNumber when possible https://bugs.webkit.org/show_bug.cgi?id=55326 Currently, randomNumber does a bunch of platform-specific work that to get a cryptographic randomness when available. Instead, we should use cryptographicallyRandomNumber, which abstracts this work. Unfortunately, we can't remove all of the WTF::randomNumber implementation because not every port has access to cryptographically random numbers. * wtf/RandomNumber.cpp: (WTF::randomNumber): 2011-02-27 Benjamin Poulain <ikipou@gmail.com> Reviewed by Darin Adler. Eliminate DeprecatedPtrList from RenderBlock https://bugs.webkit.org/show_bug.cgi?id=54972 Add methods find() and contains() using an adaptor to ListHashSet. Those method are like the one of HashSet, they allow to find objects based on a different key than the one used to define the set. Add convenience methods for direct access to the head and tail of the list. Those methods are providing similar API/behavior as Vector. * wtf/ListHashSet.h: (WTF::::first): (WTF::::last): (WTF::::removeLast): (WTF::ListHashSetTranslatorAdapter::hash): (WTF::ListHashSetTranslatorAdapter::equal): (WTF::::find): (WTF::::contains): 2011-02-26 Patrick Gansterer <paroga@webkit.org> Reviewed by Andreas Kling. Add support for DragonFly BSD https://bugs.webkit.org/show_bug.cgi?id=54407 DragonFly BSD is based on FreeBSD, so handle it like FreeBSD. * wtf/Platform.h: 2011-02-26 Adam Barth <abarth@webkit.org> Reviewed by Dimitri Glazkov. JavaScriptCore should use the xcconfig file instead of importing that information into GYP https://bugs.webkit.org/show_bug.cgi?id=55282 Technically, this breaks the build because I had removed one of the warnings in this config file, but this change seems like an improvement. * gyp/JavaScriptCore.gyp: 2011-02-26 Thouraya ANDOLSI <thouraya.andolsi@st.com> Reviewed by Nikolas Zimmermann. SH4 JIT SUPPORT https://bugs.webkit.org/show_bug.cgi?id=44329 Provide an ExecutableAllocater::cacheFlush() implementation for Linux/SH4. * jit/ExecutableAllocator.h: (JSC::ExecutableAllocator::cacheFlush): 2011-02-25 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r79627. http://trac.webkit.org/changeset/79627 https://bugs.webkit.org/show_bug.cgi?id=55274 broke worker tests (Requested by olliej on #webkit). * JavaScriptCore.exp: * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: * runtime/Heap.cpp: (JSC::Heap::allocate): * runtime/Heap.h: * runtime/JSCell.h: (JSC::JSCell::JSCell::operator new): (JSC::JSCell::MarkedSpace::sizeClassFor): (JSC::JSCell::MarkedSpace::allocate): * runtime/MarkedBlock.h: * runtime/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::allocateBlock): (JSC::MarkedSpace::reset): * runtime/MarkedSpace.h: (JSC::MarkedSpace::SizeClass::SizeClass): 2011-02-25 Michael Saboff <msaboff@apple.com> Reviewed by Darin Adler. Leak in JSParser::Scope of ScopeLabelInfo Vector https://bugs.webkit.org/show_bug.cgi?id=55249 Changed m_labels to be an OwnPtr<>. Added VectorTraits and Scope copy constructor to support this change. * parser/JSParser.cpp: (JSC::JSParser::Scope::~Scope): 2011-02-25 Fumitoshi Ukai <ukai@chromium.org> Reviewed by Adam Barth. WebSocket uses insecure random numbers https://bugs.webkit.org/show_bug.cgi?id=54714 * JavaScriptCore.exp: Export WTF::cryptographicallyRandomNumber() 2011-02-25 Patrick Gansterer <paroga@webkit.org> Reviewed by Adam Roben. Move timeBeginPeriod into OS(WINDOWS) section https://bugs.webkit.org/show_bug.cgi?id=55247 * jsc.cpp: (main): timeBeginPeriod is available on all Windows versions and not compiler specific. 2011-02-25 Patrick Gansterer <paroga@webkit.org> Unreviewed WinCE build fix for r79695. * jsc.cpp: (main): SetErrorMode isn't available on WinCE. 2011-02-25 Adam Roben <aroben@apple.com> Work around Cygwin's crash-suppression behavior Cygwin calls ::SetErrorMode(SEM_FAILCRITICALERRORS), which any processes it launches will inherit. This is bad for testing/debugging, as it causes the post-mortem debugger not to be invoked. (Cygwin does this because it makes crashes more UNIX-y.) We reset the error mode when our test apps launch to work around Cygwin's behavior. Fixes <http://webkit.org/b/55222> Test apps crash silently (without invoking post-mortem debugger) when launched from Cygwin 1.7 Reviewed by Darin Adler. * API/tests/testapi.c: Added a now-needed #include. (main): * jsc.cpp: (main): Call ::SetErrorMode(0) to undo Cygwin's folly. * JavaScriptCore.vcproj/testapi/testapiCommon.vsprops: Define NOMINMAX like many of our other projects do so that windows.h won't define min/max macros that interfere with std::numeric_limits<T>::min/max. 2011-02-24 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. Add GYP project for JavaScriptCore https://bugs.webkit.org/show_bug.cgi?id=55027 Again, this GYP files is very rough, but it succeeds in building JavaScriptCore. There's a lot more work to do here, especially in the area of sharing with JavaScriptGlue.gyp. This patch is more of a checkpoint so that other folks can help out if they wish. * gyp: Added. * gyp/JavaScriptCore.gyp: Added. * gyp/generate-derived-sources.sh: Added. 2011-02-24 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. Add missing files to JavaScriptCore.gypi https://bugs.webkit.org/show_bug.cgi?id=55193 I forgot to add mm files in my previous patch. * JavaScriptCore.gyp/JavaScriptCore.gyp: * JavaScriptCore.gypi: 2011-02-24 Adam Barth <abarth@webkit.org> Reviewed by Eric Seidel. Remove unused parameter name in GCActivityCallback.cpp https://bugs.webkit.org/show_bug.cgi?id=55194 This change is not strictly required for the GYP-based build system, but I noticed this error when working on the new build system. * runtime/GCActivityCallback.cpp: (JSC::DefaultGCActivityCallback::DefaultGCActivityCallback): 2011-02-24 James Robinson <jamesr@chromium.org> Reviewed by Darin Fisher. Add a USE() macro to control use of the built-in UTF8 codec https://bugs.webkit.org/show_bug.cgi?id=55189 Defaults USE(BUILTIN_UTF8_CODEC) to true for all platforms except chromium, which controls the flag via features.gypi. * wtf/Platform.h: 2011-02-24 Geoffrey Garen <ggaren@apple.com> Reviewed by Darin Adler. Variable-sized allocation (but still capped at 64 bytes) https://bugs.webkit.org/show_bug.cgi?id=55159 SunSpider reports no change. * JavaScriptCore.exp: Some day, I hope not to have to edit this file. * runtime/Heap.cpp: (JSC::Heap::allocateSlowCase): Split allocation into a fast and slow case, so the fast case can inline size class selection and turn it into a compile-time constant. Changed the collect-on-every allocation debugging switch to collect only on every slow allocation, so you can still flip the switch without recompiling the world. This may also be preferable for debugging purposes, since collecting after every single allocation can be unusably slow, and can mask problems by running destructors early. * runtime/Heap.h: Ditto. * runtime/JSCell.h: (JSC::JSCell::MarkedSpace::sizeClassFor): (JSC::JSCell::Heap::allocate): (JSC::JSCell::JSCell::operator new): The inlining mentioned above. * runtime/MarkedBlock.h: Dropped the block size from 256KB to 16KB. With multiple size classes, allocating a full 256KB for the first allocation in a given class can be pathologically wasteful. (8KB, or 4KB Mac and 8KB Windows, would be even better, but that seems to be a peformance regression for now.) * runtime/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::allocateBlock): (JSC::MarkedSpace::reset): There's more than one size class now, and its cell size is not constant. * runtime/MarkedSpace.h: (JSC::MarkedSpace::SizeClass::SizeClass): Ditto. 2011-02-23 Oliver Hunt <oliver@apple.com> Reviewed by Geoffrey Garen. Make WeakGCMap use new handle infrastructure https://bugs.webkit.org/show_bug.cgi?id=55100 Remove old WeakGCMap implementation and move over to new handle based logic. This has a number of benefits, most notably it makes a WeakGCMap always reflect the true state of the world by as all entries are removed at the first gc cycle that makes them dead. This allows us to get rid of code in a wide variety of objects where the only purpose was to remove themselves from maps. It also means that we no longer need to have special "unchecked" versions of any functions on WeakGCMap. Alas in order to maintain compatibility with the JSWeakObjectMapClear API it is still necessary to have an api that resembles uncheckedRemove, this is now deprecatedRemove and will be dealt with in a later patch. In order to get correct semantics in WeakGCMap we need more contextual information in the finalizer, so we've added an abstract class based finaliser and a context parameter to the calls. The new an improved WeakGCMap also results in sigificantly more churn in the weak handle lists so exposed some potential problems during the post mark phase which have been rectified as well. * API/JSWeakObjectMapRefPrivate.cpp: * API/JSWeakObjectMapRefPrivate.h: * runtime/Heap.cpp: (JSC::Heap::globalObjectCount): (JSC::Heap::protectedGlobalObjectCount): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: * runtime/JSGlobalObject.cpp: (JSC::JSGlobalObject::~JSGlobalObject): (JSC::JSGlobalObject::init): * runtime/WeakGCMap.h: (JSC::WeakGCMap::iterator::iterator): (JSC::WeakGCMap::iterator::get): (JSC::WeakGCMap::iterator::getSlot): (JSC::WeakGCMap::iterator::operator++): (JSC::WeakGCMap::iterator::operator==): (JSC::WeakGCMap::iterator::operator!=): (JSC::WeakGCMap::WeakGCMap): (JSC::WeakGCMap::isEmpty): (JSC::WeakGCMap::clear): (JSC::WeakGCMap::get): (JSC::WeakGCMap::getSlot): (JSC::WeakGCMap::set): (JSC::WeakGCMap::take): (JSC::WeakGCMap::size): (JSC::WeakGCMap::deprecatedRemove): (JSC::WeakGCMap::begin): (JSC::WeakGCMap::end): (JSC::WeakGCMap::~WeakGCMap): (JSC::WeakGCMap::finalize): * runtime/WeakGCPtr.h: (JSC::WeakGCPtr::WeakGCPtr): (JSC::WeakGCPtr::set): 2011-02-24 Oliver Hunt <oliver@apple.com> Reviewed by Gavin Barraclough. Make weaklist processing deal with weak handles being removed during the iteration https://bugs.webkit.org/show_bug.cgi?id=55105 It is possible for the handle heap to end up in a broken state if a handle's finalizer removes either the current or next handle to be visited during the post-gc cleanup. This patch removes that problem by allowing the deallocate(Node*) routine to update the iterator if it is called during finalization. * collector/handles/HandleHeap.cpp: (JSC::HandleHeap::HandleHeap): (JSC::HandleHeap::updateAfterMark): (JSC::HandleHeap::clearWeakPointers): (JSC::HandleHeap::writeBarrier): (JSC::HandleHeap::protectedGlobalObjectCount): * collector/handles/HandleHeap.h: (JSC::Finalizer::~Finalizer): (JSC::HandleHeap::getFinalizer): (JSC::HandleHeap::deallocate): (JSC::HandleHeap::makeWeak): (JSC::HandleHeap::makeSelfDestroying): (JSC::HandleHeap::Node::Node): (JSC::HandleHeap::Node::setFinalizer): (JSC::HandleHeap::Node::finalizer): (JSC::HandleHeap::Node::finalizerContext): * interpreter/RegisterFile.cpp: (JSC::RegisterFile::setGlobalObject): (JSC::GlobalObjectNotifier::finalize): (JSC::RegisterFile::globalObjectCollectedNotifier): * interpreter/RegisterFile.h: (JSC::RegisterFile::RegisterFile): * runtime/Heap.cpp: (JSC::Heap::destroy): * runtime/WeakGCPtr.h: (JSC::WeakGCPtr::WeakGCPtr): (JSC::WeakGCPtr::set): 2011-02-24 Michael Saboff <msaboff@apple.com> Reviewed by Oliver Hunt. PatternAlternative leaked in YarrPatternConstructor::atomParenthesesEnd() https://bugs.webkit.org/show_bug.cgi?id=55156 Added code to delete unneeded PatternAlternative after it is removed from m_alternatives Vector. * yarr/YarrPattern.cpp: (JSC::Yarr::YarrPatternConstructor::atomParenthesesEnd): 2011-02-24 Eric Seidel <eric@webkit.org> Reviewed by Anders Carlsson. VectorBuffer should not call malloc(0) https://bugs.webkit.org/show_bug.cgi?id=55091 Turns out the malloc() call which was so hot in: https://bugs.webkit.org/show_bug.cgi?id=55005 was actually just malloc(0). We shouldn't be calling malloc(0) anyway, since there is no need to and it might actually do work on some systems. I believe on Mac it ends up taking the standard spinlocks (expensive) and the code on Brew actually does a malloc(1) instead. Neither is desirable. * wtf/Vector.h: (WTF::VectorBufferBase::allocateBuffer): (WTF::VectorBufferBase::tryAllocateBuffer): 2011-02-24 Patrick Gansterer <paroga@webkit.org> Reviewed by Darin Adler. Remove obsolete PLATFORM(CI) https://bugs.webkit.org/show_bug.cgi?id=55082 * wtf/Platform.h: 2011-02-24 Martin Robinson <mrobinson@igalia.com> Reviewed by Xan Lopez. [GTK] Remove the GFile GOwnPtr specialization https://bugs.webkit.org/show_bug.cgi?id=55154 Remove the GFile specialization of GOwnPtr. It's sufficient to use GRefPtr to track GFiles since they are just regular reference-counted GObjects. * wtf/gobject/GOwnPtr.cpp: Remove GFile specialization. * wtf/gobject/GOwnPtr.h: Ditto. 2011-02-24 Patrick Gansterer <paroga@webkit.org> Reviewed by Eric Seidel. Rename PLATFORM(SKIA) to USE(SKIA) https://bugs.webkit.org/show_bug.cgi?id=55090 * wtf/Platform.h: 2011-02-24 Patrick Gansterer <paroga@webkit.org> Reviewed by Alexey Proskuryakov. Remove pthreads dependecy for JSLock https://bugs.webkit.org/show_bug.cgi?id=54832 JSLock is only needed to support an obsolete execution model where JavaScriptCore automatically protected against concurrent access from multiple threads. So it's safe to disable it on non-mac platforms where we don't have native pthreads. * runtime/JSLock.cpp: 2011-02-24 Chao-ying Fu <fu@mips.com> Reviewed by Eric Seidel. Fix MIPS build with new patchOffsetPut/GetByIdPropertyMapOffset1/2 values https://bugs.webkit.org/show_bug.cgi?id=54997 * jit/JIT.h: * jit/JITStubs.cpp: (JSC::JITThunks::JITThunks): 2011-02-24 Andras Becsi <abecsi@webkit.org> Reviewed by Laszlo Gombos. [Qt] MinGW build fails to link https://bugs.webkit.org/show_bug.cgi?id=55050 Prepend the libraries of subcomponents instead of appending them to fix the library order according to the dependency of the libraries * JavaScriptCore.pri: rename addJavaScriptCore to prependJavaScriptCore * jsc.pro: ditto 2011-02-24 Eric Seidel <eric@webkit.org> Reviewed by Adam Barth. Deque<T> should support inline capacity https://bugs.webkit.org/show_bug.cgi?id=55032 The title says it all. There are currently no places which use this code yet, however it's been tested in conjunction with code for bug 55005. This also adds an ASSERT that capacity is never 1. If you were able to set the capacity equal to 1, the Deque would just get confused and happily append your item but still think it had size 0. * wtf/Deque.h: (WTF::DequeIterator::DequeIterator): (WTF::DequeConstIterator::DequeConstIterator): (WTF::DequeReverseIterator::DequeReverseIterator): (WTF::DequeConstReverseIterator::DequeConstReverseIterator): (WTF::::checkValidity): (WTF::::checkIndexValidity): (WTF::::invalidateIterators): (WTF::::Deque): (WTF::deleteAllValues): (WTF::::operator): (WTF::::destroyAll): (WTF::::~Deque): (WTF::::swap): (WTF::::clear): (WTF::::findIf): (WTF::::expandCapacityIfNeeded): (WTF::::expandCapacity): (WTF::::takeFirst): (WTF::::append): (WTF::::prepend): (WTF::::removeFirst): (WTF::::remove): (WTF::::addToIteratorsList): (WTF::::removeFromIteratorsList): (WTF::::DequeIteratorBase): (WTF::::~DequeIteratorBase): (WTF::::isEqual): (WTF::::increment): (WTF::::decrement): (WTF::::after): (WTF::::before): * wtf/Vector.h: 2011-02-22 Adam Barth <abarth@webkit.org> Reviewed by Ojan Vafai. Add missing files to JavaScriptCore.gypi https://bugs.webkit.org/show_bug.cgi?id=55020 gypi files are supposed to list every file under the sun. This patch adds some missing files and sorts the rest. * JavaScriptCore.gypi: 2011-02-23 Geoffrey Garen <ggaren@apple.com> Reviewed by Darin Adler. Refactored MarkedSpace to operate in terms of size classes https://bugs.webkit.org/show_bug.cgi?id=55106 SunSpider reports no change. * runtime/JSCell.h: (JSC::JSCell::MarkedSpace::sizeClassFor): (JSC::JSCell::MarkedSpace::allocate): Delegate allocation based on size class. Since these functions are inline, the compiler can constant fold them. * runtime/MarkedBlock.h: (JSC::MarkedBlock::cellSize): (JSC::MarkedBlock::size): Factored out a cellSize() helper. * runtime/MarkedSpace.cpp: (JSC::MarkedSpace::allocateBlock): (JSC::MarkedSpace::allocateFromSizeClass): (JSC::MarkedSpace::shrink): (JSC::MarkedSpace::reset): * runtime/MarkedSpace.h: (JSC::MarkedSpace::SizeClass::SizeClass): (JSC::MarkedSpace::SizeClass::reset): Changed to operate in terms of abstract SizeClass objects, which are independent linked lists of blocks of a certain size class, instead of a single m_heap object. 2011-02-23 Adam Barth <abarth@webkit.org> Reviewed by James Robinson. [Chromium] Use WebKitClient for OSRandomSource instead of trying to talk to the file system in the sandbox https://bugs.webkit.org/show_bug.cgi?id=55093 Exclude OSRandomSource.cpp from the Chromium build. This function is implemented in WebKit/chromium/src instead. * JavaScriptCore.gyp/JavaScriptCore.gyp: 2011-02-23 Oliver Hunt <oliver@apple.com> Roll out r64156 as it introduces incorrect behaviour. * runtime/JSByteArray.h: (JSC::JSByteArray::setIndex): 2011-02-23 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Moved the "nextAtom" allocation pointer into MarkedBlock for better encapsulation https://bugs.webkit.org/show_bug.cgi?id=55079 SunSpider reports no change. * runtime/Heap.cpp: (JSC::Heap::reset): Moved Zombie sweeping here, up from MarkedSpace, since we want Heap to logically control MarkedSpace. MarkedSpace should never choose to sweep itself. * runtime/JSCell.h: (JSC::JSCell::MarkedBlock::allocate): Updated for nextAtom becoming a member of MarkedBlock. No need to reset nextAtom to firstAtom() when we reach the end of a block, since there's now an explicit reset pass during GC. * runtime/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): * runtime/MarkedBlock.h: (JSC::MarkedBlock::reset): Added the nextAtom data member, and reordered some data members to improve cache locality. * runtime/MarkedSpace.cpp: (JSC::MarkedSpace::MarkedSpace): (JSC::MarkedSpace::allocate): (JSC::MarkedSpace::reset): * runtime/MarkedSpace.h: (JSC::CollectorHeap::CollectorHeap): Removed nextAtom, and added an explicit reset pass. 2011-02-23 James Robinson <jamesr@chromium.org> Unreviewed, rolling out r79428. http://trac.webkit.org/changeset/79428 https://bugs.webkit.org/show_bug.cgi?id=54714 Does not work in the Chromium sandbox * JavaScriptCore.exp: 2011-02-23 Adam Roben <aroben@apple.com> Fix an off-by-one error in JSC::appendSourceToError Looks like this bug has been around since the code was first added in r35245. Fixes <http://webkit.org/b/55052> <rdar://problem/9043512> Crash in JSC::appendSourceToError when running fast/dom/objc-big-method-name.html on Windows with full page heap enabled Reviewed by Darin Adler. * interpreter/Interpreter.cpp: (JSC::appendSourceToError): When trimming whitespace off the end of the string, examine the character at stop-1 rather than at stop. At this point in the code, stop represents the index just past the end of the characters we care about, and can even be just past the end of the entire data buffer. 2011-02-23 Geoffrey Garen <ggaren@apple.com> Reviewed by Darin Adler. Rolled back in r79367 with SnowLeopard Release bot crash fixed. https://bugs.webkit.org/show_bug.cgi?id=54999 The crash was caused by failure to update the "nextBlock" pointer when removing a block from the list while shrinking. The fix is to update the "nextBlock" pointer. This crash was very rare because it only happened in cases where the very first block in the heap contained no marked cells. 2011-02-23 Dan Bernstein <mitz@apple.com> Reviewed by Gavin Barraclough. Include frame numbers in backtraces. https://bugs.webkit.org/show_bug.cgi?id=55060 * wtf/Assertions.cpp: 2011-02-23 Xan Lopez <xlopez@igalia.com> Reviewed by Gavin Barraclough. latest jsc for armv7 crashes in sunspider tests https://bugs.webkit.org/show_bug.cgi?id=54667 Update JIT offset values in ARMv7 after r78732. Fixes crashes in SunSpider and JavaScript tests. * jit/JIT.h: update values. 2011-02-23 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r79418. http://trac.webkit.org/changeset/79418 https://bugs.webkit.org/show_bug.cgi?id=55043 "breaks shlib linux build" (Requested by morrita on #webkit). * JavaScriptCore.gyp/JavaScriptCore.gyp: * JavaScriptCore.gypi: 2011-02-23 Patrick Gansterer <paroga@webkit.org> Reviewed by Alexey Proskuryakov. Use DEFINE_STATIC_LOCAL for ignoreSetMutex in Structure.cpp https://bugs.webkit.org/show_bug.cgi?id=54831 * runtime/InitializeThreading.cpp: (JSC::initializeThreadingOnce): * runtime/Structure.cpp: (JSC::ignoreSetMutex): (JSC::Structure::Structure): (JSC::Structure::~Structure): (JSC::Structure::initializeThreading): * runtime/Structure.h: 2011-02-23 Patrick Gansterer <paroga@webkit.org> Reviewed by Darin Adler. Rename PLATFORM(CF) to USE(CF) https://bugs.webkit.org/show_bug.cgi?id=53540 * runtime/DatePrototype.cpp: * runtime/GCActivityCallbackCF.cpp: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * wtf/CurrentTime.cpp: * wtf/Platform.h: * wtf/text/AtomicString.h: * wtf/text/StringImpl.h: (WTF::StringImpl::computeHash): * wtf/text/WTFString.h: * wtf/unicode/icu/CollatorICU.cpp: (WTF::Collator::userDefault): 2011-02-23 Fumitoshi Ukai <ukai@chromium.org> Unreviewed build fix for Windows. WebSocket uses insecure random numbers https://bugs.webkit.org/show_bug.cgi?id=54714 * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Export WTF::cryptographicallyRandomNumber() 2011-02-23 Fumitoshi Ukai <ukai@chromium.org> Reviewed by Adam Barth. WebSocket uses insecure random numbers https://bugs.webkit.org/show_bug.cgi?id=54714 * JavaScriptCore.exp: Export WTF::cryptographicallyRandomNumber() 2011-02-22 Adam Barth <abarth@webkit.org> Reviewed by Ojan Vafai. Add missing files to JavaScriptCore.gypi https://bugs.webkit.org/show_bug.cgi?id=55020 gypi files are supposed to list every file under the sun. This patch adds some missing files and sorts the rest. * JavaScriptCore.gypi: 2011-02-22 Sheriff Bot <webkit.review.bot@gmail.com> Unreviewed, rolling out r79367. http://trac.webkit.org/changeset/79367 https://bugs.webkit.org/show_bug.cgi?id=55012 all layout tests are crashing on Snow Leopard (Requested by rniwa on #webkit). * GNUmakefile.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: * runtime/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): * runtime/MarkedBlock.h: * runtime/MarkedSpace.cpp: (JSC::MarkedSpace::destroy): (JSC::MarkedSpace::allocateBlock): (JSC::MarkedSpace::freeBlock): (JSC::MarkedSpace::allocate): (JSC::MarkedSpace::shrink): (JSC::MarkedSpace::reset): * runtime/MarkedSpace.h: (JSC::CollectorHeap::collectorBlock): * wtf/CMakeLists.txt: * wtf/DoublyLinkedList.h: Removed. 2011-02-22 Geoffrey Garen <ggaren@apple.com> Reviewed by Oliver Hunt. Manage MarkedBlocks in a linked list instead of a vector, so arbitrary removal is O(1) https://bugs.webkit.org/show_bug.cgi?id=54999 SunSpider reports no change. * GNUmakefile.am: * JavaScriptCore.gypi: * JavaScriptCore.vcproj/WTF/WTF.vcproj: * JavaScriptCore.xcodeproj/project.pbxproj: So many build systems, so little time. * wtf/CMakeLists.txt: * runtime/MarkedBlock.cpp: (JSC::MarkedBlock::MarkedBlock): * runtime/MarkedBlock.h: (JSC::MarkedBlock::setPrev): (JSC::MarkedBlock::setNext): (JSC::MarkedBlock::prev): (JSC::MarkedBlock::next): Added linked list data members and accessors. * runtime/MarkedSpace.cpp: (JSC::MarkedSpace::destroy): (JSC::MarkedSpace::allocateBlock): Stop using vector, since it doesn't exist anymore. (JSC::MarkedSpace::freeBlocks): New helper function for updating relevant data structures when freeing blocks. (JSC::MarkedSpace::allocate): Updated for nextBlock being a pointer and not a vector index. (JSC::MarkedSpace::shrink): Construct a temporary list of empties and then free them, to avoid modifying our hash table while iterating it. This wasn't a concern before because we were using indirect array indexing, not direct pointer indexing. (JSC::MarkedSpace::reset): Updated for nextBlock being a pointer and not a vector index. * runtime/MarkedSpace.h: (JSC::CollectorHeap::CollectorHeap): Changed data type from vector to linked list. * wtf/DoublyLinkedList.h: Added. New linked list class. (WTF::::DoublyLinkedList): (WTF::::isEmpty): (WTF::::head): (WTF::::append): (WTF::::remove): 2011-02-22 Gavin Barraclough <barraclough@apple.com> Windows build fix. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-02-22 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 54988 - Re-create StructureTransitionTable class, encapsulate transition table The Structure class keeps a table of transitions to derived Structure types. Since this table commonly contains a single entry we employ an optimization where instead of holding a map, we may hold a pointer directly to a single instance of the mapped type. We use an additional bit of data to flag whether the pointer is currently pointing to a table of transitions, or a singleton transition. Previously we had commonly used a pattern of storing data in the low bits of pointers, but had moved away from this since it causes false leaks to be reported by the leaks tool. However in this case, the entries in the map are weak links - this pointer will never be responsible for keeping an object alive. As such we can use this approach provided that the bit is set when a table is not in use (otherwise the table would appear to be leaked). Additionally, the transition table currently allows two entries to exist for a given key - one specialized to a particular value, and one not specialized. This is unnecessary, wasteful, and a little inconsistent. (If you create an entry for a specialized value, then a non-specialized entry, both will exist. If you create an entry for a non-specialized value, then try to create a specialized entry, only a non-specialized form will be allowed.) This shows a small progression on v8. * JavaScriptCore.exp: * runtime/JSObject.h: (JSC::JSObject::putDirectInternal): * runtime/Structure.cpp: (JSC::StructureTransitionTable::contains): (JSC::StructureTransitionTable::get): (JSC::StructureTransitionTable::remove): (JSC::StructureTransitionTable::add): (JSC::Structure::dumpStatistics): (JSC::Structure::Structure): (JSC::Structure::~Structure): (JSC::Structure::addPropertyTransitionToExistingStructure): (JSC::Structure::addPropertyTransition): * runtime/Structure.h: (JSC::Structure::get): * runtime/StructureTransitionTable.h: (JSC::StructureTransitionTable::Hash::hash): (JSC::StructureTransitionTable::Hash::equal): (JSC::StructureTransitionTable::HashTraits::emptyValue): (JSC::StructureTransitionTable::HashTraits::constructDeletedValue): (JSC::StructureTransitionTable::HashTraits::isDeletedValue): (JSC::StructureTransitionTable::StructureTransitionTable): (JSC::StructureTransitionTable::~StructureTransitionTable): (JSC::StructureTransitionTable::isUsingSingleSlot): (JSC::StructureTransitionTable::map): (JSC::StructureTransitionTable::setMap): (JSC::StructureTransitionTable::singleTransition): (JSC::StructureTransitionTable::setSingleTransition): 2011-02-22 Andras Becsi <abecsi@webkit.org> Reviewed by Laszlo Gombos. [Qt] Redesign the build system https://bugs.webkit.org/show_bug.cgi?id=51339 Part 2. Build WebCore as a static library, compile the WebKit API and WebKit2 API in a final step and link to WebKit2, WebCore and JSC libraries to fix linking issues resulting from stripped away symbols. * JavaScriptCore.pri: Remove the workaround. 2011-02-21 Adam Roben <aroben@apple.com> Fix linker warning on Windows r79135 tried to export JSObject::s_info by adding it to JavaScriptCore.def. This is the obvious way (since it's how we export functions), but unfortunately it doesn't work correct. r79222 made us export it the right way (using the JS_EXPORTDATA macro), but forgot to remove it from JavaScriptCore.def. This caused us to get linker warnings about exporting the symbol multiple times. Rubber-stamped by Anders Carlsson. * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: Removed JSObject::s_info. 2011-02-21 Brian Weinstein <bweinstein@apple.com> Reviewed by Adam Roben. WebResourceCacheManager should be responsible for managing the CFURLCache as well as the WebCore memory cache. https://bugs.webkit.org/show_bug.cgi?id=54886 Part of <rdar://problem/8971738> Add a new use flag for using the CFURLCache. * wtf/Platform.h: 2011-02-21 Xan Lopez <xlopez@igalia.com> Reviewed by Gavin Barraclough. Use ASSERT_JIT_OFFSET in JITPropertyAccess32_64.cpp https://bugs.webkit.org/show_bug.cgi?id=54901 * jit/JIT.h: swap actual and expected values in message, they were reversed. * jit/JITCall32_64.cpp: (JSC::JIT::compileOpCall): use ASSERT_JIT_OFFSET instead of a simple ASSERT. * jit/JITPropertyAccess32_64.cpp: (JSC::JIT::emit_op_method_check): ditto. (JSC::JIT::compileGetByIdHotPath): ditto. (JSC::JIT::compileGetByIdSlowCase): ditto. (JSC::JIT::emit_op_put_by_id): ditto. 2011-02-21 Gavin Barraclough <barraclough@apple.com> Ruber stamped by Sam Weinig Bug 54899 - Math.LOG10E should be 0.4342944819032518 This value is quoted in section 15.8.1.5 of the spec. * runtime/MathObject.cpp: (JSC::MathObject::MathObject): 2011-02-21 Gavin Barraclough <barraclough@apple.com> Reviewed by Sam Weinig. Bug 54894 - Make inheritance structure described by ClassInfo match C++ class hierarchy. The ClassInfo objects describe an inheritance hierarchy, with each ClassInfo instance containing a pointer to its parent class. These links should reflect the inheritance hierarchy of C++ classes below JSObject. For the large part it does, but in some cases entries in the C++ hierarchy are skipped over. This presently likely doesn't matter, since intervening C++ classes may not have ClassInfo - but would be a potential bug were ClassInfo were to be added. * API/JSCallbackConstructor.cpp: * API/JSCallbackFunction.cpp: * API/JSCallbackObjectFunctions.h: * runtime/Arguments.h: * runtime/ArrayPrototype.cpp: * runtime/BooleanObject.cpp: * runtime/DateInstance.cpp: * runtime/DatePrototype.cpp: * runtime/ErrorInstance.cpp: * runtime/InternalFunction.cpp: * runtime/JSActivation.cpp: * runtime/JSArray.cpp: * runtime/JSFunction.cpp: * runtime/JSONObject.cpp: * runtime/JSObject.h: * runtime/JSZombie.h: * runtime/MathObject.cpp: * runtime/NativeErrorConstructor.cpp: * runtime/NumberConstructor.cpp: * runtime/NumberObject.cpp: * runtime/RegExpConstructor.cpp: * runtime/RegExpObject.cpp: * runtime/StringObject.cpp: * runtime/StringPrototype.cpp: 2011-02-21 Adam Roben <aroben@apple.com> Export JSObject::s_info from JavaScriptCore.dll This matches what we do for all other ClassInfo objects that WebCore needs access to. Fixes <http://webkit.org/b/54881> REGRESSION (r79132): Lots of tests crashing in JSCell::inherits on Windows Reviewed by Sam Weinig. * runtime/JSObject.h: Added JS_EXPORTDATA to s_info. 2011-02-21 Kristian Amlie <kristian.amlie@nokia.com> Reviewed by Andreas Kling. Switched to compiler based detection, where the error actually is. It is not the platform that needs the workaround, it is the compiler. 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: 2011-02-20 Gavin Barraclough <barraclough@apple.com> Reviewed by Oliver Hunt. https://bugs.webkit.org/show_bug.cgi?id=54839 Remove PrototypeFunction, NativeFunctionWrapper, and GlobalEvalFunction. Historically, Native functions used to be represented by PrototypeFunctions, however since introducing call optimizations to the JIT this has used JSFunctions for host calls too. At the point this change was made, the interpreter continued to use PrototypeFunctions, however since fallback from the JIT to interpreter was introduced the interpreter has had to be able to run using host functions represented using JSFunctions - leading to an unnecessary and redundant divergence in behaviour between interpreter only builds, and situations where the JIT has fallen back to interpreting. NativeFunctionWrapper only existed to select between PrototypeFunction and JSFunction for wrappers for host functions, and as such can also be removed. GlobalEvalFunction is a redundant wrapper that happens to be derived from PrototypeFunction. It existed to hold a reference to the global object - but since all functions how derive from JSObjectWithGlobalObject, this no longer requires an additional class to provide this functionality. * JavaScriptCore.JSVALUE32_64only.exp: * JavaScriptCore.JSVALUE64only.exp: * JavaScriptCore.xcodeproj/project.pbxproj: Removed symbols / references to files. * runtime/GlobalEvalFunction.cpp: Removed. * runtime/GlobalEvalFunction.h: Removed. * runtime/NativeFunctionWrapper.h: Removed. * runtime/PrototypeFunction.cpp: Removed. * runtime/PrototypeFunction.h: Removed. Removed. * runtime/Executable.cpp: (JSC::NativeExecutable::~NativeExecutable): * runtime/Executable.h: (JSC::NativeExecutable::create): (JSC::NativeExecutable::NativeExecutable): (JSC::JSFunction::nativeFunction): * runtime/JSFunction.cpp: (JSC::callHostFunctionAsConstructor): (JSC::JSFunction::JSFunction): (JSC::JSFunction::getCallData): * runtime/JSFunction.h: * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::getHostFunction): * runtime/JSGlobalData.h: (JSC::JSGlobalData::getCTIStub): Added interpreter-friendly constructors for NativeExecutables. * bytecompiler/BytecodeGenerator.cpp: * interpreter/Interpreter.cpp: * jit/JITStubs.cpp: * jsc.cpp: * runtime/ArrayConstructor.cpp: * runtime/BooleanPrototype.cpp: * runtime/BooleanPrototype.h: * runtime/CallData.h: * runtime/DateConstructor.cpp: * runtime/DateConstructor.h: * runtime/ErrorPrototype.cpp: * runtime/ErrorPrototype.h: * runtime/FunctionPrototype.cpp: * runtime/FunctionPrototype.h: * runtime/JSGlobalObject.cpp: * runtime/JSGlobalObject.h: * runtime/JSGlobalObjectFunctions.cpp: * runtime/Lookup.cpp: * runtime/NumberPrototype.cpp: * runtime/NumberPrototype.h: * runtime/ObjectConstructor.cpp: * runtime/ObjectConstructor.h: * runtime/ObjectPrototype.cpp: * runtime/ObjectPrototype.h: * runtime/RegExpPrototype.cpp: * runtime/RegExpPrototype.h: * runtime/SmallStrings.h: * runtime/StringConstructor.cpp: * runtime/StringConstructor.h: Removed use of redundant classes. 2011-02-19 Laszlo Gombos <laszlo.1.gombos@nokia.com> Unreviewed build fix for Symbian. [Symbian] Revert the removal of linking against hal after r79126. Dependency on the hal library can not be removed as it is still used (e.g. in MarkStackSymbian.cpp). * JavaScriptCore.pri: 2011-02-19 Gavin Barraclough <barraclough@apple.com> Interpreter build fix. * runtime/ArrayConstructor.cpp: * runtime/BooleanPrototype.cpp: * runtime/DateConstructor.cpp: * runtime/ErrorPrototype.cpp: * runtime/FunctionPrototype.cpp: * runtime/Lookup.cpp: * runtime/NumberPrototype.cpp: * runtime/ObjectConstructor.cpp: * runtime/ObjectPrototype.cpp: * runtime/RegExpPrototype.cpp: * runtime/StringConstructor.cpp: 2011-02-19 Gavin Barraclough <barraclough@apple.com> Build fix!! * JavaScriptCore.exp: 2011-02-19 Gavin Barraclough <barraclough@apple.com> Windows build fix!! * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-02-19 Gavin Barraclough <barraclough@apple.com> Windows build fix! * JavaScriptCore.vcproj/JavaScriptCore/JavaScriptCore.def: 2011-02-19 Gavin Barraclough <barraclough@apple.com> Build fix! * JavaScriptCore.exp: 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 ==