Consolidate the way WTF_USE_PTHREADS is enabled

[WebKit-https.git] / Source / WTF / ChangeLog

2013-02-10  Laszlo Gombos  <l.gombos@samsung.com>

        Consolidate the way WTF_USE_PTHREADS is enabled
        https://bugs.webkit.org/show_bug.cgi?id=108191

        Reviewed by Benjamin Poulain.

        Define WTF_USE_PTHREADS to 1 on all OS(UNIX) environments.

        * WTF.gyp/WTF.gyp: Remove duplicated definition of WTF_USE_PTHREADS.
        * wtf/Platform.h:

2013-02-10  Laszlo Gombos  <l.gombos@samsung.com>

        Refactor the way HAVE_XXX macros are set
        https://bugs.webkit.org/show_bug.cgi?id=108132

        Reviewed by Benjamin Poulain.

        OS(WINDOWS) and OS(UNIX) are so broadly defined that for each
        builds exactly one of them is enabled. Use this assumption to
        cleanup Platform.h.

        * wtf/Platform.h:

2013-02-10  Jae Hyun Park  <jae.park08@gmail.com>

        Rename ENABLE(GLIB_SUPPORT) to USE(GLIB)
        https://bugs.webkit.org/show_bug.cgi?id=104266

        Reviewed by Philippe Normand.

        Using USE(GLIB) instead of ENABLE(GLIB_SUPPORT) is more consistent with
        the existing macro naming conventions.

        From Platform.h
        USE() - use a particular third-party library or optional OS service
        ENABLE() - turn on a specific feature of WebKit

        * WTF.pri:
        * wtf/Platform.h:
        * wtf/gobject/GOwnPtr.cpp:
        * wtf/gobject/GOwnPtr.h:
        * wtf/gobject/GRefPtr.cpp:
        * wtf/gobject/GRefPtr.h:

2013-02-08  Michael Saboff  <msaboff@apple.com>

        ARM_NEON Inline Assembly for copyLCharsFromUCharSource() inefficient for aligned destinations
        https://bugs.webkit.org/show_bug.cgi?id=109335

        Reviewed by Filip Pizlo.

        Change a "do while" to a "while" so that we don't copy single characters to align the 
        destination when it is already aligned.

        * wtf/text/ASCIIFastPath.h:
        (WTF::copyLCharsFromUCharSource):

2013-02-08  Jer Noble  <jer.noble@apple.com>

        Bring WebKit up to speed with latest Encrypted Media spec.
        https://bugs.webkit.org/show_bug.cgi?id=97037

        Reviewed by Eric Carlson.

        Define the ENABLE_ENCRYPTED_MEDIA_V2 setting.

        * wtf/Platform.h:

2013-02-08  Adam Barth  <abarth@webkit.org>

        Use WeakPtrs to communicate between the HTMLDocumentParser and the BackgroundHTMLParser
        https://bugs.webkit.org/show_bug.cgi?id=107190

        Reviewed by Eric Seidel.

        Add the ability to create an unbound weak reference. This facility lets
        you start sending messages to a WeakPtr on another thread before the
        object backing the WeakPtr has actually been created.

        * wtf/WeakPtr.h:
        (WTF::WeakReference::createUnbound):
        (WTF::WeakReference::bindTo):
        (WeakReference):
        (WTF::WeakReference::WeakReference):
        (WTF::WeakPtr::WeakPtr):
        (WeakPtr):
        (WTF::WeakPtrFactory::WeakPtrFactory):
        (WeakPtrFactory):
        (WTF::WeakPtrFactory::revokeAll):

2013-02-08  Martin Robinson  <mrobinson@igalia.com>

        [GTK] Add an experimental gyp build
        https://bugs.webkit.org/show_bug.cgi?id=109003

        Reviewed by Gustavo Noronha Silva.

        * WTF.gyp/WTF.gyp: Filter out MetaAllocator.(cpp/h) from the Chromium
        build. It's only necessary for GTK+.
        * WTF.gypi: Add MetaAllocator to the build for WebKitGTK+.

2013-02-06  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: Native Memory Instrumentation: reportBaseAddress needs to be called after the reportNode. So it may reuse the node index for the real address.
        https://bugs.webkit.org/show_bug.cgi?id=109051

        Reviewed by Yury Semikhatsky.

        * wtf/MemoryInstrumentation.cpp:
        (WTF::MemoryInstrumentation::WrapperBase::processPointer):

2013-02-07  David Kilzer  <ddkilzer@apple.com>

        Fix #endif comment from r142163 and r142183

        * wtf/MainThread.h:
        (WTF): s/PLATFORM/USE/

2013-02-07  Benjamin Poulain  <bpoulain@apple.com>

        Fix an incorrect comment from r142163

        Unreviewed.

        * wtf/MainThread.h:
        I accidentally copied PLATFORM(IOS) instead of USE(WEB_THREAD) for the #ifdef closing.

2013-02-07  Benjamin Poulain  <bpoulain@apple.com>

        Upstream iOS isWebThread() and isUIThread()
        https://bugs.webkit.org/show_bug.cgi?id=109130

        Reviewed by Sam Weinig.

        On iOS, it is sometimes necessary to differenciate the thread running WebCore,
        and the thread running the UI. This patch upstream those functions.

        * wtf/MainThread.cpp:
        * wtf/MainThread.h:
        Disable the legacy initializer as it is incorrect when using the WebThread to run WebCore.
        (WTF::isWebThread):
        (WTF::isUIThread):
        Return true when the current thread is the Web/UI thread.

        * wtf/mac/MainThreadMac.mm:
        (WTF::isUIThread):
        (WTF::isWebThread):

        * wtf/text/AtomicString.cpp:
        (WTF::AtomicStringTable::create):
        Use the newly added methods.

2013-02-07  Caio Marcelo de Oliveira Filho  <caio.oliveira@openbossa.org>

        [Qt] Fix build without 3D_GRAPHICS
        https://bugs.webkit.org/show_bug.cgi?id=109194

        Reviewed by Noam Rosenthal.

        Now that Coordinated Graphics was moved to WebCore, we need to explicitly enable
        it when we have 3D_GRAPHICS. This dependency was implicitly by the fact that
        3D_GRAPHICS is a dependency of WebKit2 and Coordinated Graphics was only
        available there. This should fix build for Qt SH4 Linux.

        * wtf/Platform.h:

2013-02-06  Zan Dobersek  <zdobersek@igalia.com>

        [WTFURL] Fix erroneous header inclusions in WTFURL code
        https://bugs.webkit.org/show_bug.cgi?id=109040

        Reviewed by Benjamin Poulain.

        Include WTFURL API headers from files in Source/WTF/wtf/url/src
        by specifying their relative path to Source/WTF. 

        * GNUmakefile.am: The changes make it possible to compile the WTF library
        without specifying both Source/WTF/wtf/url/api and Source/WTF/wtf/url/src
        as inclusion directories in CPPFLAGS, so remove these two entries.
        * wtf/url/src/RawURLBuffer.h:
        * wtf/url/src/URLCanon.h: Include the URLParse.h header by specifying
        only the base name as it's located in the same directory.
        * wtf/url/src/URLCanonQuery.cpp:
        * wtf/url/src/URLUtil.h:

2013-02-06  Tony Gentilcore  <tonyg@chromium.org>

        Call XSSAuditor's didBlockScript() for the threaded HTML parser
        https://bugs.webkit.org/show_bug.cgi?id=108726

        Reviewed by Adam Barth.

        This patch adds isSafeToSendToAnotherThread() methods to CString, String, ParsedURL and URLString.
        These methods check to ensure there are 0 or 1 references.

        * wtf/text/CString.cpp:
        (WTF::CString::isSafeToSendToAnotherThread): Added.
        (WTF):
        * wtf/text/CString.h:
        (CString):
        * wtf/text/WTFString.cpp:
        (WTF::String::isSafeToSendToAnotherThread): Added.
        (WTF):
        * wtf/text/WTFString.h:
        (String):
        * wtf/url/api/ParsedURL.h:
        (WTF::ParsedURL::isSafeToSendToAnotherThread): Added.
        * wtf/url/api/URLString.h:
        (WTF::URLString::isSafeToSendToAnotherThread): Added.

2013-02-06  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: Native Memory Instrumentation: assign class name to the heap graph node automatically
        https://bugs.webkit.org/show_bug.cgi?id=107262

        Reviewed by Yury Semikhatsky.

        We need a way to calculate class name for a pointer automatically.
        Otherwise we need to write className manually in all the instrumentation methods.
        And for all reported but not instrumented classes.

        C++ can do that for us with help of typeid but unfortunatelly it requires rtti.
        There is another way to do that. C++ preprocessor provides a define which has a function name.

        For g++ and clang it is __PRETTY_FUNCTION__.
        For MSVC it is __FUNCTION__.
        The content of the string is a function signature.
        We can use it because it has the name of the template argument.
        The format is sligthly different. That's why I made two different parsers.
        One for MSVC the other for GCC, Clang etc.
        The other problem is the resulting binary size.
        I made very simple function that does the only thing, returns the smallest possible function signature.
        Unfortunatelly MSVC doesn't generate template argument name for functions.
        It does this only for classes.

        * wtf/MemoryInstrumentation.cpp:
        (WTF):
        (WTF::className):
        (WTF::MemoryClassInfo::callReportObjectInfo):
        (WTF::MemoryClassInfo::init):
        * wtf/MemoryInstrumentation.h:
        (WTF):
        (WTF::FN::fn):
        (WTF::fn):
        (WTF::MemoryClassInfo::MemoryClassInfo):
        (MemoryClassInfo):
        (WTF::::reportObjectMemoryUsage):

2013-02-05  Mark Lam  <mark.lam@apple.com>

        Fix EnumClass so that it can be used with switch statements.
        https://bugs.webkit.org/show_bug.cgi?id=109004.

        Reviewed by Sam Weinig.

        * wtf/EnumClass.h:
        (WTF::EnumClass::operator==):
        (WTF::EnumClass::operator!=):
        (WTF::EnumClass::operator<):
        (WTF::EnumClass::operator<=):
        (WTF::EnumClass::operator>):
        (WTF::EnumClass::operator>=):
        (EnumClass):
        (WTF::EnumClass::operator Value):

2013-02-05  Oliver Hunt  <oliver@apple.com>

        Disable TCMalloc hardening as it's breaking leaks.

        Reviewed by Gavin Barraclough.

        * wtf/FastMalloc.cpp:

2013-02-05  Mark Lam  <mark.lam@apple.com>

        Rename ENUM_CLASS_BEGIN() macro to ENUM_CLASS(), and make DatabaseType a strong enum.
        https://bugs.webkit.org/show_bug.cgi?id=108988.

        Reviewed by Alexey Proskuryakov.

        * wtf/EnumClass.h:

2013-02-05  Roger Fong  <roger_fong@apple.com>

        Unreviewed build fix.

        * wtf/FastMalloc.cpp:

2013-02-05  Benjamin Poulain  <bpoulain@apple.com>

        Tidy up StackBounds
        https://bugs.webkit.org/show_bug.cgi?id=108889

        Reviewed by Ryosuke Niwa.

        * wtf/StackBounds.h:
        (StackBounds):
        (WTF::StackBounds::isSafeToRecurse):
        (WTF::StackBounds::size):
        Adopt a more conventional style for a multiline branch.

        (WTF::StackBounds::StackBounds):
        (WTF::StackBounds::current):
        (WTF::StackBounds::recursionLimit):
        Make those method private.

        Making the constructor private ensure initialize() is alwasy called on any StackBounds.

2013-02-05  Zan Dobersek  <zdobersek@igalia.com>

        [WTFURL] Comparison between signed and unsigned integer expressions in URLUtil.cpp
        https://bugs.webkit.org/show_bug.cgi?id=108955

        Reviewed by Benjamin Poulain.

        * wtf/url/src/URLUtil.cpp:
        (URLUtilities): Make the counter variable a signed integer to get rid of the warning.

2013-02-05  Benjamin Poulain  <bpoulain@apple.com>

        Rationalize the use of iOS/Mac #defines in Assertions.cpp
        https://bugs.webkit.org/show_bug.cgi?id=108870

        Reviewed by David Kilzer.

        * wtf/Assertions.cpp: Instead of using PLATFORM(MAC) and assume it works for iOS and OS X,
        use CF as the guard for CFString. Similarily, USE_APPLE_SYSTEM_LOG guards code using ASL.

2013-02-05  Benjamin Poulain  <bpoulain@apple.com>

        Make StringBuilder::toAtomicString() consistent with StringBuilder::toString() for strings of length zero
        https://bugs.webkit.org/show_bug.cgi?id=108894

        Reviewed by Andreas Kling.

        * wtf/text/StringBuilder.h:
        (WTF::StringBuilder::toAtomicString): The function was returning the nullAtom for strings of length zero.
        This is inconsistent with StringBuilder::toString() which always return an empty string.

        This patch unifies the behavior.

2013-02-04  Mark Hahnenberg  <mhahnenberg@apple.com>

        Structure::m_outOfLineCapacity is unnecessary
        https://bugs.webkit.org/show_bug.cgi?id=108206

        Reviewed by Darin Adler.

        Simplifying the utility functions that we use since we don't need a 
        bunch of fancy templates for this one specific call site.

        * wtf/MathExtras.h:
        (WTF::roundUpToPowerOfTwo):

2013-02-05  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r141905.
        http://trac.webkit.org/changeset/141905
        https://bugs.webkit.org/show_bug.cgi?id=108963

        "Broke mac build" (Requested by tonyg-cr on #webkit).

        * wtf/text/CString.cpp:
        * wtf/text/CString.h:
        * wtf/text/WTFString.cpp:
        * wtf/text/WTFString.h:
        (String):
        * wtf/url/api/ParsedURL.h:
        (ParsedURL):
        * wtf/url/api/URLString.h:

2013-02-05  Tony Gentilcore  <tonyg@chromium.org>

        Call XSSAuditor's didBlockScript() for the threaded HTML parser
        https://bugs.webkit.org/show_bug.cgi?id=108726

        Reviewed by Adam Barth.

        This patch adds isSafeToSendToAnotherThread() methods to CString, String, ParsedURL and URLString.
        These methods check to ensure there are 0 or 1 references.

        * wtf/text/CString.cpp:
        (WTF::CString::isSafeToSendToAnotherThread): Added.
        (WTF):
        * wtf/text/CString.h:
        (CString):
        * wtf/text/WTFString.cpp:
        (WTF::String::isSafeToSendToAnotherThread): Added.
        (WTF):
        * wtf/text/WTFString.h:
        (String):
        * wtf/url/api/ParsedURL.h:
        (WTF::ParsedURL::isSafeToSendToAnotherThread): Added.
        * wtf/url/api/URLString.h:
        (WTF::URLString::isSafeToSendToAnotherThread): Added.

2013-02-04  Benjamin Poulain  <bpoulain@apple.com>

        Build fix for AtomicString on iOS

        Unreviewed. The commit r141812 rely on isUIThread(), there is no such things
        in the tree right now. Use pthread_main_np() until more thread handling is upstreamed.

        * wtf/text/AtomicString.cpp:
        (WTF::AtomicStringTable::create):

2013-02-04  Abhishek Arya  <inferno@chromium.org>

        Add ASSERT_WITH_SECURITY_IMPLICATION to detect out of bounds access
        https://bugs.webkit.org/show_bug.cgi?id=108668

        Reviewed by Eric Seidel.

        * wtf/AVLTree.h:
        (WTF::AVLTreeDefaultBSet::operator[]):
        * wtf/BitArray.h:
        (WTF::BitArray::set):
        (WTF::BitArray::get):
        * wtf/FastBitVector.h:
        (WTF::FastBitVector::set):
        (WTF::FastBitVector::clear):
        (WTF::FastBitVector::get):
        * wtf/FixedArray.h:
        (WTF::FixedArray::operator[]):
        * wtf/RefCountedArray.h:
        (WTF::RefCountedArray::at):
        * wtf/TypedArrayBase.h:
        (WTF::TypedArrayBase::item):
        * wtf/text/StringBuffer.h:
        (WTF::StringBuffer::operator[]):
        * wtf/text/StringBuilder.h:
        (WTF::StringBuilder::operator[]):
        * wtf/text/StringImpl.h:
        (WTF::StringImpl::operator[]):

2013-02-04  Benjamin Poulain  <benjamin@webkit.org>

        Upstream iOS's AtomicString
        https://bugs.webkit.org/show_bug.cgi?id=108139

        Reviewed by David Kilzer.

        On iOS, WebCore can be executed from two different threads. To maintain consistency,
        a few changes had been made:
        -The main UI thread and the WebThread share the same AtomicStringTable.
        -A spin lock is needed before any access to prevent any concurrent modification of the string table.
         The spin lock also prevent race on the static initialization of the shared table.

        * wtf/Platform.h:
        Introduce a new USE(WEB_THREAD) to scope changes related to iOS Web Thread.
        * wtf/text/AtomicString.cpp:
        (AtomicStringTableLocker):
        (WTF::AtomicStringTableLocker::AtomicStringTableLocker):
        (WTF::AtomicStringTableLocker::~AtomicStringTableLocker):
        (WTF::AtomicStringTable::create):
        wtfThreadData() is not necessarily inlined on ARM. When it is not inlined, the old code
        causes two call to the function. Instead, we can keep the value in register and pass it
        to AtomicStringTable::create().
        (WTF::stringTable):
        (WTF::addToStringTable):
        (WTF::AtomicString::addSlowCase):
        (WTF::AtomicString::find):
        (WTF::AtomicString::remove):

2013-02-04  Martin Robinson  <mrobinson@igalia.com>

        Fix GTK+ 'make dist' in preparation for the 1.11.5 release.

        * GNUmakefile.list.am:

2013-02-04  David Kilzer  <ddkilzer@apple.com>

        Sort WTF Xcode project file

        * WTF.xcodeproj/project.pbxproj:

2013-02-02  Andreas Kling  <akling@apple.com>

        Vector should consult allocator about ideal size when choosing capacity.
        <http://webkit.org/b/108410>
        <rdar://problem/13124002>

        Reviewed by Benjamin Poulain.

        Added WTF::fastMallocGoodSize(), a workalike/wrapper for OS X's malloc_good_size().
        It returns the actual size of the block that will get allocated for a given byte size.

        Vector's internal buffer now checks with the allocator if the resulting allocation
        could actually house more objects and updates its capacity to make use of the space.

        * wtf/Deque.h:
        (WTF::::expandCapacity):
        * wtf/FastMalloc.cpp:
        (WTF::fastMallocGoodSize):
        * wtf/FastMalloc.h:
        * wtf/Vector.h:
        (WTF::VectorBufferBase::allocateBuffer):
        (WTF::VectorBufferBase::tryAllocateBuffer):
        (WTF::VectorBufferBase::reallocateBuffer):

2013-02-02  Mark Rowe  <mrowe@apple.com>

        <http://webkit.org/b/108745> WTF shouldn't use a script build phase to detect the presence of headers when the compiler can do it for us

        Reviewed by Sam Weinig.

        * WTF.xcodeproj/project.pbxproj: Remove the script phase that used to generate a header file
        containing information about whether certain header files exist on the system.
        * wtf/FastMalloc.cpp: Use Clang's __has_include to detect whether the header exists before including it.

2013-02-01  Roger Fong  <roger_fong@apple.com>

        Unreviewed. WTF VS2010 project cleanup.

        * WTF.vcxproj/WTF.vcxproj:
        * WTF.vcxproj/WTF.vcxproj.filters:
        * WTF.vcxproj/WTFPreLink.cmd: Removed.

2013-02-01  Zan Dobersek  <zdobersek@igalia.com>

        [GTK] Add WTFURL source files to the build
        https://bugs.webkit.org/show_bug.cgi?id=108215

        Reviewed by Benjamin Poulain.

        The WTFURL implementation sources are all still guarded by the USE(WTFURL) guard,
        meaning that the GTK port still uses the default KURL backend. To use the WTFURL
        backend instead, one would have to define WTF_USE_WTFURL in Platform.h.

        * GNUmakefile.am: List the directories from which source headers will be included.
        Plenty of WTFURL code currently just includes the required header by name rather
        than specifying the header path as relative to Source/WTF. In the future all the inclusions
        should probaby be changed to include the header through the path relative to Source/WTF.
        * GNUmakefile.list.am: Add build targets for the WTFURL source files.
        * wtf/url/api/ParsedURL.cpp: Specify the complete path to the required headers and
        reorder the inclusions.
        * wtf/url/src/URLCanon.h: Ditto.

2013-01-18  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: Native Memory Instrumentation: provide edge names and class names for WTF containers and strings
        https://bugs.webkit.org/show_bug.cgi?id=107303

        Reviewed by Yury Semikhatsky.

        I'd like to use ValueType[] as className for the container data members
        because class names of template parameters already present in the container class names.

        * wtf/MemoryInstrumentationArrayBufferView.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashCountedSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashMap.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationListHashSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationSequence.h:
        * wtf/MemoryInstrumentationString.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationVector.h:
        (WTF::reportMemoryUsage):

2013-02-01  Geoffrey Garen  <ggaren@apple.com>

        Added TriState to WTF and started using it in one place
        https://bugs.webkit.org/show_bug.cgi?id=108628

        Reviewed by Beth Dakin.

        Useful for expressing "maybe" conditions in a boolean context.

        * WTF.xcodeproj/project.pbxproj:
        * wtf/TriState.h: Added.
        (WTF):

2013-02-01  Patrick Gansterer  <paroga@webkit.org>

        Build fix for WinCE after r137709
        https://bugs.webkit.org/show_bug.cgi?id=105767

        Do not pass functions as const references to
        templated arguments to make the compiler happy.

        * wtf/StdLibExtras.h:
        (WTF::binarySearch):
        (WTF::tryBinarySearch):
        (WTF::approximateBinarySearch):

2013-01-31  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: Native Memory Instrumentation: replace nodeName argument with className
        https://bugs.webkit.org/show_bug.cgi?id=107278

        Reviewed by Yury Semikhatsky.

        I replaced nodeName with className because we newer report node name for private and raw buffers
        but need to report their class names.

        * wtf/MemoryInstrumentation.cpp:
        (WTF::MemoryInstrumentation::reportLinkToBuffer):
        (WTF::MemoryClassInfo::addRawBuffer):
        (WTF::MemoryClassInfo::addPrivateBuffer):
        * wtf/MemoryInstrumentation.h:
        (WTF::MemoryInstrumentation::addRawBuffer):
        (MemoryClassInfo):

2013-01-31  Jessie Berlin  <jberlin@apple.com>

        Rolling out r141407 because it is causing crashes under
        WTF::TCMalloc_Central_FreeList::FetchFromSpans() in Release builds.

        * wtf/Deque.h:
        (WTF::::expandCapacity):
        * wtf/FastMalloc.cpp:
        * wtf/FastMalloc.h:
        (WTF):
        * wtf/Vector.h:
        (WTF::VectorBufferBase::allocateBuffer):
        (WTF::VectorBufferBase::tryAllocateBuffer):
        (WTF::VectorBufferBase::reallocateBuffer):

2013-01-31  Mark Lam  <mark.lam@apple.com>

        Abstraction for hiding enum class.
        https://bugs.webkit.org/show_bug.cgi?id=108533

        Reviewed by Anders Carlsson.

        * wtf/Compiler.h:
        * wtf/EnumClass.h: Copied from Source/WTF/wtf/TypeSafeEnum.h.
        (WTF::EnumClass::EnumClass):
        (WTF::EnumClass::operator==):
        (WTF::EnumClass::operator!=):
        (WTF::EnumClass::operator<):
        (WTF::EnumClass::operator<=):
        (WTF::EnumClass::operator>):
        (WTF::EnumClass::operator>=):
        * wtf/TypeSafeEnum.h: Removed.

2013-01-31  Andreas Kling  <akling@apple.com>

        Vector should consult allocator about ideal size when choosing capacity.
        <http://webkit.org/b/108410>
        <rdar://problem/13124002>

        Reviewed by Benjamin Poulain.

        Added WTF::fastMallocGoodSize(), a workalike/wrapper for OS X's malloc_good_size().
        It returns the actual size of the block that will get allocated for a given byte size.

        Vector's internal buffer now checks with the allocator if the resulting allocation
        could actually house more objects and updates its capacity to make use of the space.

        * wtf/Deque.h:
        (WTF::::expandCapacity):
        * wtf/FastMalloc.cpp:
        (WTF::fastMallocGoodSize):
        * wtf/FastMalloc.h:
        * wtf/Vector.h:
        (WTF::VectorBufferBase::allocateBuffer):
        (WTF::VectorBufferBase::tryAllocateBuffer):
        (WTF::VectorBufferBase::reallocateBuffer):

2013-01-30  Christophe Dumez  <christophe.dumez@intel.com>

        Add a StringTypeAdapter for ASCIILiteral
        https://bugs.webkit.org/show_bug.cgi?id=108338

        Reviewed by Benjamin Poulain.

        Add StringTypeAdapter for ASCIILiteral type so that concatenation of an
        ASCIILiteral and a String using + operator is efficiently handled.

        * wtf/text/StringConcatenate.h:
        * wtf/text/StringOperators.h:
        (WTF::operator+): Inline some of the operator+ functions that were not
        yet.
        (WTF):

2013-01-30  David Kilzer  <ddkilzer@apple.com>

        Upstream iOS build file changes for WTF
        <http://webkit.org/b/108221>

        Reviewed by Mark Rowe.

        * Configurations/Base.xcconfig:
        - Import iOS.xcconfig.
        - Remove VALID_ARCHS.  Modern Xcodes define these correctly.
        - Make HEADER_SEARCH_PATHS work with iOS Simulator builds.
        - Add SUPPORTED_PLATFORMS so both Mac and iOS SDKs appear in
          schemes.
        - Define INSTALL_PATH when building for macosx SDK.
        * Configurations/CopyWTFHeaders.xcconfig:
        - Make PRIVATE_HEADERS_FOLDER_PATH work with iOS Simulator
          builds.
        * Configurations/DebugRelease.xcconfig:
        - Simplify ARCHS.  This works correctly on 32-bit-only SDKs.
        * Configurations/WTF.xcconfig:
        - Fix installation directory for iOS Simulator builds by
          defining INSTALL_PATH_ACTUAL.
        * Configurations/iOS.xcconfig: Add.
        * WTF.xcodeproj/project.pbxproj:
        - Add iOS.xcconfig to the project.

2013-01-29  Mark Hahnenberg  <mhahnenberg@apple.com>

        Structure::m_outOfLineCapacity is unnecessary
        https://bugs.webkit.org/show_bug.cgi?id=108206

        Reviewed by Geoffrey Garen.

        We're adding a new function that gives us the ability to round a value up to the next power of a certain base.

        * wtf/MathExtras.h:
        (WTF::roundUpToPowerOf):

2013-01-30  Zeno Albisser  <zeno@webkit.org>

        [Qt] Fix Qt/Mac build after r141024 and r141037
        https://bugs.webkit.org/show_bug.cgi?id=108318

        Reviewed by Kentaro Hara.

        * wtf/Functional.h:
            Make sure Block.h is included and its functionality
            is enabled for Qt on Mac.

2013-01-30  Simon Hausmann  <simon.hausmann@digia.com>

        [Qt] Remove QT4_UNICODE related code paths
        https://bugs.webkit.org/show_bug.cgi?id=108316

        Reviewed by Kenneth Rohde Christiansen.

        Get rid of QT4_UNICODE and any related code paths. The Qt port
        requires Qt5 and ICU these days.

        * WTF.gypi:
        * WTF.pro:
        * wtf/unicode/Unicode.h:
        * wtf/unicode/qt4/UnicodeQt4.h: Removed.

2013-01-29  Mark Lam  <mark.lam@apple.com>

        Introducing WTF::TypeSafeEnum and DatabaseError.
        https://bugs.webkit.org/show_bug.cgi?id=108279.

        Reviewed by Geoffrey Garen.

        * WTF.xcodeproj/project.pbxproj:
        * wtf/TypeSafeEnum.h: Added.
        (WTF):
        (TypeSafeEnum):
        (WTF::TypeSafeEnum::TypeSafeEnum):
        (WTF::TypeSafeEnum::value):
        (WTF::TypeSafeEnum::operator==):
        (WTF::TypeSafeEnum::operator!=):
        (WTF::TypeSafeEnum::operator<):
        (WTF::TypeSafeEnum::operator<=):
        (WTF::TypeSafeEnum::operator>):
        (WTF::TypeSafeEnum::operator>=):

2013-01-29  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r140983.
        http://trac.webkit.org/changeset/140983
        https://bugs.webkit.org/show_bug.cgi?id=108277

        Unfortunately, this API has one last client (Requested by
        abarth on #webkit).

        * wtf/Platform.h:

2013-01-28  Benjamin Poulain  <benjamin@webkit.org>

        String constructed from Literals should be non-empty
        https://bugs.webkit.org/show_bug.cgi?id=108103

        Reviewed by Eric Carlson.

        For efficiency, the construction from literal only works with valid non-empty strings.
        One of the constructor was changed to fix an error from HTMLMediaElement.

        This patch replaces the branch with an assertions.

        * wtf/text/StringImpl.cpp:
        (WTF::StringImpl::createFromLiteral):

2013-01-28  David Kilzer  <ddkilzer@apple.com>

        BUILD FIX: Platform.h:1212:22: error: '__MAC_OS_X_VERSION_MIN_REQUIRED' is not defined, evaluates to 0 [-Werror,-Wundef]

        This fixes the following build error introduced in r140366 for
        Bug 107098:

            Platform.h:1212:22: error: '__MAC_OS_X_VERSION_MIN_REQUIRED' is not defined, evaluates to 0 [-Werror,-Wundef]
            #if PLATFORM(MAC) && __MAC_OS_X_VERSION_MIN_REQUIRED >= 1080 && !PLATFORM(IOS)
                                 ^
            1 error generated.

        * wtf/Platform.h: Check !PLATFORM(IOS) first so that iOS builds
        do not try to evaluate __MAC_OS_X_VERSION_MIN_REQUIRED.

2013-01-28  Adam Barth  <abarth@webkit.org>

        Remove webkitNotifications.createHTMLNotification
        https://bugs.webkit.org/show_bug.cgi?id=107598

        Reviewed by Benjamin Poulain.

        * wtf/Platform.h:

2013-01-27  Laszlo Gombos  <l.gombos@samsung.com>

        [ANDROID] Fix a typo
        https://bugs.webkit.org/show_bug.cgi?id=108046

        Reviewed by Adam Barth.

        HAVE_NMAP -> HAVE_MMAP.

        * wtf/Platform.h:

2013-01-27  Zoltan Arvai  <zarvai@inf.u-szeged.hu>

        Fixing atomicIncrement implementation for Windows by dropping support before XP SP2.
        https://bugs.webkit.org/show_bug.cgi?id=106740

        Reviewed by Benjamin Poulain.

        Adding int64_t type atomicIncrement and atomicDecrement implementations for Windows
        into Atomics.h required by WebKit2 after r139514. Separating WinCE implementation
        that does not support WebKit2 and has no support for 64 bit interlocked methods.

        Increasing WINVER and _WIN32_WINNT to XP SP2, because the 64 bit type interlocked methods
        are not supported on previous versions on 32 bit target.

        * config.h:
        * wtf/Atomics.h:
        (WTF):
        (WTF::atomicIncrement):
        (WTF::atomicDecrement):

2013-01-26  Andras Becsi  <andras.becsi@digia.com>

        Unreviewed fix after r140451 to make GIT-SVN repositories happy.

        * WTF.vcproj/WTF.sln:

2013-01-26  Justin Schuh  <jschuh@chromium.org>

        [CHROMIUM] Suppress more c4267 build warnings for Win64 targets
        https://bugs.webkit.org/show_bug.cgi?id=107993

        Reviewed by Abhishek Arya.

        * WTF.gyp/WTF.gyp:

2013-01-25  Eric Seidel  <eric@webkit.org>

        Support 4 and 5 argument bound static functions
        https://bugs.webkit.org/show_bug.cgi?id=107973

        Reviewed by Anders Carlsson.

        Yummy copy/paste template code!
        I'm about to use this in a BackgroundHTMLParser patch, but figured this should be landed separately.

        * wtf/Functional.h:
        (WTF):
        (WTF::R):

2013-01-24  Martin Robinson  <mrobinson@igalia.com>

        Abstract the logic for appending a UChar32 onto StringBuilder
        https://bugs.webkit.org/show_bug.cgi?id=107505

        Reviewed by Darin Adler.

        * wtf/text/StringBuilder.h:
        (WTF::StringBuilder::append): Added a method for appending a UChar32 to a StringBuilder.

2013-01-24  Brent Fulgham  <bfulgham@webkit.org>

        Stipulate build order between WTFGenerated and WTF
        https://bugs.webkit.org/show_bug.cgi?id=107844

        Reviewed by Tim Horton.

        * WTF.vcxproj/WTF.vcxproj: Use VS2010 project dependency
        declaration so WTF depends on a completed WTFGenerated build.

2013-01-23  Abhishek Arya  <inferno@chromium.org>

        Add support for ASSERT_WITH_SECURITY_IMPLICATION.
        https://bugs.webkit.org/show_bug.cgi?id=107699

        Reviewed by Eric Seidel.

        * wtf/Assertions.h: Add ASSERT_WITH_SECURITY_IMPLICATION to
        indicate possible security vulnerabily and enable it by default
        in fuzzing builds.
        * wtf/Vector.h: Use ASSERT_WITH_SECURITY_IMPLICATION for
        bounds check on [] operator.

2013-01-23  Tony Chang  <tony@chromium.org>

        Unreviewed, set svn:eol-style to CRLF on Windows .sln files.

        * WTF.vcproj/WTF.sln: Added property svn:eol-style.

2013-01-23  Adam Barth  <abarth@webkit.org>

        BackgroundHTMLParser::sendTokensToMainThread should use bind
        https://bugs.webkit.org/show_bug.cgi?id=107637

        Reviewed by Eric Seidel.

        * wtf/Functional.h:
            - I had to re-work the approach to validating WeakPtr |this|
              arguments a bit. Previously you couldn't pass a WeakPtr as a
              non-|this| argument to a function because we would try to unwrap
              it into a raw pointer.
        * wtf/WeakPtr.h:
        (WTF::WeakPtrFactory::revokeAll):
        (WeakPtrFactory):
            - Let clients revoke all outstanding WeakPtrs without needing to
              destroy the WeakPtrFactory.
        * wtf/chromium/MainThreadChromium.cpp:
        (WTF::callFunctionObject):
        (WTF::callOnMainThread):
            - Implement callOnMainThread for Function objects. The non-Chromium
              implementation of MainThread.cpp already implements this feature.

2013-01-23  Justin Schuh  <jschuh@chromium.org>

        [CHROMIUM] Suppress c4267 build warnings in WTF for Win64 targets
        https://bugs.webkit.org/show_bug.cgi?id=107721

        Reviewed by Abhishek Arya.

        * WTF.gyp/WTF.gyp:

2013-01-23  Oliver Hunt  <oliver@apple.com>

        Add RELEASE_ASSERT (and a few friends)
        https://bugs.webkit.org/show_bug.cgi?id=107725

        Reviewed by Filip Pizlo.

        Adding a few macros to make release mode assertions easier and more readable.
        Also makes same assertions more useful in debug builds.

        * wtf/Assertions.h:

2013-01-22  Joshua Bell  <jsbell@chromium.org>

        Prevent race condition during Worker shutdown
        https://bugs.webkit.org/show_bug.cgi?id=107577

        Reviewed by Dmitry Titov.

        Add MessageQueue::appendAndKill() which wraps those two steps with a mutex so other
        threads can't sneak a message in between.

        * wtf/MessageQueue.h: Added appendAndKill() method.

2013-01-22  Roger Fong  <roger_fong@apple.com>

        WTF project files and property sheets for getting WebKit to compile in VS2010.
        https://bugs.webkit.org/show_bug.cgi?id=106986

        Reviewed by Timothy Horton and Brent Fulgham.

        * WTF.vcproj/WTF.sln:
        * WTF.vcxproj: Added.
        * WTF.vcxproj/WTF.vcxproj: Added.
        * WTF.vcxproj/WTF.vcxproj.filters: Added.
        * WTF.vcxproj/WTF.vcxproj.user: Added.
        * WTF.vcxproj/WTFCommon.props: Added.
        * WTF.vcxproj/WTFDebug.props: Added.
        * WTF.vcxproj/WTFGenerated.make: Added.
        * WTF.vcxproj/WTFGenerated.vcxproj: Added.
        * WTF.vcxproj/WTFGenerated.vcxproj.filters: Added.
        * WTF.vcxproj/WTFGenerated.vcxproj.user: Added.
        * WTF.vcxproj/WTFGeneratedCommon.props: Added.
        * WTF.vcxproj/WTFGeneratedDebug.props: Added.
        * WTF.vcxproj/WTFGeneratedRelease.props: Added.
        * WTF.vcxproj/WTFPostBuild.cmd: Added.
        * WTF.vcxproj/WTFPreBuild.cmd: Added.
        * WTF.vcxproj/WTFPreLink.cmd: Added.
        * WTF.vcxproj/WTFRelease.props: Added.
        * WTF.vcxproj/build-generated-files.sh: Added.
        * WTF.vcxproj/copy-files.cmd: Added.
        * WTF.vcxproj/work-around-vs-dependency-tracking-bugs.py: Added.
        * config.h:

2013-01-22  David Kilzer  <ddkilzer@apple.com>

        Fix DateMath.cpp to compile with -Wshorten-64-to-32
        <http://webkit.org/b/107503>

        Reviewed by Darin Adler.

        Fixes the following build errors with -Wshorten-64-to-32:

            DateMath.cpp:742:47: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                if (month == 2 && day > 28 && !isLeapYear(year))
                                               ~~~~~~~~~~ ^~~~
            DateMath.cpp:757:48: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                double dateSeconds = ymdhmsToSeconds(year, month, day, hours, minutes, seconds) - timeZoneSeconds;
                                     ~~~~~~~~~~~~~~~       ^~~~~
            DateMath.cpp:757:55: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                double dateSeconds = ymdhmsToSeconds(year, month, day, hours, minutes, seconds) - timeZoneSeconds;
                                     ~~~~~~~~~~~~~~~              ^~~
            DateMath.cpp:757:60: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                double dateSeconds = ymdhmsToSeconds(year, month, day, hours, minutes, seconds) - timeZoneSeconds;
                                     ~~~~~~~~~~~~~~~                   ^~~~~
            DateMath.cpp:757:67: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                double dateSeconds = ymdhmsToSeconds(year, month, day, hours, minutes, seconds) - timeZoneSeconds;
                                     ~~~~~~~~~~~~~~~                          ^~~~~~~
            DateMath.cpp:996:59: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                                offset = ((o / 100) * 60 + (o % 100)) * sgn;
                                       ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~
            DateMath.cpp:998:37: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                                offset = o * 60 * sgn;
                                       ~ ~~~~~~~^~~~~
            DateMath.cpp:1005:40: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                            offset = (o * 60 + o2) * sgn;
                                   ~ ~~~~~~~~~~~~~~^~~~~
            DateMath.cpp:1041:40: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                return ymdhmsToSeconds(year, month + 1, day, hour, minute, second) * msPerSecond;
                       ~~~~~~~~~~~~~~~       ~~~~~~^~~
            DateMath.cpp:1041:45: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                return ymdhmsToSeconds(year, month + 1, day, hour, minute, second) * msPerSecond;
                       ~~~~~~~~~~~~~~~                  ^~~
            DateMath.cpp:1041:50: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                return ymdhmsToSeconds(year, month + 1, day, hour, minute, second) * msPerSecond;
                       ~~~~~~~~~~~~~~~                       ^~~~
            DateMath.cpp:1041:56: error: implicit conversion loses integer precision: 'long' to 'int' [-Werror,-Wshorten-64-to-32]
                return ymdhmsToSeconds(year, month + 1, day, hour, minute, second) * msPerSecond;
                       ~~~~~~~~~~~~~~~                             ^~~~~~
            12 errors generated.

        * wtf/DateMath.cpp:
        (WTF::ymdhmsToSeconds): Change year argument from long to int.
        Change mon, day, hour, minute arguments from int to long.
        (WTF::parseInt): Add.  Identical to parseLong but bounds checks
        for type int.
        (WTF::parseLong): Switch to std::numeric_limits<long> instead of
        macros.
        (WTF::parseES5DatePortion): Change year argument from long to
        int.
        (WTF::parseES5DateFromNullTerminatedCharacters): Change year
        local variable from long to int.
        (WTF::parseDateFromNullTerminatedCharacters): Change year and
        offset local variables from long to int.  Switch from using
        parseLong() to parseInt() as needed.  Ditto for labs() to abs().
        Add overflow check when switching from "MM/DD/YYYY" to
        "YYYY/MM/DD" parsing.

2013-01-17  Andy Estes  <aestes@apple.com>

        Add a USE() macro for content filtering code
        https://bugs.webkit.org/show_bug.cgi?id=107098

        Reviewed by Mark Rowe.

        * wtf/Platform.h: Define WTF_USE_CONTENT_FILTERING on Mac platforms
        more recent than Lion.

2013-01-18  Laszlo Gombos  <l.gombos@samsung.com>

        Remove unnecessary PLATFORM() tests
        https://bugs.webkit.org/show_bug.cgi?id=107360

        Reviewed by Eric Seidel.

        PLATFORM(WIN), PLATFORM(CHROMIUM) and PLATFORM(QT) are mutually
        exclusive. Remove redundant PLATFORM() tests based on this
        invariant.

        * wtf/Platform.h:

2013-01-18  Laszlo Gombos  <l.gombos@samsung.com>

        Remove PLATFORM(TORCHMOBILE) from Assertions.h
        https://bugs.webkit.org/show_bug.cgi?id=107355

        Reviewed by Eric Seidel.

        This appears to be the only place where PLATFORM(TORCHMOBILE) is
        used. I don't believe this ifdef is needed anymore.

        * wtf/Assertions.h:

2013-01-18  Michael Saboff  <msaboff@apple.com>

        Refactor isPowerOf2() and add getLSBSet()
        https://bugs.webkit.org/show_bug.cgi?id=107306

        Reviewed by Filip Pizlo.

        Moved runtime/PropertyMapHashTable.h:PowerOf2() to new hasOneBitSet() and added getLSBSet().

        * wtf/MathExtras.h:
        (hasOneBitSet):
        (getLSBSet):

2013-01-18  David Kilzer  <ddkilzer@apple.com>

        Fix WTF::copyLCharsFromUCharSource() to compile with -Wshorten-64-to-32
        <http://webkit.org/b/107227>

        Reviewed by Benjamin Poulain.

        Fixes the following build error:

            ASCIIFastPath.h:117:59: error: implicit conversion loses integer precision: 'unsigned long' to 'const unsigned int' [-Werror,-Wshorten-64-to-32]
                    const unsigned endLength = length - ucharsPerLoop + 1;
                                   ~~~~~~~~~   ~~~~~~~~~~~~~~~~~~~~~~~^~~
            1 error generated.

        * wtf/text/ASCIIFastPath.h:
        (WTF::copyLCharsFromUCharSource): Change local variables from
        unsigned to size_t.

2013-01-14  Dominik Röttsches  <dominik.rottsches@intel.com>

        [EFL] Update freetype in jhbuild to 2.4.11 and activate subpixel layout
        https://bugs.webkit.org/show_bug.cgi?id=106774

        In order to fix a long standing linespacing/font ascent & descent issue
        we found that FreeType commit b0962ac34e660 solves that problem.
        Let's update FreeType accordingly, but bump it to 2.4.11 due to the
        maintainer recommending several security updates after that said commit.

        Also, in order to avoid extra rebaselining, let's activate subpixel
        layout at the same time.

        Reviewed by Martin Robinson.

        * wtf/Platform.h: Activate Subpixel Layout for EFL.

2013-01-17  Adam Barth  <abarth@webkit.org>

        Teach Functional.h about WeakPtr
        https://bugs.webkit.org/show_bug.cgi?id=107105

        Reviewed by Anders Carlsson.

        A common pattern in cross-thread communication is to call member
        functions of an object on a remote thread. If the caller's reference to
        the object on the remote thread is a WeakPtr, the caller usually wants
        to validate that the object still exists when the call actually takes
        place.

        It's possible to do this manually for every cross-thread call, but that
        is tiresome and error prone. Instead, we can teach bind to validate
        WeakPtr arguments when passed as the "this" parameter to a member
        function.

        * wtf/Functional.h:
        (WTF::ParamStorageTraits::validate):

2013-01-17  David Kilzer  <ddkilzer@apple.com>

        Fix vprintf_stderr_common() to compile with -Wshorten-64-to-32
        <http://webkit.org/b/107126>

        Reviewed by Darin Adler.

        Fixes the following build error:

            Assertions.cpp:92:22: error: implicit conversion loses integer precision: 'CFIndex' (aka 'long') to 'int' [-Werror,-Wshorten-64-to-32]
                    int length = CFStringGetMaximumSizeForEncoding(CFStringGetLength(str), kCFStringEncodingUTF8);
                        ~~~~~~   ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
            1 error generated.

        * wtf/Assertions.cpp:
        (vprintf_stderr_common): Use CFIndex type instead of int for the
        return value of CFStringGetMaximumSizeForEncoding().

2013-01-17  Zan Dobersek  <zdobersek@igalia.com>

        [Autotools] Unify WTF sources list regardless of the target OS, Unicode backend
        https://bugs.webkit.org/show_bug.cgi?id=105886

        Reviewed by Gustavo Noronha Silva.

        Rather than including files in the sources list in an OS-specific way,
        guard specific files with OS(UNIX|WINDOWS), USE((GLIB|ICU)_UNICODE) guards
        and include all of them in the list.

        Proper comments for namespace and #if guard closings are also added or adjusted
        where necessary in the affected files.

        * GNUmakefile.list.am:
        * wtf/OSAllocatorPosix.cpp: Add OS(UNIX) guards.
        * wtf/OSAllocatorWin.cpp: Add OS(WINDOWS) guards.
        * wtf/ThreadSpecificWin.cpp: Ditto.
        * wtf/ThreadingWin.cpp: Ditto.
        * wtf/unicode/glib/UnicodeGLib.cpp: Add USE(GLIB_UNICODE) guards.
        (WTF):
        * wtf/unicode/glib/UnicodeGLib.h: Ditto.
        (WTF):
        * wtf/unicode/icu/CollatorICU.cpp: Style changes.
        * wtf/unicode/icu/UnicodeIcu.h: Add USE(ICU_UNICODE) guards.
        (WTF):
        * wtf/win/OwnPtrWin.cpp: Add OS(WINDOWS) guards.

2013-01-16  Benjamin Poulain  <benjamin@webkit.org>

        Use GCC's implementation of atomicIncrement/Decrement on Mac
        https://bugs.webkit.org/show_bug.cgi?id=106976

        Reviewed by Filip Pizlo.

        * wtf/Atomics.h:
        GCC and LLVM have builtin for atomic ADD and SUB: __sync_add_and_fetch,
        __sync_sub_and_fetch.

        Using them let the compiler just generate the atomic operations inline
        instead of generating a function call to LibC. It also simplify the
        code a bit.

2013-01-15  Adam Barth  <abarth@webkit.org>

        Generalize DocumentWeakReference into WTF::WeakPtr
        https://bugs.webkit.org/show_bug.cgi?id=106854

        Reviewed by Darin Adler.

        This patch adds a simple WeakPtr object to WTF. To use WeakPtr, objects
        must hold a WeakPtrFactory member variable, which clears the WeakPtrs
        when the object is destructed. The underlying storage is a
        ThreadSafeRefCounted "WeakReference" object, which is shared by all the
        WeakPtrs.

        WeakPtr is a generalization of DocumentWeakReference, which is used to
        check whether the Document object is alive when tasks arive on the main
        thread from worker threads. We plan to use a similar pattern in the
        threaded HTML parser, but we want to send tasks to objects other than
        Document.

        * GNUmakefile.list.am:
        * WTF.gypi:
        * WTF.pro:
        * WTF.vcproj/WTF.vcproj:
        * WTF.xcodeproj/project.pbxproj:
        * wtf/CMakeLists.txt:
        * wtf/WeakPtr.h: Added.
        (WeakReference):
            - The ThreadSafeRefCounted object shared by all the WeakPtr
              instances. This patch uses ThreadSafeRefCounted so that we can
              use WeakPtrs to cancel cross-thread messages.
        (WTF::WeakReference::create):
        (WTF::WeakReference::get):
        (WTF::WeakReference::clear):
            - When the object is destroyed, WeakPtrFactory calls this function
              to clear all the WeakPtrs.
        (WTF::WeakReference::WeakReference):
        (WTF::WeakPtr::WeakPtr):
        (WTF::WeakPtr::get):
            - We might want to add implicit conversions and Boolean operators
              in the future, but I've kept this class simple for now.
        (WTF::WeakPtrFactory::WeakPtrFactory):
        (WTF::WeakPtrFactory::~WeakPtrFactory):
        (WTF::WeakPtrFactory::createWeakPtr):

2013-01-12  Csaba Osztrogonác  <ossy@webkit.org>

        Use __sync_add_and_fetch instead of __gnu_cxx::__exchange_and_add
        https://bugs.webkit.org/show_bug.cgi?id=106729

        After r139514 we need atomicIncrement(int64_t volatile*) for all platform. Now the
        GCC implementation of atomicIncrement() is based on __gnu_cxx::__exchange_and_add,
        which doesn't support int64_t type, but __sync_add_and_fetch does.

        Reviewed by Benjamin Poulain.

        * wtf/Atomics.h:
        (WTF::atomicIncrement):
        (WTF::atomicDecrement):

2013-01-11  Filip Pizlo  <fpizlo@apple.com>

        Add WTF_EXPORT_PRIVATE to printInternal() methods of PrintStream.h

        Rubber stamped by Mark Hahnenberg.
        
        This will make it easier to use dataLog() from WebCore.

        * wtf/PrintStream.h:
        (WTF):

2013-01-09  Antti Koivisto  <antti@apple.com>

        Release FastMalloc thread caches on memory warning
        https://bugs.webkit.org/show_bug.cgi?id=106471
        
        Reviewed by Geoff Garen.

        Use TCMalloc_ThreadCache::Cleanup() instead of calling Scavenge() twice. This releases all the memory
        and looks nicer too.

        * wtf/FastMalloc.cpp:
        (WTF::releaseFastMallocFreeMemory):

2013-01-09  Carlos Garcia Campos  <cgarcia@igalia.com>

        Unreviewed. Fix make distcheck.

        * GNUmakefile.list.am: Add missing header.

2013-01-07  Filip Pizlo  <fpizlo@apple.com>

        Rationalize closure call heuristics and profiling
        https://bugs.webkit.org/show_bug.cgi?id=106270

        Reviewed by Oliver Hunt.
        
        Add a macro to add a method to a class that returns a dumper. Allows you to have
        secondary dump() methods for dumping either more or less information.

        * wtf/PrintStream.h:
        (WTF):

2013-01-07  Ryosuke Niwa  <rniwa@webkit.org>

        Sorted the xcodeproj file.

        * WTF.xcodeproj/project.pbxproj:

2013-01-07  Filip Pizlo  <fpizlo@apple.com>

        Unreviewed, it should be possible to build JSC on ARM.

        * wtf/FastMalloc.cpp:
        (WTF::TCMalloc_PageHeap::IncrementalScavenge):

2013-01-07  Balazs Kilvady  <kilvadyb@homejinni.com>

        MIPS LLInt implementation.
        https://bugs.webkit.org/show_bug.cgi?id=99706

        Reviewed by Filip Pizlo.

        LLInt implementation for MIPS.

        * wtf/Platform.h:

2013-01-05  Jonathan Liu  <net147@gmail.com>

        Only enable MinGW-w64 pow() workaround if needed
        https://bugs.webkit.org/show_bug.cgi?id=106099

        Reviewed by Filip Pizlo.

        The pow() workaround is no longer needed in the latest version
        of MinGW-w64.

        * wtf/MathExtras.h:

2013-01-05  Jonathan Liu  <net147@gmail.com>

        Fix compilation of MathExtras.h with MinGW-w64
        https://bugs.webkit.org/show_bug.cgi?id=106105

        Reviewed by Simon Hausmann.

        The isfinite and isinf functions are required by wtf_pow but
        not defined until after wtf_pow. Move wtf_pow to after
        "using std::isfinite" and "using std::isinf" to fix compilation.

        * wtf/MathExtras.h:
        (wtf_pow):

2013-01-03  Filip Pizlo  <fpizlo@apple.com>

        CallLinkStatus should be aware of closure calls, and the DFG bytecode parser should use that as its sole internal notion of how to optimize calls
        https://bugs.webkit.org/show_bug.cgi?id=106027

        Reviewed by Mark Hahnenberg.
        
        I got tired of the various idioms for printing a list of things with comma in between, so I wrote a helper.

        * WTF.xcodeproj/project.pbxproj:
        * wtf/CommaPrinter.h: Added.
        (WTF):
        (CommaPrinter):
        (WTF::CommaPrinter::CommaPrinter):
        (WTF::CommaPrinter::dump):

2013-01-02  Simon Hausmann  <simon.hausmann@digia.com>

        [MinGW-w64] Centralize workaround for pow() implementation
        https://bugs.webkit.org/show_bug.cgi?id=105925

        Reviewed by Sam Weinig.

        As suggested by Sam, move the MinGW-w64 workaround into MathExtras.h
        away from the JSC usage.

        * wtf/MathExtras.h:
        (wtf_pow):

2013-01-02  Adam Barth  <abarth@webkit.org>

        WTF.gypi lists Platform.h twice
        https://bugs.webkit.org/show_bug.cgi?id=105731

        Reviewed by Tony Chang.

        Once ought to be enough for everybody.

        * WTF.gypi:

2012-12-31  Gavin Barraclough  <barraclough@apple.com>

        Objective-C API for JavaScriptCore
        https://bugs.webkit.org/show_bug.cgi?id=105889

        Reviewed by Filip Pizlo.

        * wtf/WTFThreadData.cpp:
        (WTF::WTFThreadData::WTFThreadData):
        * wtf/WTFThreadData.h:
        (WTFThreadData):
            - Added m_apiData - provide convenient storage for use by the API.

2012-12-28  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: Native Memory Instrumentation: instrument not instrumented members.
        https://bugs.webkit.org/show_bug.cgi?id=105830

        Reviewed by Vsevolod Vlasov.

        In some cases we don't want to visit some class members.
        As example we decided to skip pointers to interface classes such as GraphicLayerClient.
        We could use addWeakPointer for them but it can't be used for nonpointer members.
        In the offline discussion we came to a conclusion that we need a new instrumentation
        method ignoreMember, which will be used for all the members which we won't like to visit/instrument.

        DriveBy: Also I instrumented not yet instrumented members.

        * wtf/MemoryInstrumentation.h:
        (MemoryClassInfo):
        (WTF::MemoryClassInfo::ignoreMember):

2012-12-26  Gyuyoung Kim  <gyuyoung.kim@samsung.com>

        Fix build warning in OSAllocatorPosix.cpp
        https://bugs.webkit.org/show_bug.cgi?id=105761

        Reviewed by Kentaro Hara.

        * wtf/OSAllocatorPosix.cpp:
        (WTF::OSAllocator::reserveUncommitted): Fix unused param warnings.

2012-12-24  Yury Semikhatsky  <yurys@chromium.org>

        Memory instrumentation: provide a way to mark a node as a root
        https://bugs.webkit.org/show_bug.cgi?id=105737

        Reviewed by Alexander Pavlov.

        Objects added using MemoryInstrumentation::addRootObject will be marked as roots.

        * wtf/MemoryInstrumentation.cpp:
        (WTF::MemoryInstrumentation::WrapperBase::process):
        (WTF):
        (WTF::MemoryInstrumentation::WrapperBase::processPointer):
        (WTF::MemoryInstrumentation::WrapperBase::processRootObjectRef):
        * wtf/MemoryInstrumentation.h:
        (WrapperBase):
        * wtf/MemoryObjectInfo.h: added root marker which can be used by the heap graph builder.
        (WTF::MemoryObjectInfo::MemoryObjectInfo):
        (WTF::MemoryObjectInfo::isRoot):
        (WTF::MemoryObjectInfo::markAsRoot):
        (MemoryObjectInfo):

2012-12-24  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: Native Memory Instrumentation: propagate member type as edge type to the serialized heap graph.
        https://bugs.webkit.org/show_bug.cgi?id=105725

        Reviewed by Yury Semikhatsky.

        MemoryOwningType was renamed to MemberType.
        Source argument were removed from reportEdge, reportLeaf and other edge related methods because it is not necessary.
        MemberType argument was propagated from MemoryInstrumentation down to HeapGraphSerializer.

        The changes covered by tests in TestWebKitAPI.

        * wtf/MemoryInstrumentation.cpp:
        (WTF::MemoryInstrumentation::reportEdge):
        (WTF::MemoryInstrumentation::reportLinkToBuffer):
        (WTF::MemoryClassInfo::addPrivateBuffer):
        * wtf/MemoryInstrumentation.h:
        (MemoryInstrumentationClient):
        (MemoryInstrumentation):
        (WTF::MemoryInstrumentation::addRawBuffer):
        (WTF::MemoryInstrumentation::MemberTypeTraits::addObject):
        (WTF::MemoryInstrumentation::addObjectImpl):

2012-12-21  Ilya Tikhonovsky  <loislo@chromium.org>

        Unreviewed. Another try to fix Apple Win Release build.

        * wtf/FastMalloc.cpp:
        (WTF::ClassIndex):

2012-12-21  Ilya Tikhonovsky  <loislo@chromium.org>

        Unreviewed compilation fix for Apple Win Release after r138398.

        * wtf/FastMalloc.cpp:
        (WTF::ClassIndex):

2012-12-21  Oliver Hunt  <oliver@apple.com>

        Further harden FastMalloc
        https://bugs.webkit.org/show_bug.cgi?id=105656

        Reviewed by Gavin Barraclough.

        This increases the degree to which we harden the FastMalloc
        linked lists.  We now also mask the previous and next pointers
        in the doubly linked list implementation.  I've also made
        the masking itself somewhat more complex without a measurable
        cost.  We still use ASLR to provide some general entropy, but
        we blind the pointers against each nodes 'this' pointer.

        * wtf/FastMalloc.cpp:
        (WTF::ClassIndex):
        (WTF::SLL_Next):
        (WTF::SLL_SetNext):
        (WTF::Span::next):
        (WTF::Span::prev):
        (WTF::Span::setNext):
        (WTF::Span::setPrev):
        (Span):
          As Span now has to do masking on the next and previous pointers,
          I've updated the code to use accessors instead.
        (WTF::DLL_Init):
        (WTF::DLL_Remove):
        (WTF::DLL_IsEmpty):
        (WTF::DLL_Length):
        (WTF::DLL_Prepend):
        (WTF::TCMalloc_Central_FreeList::enumerateFreeObjects):
        (WTF::TCMalloc_PageHeap::scavenge):
        (WTF::TCMalloc_PageHeap::New):
        (WTF::TCMalloc_PageHeap::AllocLarge):
        (WTF::TCMalloc_PageHeap::ReturnedBytes):
        (WTF::TCMalloc_PageHeap::Check):
        (WTF::TCMalloc_PageHeap::CheckList):
        (WTF::TCMalloc_PageHeap::ReleaseFreeList):
        (WTF::TCMalloc_Central_FreeList::FetchFromSpans):

2012-12-20  Ryuan Choi  <ryuan.choi@samsung.com>

        [EFL] Build break with latest EFL libraries.
        https://bugs.webkit.org/show_bug.cgi?id=104827

        Reviewed by Laszlo Gombos.

        The eo EFL package is introduced and evas and ecore use it since 1.8.
        While introducing Eo, EFL changed several structures of Evas and Ecore
        from own specific class to Eo.

        So, this patch adds FindEo and changes declaration of Evas, Evas_Object,
        Ecore_Timer to build with latest EFL libraries.

        * wtf/OwnPtrCommon.h: Modified declaration of Evas_Object and Ecore_Timer.
        * wtf/PlatformEfl.cmake: Includes eo EFL library.
        * wtf/efl/RefPtrEfl.h: Modified declaration of Evas_Object.

2012-12-20  Oliver Hunt  <oliver@apple.com>

        Harden pointers in FastMalloc's singly linked list implementation
        https://bugs.webkit.org/show_bug.cgi?id=105571

        Reviewed by Gavin Barraclough.

        Add simple xor based hardening of the next pointer in the
        fast malloc singly linked list implementation.  We rely on
        ASLR to introduce the address randomness we want for the mask.
        Happily this produces a very low cost random value to use.

        * wtf/FastMalloc.cpp:
        (WTF):
        (WTF::SLL_Next):
        (WTF::SLL_SetNext):

2012-12-19  Oliver Hunt  <oliver@apple.com>

        StringImpl isolatedCopy unnecessarily copies text-segment character data
        https://bugs.webkit.org/show_bug.cgi?id=105376

        Reviewed by Anders Carlsson.

        This patch adds a new (private) helper to StringImpl that tests whether the StringImpl
        is backed by an ASCII literal.  This allows isolatedCopy() to safely use the createFromLiteral
        constructor rather than making an unnecessary copy.

        * wtf/text/StringImpl.h:
        (StringImpl):
        (WTF::StringImpl::isASCIILiteral):
        (WTF::StringImpl::isolatedCopy):

2012-12-19  Oliver Hunt  <oliver@apple.com>

        WTF String from ASCIILiteral fails to correctly handle empty strings.
        https://bugs.webkit.org/show_bug.cgi?id=105453

        Reviewed by Anders Carlsson.

        When we have an zero length literal we should return the empty StringImpl.

        * wtf/text/StringImpl.cpp:
        (WTF::StringImpl::createFromLiteral):

2012-12-18  Yury Semikhatsky  <yurys@chromium.org>

        Web Inspector: show cached images under MemoryCache -> Images section
        https://bugs.webkit.org/show_bug.cgi?id=105261

        Reviewed by Pavel Feldman.

        Use first provided name and class name. If one of the ancestors tries to overwrite
        it just ignore the call.

        * wtf/MemoryObjectInfo.h:
        (WTF::MemoryObjectInfo::setClassName):
        (WTF::MemoryObjectInfo::setName):

2012-12-18  Uli Schlachter  <psychon@znc.in>

        Implement uncommitted memory for Linux.
        https://bugs.webkit.org/show_bug.cgi?id=65766

        Reviewed by Simon Hausmann.

        The old approach used MAP_NORESERVE to allocate address space without
        committing it. However, that flag gets ignored if
        /proc/sys/vm/overcommit_memory is set to 2. The new approach uses a
        mapping with PROT_NONE. This works because mappings which aren't even
        readable don't get accounted as committed on Linux.

        * wtf/OSAllocatorPosix.cpp:
        (WTF::OSAllocator::reserveUncommitted):
        (WTF::OSAllocator::reserveAndCommit):
        (WTF::OSAllocator::commit):
        (WTF::OSAllocator::decommit):

2012-12-17  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: Native Memory Instrumentation: MemoryInstrumentation doesn't detect reportMemoryUsage method defined in a base class.
        https://bugs.webkit.org/show_bug.cgi?id=105026

        Reviewed by Yury Semikhatsky.

        Old SFINAE test was replaced with new one based on this article:
        http://stackoverflow.com/questions/1966362/sfinae-to-check-for-inherited-member-functions

        * wtf/MemoryInstrumentation.h:
        (MemoryInstrumentation):
        (yes):
        (IsInstrumented):
        (no):
        (WTF::MemoryInstrumentation::IsInstrumented::BaseMixin::reportMemoryUsage):
        (WTF::MemoryInstrumentation::selectInstrumentationMethod):
        (InstrumentationSelector):
        (WTF):
        (WTF::::reportObjectMemoryUsage):

2012-12-17  Ilya Tikhonovsky  <loislo@chromium.org>

        Unreviewed, rolling out r137892.
        http://trac.webkit.org/changeset/137892
        https://bugs.webkit.org/show_bug.cgi?id=105026

        it broke compilation on windows

        * wtf/MemoryInstrumentation.h:
        (WTF::MemoryInstrumentation::selectInstrumentationMethod):
        (MemoryInstrumentation):
        (WTF::MemoryInstrumentation::reportObjectMemoryUsage):
        (WTF::reportMemoryUsage):

2012-12-14  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: Native Memory Instrumentation: MemoryInstrumentation doesn't detect reportMemoryUsage method defined in a base class.
        https://bugs.webkit.org/show_bug.cgi?id=105026

        Reviewed by Yury Semikhatsky.

        Old SFINAE test was replaced with new one based on this article:
        http://stackoverflow.com/questions/1966362/sfinae-to-check-for-inherited-member-functions

        * wtf/MemoryInstrumentation.h:
        (MemoryInstrumentation):
        (yes):
        (IsInstrumented):
        (no):
        (WTF::MemoryInstrumentation::IsInstrumented::BaseMixin::reportMemoryUsage):
        (WTF::MemoryInstrumentation::selectInstrumentationMethod):
        (InstrumentationSelector):
        (WTF):
        (WTF::::reportObjectMemoryUsage):

2012-12-14  Yury Semikhatsky  <yurys@chromium.org>

        Web Inspector: add data grid for exploring native heap graph
        https://bugs.webkit.org/show_bug.cgi?id=105012

        Reviewed by Pavel Feldman.

        Do not report edges with null target.

        * wtf/MemoryInstrumentation.h:
        (WTF::MemoryInstrumentation::addObjectImpl):

2012-12-13  Filip Pizlo  <fpizlo@apple.com>

        Attempt to rationalize and simplify WTF::binarySearch
        https://bugs.webkit.org/show_bug.cgi?id=104890

        Reviewed by Maciej Stachowiak.

        Binary search now has three modes:

        The default: assert that the key was found, but return an adjacent element if it
        wasn't found, if asserts are turned off.
        
        tryBinarySearch: return 0 if the key wasn't found.
        
        approximateBinarySearch: if the key is not found, return an adjacent element (either
        the left or right; no guarantee which).
        
        This also reduces the number of variants of binarySearch. The functor variant is now
        gone because binarySearch now always uses a functor for the key extractor. The
        generic variant is now gone because binarySearch always expects an array type that
        can do operator[].

        * wtf/StdLibExtras.h:
        (WTF::binarySearchImpl):
        (WTF::binarySearch):
        (WTF::tryBinarySearch):
        (WTF::approximateBinarySearch):

2012-12-13  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: Native Memory Instrumentation: do not validate pointers to objects in RenderArena agains tcmalloc data.
        https://bugs.webkit.org/show_bug.cgi?id=104903

        Reviewed by Yury Semikhatsky.

        This is a simplest approach to do that with zero performance overhead.

        * wtf/MemoryInstrumentation.cpp:
        (WTF::MemoryClassInfo::setCustomAllocation):
        (WTF):
        * wtf/MemoryInstrumentation.h:
        (MemoryClassInfo):
        * wtf/MemoryObjectInfo.h:
        (WTF::MemoryObjectInfo::MemoryObjectInfo):
        (WTF::MemoryObjectInfo::customAllocation):
        (WTF::MemoryObjectInfo::setCustomAllocations):
        (MemoryObjectInfo):

2012-12-12  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: Native Memory Instrumentation: remove fake root MemoryObjectInfo.
        https://bugs.webkit.org/show_bug.cgi?id=104796

        Reviewed by Yury Semikhatsky.

        It was not a good idea to introduce a fake root MemoryObjectInfo.
        It makes a problem when we visit an object without its own MemoryObjectType.

        Example: RenderBox has a global pointer to a hash map.
        HashMap doesn't have its own object type because it is a generic container.
        It will inherit object type from the fake root memory object info.
        The same could happen for another container in another class with other MemoryObjectType.

        This fact forces me to create custom process method for root objects
        because they need to have their own MemoryObjectInfo with customisable memory object type.

        Drive by fix: InstrumentedPointer* was replaced with Wrapper* because actually it is using
        for instrumented and not instrumented object classes.

        * wtf/MemoryInstrumentation.cpp:
        (WTF::MemoryInstrumentation::MemoryInstrumentation):
        (WTF::MemoryInstrumentation::WrapperBase::WrapperBase):
        (WTF::MemoryInstrumentation::WrapperBase::process):
        (WTF::MemoryInstrumentation::WrapperBase::processRootObjectRef):
        (WTF):
        * wtf/MemoryInstrumentation.h:
        (WTF::MemoryInstrumentation::addRootObject):
        (MemoryInstrumentation):
        (WrapperBase):
        (WTF::MemoryInstrumentation::WrapperBase::~WrapperBase):
        (Wrapper):
        (OwningTraits):
        (WTF::MemoryInstrumentation::OwningTraits::addRootObject):
        (WTF::MemoryInstrumentation::addObjectImpl):
        (WTF::::Wrapper):
        (WTF::::callReportMemoryUsage):

2012-12-12  Sergio Martins  <sergio.martins@kdab.com>

        Fix QNX Qt build by enabling JIT.
        https://bugs.webkit.org/show_bug.cgi?id=102794

        Reviewed by Simon Hausmann.

        JIT is working, so enable it. Otherwise LLINT gets enabled, and
        that doesn't compile.

        YARR_JIT is crashing at this time though, so disable it.

        * wtf/Platform.h:

2012-12-11  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: Native Memory Instrumentation: simplify Sequences processing code.
        https://bugs.webkit.org/show_bug.cgi?id=104777

        Reviewed by Yury Semikhatsky.

        Redesign reportMemoryUsage for the sequences.
        With the old schema we couldn't instrument HashMap<key, WebCore::LayoutUnit>
        because LayoutUnit has overloaded conversion operators for double, int, etc.
        I wrote an empty SequenceMemoryInstrumentationTraits for LayoutUnit class
        but it doesn't help because the root of problem was hidden in the free template function reportSequenceMemoryUsage.
        I could overload reportSequenceMemoryUsage too but in this case I need to fully specify all the types
        for the iterators like HashMap<.....>::const_iterator. Too many letters.
        I found a better solution. Free reportSequenceMemoryUsage can be converted into default implementation of static
        reportMemoryUsage traits method. In this case the types of iterators will be inferred automaticaly. The old default
        implementation becomes a custom implementation for the user defined types.

        * wtf/MemoryInstrumentationHashCountedSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashMap.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationListHashSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationSequence.h:
        (WTF):
        (WTF::SequenceMemoryInstrumentationTraits::reportMemoryUsage):
        (SequenceMemoryInstrumentationTraits):
        * wtf/MemoryInstrumentationVector.h:
        (WTF::reportMemoryUsage):

2012-12-10  Yury Semikhatsky  <yurys@chromium.org>

        Memory instrumentation: make sure each edge is reported only once
        https://bugs.webkit.org/show_bug.cgi?id=104630

        Reviewed by Pavel Feldman.

        Make sure that outgoing edges are reported only once if we come to the same
        object twice: first by an address of a base class and then by a real address
        of the object.

        * wtf/MemoryInstrumentation.cpp:
        (WTF::MemoryInstrumentation::MemoryInstrumentation):
        (WTF::MemoryInstrumentation::reportLinkToBuffer):
        (WTF::MemoryInstrumentation::InstrumentedPointerBase::InstrumentedPointerBase):
        (WTF::MemoryInstrumentation::InstrumentedPointerBase::process):
        (WTF::MemoryClassInfo::init):
        (WTF::MemoryClassInfo::addRawBuffer):
        (WTF::MemoryClassInfo::addPrivateBuffer):
        * wtf/MemoryInstrumentation.h:
        (InstrumentedPointerBase):
        (MemoryInstrumentation):
        (InstrumentedPointer):
        (WTF::MemoryClassInfo::MemoryClassInfo):
        (WTF::MemoryClassInfo::addMember):
        (MemoryClassInfo):
        (WTF::::InstrumentedPointer):
        (WTF::::callReportMemoryUsage):
        * wtf/MemoryObjectInfo.h:
        (WTF::MemoryObjectInfo::MemoryObjectInfo):
        (WTF::MemoryObjectInfo::firstVisit):
        (WTF::MemoryObjectInfo::className):
        (WTF::MemoryObjectInfo::name):
        (WTF::MemoryObjectInfo::setAlreadyVisited):
        (MemoryObjectInfo):

2012-12-10  Benjamin Poulain  <benjamin@webkit.org>

        Add convenience methods to use ListHashSet for a LRU cache
        https://bugs.webkit.org/show_bug.cgi?id=104499

        Reviewed by Sam Weinig.

        ListHashSet is a great abstract data type to implement caches.
        Unfortunately, the class was missing methods to do that easily and
        efficiently. This patch fixes that.

        The names appendOrMoveToLast() and prependOrMoveToFirst() were chosen
        in favor of append()/prepend() to ensure they are not used in place of
        add() by accident.

        * wtf/ListHashSet.h:
        (ListHashSet):
        (WTF::::removeFirst):
        (WTF::::appendOrMoveToLast):
        (WTF::::prependOrMoveToFirst):
        (WTF::::unlink):
        (WTF::::unlinkAndDelete):
        (WTF::::prependNode):

2012-12-10  Eric Carlson  <eric.carlson@apple.com>

        Add support for in-band text tracks to Mac port
        https://bugs.webkit.org/show_bug.cgi?id=103663

        Reviewed by Sam Weinig.

        * wtf/Platform.h: Define HAVE_AVFOUNDATION_TEXT_TRACK_SUPPORT.

2012-12-08  Patrick Gansterer  <paroga@webkit.org>

        Compiling WTF outside of WebKit failed on Windows
        https://bugs.webkit.org/show_bug.cgi?id=100201

        Reviewed by Darin Adler.

        Explicitly use the Unicode variants of the Windows API to
        allow building the WTF library without the UNICODE define.

        * wtf/win/MainThreadWin.cpp:
        (WTF::initializeMainThreadPlatform):

2012-12-08  Gustavo Noronha Silva  <gns@gnome.org>

        Unreviewed trivial space fix to silence an automake warning.

        * GNUmakefile.list.am:

2012-12-05  Halton Huo  <halton.huo@intel.com>

        [CMake] Unify coding style for CMake files
        https://bugs.webkit.org/show_bug.cgi?id=103605

        Reviewed by Laszlo Gombos.

        Update cmake files(.cmake, CMakeLists.txt) with following style rules:
        1. Indentation
        1.1 Use spaces, not tabs.
        1.2 Four spaces as indent.
        2. Spacing
        2.1 Place one space between control statements and their parentheses.
            For eg, if (), else (), elseif (), endif (), foreach (),
            endforeach (), while (), endwhile (), break ().
        2.2 Do not place spaces between function and macro statements and
            their parentheses. For eg, macro(), endmacro(), function(),
            endfunction().
        2.3 Do not place spaces between a command or function or macro and its
            parentheses, or between a parenthesis and its content. For eg,
            message("testing") not message( "testing") or message ("testing" )
        2.4 No space at line ending.
        3. Lowercase when call commands macros and functions. For eg,
           add_executable() not ADD_EXECUTABLE(), set() not SET().

        * CMakeLists.txt:
        * wtf/CMakeLists.txt:
        * wtf/PlatformBlackBerry.cmake:
        * wtf/PlatformEfl.cmake:
        * wtf/PlatformWinCE.cmake:

2012-12-04  Jon Lee  <jonlee@apple.com>

        Extend StringHasher to take a stream of characters
        https://bugs.webkit.org/show_bug.cgi?id=104076
        <rdar://problem/12811887>

        Reviewed by Darin Adler.

        We essentially move computeHash() to a member function of StringHasher. This allows clients of the
        hasher to append strings to the general hash, without having to first concatenate the strings.

        * wtf/StringHasher.h:
        (WTF::StringHasher::addCharacters): Same as computeHash(), except also take into account the possibility
        that there is a pending character to hash. If this is the case, we align the word boundary by consuming the
        first character in data, adjust the length, move the data pointer forward one, and update the hasher's
        pending character state. Also add an early return so that we are guaranteed at least one character in case
        we run into this situation.
        (WTF::StringHasher::computeHash): Refactor into addCharacters().

2012-12-04  Yury Semikhatsky  <yurys@chromium.org>

        Web Inspector: simplify reportMemoryUsage signature
        https://bugs.webkit.org/show_bug.cgi?id=104028

        Reviewed by Vsevolod Vlasov.

        Removed const& modifier from instrumented pointer type. Was const T* const&,
        now it is const T*

        * wtf/MemoryInstrumentation.cpp:
        (WTF::MemoryClassInfo::addRawBuffer):
        * wtf/MemoryInstrumentation.h:
        (MemoryInstrumentation):
        (WTF::MemoryInstrumentation::selectInstrumentationMethod):
        (WTF::MemoryInstrumentation::reportObjectMemoryUsage):
        (MemoryClassInfo):
        (WTF::reportMemoryUsage):
        (WTF::MemoryInstrumentation::addObjectImpl):
        (WTF):
        * wtf/MemoryInstrumentationArrayBufferView.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashCountedSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashMap.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationListHashSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationParsedURL.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationString.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationVector.h:
        (WTF::reportMemoryUsage):

2012-12-01  Filip Pizlo  <fpizlo@apple.com>

        JSC should be able to report profiling data associated with the IR dumps and disassembly
        https://bugs.webkit.org/show_bug.cgi?id=102999

        Reviewed by Gavin Barraclough.

        Made some minor changes to support the new profiler. FileOutputStream now has an
        open() method, and DataLog uses it. StringPrintStream has a reset() method, which
        allows you to reuse the same StringPrintStream for creating multiple strings.
        SegmentedVector now has a const operator[]. And, WTFString now can do fromUTF8() on
        a CString directly.

        * wtf/DataLog.cpp:
        (WTF::initializeLogFileOnce):
        * wtf/FilePrintStream.cpp:
        (WTF::FilePrintStream::open):
        (WTF):
        * wtf/FilePrintStream.h:
        * wtf/SegmentedVector.h:
        (WTF::SegmentedVector::at):
        (SegmentedVector):
        (WTF::SegmentedVector::operator[]):
        * wtf/StringPrintStream.cpp:
        (WTF::StringPrintStream::reset):
        (WTF):
        * wtf/StringPrintStream.h:
        (StringPrintStream):
        * wtf/text/WTFString.cpp:
        (WTF::String::fromUTF8):
        (WTF):
        * wtf/text/WTFString.h:
        (String):

2012-12-04  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r136597.
        http://trac.webkit.org/changeset/136597
        https://bugs.webkit.org/show_bug.cgi?id=104065

        broke Windows compilation (Requested by yurys on #webkit).

        * wtf/MemoryInstrumentation.cpp:
        (WTF::MemoryClassInfo::addRawBuffer):
        * wtf/MemoryInstrumentation.h:
        (MemoryInstrumentation):
        (WTF::MemoryInstrumentation::selectInstrumentationMethod):
        (WTF::MemoryInstrumentation::reportObjectMemoryUsage):
        (MemoryClassInfo):
        (WTF::reportMemoryUsage):
        (WTF::MemoryInstrumentation::addObjectImpl):
        (WTF):
        * wtf/MemoryInstrumentationArrayBufferView.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashCountedSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashMap.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationListHashSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationParsedURL.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationString.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationVector.h:
        (WTF::reportMemoryUsage):

2012-12-04  Yury Semikhatsky  <yurys@chromium.org>

        Web Inspector: simplify reportMemoryUsage signature
        https://bugs.webkit.org/show_bug.cgi?id=104028

        Reviewed by Vsevolod Vlasov.

        Removed const& modifier from instrumented pointer type. Was const T* const&,
        now it is const T*

        * wtf/MemoryInstrumentation.cpp:
        (WTF::MemoryClassInfo::addRawBuffer):
        * wtf/MemoryInstrumentation.h:
        (MemoryInstrumentation):
        (WTF::MemoryInstrumentation::selectInstrumentationMethod):
        (WTF::MemoryInstrumentation::reportObjectMemoryUsage):
        (MemoryClassInfo):
        (WTF::reportMemoryUsage):
        (WTF::MemoryInstrumentation::addObjectImpl):
        (WTF):
        * wtf/MemoryInstrumentationArrayBufferView.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashCountedSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashMap.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationHashSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationListHashSet.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationParsedURL.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationString.h:
        (WTF::reportMemoryUsage):
        * wtf/MemoryInstrumentationVector.h:
        (WTF::reportMemoryUsage):

2012-12-03  Yury Semikhatsky  <yurys@chromium.org>

        Web Inspector: NMI introduce heap graph recording API
        https://bugs.webkit.org/show_bug.cgi?id=103232

        Reviewed by Vsevolod Vlasov.

        MemoryInstrumentationClient is extended with several methods for reporting
        heap graph vertices and edges. Edge source is always idenfied by heap object
        address. Target node might be either a destination object address(either real
        or adjusted if the edge was retrieved from a pointer to a base class) or a
        MemoryObjectInfo in case of internal buffers.

        * wtf/MemoryInstrumentation.cpp:
        (WTF::MemoryInstrumentation::reportEdge):
        (WTF):
        (WTF::MemoryInstrumentation::reportLinkToBuffer): report edge between the object
        and a buffer which it owns.
        (WTF::MemoryInstrumentation::InstrumentedPointerBase::process):
        (WTF::MemoryClassInfo::addRawBuffer):
        (WTF::MemoryClassInfo::addPrivateBuffer):
        * wtf/MemoryInstrumentation.h:
        (MemoryInstrumentationClient):
        (WTF::MemoryInstrumentationClient::reportNode):
        (WTF::MemoryInstrumentationClient::reportEdge):
        (WTF::MemoryInstrumentationClient::reportLeaf):
        (WTF::MemoryInstrumentationClient::reportBaseAddress): if we found an object by a pointer
        to base class it may have different value than real object address. Report both real and
        adjusted addresses so that they both can be used for identifying corresponding heap graph
        node.
        (WTF::MemoryInstrumentation::addRootObject):
        (MemoryInstrumentation):
        (WTF::MemoryInstrumentation::addObject):
        (WTF::MemoryInstrumentation::addRawBuffer):
        (WTF::MemoryInstrumentation::OwningTraits::addObject):
        (WTF::MemoryClassInfo::addMember):
        (MemoryClassInfo):
        (WTF::MemoryInstrumentation::addObjectImpl):
        * wtf/MemoryObjectInfo.h: added methods for repoting class name and some name string.
        (WTF::MemoryObjectInfo::setClassName):
        (WTF::MemoryObjectInfo::setName):
        (MemoryObjectInfo):

2012-12-03  Filip Pizlo  <fpizlo@apple.com>

        Replace JSValue::description() with JSValue::dump(PrintStream&)
        https://bugs.webkit.org/show_bug.cgi?id=103866

        Reviewed by Darin Adler.

        Make it easier to get a String from a StringPrintStream.

        * wtf/StringPrintStream.cpp:
        (WTF::StringPrintStream::toString):
        * wtf/StringPrintStream.h:
        (StringPrintStream):
        (WTF::toString):

2012-12-02  Filip Pizlo  <fpizlo@apple.com>

        It should be possible to build and run with DFG_ENABLE(PROPAGATION_VERBOSE)
        https://bugs.webkit.org/show_bug.cgi?id=103848

        Reviewed by Sam Weinig.

        Increase the variadicity of dataLog() and print(), and add a pointerDump()
        which allows passing a pointer to something that has a printInternal(). If
        the pointer is null, it'll print "(null)"; otherwise it'll invoke the
        printInternal() function.

        * wtf/DataLog.h:
        (WTF):
        (WTF::dataLog):
        * wtf/PrintStream.h:
        (PrintStream):
        (WTF::PrintStream::print):
        (WTF):
        (PointerDump):
        (WTF::PointerDump::PointerDump):
        (WTF::PointerDump::dump):
        (WTF::pointerDump):

2012-11-30  Simon Hausmann  <simon.hausmann@digia.com>, Pierre Rossi  <pierre.rossi@digia.com>

        [Qt] Separate Qt WebKit into Qt WebKit and Qt WebKit Widgets
        https://bugs.webkit.org/show_bug.cgi?id=99314

        Reviewed by Tor Arne Vestbø.

        This big change separates QtWebKit into QtWebKit and QtWebKitWidgets as
        shared libraries.

        It's a big refactoring that mostly involves moving WebCore dependent
        code into QtWebKit and accessing it through exported QWebFrameAdapter
        and QWebPageAdapter classes.


        * wtf/text/WTFString.h:
        (String): Export the QString conversions also as WTF_STRING_API.

2012-11-29  Filip Pizlo  <fpizlo@apple.com>

        It should be easy to find code blocks in debug dumps
        https://bugs.webkit.org/show_bug.cgi?id=103623

        Reviewed by Geoffrey Garen.

        Changed RawPointer to accept both const void* and void*, and use the former internally.
        
        Cleaned up SHA1 so that the functionality already used internally for self-testing is
        available via the API. This includes addBytes(CString) and computing hex digests.

        * wtf/RawPointer.h:
        (WTF::RawPointer::RawPointer):
        (RawPointer):
        (WTF::RawPointer::value):
        * wtf/SHA1.cpp:
        (WTF::expectSHA1):
        (WTF::SHA1::hexDigest):
        (WTF::SHA1::computeHexDigest):
        * wtf/SHA1.h:
        (WTF::SHA1::addBytes):

2012-11-29  Patrick Gansterer  <paroga@webkit.org>

        Build fix for WinCE after r136096.

        * wtf/StringPrintStream.cpp: Added missing include file.

2012-11-28  Filip Pizlo  <fpizlo@apple.com>

        SpeculatedType dumping should not use the static char buffer[thingy] idiom
        https://bugs.webkit.org/show_bug.cgi?id=103584

        Reviewed by Michael Saboff.

        Added a StringPrintStream, and made it easy to create dumpers for typedefs to primitives.

        * GNUmakefile.list.am:
        * WTF.gypi:
        * WTF.pro:
        * WTF.vcproj/WTF.vcproj:
        * WTF.xcodeproj/project.pbxproj:
        * wtf/CMakeLists.txt:
        * wtf/PrintStream.cpp:
        (WTF::dumpCharacter):
        * wtf/PrintStream.h:
        (WTF::printInternal):
        * wtf/StringPrintStream.cpp: Added.
        (WTF::StringPrintStream::StringPrintStream):
        (WTF::StringPrintStream::~StringPrintStream):
        (WTF::StringPrintStream::vprintf):
        (WTF::StringPrintStream::toCString):
        (WTF::StringPrintStream::increaseSize):
        * wtf/StringPrintStream.h: Added.
        (StringPrintStream):
        (WTF::toCString):

2012-11-28  Michael Saboff  <msaboff@apple.com>

        Update String Stats for recent dataLog changes and add summary
        https://bugs.webkit.org/show_bug.cgi?id=103583

        Reviewed by Filip Pizlo.

        Updated calls to dataLog() to dataLogF() as a results of r135469.
        Added total savings from 8 bit strings in bytes and as a percentage.

        * wtf/text/StringImpl.cpp:
        (WTF::StringStats::printStats):

2012-11-28  Roger Fong  <roger_fong@apple.com>

        Make DataLog work/compile properly on Windows.
        https://bugs.webkit.org/show_bug.cgi?id=103544

        Reviewed by Filip Pizlo.

        * wtf/DataLog.cpp:
        (WTF::initializeLogFileOnce):

2012-11-28  Filip Pizlo  <fpizlo@apple.com>

        It should be possible to say dataLog("count = ", count, "\n") instead of dataLogF("count = %d\n", count)
        https://bugs.webkit.org/show_bug.cgi?id=103009

        Reviewed by Michael Saboff.

        Added the ability to use out.print(...) and dataLog(...) variadically and with
        WTF::printInternal(PrintStream&, const T&) overloads for any type T that you want
        to be able to pass as an argument to out.print() or dataLog(). Also added one
        example class for doing this: RawPointer, which can be used to force a pointer
        to be printed as "%p" rather than matching any overloads that you might want to
        introduce.

        * GNUmakefile.list.am:
        * WTF.gypi:
        * WTF.pro:
        * WTF.vcproj/WTF.vcproj:
        * WTF.xcodeproj/project.pbxproj:
        * wtf/CMakeLists.txt:
        * wtf/DataLog.h:
        (WTF):
        (WTF::dataLog):
        * wtf/PrintStream.cpp:
        (WTF::printInternal):
        (WTF):
        * wtf/PrintStream.h:
        (PrintStream):
        (WTF::PrintStream::print):
        (WTF):
        * wtf/RawPointer.h: Added.
        (WTF):
        (RawPointer):
        (WTF::RawPointer::RawPointer):
        (WTF::RawPointer::value):

2012-11-27  Filip Pizlo  <fpizlo@apple.com>

        Convert some remaining uses of FILE* to PrintStream&.

        Rubber stamped by Mark Hahnenberg.

        * wtf/BitVector.cpp:
        (WTF::BitVector::dump):
        * wtf/BitVector.h:
        (BitVector):

2012-11-27  Pratik Solanki  <psolanki@apple.com>

        objc/objc-runtime.h does not exist on all PLATFORM(MAC)
        https://bugs.webkit.org/show_bug.cgi?id=101780

        Reviewed by Brent Fulgham.

        Clean up header includes so we don't include objc/objc-runtime.h.

        * wtf/Functional.h:
        * wtf/ObjcRuntimeExtras.h: Include objc/message.h here so we don't rely on proper ordering
        of includes in files that include this header.

2012-11-27  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r135828.
        http://trac.webkit.org/changeset/135828
        https://bugs.webkit.org/show_bug.cgi?id=103405

        [Chromium] Multiple ASAN errors after a WebKit roll due to
        mismatched malloc/delete (not free) resulting from
        StringImpl::deref() (Requested by apavlov on #webkit).

        * wtf/FastAllocBase.h:
        * wtf/Platform.h:

2012-11-26  Adam Barth  <abarth@webkit.org>

        [Chromium] fastMalloc has an extra branch on Windows
        https://bugs.webkit.org/show_bug.cgi?id=103027

        Reviewed by Eric Seidel.

        There's no need to override the new/delete operators on non-Mac
        platform because:

        1) We use the system malloc anyway.
        2) We've modified the system malloc to crash in out-of-memory conditions.

        This patch removes a branch (and a call) from malloc, which will
        hopefully improve performance. I haven't measured the performance
        characteristics of this patch, but I will look at the graphs closely
        when landing.

        * wtf/FastAllocBase.h:
        * wtf/Platform.h:

2012-11-26  Filip Pizlo  <fpizlo@apple.com>

        DataLog to a file should work if there are multiple processes using WTF
        https://bugs.webkit.org/show_bug.cgi?id=103323

        Reviewed by Mark Hahnenberg.

        Whereas before DataLog would open a file with the name you specified, now it'll open a file with the
        name plus the PID appended to it. So if you are dealing with multiple processes running with DataLog
        to a file enabled, you'll get multiple separate log files.

        * wtf/DataLog.cpp:
        (WTF::initializeLogFileOnce):

2012-11-26  Zeno Albisser  <zeno@webkit.org>

        [Qt] Fix the LLInt build on Mac
        https://bugs.webkit.org/show_bug.cgi?id=97587

        Reviewed by Simon Hausmann.

        * wtf/InlineASM.h:
            Use OS(DARWIN) instead of PLATFORM(MAC),
            in order to allow Qt to use the same code.
        * wtf/Platform.h:

2012-11-26  Patrick Gansterer  <paroga@webkit.org>

        Build fix for WinCE after r135640.

        * wtf/DataLog.cpp:

2012-11-24  Adam Barth  <abarth@webkit.org>

        Chromium should use TCMalloc on Mac to go fast
        https://bugs.webkit.org/show_bug.cgi?id=102866

        Reviewed by Eric Seidel.

        This patch enables TCMalloc for some WebKit objects. It improves
        dom-modify on Mac by 5-10%.

        On non-Mac platforms, Chromium already uses TCMalloc throughout the
        project. Unfortunately, we haven't yet figured out how to turn TCMalloc
        on globally on Mac because the approach we use for other platforms makes
        some OS X APIs sad.

        The next best thing would be to enable TCMalloc for WebKit by
        overriding the global new and delete operator, as is done on apple-mac
        and other platforms. Unfortunately, we cannot use that approach either
        because the Chromium WebKit API is not memory tight.

        Fortunately, WebKit has a mechanism for selectively enabling TCMalloc
        for a selection of objects by overriding the new and delete operators
        on those objects. This patch opts chromium-mac into that scheme by
        setting the appropriate preprocessor macros.

        * WTF.gyp/WTF.gyp:
        * wtf/Platform.h:

2012-11-21  Filip Pizlo  <fpizlo@apple.com>

        Any function that can log things should be able to easily log them to a memory buffer as well
        https://bugs.webkit.org/show_bug.cgi?id=103000

        Reviewed by Sam Weinig.

        We have a number of places where we pass around a FILE* as a target to which to print
        some logging information. But the purpose of passing FILE* instead of always assuming
        that we should dump to stderr is that it may be sometimes useful to send the logging
        information elsewhere. Unfortunately, FILE* isn't quite powerful enough: it's combersome
        to use it to send logging to a string, for example.
        
        We could get around this by using <iostream> and <sstream>, but so far this aspect of
        C++ has not been part of the WebKit coding conventions. Personally I find <iostream>
        awkward due to its abuse of operator overloading.
        
        So this patch introduces the PrintStream abstract class, which offers printf-like
        functionality while completely abstracting the destination and mechanism of the printing
        output. It would be trivial to implement a StringPrintStream, for example. This will feed
        into work on https://bugs.webkit.org/show_bug.cgi?id=102999.
        
        This also sets us up for creating templatized print() and println() methods that will
        allow us to say things like out.print("count = ", count, "\n"), but that is the topic
        of https://bugs.webkit.org/show_bug.cgi?id=103009.
        
        This patch also changes dataLog() to use FilePrintStream internally, and WTF::dataFile()
        now returns a FilePrintStream&. Any previous users of WTF::dataFile() have been changed
        to expect a PrintStream&.

        * GNUmakefile.list.am:
        * WTF.pro:
        * WTF.vcproj/WTF.vcproj:
        * WTF.xcodeproj/project.pbxproj:
        * wtf/CMakeLists.txt:
        * wtf/DataLog.cpp:
        (WTF):
        (WTF::initializeLogFileOnce):
        (WTF::initializeLogFile):
        (WTF::dataFile):
        (WTF::dataLogV):
        (WTF::dataLogString):
        * wtf/DataLog.h:
        (WTF):
        * wtf/FilePrintStream.cpp: Added.
        (WTF):
        (WTF::FilePrintStream::FilePrintStream):
        (WTF::FilePrintStream::~FilePrintStream):
        (WTF::FilePrintStream::vprintf):
        (WTF::FilePrintStream::flush):
        * wtf/FilePrintStream.h: Added.
        (WTF):
        (FilePrintStream):
        (WTF::FilePrintStream::file):
        * wtf/PrintStream.cpp: Added.
        (WTF):
        (WTF::PrintStream::PrintStream):
        (WTF::PrintStream::~PrintStream):
        (WTF::PrintStream::printf):
        (WTF::PrintStream::print):
        (WTF::PrintStream::println):
        (WTF::PrintStream::flush):
        (WTF::print):
        * wtf/PrintStream.h: Added.
        (WTF):
        (PrintStream):
        (WTF::print):
        (WTF::println):

2012-11-23  Robert Kroeger  <rjkroege@chromium.org>

        Remove unused ScrollByPixelVelocity
        https://bugs.webkit.org/show_bug.cgi?id=102840

        Reviewed by Sam Weinig.

        The GESTURE_ANIMATION feature turns on code in WebCore is unused.
        Remove it.

        No new tests: code removal/cleanup.

        * wtf/Platform.h:

2012-11-23  Laszlo Gombos  <l.gombos@samsung.com>

        [EFL] Define WTF_PLATFORM_EFL in Platform.h
        https://bugs.webkit.org/show_bug.cgi?id=101482

        Reviewed by Kenneth Rohde Christiansen.

        Define WTF_PLATFORM_EFL in Platform.h to be consistent with 
        other ports.

        * wtf/Platform.h:

2012-11-16  Yury Semikhatsky  <yurys@chromium.org>

        Memory instrumentation: extract MemoryObjectInfo declaration into a separate file
        https://bugs.webkit.org/show_bug.cgi?id=102510

        Reviewed by Pavel Feldman.

        Moved MemoryObjectInfo into separate header. Moved definition of MemoryInstrumentation
        methods that depend on MemoryObjectInfo declaration into MemoryInstrumentation.cpp to
        make MemoryInstrumentation require only forward declaration of MemoryObjectInfo.

        * GNUmakefile.list.am:
        * WTF.gypi:
        * WTF.pro:
        * WTF.vcproj/WTF.vcproj:
        * wtf/CMakeLists.txt:
        * wtf/MemoryInstrumentation.cpp: Added.
        (WTF):
        (WTF::MemoryInstrumentation::MemoryInstrumentation):
        (WTF::MemoryInstrumentation::~MemoryInstrumentation):
        (WTF::MemoryInstrumentation::getObjectType): this method allows to get object type without
        pulling in MemoryObjectInfo.h and all its dependencies.
        (WTF::MemoryInstrumentation::callReportObjectInfo):
        (WTF::MemoryInstrumentation::InstrumentedPointerBase::InstrumentedPointerBase):
        (WTF::MemoryInstrumentation::InstrumentedPointerBase::process):
        (WTF::MemoryClassInfo::init):
        (WTF::MemoryClassInfo::addRawBuffer):
        (WTF::MemoryClassInfo::addPrivateBuffer):
        * wtf/MemoryInstrumentation.h:
        (MemoryInstrumentation):
        (WTF::MemoryInstrumentation::addRootObject):
        (InstrumentedPointerBase):
        (WTF::MemoryInstrumentation::reportObjectMemoryUsage):
        (InstrumentedPointer):
        (WTF::MemoryInstrumentation::addObject): we now pass owner's MemoryObjectInfo in all places
        where we report objects pointed by the owner.
        (WTF::MemoryInstrumentation::OwningTraits::addObject):
        (WTF::MemoryClassInfo::MemoryClassInfo):
        (WTF::MemoryClassInfo::addMember):
        (MemoryClassInfo):
        (WTF::MemoryInstrumentation::addObjectImpl):
        (WTF::::InstrumentedPointer):
        (WTF::::callReportMemoryUsage):
        * wtf/MemoryObjectInfo.h: Added.
        (WTF):
        (MemoryObjectInfo):
        (WTF::MemoryObjectInfo::MemoryObjectInfo):
        (WTF::MemoryObjectInfo::objectType):
        (WTF::MemoryObjectInfo::objectSize):
        (WTF::MemoryObjectInfo::reportedPointer):
        (WTF::MemoryObjectInfo::memoryInstrumentation):
        (WTF::MemoryObjectInfo::reportObjectInfo):

2012-11-23  Krzysztof Czech  <k.czech@samsung.com>

        [EFL] Platform support for Accessibility feature.
        https://bugs.webkit.org/show_bug.cgi?id=100848

        Reviewed by Gyuyoung Kim.

        Enable HAVE(ACCESSIBILITY) ifdefs so that EFL port can use it.

        * wtf/Platform.h:

2012-11-22  Michael Saboff  <msaboff@apple.com>

        HTML integer parsing functions don't natively handle 8 bit strings
        https://bugs.webkit.org/show_bug.cgi?id=102997

        Reviewed by Filip Pizlo.

        Added exports to the LChar* versions of charactersToIntStrict() and charactersToUIntStrict()
        to support the changes made to parseHTMLInteger() and parseHTMLNonNegativeInteger().

        * wtf/text/WTFString.h:
        (WTF::charactersToIntStrict): Added export
        (WTF::charactersToUIntStrict): Added export

2012-11-21  Filip Pizlo  <fpizlo@apple.com>

        Rename dataLog() and dataLogV() to dataLogF() and dataLogFV()
        https://bugs.webkit.org/show_bug.cgi?id=103001

        Rubber stamped by Dan Bernstein.

        * wtf/DataLog.cpp:
        (WTF::dataLogFV):
        (WTF::dataLogF):
        (WTF::dataLogFString):
        * wtf/DataLog.h:
        (WTF):
        * wtf/HashTable.cpp:
        (WTF::HashTableStats::dumpStats):
        * wtf/HashTable.h:
        (WTF::HashTable::Stats::dumpStats):
        * wtf/MetaAllocator.cpp:
        (WTF::MetaAllocator::dumpProfile):
        * wtf/StackStats.cpp:
        (WTF::StackStats::initialize):
        (WTF::StackStats::PerThreadStats::PerThreadStats):
        (WTF::StackStats::CheckPoint::CheckPoint):
        (WTF::StackStats::CheckPoint::~CheckPoint):
        (WTF::StackStats::probe):
        (WTF::StackStats::LayoutCheckPoint::LayoutCheckPoint):
        * wtf/text/WTFString.cpp:
        (String::show):

2012-11-21  Anthony G. Basile  <blueness@gentoo.org>

        Source/WTF/wtf/Assertions.cpp: fix build on a uClibc system
        https://bugs.webkit.org/show_bug.cgi?id=102946

        Reviewed by Tony Chang.

        Webkit-Gtk fails to build on a uClibc system because
        Assertions.cpp assumes that all Linux systems have execinfo.h
        and provide backtrace(). This is not necessarily the case for
        uClibc which can be configured without these.  This patch
        refines the check for OS(LINUX) to prevent this breakage.
        Originally reported at https://bugs.gentoo.org/441674

        * wtf/Assertions.cpp:

2012-11-19  Filip Pizlo  <fpizlo@apple.com>

        DFG should be able to cache closure calls
        https://bugs.webkit.org/show_bug.cgi?id=102662

        Reviewed by Gavin Barraclough.

        Added support to the meta allocator for easily querying whether an address falls within
        a certain allocated chunk. Also added a useful debug routine to SentinelLinkedList,
        which can be used to check if a node is on a particular list.

        * wtf/MetaAllocatorHandle.h:
        (MetaAllocatorHandle):
        (WTF::MetaAllocatorHandle::containsIntegerAddress):
        (WTF::MetaAllocatorHandle::contains):
        * wtf/SentinelLinkedList.h:
        (SentinelLinkedList):
        (WTF::::isOnList):
        (WTF):

2012-11-19  Laszlo Gombos  <l.gombos@samsung.com>

        Remove ReadWriteLock
        https://bugs.webkit.org/show_bug.cgi?id=101637

        Reviewed by Darin Adler.

        Remove ReadWriteLock as it does not seems to be used.

        * wtf/Platform.h: Remove the definition of HAVE_PTHREAD_RWLOCK.

        * wtf/ThreadingPrimitives.h: Remove the PlatformReadWriteLock type 
        and the ReadWriteLock class.

        * wtf/ThreadingPthreads.cpp: Remove the implementation of
        the ReadWriteLock class using pthreads.

2012-11-19  Brady Eidson  <beidson@apple.com>

        Add 64-bit specializations for atomicIncrement and atomicDecrement
        https://bugs.webkit.org/show_bug.cgi?id=102702

        Reviewed by Eric Carlson.

        * wtf/Atomics.h:
        (WTF::atomicIncrement): Add A 64-bit version for Darwin.
        (WTF::atomicDecrement): Ditto.

2012-11-18  Laszlo Gombos  <l.gombos@samsung.com>

        Remove non-existent directories from the make system
        https://bugs.webkit.org/show_bug.cgi?id=102632

        Reviewed by Adam Barth.

        Remove (non-existent) symbian references from the exclude list in gyp project files.

        * WTF.gyp/WTF.gyp:

2012-11-16  Michael Saboff  <msaboff@apple.com>

        String::fromUTF8() should take advantage of the ASCII check in convertUTF8ToUTF16()
        https://bugs.webkit.org/show_bug.cgi?id=100577

        Reviewed by Oliver Hunt.

        Passed in ASCII flag to convertUTF8ToUTF16() and if try, create an 8 bit string from the original arguments.
        Relanding after fix to https://bugs.webkit.org/show_bug.cgi?id=102482.

        * wtf/text/WTFString.cpp:
        (WTF::String::fromUTF8):

2012-11-15  Yury Semikhatsky  <yurys@chromium.org>

        Memory instrumentation: add code for reporting stack traces of unknown instrumented objects
        https://bugs.webkit.org/show_bug.cgi?id=102384

        Reviewed by Pavel Feldman.

        Added an option to collect stack traces for instrumented pointers so that they
        can be printed in case the check failed for the pointer. This code is hidden
        behind a define.

        * wtf/MemoryInstrumentation.h:
        (MemoryInstrumentationClient):
        (WTF::MemoryInstrumentation::checkCountedObject): the method now returns false
        in case the check has failed.
        (InstrumentedPointer):
        (WTF::::InstrumentedPointer):
        (WTF):
        (WTF::::process):

2012-11-15  Mark Hahnenberg  <mhahnenberg@apple.com>

        Windows Fibers can corrupt the cached StackBounds
        https://bugs.webkit.org/show_bug.cgi?id=102411

        Reviewed by Geoffrey Garen.

        Windows has support for something called fibers, which are like lightweight versions of 
        threads. Multiple fibers can run within the context of a single thread and they have access 
        to the same thread local storage but have different stacks. If we create a new JSGlobalContext 
        on one fiber, then switch to another fiber and create a JSGlobalContext there, we will call 
        initializeThreading() once for each new JSGlobalContext created. However, since these fibers 
        are technically running inside the same thread, they will clobber each other's wtfThreadData(), 
        which is stored using thread local storage. This can lead to corruption of the WTFThreadData 
        structure for the fibers other than the last one to create a new JSGlobalContext, including 
        the StackBounds data structure which is used during conservative scanning, among other things. 
        This can lead to crashes during garbage collection on Windows if fibers are used.

        A quick fix would be to always get a fresh StackBounds data structure when asking for it 
        instead of using the cached version from the thread local storage. There is a larger problem 
        in that these fibers can corrupt other WebKit data that uses thread local storage. We'll leave 
        those theoretical fixes for future theoretical bugs.

        * wtf/WTFThreadData.h:
        (WTF::WTFThreadData::stack): We now refresh the m_stackBounds field whenever somebody asks for 
        the StackBounds.

2012-11-15  Maciej Stachowiak  <mjs@apple.com>

        Fix an erroneous comment about the operators required by binarySearch
        https://bugs.webkit.org/show_bug.cgi?id=102406

        Reviewed by Anders Carlsson.

        * wtf/StdLibExtras.h: binarySearch needs '==' and '<', not '--', '<' and '>'.

2012-11-14  Michael Saboff  <msaboff@apple.com>

        String::append() should handle two 8 bit strings without converting both to 16 bits
        https://bugs.webkit.org/show_bug.cgi?id=102286

        Reviewed by Oliver Hunt.

        If both strings are 8 bit, then allocate and copy to a new 8 bit string.  Since most strings are
        8 bit, this will save up to 3x the resulting string length in bytes.  2x is due to the possible
        surviving 16 bit source string upconversion and 1x for the 16 bit result now being 8 bit.

        * wtf/text/WTFString.cpp:
        (WTF::String::append):

2012-11-14  Csaba Osztrogonác  <ossy@webkit.org>

        [Qt][ARM] Enable the DFG JIT on ARMv7(Thumb2)
        https://bugs.webkit.org/show_bug.cgi?id=101747

        Reviewed by Simon Hausmann.

        * wtf/Platform.h:

2012-11-13  Christophe Dumez  <christophe.dumez@intel.com>

        Use prefix form of increment / decrement operators in WTF String classes when possible
        https://bugs.webkit.org/show_bug.cgi?id=101859

        Reviewed by Benjamin Poulain.

        Use prefix form of increment / decrement operators whenever possible in
        WTF String classes as this seems to be the convention in WebKit.

        * wtf/text/ASCIIFastPath.h:
        (WTF::copyLCharsFromUCharSource):
        * wtf/text/AtomicString.cpp:
        (WTF::AtomicString::add):
        * wtf/text/Base64.cpp:
        (WTF::base64Encode):
        (WTF::base64DecodeInternal):
        * wtf/text/StringBuilder.cpp:
        (WTF::StringBuilder::allocateBufferUpConvert):
        * wtf/text/StringConcatenate.h:
        * wtf/text/StringImpl.cpp:
        (WTF::StringImpl::getData16SlowCase):
        (WTF::StringImpl::upconvertCharacters):
        (WTF::StringImpl::containsOnlyWhitespace):
        (WTF::StringImpl::lower):
        (WTF::StringImpl::upper):
        (WTF::StringImpl::foldCase):
        (WTF::StringImpl::stripMatchedCharacters):
        (WTF::StringImpl::removeCharacters):
        (WTF::StringImpl::simplifyMatchedCharactersToSpace):
        (WTF::reverseFindInner):
        (WTF::reverseFindIgnoringCaseInner):
        (WTF::StringImpl::replace):
        (WTF::StringImpl::createWithTerminatingNullCharacter):
        * wtf/text/StringImpl.h:
        (WTF::codePointCompare):
        * wtf/text/WTFString.cpp:
        (WTF::String::String):
        (WTF::toIntegralType):

2012-11-13  Ryosuke Niwa  <rniwa@webkit.org>

        Build fix after r134490.

        * wtf/FastMalloc.cpp:

2012-11-13  Brent Fulgham  <bfulgham@webkit.org> and Alex Christensen <alex.christensen@flexsim.com>

        Fix FastMalloc.cpp compile error for MSVC in 64-bit.
        https://bugs.webkit.org/show_bug.cgi?id=88344

        Reviewed by Ryosuke Niwa.

        MSVC will not compile array declarations of zero size. The existing
        padding math for TCMalloc_Central_FreeListPadded would evaluate
        to zero on 64-bit machines, preventing the compile from finishing.

        * wtf/FastMalloc.cpp:
        (TCMalloc_Central_FreeListPadded_Template): Add new template (and
        zero-size specialization) to provide proper behavior under 64-bit
        Windows build.

2012-11-13  Brent Fulgham <bfulgham@webkit.org> and Alex Christensen  <alex.christensen@flexsim.com>

        FastMalloc.cpp needs to be reordered before padding bug can be fixed
        https://bugs.webkit.org/show_bug.cgi?id=89366

        Note: This file violates Style rules.  This change only moves
        the TCMalloc_Central_FreeList class declaration and the various
        FastMallocZone methods earlier in the file.

        Reviewed by Ryosuke Niwa.

        * wtf/FastMalloc.cpp:
        Reordered definitions to prepare for adding specialized template
        (WTF):
        (TCMalloc_Central_FreeList):
        (WTF::TCMalloc_Central_FreeList::length):
        (WTF::TCMalloc_Central_FreeList::tc_length):
        (WTF::TCMalloc_Central_FreeList::enumerateFreeObjects):
        (TCMalloc_Central_FreeListPadded):
        (FastMallocZone):
        (WTF::FastMallocZone::goodSize):
        (WTF::FastMallocZone::check):
        (WTF::FastMallocZone::print):
        (WTF::FastMallocZone::log):
        (WTF::FastMallocZone::forceLock):
        (WTF::FastMallocZone::forceUnlock):
        (WTF::FastMallocZone::statistics):
        (WTF::FastMallocZone::zoneValloc):
        (WTF::FastMallocZone::zoneDestroy):
        (WTF::KernelSupportsTLS):
        (WTF::CheckIfKernelSupportsTLS):
        (TCMalloc_ThreadCache):

2012-11-11  Kenichi Ishibashi  <bashi@chromium.org>

        WTFString::utf8() should have a mode of conversion to use replacement character
        https://bugs.webkit.org/show_bug.cgi?id=101678

        Reviewed by Alexander Pavlov.

        Introduce conversion mode to String::utf8().
        There are three conversion modes; lenient mode, strict mode, and
        "replacing unpaired surrogates with the replacement character" (replacement) mode.
        Lenient mode converts unpaired surrogates. Strict mode fails when there is an unpaired
        surrogates and returns CString(). Replacement mode replaces unpaired surrogates with
        the replacement character(U+FFFD). Replacement mode implements the algorithm defined at
        http://dev.w3.org/2006/webapi/WebIDL/#dfn-obtain-unicode. WebSocket::send() requires
        this algorithm to encode a string to utf-8.

        * wtf/text/WTFString.cpp:
        (WTF::String::utf8): Changed to take ConversionMode as the argument.
        * wtf/text/WTFString.h:
        (String):

2012-11-09  Alexei Filippov  <alph@chromium.org>

        Web Inspector: Fix heap snapshots counted several times by NMI
        https://bugs.webkit.org/show_bug.cgi?id=101085

        The fix moves snapshots size counting to V8PerIsolateData class. As long
        as it has one instance per isolate its reportMemoryUsage should be
        called once per isolate.

        Reviewed by Yury Semikhatsky.

        * wtf/MemoryInstrumentation.h:
        (WTF::MemoryClassInfo::addPrivateBuffer):

2012-11-08  Michael Saboff  <msaboff@apple.com>

        HashAndUTF8CharactersTranslator should create an 8 bit string if possible
        https://bugs.webkit.org/show_bug.cgi?id=101515

        Reviewed by Darin Adler.

        Added isAllASCII flag that's passed to convertUTF8ToUTF16().  If on return it is true, create an
        8 bit string, otherwise use the 16 bit string.

        * wtf/text/AtomicString.cpp:
        (WTF::HashAndUTF8CharactersTranslator::translate):

2012-11-08  Simon Hausmann  <simon.hausmann@digia.com>

        [Qt] Fix build with MSVC 2012 and Angle
        https://bugs.webkit.org/show_bug.cgi?id=101588

        Reviewed by Tor Arne Vestbø.

        Angle includes STL's <memory> header file, which with MSVC 2012 includes stdint.h. MSVC 2012
        does ship stdint.h, but it's lacking other headers such as inttypes.h.

        So for the rest of WebKit we have to continue to use our own versions of these files from
        JavaScriptCore/os-win32. But for Angle we are just including a shipped STL header file (memory)
        and so it's up to the compiler to make that work, i.e. using the stdint.h it ships.

        This patch moves the addition of JavaScriptCore/os-win32 out of default_post.prf, so that it
        doesn't apply to each and every target anymore. In particular it won't apply to Angle anymore,
        because its presence causes a build issue where due to the addition of os-win32/ we end up
        including our own stdint.h but are lacking WTF/ to be in the include search path (which our own
        stdint.h requires). So out of default_post.prf and into WTF.pri, our own wrappers are now only
        used where WTF is also needed, and since os-win32/stdint.h depends on wtf/Platform.h, it seems
        like a logical location.

        * WTF.pri:

2012-11-07  Hans Wennborg  <hans@chromium.org>

        Fix asm operand type for weakCompareAndSwap on ARM_THUMB2
        https://bugs.webkit.org/show_bug.cgi?id=101238

        Reviewed by Benjamin Poulain.

        'result' was a bool, but the asm was expecting a 32-bit register. A
        recent version of Clang warned about this:

          WebKit/Source/WTF/wtf/Atomics.h:163:34: error: the size being stored
          is truncated, use a modifier to specify the size [-Werror,-Wasm-operand-widths]

        This patch fixes it by making 'result' an unsigned. It does not change
        the functionality in practice ('result' would be in a 32-bit register
        anyway), but it makes the code more correct and makes the warning go
        away.

        Also make 'result' for the X86 version an 'unsigned char' to make it
        more clear that it is a byte.

        * wtf/Atomics.h:
        (WTF::weakCompareAndSwap):

2012-11-07  Remy Demarest  <rdemarest@apple.com>

        Function adoptNS and adoptCF should not generate memory leak diagnostic with Clang Static Analyzer.
        https://bugs.webkit.org/show_bug.cgi?id=101420

        Reviewed by Benjamin Poulain.

        Add attributes to the adoptNS and adoptCF function arguments to remove Clang Static Analyzer diagnotics.
        Define CF_RELEASES_ARGUMENT and NS_RELEASES_ARGUMENT if not available, use them in function declarations.

        * wtf/RetainPtr.h:
        (WTF::adoptCF): Adds CF_RELEASES_ARGUMENT attribute to the argument.
        (WTF::adoptNS): Adds NS_RELEASES_ARGUMENT attribute to the argument.

2012-11-07  Christophe Dumez  <christophe.dumez@intel.com>

        Add replaceWithLiteral() method to WTF::String
        https://bugs.webkit.org/show_bug.cgi?id=101257

        Reviewed by Benjamin Poulain.

        Add replaceWithLiteral() method to WTF::String that takes
        replacement string as a literal to avoid uselessly constructing
        a String object.

        * wtf/text/StringImpl.cpp:
        (WTF::StringImpl::replace):
        (WTF):
        * wtf/text/StringImpl.h:
        (WTF::StringImpl::replace):
        (StringImpl):
        * wtf/text/WTFString.h:
        (String):
        (WTF::String::replaceWithLiteral):

2012-11-06  Michael Saboff  <msaboff@apple.com>

        StringBuilder::append(UChar) with an 8 bit quantity shouldn't change the contents to 16 bits
        https://bugs.webkit.org/show_bug.cgi?id=101421

        Reviewed by Anders Carlsson.

        If the string builder contains only 8 bit data, check if the character being appended contains
        8 bit data.  If so, append it to the 8 bit buffer instead of converting the buffer to 16 bits.

        * wtf/text/StringBuilder.cpp:
        (WTF::StringBuilder::append):
        * wtf/text/StringBuilder.h:
        (WTF::StringBuilder::append):

2012-11-06  Benjamin Poulain  <benjamin@webkit.org>

        Speed up TransformationMatrix::multiply() on modern ARM
        https://bugs.webkit.org/show_bug.cgi?id=101084

        Reviewed by Gavin Barraclough.

        * wtf/Platform.h:
        Add CPU(ARM_VFP) for detecting VFP availability.
        Add CPU(APPLE_ARMV7S) for the Apple ARMv7S architecture.

2012-11-06  Laszlo Gombos  <l.gombos@samsung.com>

        Refactor setting TEXTURE_MAPPER_GL
        https://bugs.webkit.org/show_bug.cgi?id=99758

        Reviewed by Noam Rosenthal.

        Make the rule that sets the default for USE(3D_GRAPHICS) port independent.

        If not set, set USE(TEXTURE_MAPPER_GL) when USE(TEXTURE_MAPPER) and 
        USE(3D_GRAPHICS) is enabled.

        * wtf/Platform.h:

2012-11-05  Philip Rogers  <pdr@google.com>

        Unblock SVG external references
        https://bugs.webkit.org/show_bug.cgi?id=100635

        Reviewed by Adam Barth.

        This patch reverts r132849 and r132869 because the potential XSS issue
        turned out to not be an issue after all.

        Covered by existing tests, many of which are re-whitelisted with this patch.

        * wtf/Platform.h:

2012-11-05  Dima Gorbik  <dgorbik@apple.com>

        Back out controversial changes from Bug 98665.
        https://bugs.webkit.org/show_bug.cgi?id=101244

        Reviewed by David Kilzer.

        Backing out changes from Bug 98665 until further discussions take place on rules for including Platform.h in Assertions.h.

        * wtf/Assertions.h:

2012-11-03  Alexey Proskuryakov  <ap@apple.com>

        Get rid of USE(CFURLSTORAGESESSIONS)
        https://bugs.webkit.org/show_bug.cgi?id=101131

        Reviewed by Sam Weinig.

        * wtf/Platform.h:

2012-11-02  Ilya Tikhonovsky  <loislo@chromium.org>

        Web Inspector: NMI instrument NodeRareData::Map. It uses ~250k on nytimes.com
        https://bugs.webkit.org/show_bug.cgi?id=101052

        Reviewed by Yury Semikhatsky.

        I noticed that in many cases cache structures are plain static HashMap, HashSet etc.
        MemoryAgent can visit it but instrumentation for these containers report no objectType.
        It means that addRootObject method needs to accept objectType property as an argument.
        Otherwise I would have had to create a proxy class with proper objectType.

        * wtf/MemoryInstrumentation.h:
        (WTF::MemoryInstrumentation::addRootObject):

2012-11-01  Kent Tamura  <tkent@chromium.org>

        Introduce ENABLE_DATE_AND_TIME_INPUT_TYPES, and clarify usage of other related flags
        https://bugs.webkit.org/show_bug.cgi?id=101007

        Reviewed by Kentaro Hara.

        * wtf/Platform.h:
        Add ENABLE_DATE_AND_TIME_INPUT_TYPES.  It's a union of
        ENABLE_INPUT_TYPE_{DATE,DATETIME,DATETIMELOCAL,MONTH,TIME,WEEK}.

2012-11-01  Yury Semikhatsky  <yurys@chromium.org>

        Memory instrumentation: do not call checkCountedObject with wrong pointers
        https://bugs.webkit.org/show_bug.cgi?id=100958

        Reviewed by Alexander Pavlov.

        Removed calls to checkCountedObject from places where the pointer may contain
        an address of a base class which may differ from the actual object pointer. Instead
        checkCountedObject is only called right after processing deferred pointer where
        we know real address.

        * wtf/MemoryInstrumentation.h:
        (WTF::MemoryInstrumentation::addObjectImpl):
        (WTF::::process):

2012-11-01  Alexey Proskuryakov  <ap@apple.com>

        Rename HAVE(NETWORK_CFDATA_ARRAY_CALLBACK) to USE(NETWORK_CFDATA_ARRAY_CALLBACK)

        Rubber-stamped by Joe Pecoraro.

        All CFNetwork based platforms have this now, but not all use it at the moment.

2012-11-01  Alexey Proskuryakov  <ap@apple.com>

        Fix HAVE(NETWORK_CFDATA_ARRAY_CALLBACK) build
        https://bugs.webkit.org/show_bug.cgi?id=100979

        Reviewed by Joseph Pecoraro.

        * wtf/Platform.h: Added a FIXME about HAVE(NETWORK_CFDATA_ARRAY_CALLBACK).

2012-11-01  Yury Semikhatsky  <yurys@chromium.org>

        Memory instrumentation: report memory occupied by ResourceRequest instead of its base ResourceRequestBase
        https://bugs.webkit.org/show_bug.cgi?id=100497

        Reviewed by Alexander Pavlov.

        * wtf/MemoryInstrumentation.h:
        (WTF::MemoryClassInfo::addPrivateBuffer): do not report memory usage for empty buffer.

2012-10-31  Anders Carlsson  <andersca@apple.com>

        Fix build.
        <rdar://problem/12612207>.

        Reviewed by Sam Weinig.

        * wtf/CheckedArithmetic.h:

2012-10-31  Benjamin Poulain  <bpoulain@apple.com>

        Add an optimized version of copyLCharsFromUCharSource for ARM
        https://bugs.webkit.org/show_bug.cgi?id=94886

        Reviewed by Gavin Barraclough.

        Michael Saboff added a SIMD version of copyLCharsFromUCharSource() in r125846.

        This patch a similar optimization for ARMv7 by using the interleaved load/store available
        in the NEON extension.

        The performance gains:
        -10000 characters: ~3.5 times faster.
        -20 characters (2 vectors): ~55% faster.
        -15 characters (1 vector): ~21% faster.
        -3 characters (no vector, pure overhead): ~10% slower.

        * wtf/text/ASCIIFastPath.h:
        (WTF::copyLCharsFromUCharSource):

2012-10-31  Christophe Dumez  <christophe.dumez@intel.com>

        [EFL][WK2][AC] Use smart pointers for Evas_GL types
        https://bugs.webkit.org/show_bug.cgi?id=100745

        Reviewed by Kenneth Rohde Christiansen.

        Make OwnPtr usable with Evas_GL type to avoid handling
        raw pointers in EFL port.

        * wtf/OwnPtrCommon.h:
        (WTF):
        * wtf/efl/OwnPtrEfl.cpp:
        (WTF):
        (WTF::deleteOwnedPtr):

2012-10-29  Anders Carlsson  <andersca@apple.com>

        Build WebKit as C++11 on Mac
        https://bugs.webkit.org/show_bug.cgi?id=100720

        Reviewed by Daniel Bates.

        * Configurations/Base.xcconfig:
        Add CLANG_CXX_LANGUAGE_STANDARD=gnu++0x.

        * wtf/text/AtomicString.cpp:
        (WTF::AtomicString::add):
        Change the type of length to be unsigned to prevent implicit type conversions in non-constant initializer lists.

        * wtf/unicode/UTF8.cpp:
        (Unicode):
        Add static_casts to prevent implicit type conversions in non-constant initializer lists.

2012-10-29  Anders Carlsson  <andersca@apple.com>

        String::createCFString should return a RetainPtr
        https://bugs.webkit.org/show_bug.cgi?id=100419

        Reviewed by Andreas Kling.

        Make String::createCFString and StringImpl::createCFString return RetainPtrs.

        * wtf/text/AtomicString.h:
        * wtf/text/StringImpl.h:
        * wtf/text/WTFString.h:

2012-10-30  Mark Rowe  <mrowe@apple.com>

        Fix WTF to not install a few header files in bogus locations.

        * WTF.xcodeproj/project.pbxproj:

2012-10-28  Mark Rowe  <mrowe@apple.com>

        Simplify Xcode configuration settings that used to vary between OS versions.

        Reviewed by Dan Bernstein.

        * Configurations/Base.xcconfig:
        * Configurations/DebugRelease.xcconfig:

2012-10-28  Mark Rowe  <mrowe@apple.com>

        Remove references to unsupported OS and Xcode versions.

        Reviewed by Anders Carlsson.

        * Configurations/Base.xcconfig:
        * Configurations/CompilerVersion.xcconfig: Removed.
        * Configurations/DebugRelease.xcconfig:
        * WTF.xcodeproj/project.pbxproj:

2012-10-29  Anders Carlsson  <andersca@apple.com>

        AtomicString(CFStringRef) shouldn't unconditionally create a StringImpl
        https://bugs.webkit.org/show_bug.cgi?id=100701

        Reviewed by Dan Bernstein.

        * wtf/text/AtomicString.h:
        (WTF::AtomicString::AtomicString):
        Change the constructors that take a CFStringRef and an NSString * to call AtomicString::add(CFStringRef)
        and remove AtomicString::createCFString.

        (WTF::AtomicString::add):
        Add new member function declaration. The definition is in a new file in WebCore, AtomicStringCF.cpp.
        Also, fix the overload of add that takes a const char* to call the right other overload instead of itself.

2012-10-29  Adam Barth  <abarth@webkit.org>

        Unreviewed. Correct my previous patch to disable external SVG
        references only on PLATFORM(CHROMIUM).

        * wtf/Platform.h:

2012-10-29  Adam Barth  <abarth@webkit.org>

        Block SVG external references pending a security review
        https://bugs.webkit.org/show_bug.cgi?id=100635

        Reviewed by Eric Seidel.

        We need to do a security review of loading external SVG references
        before we're sure that it is safe.

        * wtf/Platform.h:

2012-10-29  Michael Saboff  <msaboff@apple.com>

        String::split(UChar, Vector<String>&) shouldn't create a temporary String
        https://bugs.webkit.org/show_bug.cgi?id=100578

        Reviewed by Anders Carlsson.

        Changed split(UChar, Vector<String>&) to call split(UChar, bool, Vector<String>&) instead of creating a
        string and calling the split(String,...) version and moved it to WTFString.h.  Also moved
        split(const String& separator, Vector<String>& result) to WTFString.h.

        * wtf/text/WTFString.cpp:
        (WTF::String::split):
        * wtf/text/WTFString.h:
        (WTF::String::split):

2012-10-29  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r132736.
        http://trac.webkit.org/changeset/132736
        https://bugs.webkit.org/show_bug.cgi?id=100652

        It broke all plugin related tests on GTK and on Qt (Requested
        by Ossy on #webkit).

        * wtf/text/WTFString.cpp:
        (WTF::String::fromUTF8):

2012-10-27  Michael Saboff  <msaboff@apple.com>

        Try to create AtomicString as 8 bit where possible
        https://bugs.webkit.org/show_bug.cgi?id=100575

        Reviewed by Oliver Hunt.

        Added StringImpl::create8BitIfPossible() that first tries to create an 8 bit string.  If it finds a 16 bit character
        during processing, it calls the standard create() method.  The assumption is that this will be used on mostly 8 bit
        strings and ones that are shorter (in the tens of characters).  Changed AtomicString to use the new creation method
        for UChar based construction.

        * wtf/text/AtomicString.cpp:
        (WTF::UCharBufferTranslator::translate):
        * wtf/text/StringImpl.cpp:
        (WTF::StringImpl::create8BitIfPossible):
        * wtf/text/StringImpl.h:
        (WTF::StringImpl::create8BitIfPossible):

2012-10-27  Michael Saboff  <msaboff@apple.com>

        String::fromUTF8() should take advantage of the ASCII check in convertUTF8ToUTF16()
        https://bugs.webkit.org/show_bug.cgi?id=100577

        Reviewed by Oliver Hunt.

        Passed is ASCII flag to convertUTF8ToUTF16() and if try, create an 8 bit string from the original arguments.

        * wtf/text/WTFString.cpp:
        (WTF::String::fromUTF8):

2012-10-27  Dan Bernstein  <mitz@apple.com>

        REAL_PLATFORM_NAME build setting is no longer needed
        https://bugs.webkit.org/show_bug.cgi?id=100587

        Reviewed by Mark Rowe.

        Removed the definition of REAL_PLATFORM_NAME and replaced references to it with references
        to PLATFORM_NAME.

        * Configurations/Base.xcconfig:
        * Configurations/CompilerVersion.xcconfig:
        * Configurations/DebugRelease.xcconfig:

2012-10-26  Sheriff Bot  <webkit.review.bot@gmail.com>

        Unreviewed, rolling out r132689.
        http://trac.webkit.org/changeset/132689
        https://bugs.webkit.org/show_bug.cgi?id=100574

        Broke HashMaps containing RetainPtrs (Requested by andersca on
        #webkit).

        * wtf/RetainPtr.h:

2012-10-26  Anders Carlsson  <andersca@apple.com>

        Add an operator& to RetainPtr
        https://bugs.webkit.org/show_bug.cgi?id=100549

        Reviewed by Dan Bernstein.

        This is useful for APIs that return references using out parameters.

        * wtf/RetainPtr.h:
        (WTF::RetainPtr::operator&):

2012-10-26  Gabor Ballabas  <gaborb@inf.u-szeged.hu>

        [Qt] Enable JSC's disassembler on x86, x86_64 Linux
        https://bugs.webkit.org/show_bug.cgi?id=100386

        Reviewed by Simon Hausmann.

        It works fine on Linux x86, x86_64 just needs to be enabled in the
        QtWebKit build system.

        * wtf/Platform.h:

2012-10-25  Michael Saboff  <msaboff@apple.com>

        REGRESSION (r131836): failures in list styles tests on EFL, GTK
        https://bugs.webkit.org/show_bug.cgi?id=99824

        Reviewed by Oliver Hunt.

        Fixed ASCII check.
        Added placeholder argument for new argument added to convertUTF8ToUTF16() in http://trac.webkit.org/changeset/131836.

        * wtf/unicode/UTF8.cpp:
        (WTF::Unicode::convertUTF8ToUTF16):
        * wtf/unicode/glib/UnicodeGLib.cpp:
        (WTF::Unicode::convertCase):

2012-10-25  Christophe Dumez  <christophe.dumez@intel.com>

        [EFL][WK2] Remove some C'ism from EwkView
        https://bugs.webkit.org/show_bug.cgi?id=100370

        Reviewed by Kenneth Rohde Christiansen.

        Make OwnPtr useable for Ecore_IMF_Context to avoid
        using raw pointers in EFL port.

        * wtf/OwnPtrCommon.h:
        (WTF):
        * wtf/PlatformEfl.cmake:
        * wtf/efl/OwnPtrEfl.cpp:
        (WTF::deleteOwnedPtr):
        (WTF):

2012-09-27  Yury Semikhatsky  <yurys@chromium.org>

        Web Inspector: provide memory instrumentation for ListHashSet
        https://bugs.webkit.org/show_bug.cgi?id=97786

        Reviewed by Vsevolod Vlasov.

        Added memory instrumentation for ListHashSet.

        * GNUmakefile.list.am:
        * WTF.gypi:
        * WTF.pro:
        * WTF.vcproj/WTF.vcproj:
        * WTF.xcodeproj/project.pbxproj:
        * wtf/HashSet.h:
        (WTF):
        * wtf/ListHashSet.h:
        (ListHashSet):
        (ListHashSetNodeAllocator): changed visibility of inPool method to public
        (WTF::ListHashSetNodeAllocator::pool):
        (WTF::ListHashSetNodeAllocator::pastPool):
        (WTF::::sizeInBytes): added a method that returns size of the set in bytes including
        all its internals but not the content elements.
        (WTF):
        * wtf/MemoryInstrumentation.h: removed onsolete method. All clients were updated to
        use generic addMember instead.
        (MemoryInstrumentation):
        (WTF::MemoryClassInfo::addPrivateBuffer):
        * wtf/MemoryInstrumentationHashSet.h:
        * wtf/MemoryInstrumentationListHashSet.h: Copied from Source/WTF/wtf/MemoryInstrumentationHashSet.h.
        (WTF):
        (WTF::reportMemoryUsage):

2012-10-23  Benjamin Poulain  <benjamin@webkit.org>

        WTFURL: Implement KURL::setPort()
        https://bugs.webkit.org/show_bug.cgi?id=99898

        Reviewed by Adam Barth.

        Add an efficient implementation for remplacing the port component
        in an URL. This will be the base for replacing other components
        efficiently.

        The testing is covered by fast/dom/HTMLAnchorElement/set-href-attribute-port.html

        * wtf/url/api/ParsedURL.cpp:
        (WTF::generateNewSpecWithPort):
        (WTF::replacePortWithString):
        (WTF::ParsedURL::replacePort):
        * wtf/url/api/ParsedURL.h:
        (ParsedURL):

2012-10-22  Michael Saboff  <msaboff@apple.com>

        Regression(r131655): Crash in StringImpl::findIgnoringCase
        https://bugs.webkit.org/show_bug.cgi?id=99753

        Reviewed by Geoffrey Garen.

        Fixed to use the searchLength to take into account the start index.

        Added LayoutTests/fast/js/find-ignoring-case-regress-99753.html

        * wtf/text/StringImpl.cpp:
        (WTF::StringImpl::findIgnoringCase):

2012-10-22  Simon Hausmann  <simon.hausmann@digia.com>

        Unreviewed: Re-enable LLINT on Qt/Linux after r131932.

        * wtf/Platform.h:

2012-10-22  Yury Semikhatsky  <yurys@chromium.org>

        Web Inspector: do not double count memory of objects with multiple ancestors
        https://bugs.webkit.org/show_bug.cgi?id=99958

        Reviewed by Alexander Pavlov.

        Make sure memory occupied by objects of classes with multiple inheritance is
        not double counted.

        * wtf/MemoryInstrumentation.h:
        (WTF::MemoryObjectInfo::MemoryObjectInfo):
        (WTF::MemoryObjectInfo::reportedPointer):
        (WTF::MemoryObjectInfo::reportObjectInfo): 1) Store actual pointer to the instrumented
        object as it may differ from the original pointer by which it was reported. 2) Make
        the method non-template and calculate object size on the caller side.
        (MemoryObjectInfo):
        (WTF::MemoryInstrumentation::reportObjectMemoryUsage):
        (WTF::MemoryClassInfo::MemoryClassInfo):
        (WTF::MemoryInstrumentation::InstrumentedPointer::process): use adjusted pointer
        returned by reportMemoryUsage to check if the object has already been visited when
        the pointer differs from the orinal one(it may happen if the object was reported
        by a pointer to one of its base classes).

2012-10-20