v8/src/objects.h

5032 lines
180 KiB
C
Raw Normal View History

// Copyright 2006-2009 the V8 project authors. All rights reserved.
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
// * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
// * Redistributions in binary form must reproduce the above
// copyright notice, this list of conditions and the following
// disclaimer in the documentation and/or other materials provided
// with the distribution.
// * Neither the name of Google Inc. nor the names of its
// contributors may be used to endorse or promote products derived
// from this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#ifndef V8_OBJECTS_H_
#define V8_OBJECTS_H_
#include "builtins.h"
#include "code-stubs.h"
#include "smart-pointer.h"
#include "unicode-inl.h"
#if V8_TARGET_ARCH_ARM
#include "arm/constants-arm.h"
#endif
//
// All object types in the V8 JavaScript are described in this file.
//
// Inheritance hierarchy:
// - Object
// - Smi (immediate small integer)
// - Failure (immediate for marking failed operation)
// - HeapObject (superclass for everything allocated in the heap)
// - JSObject
// - JSArray
// - JSRegExp
// - JSFunction
// - GlobalObject
// - JSGlobalObject
// - JSBuiltinsObject
// - JSGlobalProxy
// - JSValue
// - Array
// - ByteArray
// - PixelArray
// - ExternalArray
// - ExternalByteArray
// - ExternalUnsignedByteArray
// - ExternalShortArray
// - ExternalUnsignedShortArray
// - ExternalIntArray
// - ExternalUnsignedIntArray
// - ExternalFloatArray
// - FixedArray
// - DescriptorArray
// - HashTable
// - Dictionary
// - SymbolTable
// - CompilationCacheTable
// - MapCache
// - Context
// - GlobalContext
// - String
// - SeqString
// - SeqAsciiString
// - SeqTwoByteString
// - ConsString
// - ExternalString
// - ExternalAsciiString
// - ExternalTwoByteString
// - HeapNumber
// - Code
// - Map
// - Oddball
// - Proxy
// - SharedFunctionInfo
// - Struct
// - AccessorInfo
// - AccessCheckInfo
// - InterceptorInfo
// - CallHandlerInfo
// - TemplateInfo
// - FunctionTemplateInfo
// - ObjectTemplateInfo
// - Script
// - SignatureInfo
// - TypeSwitchInfo
// - DebugInfo
// - BreakPointInfo
//
// Formats of Object*:
// Smi: [31 bit signed int] 0
// HeapObject: [32 bit direct pointer] (4 byte aligned) | 01
// Failure: [30 bit signed int] 11
// Ecma-262 3rd 8.6.1
enum PropertyAttributes {
NONE = v8::None,
READ_ONLY = v8::ReadOnly,
DONT_ENUM = v8::DontEnum,
DONT_DELETE = v8::DontDelete,
ABSENT = 16 // Used in runtime to indicate a property is absent.
// ABSENT can never be stored in or returned from a descriptor's attributes
// bitfield. It is only used as a return value meaning the attributes of
// a non-existent property.
};
namespace v8 {
namespace internal {
// PropertyDetails captures type and attributes for a property.
// They are used both in property dictionaries and instance descriptors.
class PropertyDetails BASE_EMBEDDED {
public:
PropertyDetails(PropertyAttributes attributes,
PropertyType type,
int index = 0) {
ASSERT(TypeField::is_valid(type));
ASSERT(AttributesField::is_valid(attributes));
ASSERT(IndexField::is_valid(index));
value_ = TypeField::encode(type)
| AttributesField::encode(attributes)
| IndexField::encode(index);
ASSERT(type == this->type());
ASSERT(attributes == this->attributes());
ASSERT(index == this->index());
}
// Conversion for storing details as Object*.
inline PropertyDetails(Smi* smi);
inline Smi* AsSmi();
PropertyType type() { return TypeField::decode(value_); }
bool IsTransition() {
PropertyType t = type();
ASSERT(t != INTERCEPTOR);
return t == MAP_TRANSITION || t == CONSTANT_TRANSITION;
}
bool IsProperty() {
return type() < FIRST_PHANTOM_PROPERTY_TYPE;
}
PropertyAttributes attributes() { return AttributesField::decode(value_); }
int index() { return IndexField::decode(value_); }
inline PropertyDetails AsDeleted();
static bool IsValidIndex(int index) { return IndexField::is_valid(index); }
bool IsReadOnly() { return (attributes() & READ_ONLY) != 0; }
bool IsDontDelete() { return (attributes() & DONT_DELETE) != 0; }
bool IsDontEnum() { return (attributes() & DONT_ENUM) != 0; }
bool IsDeleted() { return DeletedField::decode(value_) != 0;}
// Bit fields in value_ (type, shift, size). Must be public so the
// constants can be embedded in generated code.
class TypeField: public BitField<PropertyType, 0, 3> {};
class AttributesField: public BitField<PropertyAttributes, 3, 3> {};
class DeletedField: public BitField<uint32_t, 6, 1> {};
class IndexField: public BitField<uint32_t, 7, 31-7> {};
static const int kInitialIndex = 1;
private:
uint32_t value_;
};
// Setter that skips the write barrier if mode is SKIP_WRITE_BARRIER.
enum WriteBarrierMode { SKIP_WRITE_BARRIER, UPDATE_WRITE_BARRIER };
// PropertyNormalizationMode is used to specify whether to keep
// inobject properties when normalizing properties of a JSObject.
enum PropertyNormalizationMode {
CLEAR_INOBJECT_PROPERTIES,
KEEP_INOBJECT_PROPERTIES
};
// All Maps have a field instance_type containing a InstanceType.
// It describes the type of the instances.
//
// As an example, a JavaScript object is a heap object and its map
// instance_type is JS_OBJECT_TYPE.
//
// The names of the string instance types are intended to systematically
// mirror their encoding in the instance_type field of the map. The length
// (SHORT, MEDIUM, or LONG) is always mentioned. The default encoding is
// considered TWO_BYTE. It is not mentioned in the name. ASCII encoding is
// mentioned explicitly in the name. Likewise, the default representation is
// considered sequential. It is not mentioned in the name. The other
// representations (eg, CONS, EXTERNAL) are explicitly mentioned.
// Finally, the string is either a SYMBOL_TYPE (if it is a symbol) or a
// STRING_TYPE (if it is not a symbol).
//
// NOTE: The following things are some that depend on the string types having
// instance_types that are less than those of all other types:
// HeapObject::Size, HeapObject::IterateBody, the typeof operator, and
// Object::IsString.
//
// NOTE: Everything following JS_VALUE_TYPE is considered a
// JSObject for GC purposes. The first four entries here have typeof
// 'object', whereas JS_FUNCTION_TYPE has typeof 'function'.
#define INSTANCE_TYPE_LIST_ALL(V) \
V(SHORT_SYMBOL_TYPE) \
V(MEDIUM_SYMBOL_TYPE) \
V(LONG_SYMBOL_TYPE) \
V(SHORT_ASCII_SYMBOL_TYPE) \
V(MEDIUM_ASCII_SYMBOL_TYPE) \
V(LONG_ASCII_SYMBOL_TYPE) \
V(SHORT_CONS_SYMBOL_TYPE) \
V(MEDIUM_CONS_SYMBOL_TYPE) \
V(LONG_CONS_SYMBOL_TYPE) \
V(SHORT_CONS_ASCII_SYMBOL_TYPE) \
V(MEDIUM_CONS_ASCII_SYMBOL_TYPE) \
V(LONG_CONS_ASCII_SYMBOL_TYPE) \
V(SHORT_EXTERNAL_SYMBOL_TYPE) \
V(MEDIUM_EXTERNAL_SYMBOL_TYPE) \
V(LONG_EXTERNAL_SYMBOL_TYPE) \
V(SHORT_EXTERNAL_ASCII_SYMBOL_TYPE) \
V(MEDIUM_EXTERNAL_ASCII_SYMBOL_TYPE) \
V(LONG_EXTERNAL_ASCII_SYMBOL_TYPE) \
V(SHORT_STRING_TYPE) \
V(MEDIUM_STRING_TYPE) \
V(LONG_STRING_TYPE) \
V(SHORT_ASCII_STRING_TYPE) \
V(MEDIUM_ASCII_STRING_TYPE) \
V(LONG_ASCII_STRING_TYPE) \
V(SHORT_CONS_STRING_TYPE) \
V(MEDIUM_CONS_STRING_TYPE) \
V(LONG_CONS_STRING_TYPE) \
V(SHORT_CONS_ASCII_STRING_TYPE) \
V(MEDIUM_CONS_ASCII_STRING_TYPE) \
V(LONG_CONS_ASCII_STRING_TYPE) \
V(SHORT_EXTERNAL_STRING_TYPE) \
V(MEDIUM_EXTERNAL_STRING_TYPE) \
V(LONG_EXTERNAL_STRING_TYPE) \
V(SHORT_EXTERNAL_ASCII_STRING_TYPE) \
V(MEDIUM_EXTERNAL_ASCII_STRING_TYPE) \
V(LONG_EXTERNAL_ASCII_STRING_TYPE) \
V(LONG_PRIVATE_EXTERNAL_ASCII_STRING_TYPE) \
\
V(MAP_TYPE) \
V(HEAP_NUMBER_TYPE) \
V(FIXED_ARRAY_TYPE) \
V(CODE_TYPE) \
V(JS_GLOBAL_PROPERTY_CELL_TYPE) \
V(ODDBALL_TYPE) \
V(PROXY_TYPE) \
V(BYTE_ARRAY_TYPE) \
V(PIXEL_ARRAY_TYPE) \
/* Note: the order of these external array */ \
/* types is relied upon in */ \
/* Object::IsExternalArray(). */ \
V(EXTERNAL_BYTE_ARRAY_TYPE) \
V(EXTERNAL_UNSIGNED_BYTE_ARRAY_TYPE) \
V(EXTERNAL_SHORT_ARRAY_TYPE) \
V(EXTERNAL_UNSIGNED_SHORT_ARRAY_TYPE) \
V(EXTERNAL_INT_ARRAY_TYPE) \
V(EXTERNAL_UNSIGNED_INT_ARRAY_TYPE) \
V(EXTERNAL_FLOAT_ARRAY_TYPE) \
V(FILLER_TYPE) \
\
V(ACCESSOR_INFO_TYPE) \
V(ACCESS_CHECK_INFO_TYPE) \
V(INTERCEPTOR_INFO_TYPE) \
V(SHARED_FUNCTION_INFO_TYPE) \
V(CALL_HANDLER_INFO_TYPE) \
V(FUNCTION_TEMPLATE_INFO_TYPE) \
V(OBJECT_TEMPLATE_INFO_TYPE) \
V(SIGNATURE_INFO_TYPE) \
V(TYPE_SWITCH_INFO_TYPE) \
V(SCRIPT_TYPE) \
\
V(JS_VALUE_TYPE) \
V(JS_OBJECT_TYPE) \
V(JS_CONTEXT_EXTENSION_OBJECT_TYPE) \
V(JS_GLOBAL_OBJECT_TYPE) \
V(JS_BUILTINS_OBJECT_TYPE) \
Split window support from V8. Here is a description of the background and design of split window in Chrome and V8: https://docs.google.com/a/google.com/Doc?id=chhjkpg_47fwddxbfr This change list splits the window object into two parts: 1) an inner window object used as the global object of contexts; 2) an outer window object exposed to JavaScript and accessible by the name 'window'. Firefox did it awhile ago, here are some discussions: https://wiki.mozilla.org/Gecko:SplitWindow. One additional benefit of splitting window in Chrome is that accessing global variables don't need security checks anymore, it can improve applications that use many global variables. V8 support of split window: There are a small number of changes on V8 api to support split window: Security context is removed from V8, so does related API functions; A global object can be detached from its context and reused by a new context; Access checks on an object template can be turned on/off by default; An object can turn on its access checks later; V8 has a new object type, ApiGlobalObject, which is the outer window object type. The existing JSGlobalObject becomes the inner window object type. Security checks are moved from JSGlobalObject to ApiGlobalObject. ApiGlobalObject is the one exposed to JavaScript, it is accessible through Context::Global(). ApiGlobalObject's prototype is set to JSGlobalObject so that property lookups are forwarded to JSGlobalObject. ApiGlobalObject forwards all other property access requests to JSGlobalObject, such as SetProperty, DeleteProperty, etc. Security token is moved to a global context, and ApiGlobalObject has a reference to its global context. JSGlobalObject has a reference to its global context as well. When accessing properties on a global object in JavaScript, the domain security check is performed by comparing the security token of the lexical context (Top::global_context()) to the token of global object's context. The check is only needed when the receiver is a window object, such as 'window.document'. Accessing global variables, such as 'var foo = 3; foo' does not need checks because the receiver is the inner window object. When an outer window is detached from its global context (when a frame navigates away from a page), it is completely detached from the inner window. A new context is created for the new page, and the outer global object is reused. At this point, the access check on the DOMWindow wrapper of the old context is turned on. The code in old context is still able to access DOMWindow properties, but it has to go through domain security checks. It is debatable on how to implement the outer window object. Currently each property access function has to check if the receiver is ApiGlobalObject type. This approach might be error-prone that one may forget to check the receiver when adding new functions. It is unlikely a performance issue because accessing global variables are more common than 'window.foo' style coding. I am still working on the ARM port, and I'd like to hear comments and suggestions on the best way to support it in V8. Review URL: http://codereview.chromium.org/7366 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@540 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-10-21 19:07:58 +00:00
V(JS_GLOBAL_PROXY_TYPE) \
V(JS_ARRAY_TYPE) \
V(JS_REGEXP_TYPE) \
\
V(JS_FUNCTION_TYPE) \
#ifdef ENABLE_DEBUGGER_SUPPORT
#define INSTANCE_TYPE_LIST_DEBUGGER(V) \
V(DEBUG_INFO_TYPE) \
V(BREAK_POINT_INFO_TYPE)
#else
#define INSTANCE_TYPE_LIST_DEBUGGER(V)
#endif
#define INSTANCE_TYPE_LIST(V) \
INSTANCE_TYPE_LIST_ALL(V) \
INSTANCE_TYPE_LIST_DEBUGGER(V)
// Since string types are not consecutive, this macro is used to
// iterate over them.
#define STRING_TYPE_LIST(V) \
V(SHORT_SYMBOL_TYPE, \
SeqTwoByteString::kAlignedSize, \
short_symbol, \
ShortSymbol) \
V(MEDIUM_SYMBOL_TYPE, \
SeqTwoByteString::kAlignedSize, \
medium_symbol, \
MediumSymbol) \
V(LONG_SYMBOL_TYPE, \
SeqTwoByteString::kAlignedSize, \
long_symbol, \
LongSymbol) \
V(SHORT_ASCII_SYMBOL_TYPE, \
SeqAsciiString::kAlignedSize, \
short_ascii_symbol, \
ShortAsciiSymbol) \
V(MEDIUM_ASCII_SYMBOL_TYPE, \
SeqAsciiString::kAlignedSize, \
medium_ascii_symbol, \
MediumAsciiSymbol) \
V(LONG_ASCII_SYMBOL_TYPE, \
SeqAsciiString::kAlignedSize, \
long_ascii_symbol, \
LongAsciiSymbol) \
V(SHORT_CONS_SYMBOL_TYPE, \
ConsString::kSize, \
short_cons_symbol, \
ShortConsSymbol) \
V(MEDIUM_CONS_SYMBOL_TYPE, \
ConsString::kSize, \
medium_cons_symbol, \
MediumConsSymbol) \
V(LONG_CONS_SYMBOL_TYPE, \
ConsString::kSize, \
long_cons_symbol, \
LongConsSymbol) \
V(SHORT_CONS_ASCII_SYMBOL_TYPE, \
ConsString::kSize, \
short_cons_ascii_symbol, \
ShortConsAsciiSymbol) \
V(MEDIUM_CONS_ASCII_SYMBOL_TYPE, \
ConsString::kSize, \
medium_cons_ascii_symbol, \
MediumConsAsciiSymbol) \
V(LONG_CONS_ASCII_SYMBOL_TYPE, \
ConsString::kSize, \
long_cons_ascii_symbol, \
LongConsAsciiSymbol) \
V(SHORT_EXTERNAL_SYMBOL_TYPE, \
ExternalTwoByteString::kSize, \
short_external_symbol, \
ShortExternalSymbol) \
V(MEDIUM_EXTERNAL_SYMBOL_TYPE, \
ExternalTwoByteString::kSize, \
medium_external_symbol, \
MediumExternalSymbol) \
V(LONG_EXTERNAL_SYMBOL_TYPE, \
ExternalTwoByteString::kSize, \
long_external_symbol, \
LongExternalSymbol) \
V(SHORT_EXTERNAL_ASCII_SYMBOL_TYPE, \
ExternalAsciiString::kSize, \
short_external_ascii_symbol, \
ShortExternalAsciiSymbol) \
V(MEDIUM_EXTERNAL_ASCII_SYMBOL_TYPE, \
ExternalAsciiString::kSize, \
medium_external_ascii_symbol, \
MediumExternalAsciiSymbol) \
V(LONG_EXTERNAL_ASCII_SYMBOL_TYPE, \
ExternalAsciiString::kSize, \
long_external_ascii_symbol, \
LongExternalAsciiSymbol) \
V(SHORT_STRING_TYPE, \
SeqTwoByteString::kAlignedSize, \
short_string, \
ShortString) \
V(MEDIUM_STRING_TYPE, \
SeqTwoByteString::kAlignedSize, \
medium_string, \
MediumString) \
V(LONG_STRING_TYPE, \
SeqTwoByteString::kAlignedSize, \
long_string, \
LongString) \
V(SHORT_ASCII_STRING_TYPE, \
SeqAsciiString::kAlignedSize, \
short_ascii_string, \
ShortAsciiString) \
V(MEDIUM_ASCII_STRING_TYPE, \
SeqAsciiString::kAlignedSize, \
medium_ascii_string, \
MediumAsciiString) \
V(LONG_ASCII_STRING_TYPE, \
SeqAsciiString::kAlignedSize, \
long_ascii_string, \
LongAsciiString) \
V(SHORT_CONS_STRING_TYPE, \
ConsString::kSize, \
short_cons_string, \
ShortConsString) \
V(MEDIUM_CONS_STRING_TYPE, \
ConsString::kSize, \
medium_cons_string, \
MediumConsString) \
V(LONG_CONS_STRING_TYPE, \
ConsString::kSize, \
long_cons_string, \
LongConsString) \
V(SHORT_CONS_ASCII_STRING_TYPE, \
ConsString::kSize, \
short_cons_ascii_string, \
ShortConsAsciiString) \
V(MEDIUM_CONS_ASCII_STRING_TYPE, \
ConsString::kSize, \
medium_cons_ascii_string, \
MediumConsAsciiString) \
V(LONG_CONS_ASCII_STRING_TYPE, \
ConsString::kSize, \
long_cons_ascii_string, \
LongConsAsciiString) \
V(SHORT_EXTERNAL_STRING_TYPE, \
ExternalTwoByteString::kSize, \
short_external_string, \
ShortExternalString) \
V(MEDIUM_EXTERNAL_STRING_TYPE, \
ExternalTwoByteString::kSize, \
medium_external_string, \
MediumExternalString) \
V(LONG_EXTERNAL_STRING_TYPE, \
ExternalTwoByteString::kSize, \
long_external_string, \
LongExternalString) \
V(SHORT_EXTERNAL_ASCII_STRING_TYPE, \
ExternalAsciiString::kSize, \
short_external_ascii_string, \
ShortExternalAsciiString) \
V(MEDIUM_EXTERNAL_ASCII_STRING_TYPE, \
ExternalAsciiString::kSize, \
medium_external_ascii_string, \
MediumExternalAsciiString) \
V(LONG_EXTERNAL_ASCII_STRING_TYPE, \
ExternalAsciiString::kSize, \
long_external_ascii_string, \
LongExternalAsciiString)
// A struct is a simple object a set of object-valued fields. Including an
// object type in this causes the compiler to generate most of the boilerplate
// code for the class including allocation and garbage collection routines,
// casts and predicates. All you need to define is the class, methods and
// object verification routines. Easy, no?
//
// Note that for subtle reasons related to the ordering or numerical values of
// type tags, elements in this list have to be added to the INSTANCE_TYPE_LIST
// manually.
#define STRUCT_LIST_ALL(V) \
V(ACCESSOR_INFO, AccessorInfo, accessor_info) \
V(ACCESS_CHECK_INFO, AccessCheckInfo, access_check_info) \
V(INTERCEPTOR_INFO, InterceptorInfo, interceptor_info) \
V(CALL_HANDLER_INFO, CallHandlerInfo, call_handler_info) \
V(FUNCTION_TEMPLATE_INFO, FunctionTemplateInfo, function_template_info) \
V(OBJECT_TEMPLATE_INFO, ObjectTemplateInfo, object_template_info) \
V(SIGNATURE_INFO, SignatureInfo, signature_info) \
V(TYPE_SWITCH_INFO, TypeSwitchInfo, type_switch_info) \
V(SCRIPT, Script, script)
#ifdef ENABLE_DEBUGGER_SUPPORT
#define STRUCT_LIST_DEBUGGER(V) \
V(DEBUG_INFO, DebugInfo, debug_info) \
V(BREAK_POINT_INFO, BreakPointInfo, break_point_info)
#else
#define STRUCT_LIST_DEBUGGER(V)
#endif
#define STRUCT_LIST(V) \
STRUCT_LIST_ALL(V) \
STRUCT_LIST_DEBUGGER(V)
// We use the full 8 bits of the instance_type field to encode heap object
// instance types. The high-order bit (bit 7) is set if the object is not a
// string, and cleared if it is a string.
const uint32_t kIsNotStringMask = 0x80;
const uint32_t kStringTag = 0x0;
const uint32_t kNotStringTag = 0x80;
// If bit 7 is clear, bit 5 indicates that the string is a symbol (if set) or
// not (if cleared).
const uint32_t kIsSymbolMask = 0x20;
const uint32_t kNotSymbolTag = 0x0;
const uint32_t kSymbolTag = 0x20;
// If bit 7 is clear, bits 3 and 4 are the string's size (short, medium or
// long). These values are very special in that they are also used to shift
// the length field to get the length, removing the hash value. This avoids
// using if or switch when getting the length of a string.
const uint32_t kStringSizeMask = 0x18;
const uint32_t kShortStringTag = 0x18;
const uint32_t kMediumStringTag = 0x10;
const uint32_t kLongStringTag = 0x00;
// If bit 7 is clear then bit 2 indicates whether the string consists of
// two-byte characters or one-byte characters.
const uint32_t kStringEncodingMask = 0x4;
const uint32_t kTwoByteStringTag = 0x0;
const uint32_t kAsciiStringTag = 0x4;
// If bit 7 is clear, the low-order 2 bits indicate the representation
// of the string.
const uint32_t kStringRepresentationMask = 0x03;
enum StringRepresentationTag {
kSeqStringTag = 0x0,
kConsStringTag = 0x1,
kExternalStringTag = 0x3
};
// A ConsString with an empty string as the right side is a candidate
// for being shortcut by the garbage collector unless it is a
// symbol. It's not common to have non-flat symbols, so we do not
// shortcut them thereby avoiding turning symbols into strings. See
// heap.cc and mark-compact.cc.
const uint32_t kShortcutTypeMask =
kIsNotStringMask |
kIsSymbolMask |
kStringRepresentationMask;
const uint32_t kShortcutTypeTag = kConsStringTag;
enum InstanceType {
SHORT_SYMBOL_TYPE = kShortStringTag | kSymbolTag | kSeqStringTag,
MEDIUM_SYMBOL_TYPE = kMediumStringTag | kSymbolTag | kSeqStringTag,
LONG_SYMBOL_TYPE = kLongStringTag | kSymbolTag | kSeqStringTag,
SHORT_ASCII_SYMBOL_TYPE =
kShortStringTag | kAsciiStringTag | kSymbolTag | kSeqStringTag,
MEDIUM_ASCII_SYMBOL_TYPE =
kMediumStringTag | kAsciiStringTag | kSymbolTag | kSeqStringTag,
LONG_ASCII_SYMBOL_TYPE =
kLongStringTag | kAsciiStringTag | kSymbolTag | kSeqStringTag,
SHORT_CONS_SYMBOL_TYPE = kShortStringTag | kSymbolTag | kConsStringTag,
MEDIUM_CONS_SYMBOL_TYPE = kMediumStringTag | kSymbolTag | kConsStringTag,
LONG_CONS_SYMBOL_TYPE = kLongStringTag | kSymbolTag | kConsStringTag,
SHORT_CONS_ASCII_SYMBOL_TYPE =
kShortStringTag | kAsciiStringTag | kSymbolTag | kConsStringTag,
MEDIUM_CONS_ASCII_SYMBOL_TYPE =
kMediumStringTag | kAsciiStringTag | kSymbolTag | kConsStringTag,
LONG_CONS_ASCII_SYMBOL_TYPE =
kLongStringTag | kAsciiStringTag | kSymbolTag | kConsStringTag,
SHORT_EXTERNAL_SYMBOL_TYPE =
kShortStringTag | kSymbolTag | kExternalStringTag,
MEDIUM_EXTERNAL_SYMBOL_TYPE =
kMediumStringTag | kSymbolTag | kExternalStringTag,
LONG_EXTERNAL_SYMBOL_TYPE = kLongStringTag | kSymbolTag | kExternalStringTag,
SHORT_EXTERNAL_ASCII_SYMBOL_TYPE =
kShortStringTag | kAsciiStringTag | kSymbolTag | kExternalStringTag,
MEDIUM_EXTERNAL_ASCII_SYMBOL_TYPE =
kMediumStringTag | kAsciiStringTag | kSymbolTag | kExternalStringTag,
LONG_EXTERNAL_ASCII_SYMBOL_TYPE =
kLongStringTag | kAsciiStringTag | kSymbolTag | kExternalStringTag,
SHORT_STRING_TYPE = kShortStringTag | kSeqStringTag,
MEDIUM_STRING_TYPE = kMediumStringTag | kSeqStringTag,
LONG_STRING_TYPE = kLongStringTag | kSeqStringTag,
SHORT_ASCII_STRING_TYPE = kShortStringTag | kAsciiStringTag | kSeqStringTag,
MEDIUM_ASCII_STRING_TYPE = kMediumStringTag | kAsciiStringTag | kSeqStringTag,
LONG_ASCII_STRING_TYPE = kLongStringTag | kAsciiStringTag | kSeqStringTag,
SHORT_CONS_STRING_TYPE = kShortStringTag | kConsStringTag,
MEDIUM_CONS_STRING_TYPE = kMediumStringTag | kConsStringTag,
LONG_CONS_STRING_TYPE = kLongStringTag | kConsStringTag,
SHORT_CONS_ASCII_STRING_TYPE =
kShortStringTag | kAsciiStringTag | kConsStringTag,
MEDIUM_CONS_ASCII_STRING_TYPE =
kMediumStringTag | kAsciiStringTag | kConsStringTag,
LONG_CONS_ASCII_STRING_TYPE =
kLongStringTag | kAsciiStringTag | kConsStringTag,
SHORT_EXTERNAL_STRING_TYPE = kShortStringTag | kExternalStringTag,
MEDIUM_EXTERNAL_STRING_TYPE = kMediumStringTag | kExternalStringTag,
LONG_EXTERNAL_STRING_TYPE = kLongStringTag | kExternalStringTag,
SHORT_EXTERNAL_ASCII_STRING_TYPE =
kShortStringTag | kAsciiStringTag | kExternalStringTag,
MEDIUM_EXTERNAL_ASCII_STRING_TYPE =
kMediumStringTag | kAsciiStringTag | kExternalStringTag,
LONG_EXTERNAL_ASCII_STRING_TYPE =
kLongStringTag | kAsciiStringTag | kExternalStringTag,
LONG_PRIVATE_EXTERNAL_ASCII_STRING_TYPE = LONG_EXTERNAL_ASCII_STRING_TYPE,
MAP_TYPE = kNotStringTag,
HEAP_NUMBER_TYPE,
FIXED_ARRAY_TYPE,
CODE_TYPE,
ODDBALL_TYPE,
JS_GLOBAL_PROPERTY_CELL_TYPE,
PROXY_TYPE,
BYTE_ARRAY_TYPE,
PIXEL_ARRAY_TYPE,
EXTERNAL_BYTE_ARRAY_TYPE,
EXTERNAL_UNSIGNED_BYTE_ARRAY_TYPE,
EXTERNAL_SHORT_ARRAY_TYPE,
EXTERNAL_UNSIGNED_SHORT_ARRAY_TYPE,
EXTERNAL_INT_ARRAY_TYPE,
EXTERNAL_UNSIGNED_INT_ARRAY_TYPE,
EXTERNAL_FLOAT_ARRAY_TYPE,
FILLER_TYPE,
SMI_TYPE,
ACCESSOR_INFO_TYPE,
ACCESS_CHECK_INFO_TYPE,
INTERCEPTOR_INFO_TYPE,
SHARED_FUNCTION_INFO_TYPE,
CALL_HANDLER_INFO_TYPE,
FUNCTION_TEMPLATE_INFO_TYPE,
OBJECT_TEMPLATE_INFO_TYPE,
SIGNATURE_INFO_TYPE,
TYPE_SWITCH_INFO_TYPE,
#ifdef ENABLE_DEBUGGER_SUPPORT
DEBUG_INFO_TYPE,
BREAK_POINT_INFO_TYPE,
#endif
SCRIPT_TYPE,
JS_VALUE_TYPE,
JS_OBJECT_TYPE,
JS_CONTEXT_EXTENSION_OBJECT_TYPE,
JS_GLOBAL_OBJECT_TYPE,
JS_BUILTINS_OBJECT_TYPE,
Split window support from V8. Here is a description of the background and design of split window in Chrome and V8: https://docs.google.com/a/google.com/Doc?id=chhjkpg_47fwddxbfr This change list splits the window object into two parts: 1) an inner window object used as the global object of contexts; 2) an outer window object exposed to JavaScript and accessible by the name 'window'. Firefox did it awhile ago, here are some discussions: https://wiki.mozilla.org/Gecko:SplitWindow. One additional benefit of splitting window in Chrome is that accessing global variables don't need security checks anymore, it can improve applications that use many global variables. V8 support of split window: There are a small number of changes on V8 api to support split window: Security context is removed from V8, so does related API functions; A global object can be detached from its context and reused by a new context; Access checks on an object template can be turned on/off by default; An object can turn on its access checks later; V8 has a new object type, ApiGlobalObject, which is the outer window object type. The existing JSGlobalObject becomes the inner window object type. Security checks are moved from JSGlobalObject to ApiGlobalObject. ApiGlobalObject is the one exposed to JavaScript, it is accessible through Context::Global(). ApiGlobalObject's prototype is set to JSGlobalObject so that property lookups are forwarded to JSGlobalObject. ApiGlobalObject forwards all other property access requests to JSGlobalObject, such as SetProperty, DeleteProperty, etc. Security token is moved to a global context, and ApiGlobalObject has a reference to its global context. JSGlobalObject has a reference to its global context as well. When accessing properties on a global object in JavaScript, the domain security check is performed by comparing the security token of the lexical context (Top::global_context()) to the token of global object's context. The check is only needed when the receiver is a window object, such as 'window.document'. Accessing global variables, such as 'var foo = 3; foo' does not need checks because the receiver is the inner window object. When an outer window is detached from its global context (when a frame navigates away from a page), it is completely detached from the inner window. A new context is created for the new page, and the outer global object is reused. At this point, the access check on the DOMWindow wrapper of the old context is turned on. The code in old context is still able to access DOMWindow properties, but it has to go through domain security checks. It is debatable on how to implement the outer window object. Currently each property access function has to check if the receiver is ApiGlobalObject type. This approach might be error-prone that one may forget to check the receiver when adding new functions. It is unlikely a performance issue because accessing global variables are more common than 'window.foo' style coding. I am still working on the ARM port, and I'd like to hear comments and suggestions on the best way to support it in V8. Review URL: http://codereview.chromium.org/7366 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@540 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-10-21 19:07:58 +00:00
JS_GLOBAL_PROXY_TYPE,
JS_ARRAY_TYPE,
JS_REGEXP_TYPE,
JS_FUNCTION_TYPE,
// Pseudo-types
FIRST_NONSTRING_TYPE = MAP_TYPE,
FIRST_TYPE = 0x0,
INVALID_TYPE = FIRST_TYPE - 1,
LAST_TYPE = JS_FUNCTION_TYPE,
// Boundaries for testing the type is a JavaScript "object". Note that
// function objects are not counted as objects, even though they are
// implemented as such; only values whose typeof is "object" are included.
FIRST_JS_OBJECT_TYPE = JS_VALUE_TYPE,
LAST_JS_OBJECT_TYPE = JS_REGEXP_TYPE
};
enum CompareResult {
LESS = -1,
EQUAL = 0,
GREATER = 1,
NOT_EQUAL = GREATER
};
#define DECL_BOOLEAN_ACCESSORS(name) \
inline bool name(); \
inline void set_##name(bool value); \
#define DECL_ACCESSORS(name, type) \
inline type* name(); \
inline void set_##name(type* value, \
WriteBarrierMode mode = UPDATE_WRITE_BARRIER); \
class StringStream;
class ObjectVisitor;
struct ValueInfo : public Malloced {
ValueInfo() : type(FIRST_TYPE), ptr(NULL), str(NULL), number(0) { }
InstanceType type;
Object* ptr;
const char* str;
double number;
};
// A template-ized version of the IsXXX functions.
template <class C> static inline bool Is(Object* obj);
// Object is the abstract superclass for all classes in the
// object hierarchy.
// Object does not use any virtual functions to avoid the
// allocation of the C++ vtable.
// Since Smi and Failure are subclasses of Object no
// data members can be present in Object.
class Object BASE_EMBEDDED {
public:
// Type testing.
inline bool IsSmi();
inline bool IsHeapObject();
inline bool IsHeapNumber();
inline bool IsString();
inline bool IsSymbol();
// See objects-inl.h for more details
inline bool IsSeqString();
inline bool IsExternalString();
inline bool IsExternalTwoByteString();
inline bool IsExternalAsciiString();
inline bool IsSeqTwoByteString();
inline bool IsSeqAsciiString();
inline bool IsConsString();
inline bool IsNumber();
inline bool IsByteArray();
inline bool IsPixelArray();
inline bool IsExternalArray();
inline bool IsExternalByteArray();
inline bool IsExternalUnsignedByteArray();
inline bool IsExternalShortArray();
inline bool IsExternalUnsignedShortArray();
inline bool IsExternalIntArray();
inline bool IsExternalUnsignedIntArray();
inline bool IsExternalFloatArray();
inline bool IsFailure();
inline bool IsRetryAfterGC();
inline bool IsOutOfMemoryFailure();
inline bool IsException();
inline bool IsJSObject();
inline bool IsJSContextExtensionObject();
inline bool IsMap();
inline bool IsFixedArray();
inline bool IsDescriptorArray();
inline bool IsContext();
inline bool IsCatchContext();
inline bool IsGlobalContext();
inline bool IsJSFunction();
inline bool IsCode();
inline bool IsOddball();
inline bool IsSharedFunctionInfo();
inline bool IsJSValue();
inline bool IsStringWrapper();
inline bool IsProxy();
inline bool IsBoolean();
inline bool IsJSArray();
inline bool IsJSRegExp();
inline bool IsHashTable();
inline bool IsDictionary();
inline bool IsSymbolTable();
inline bool IsCompilationCacheTable();
inline bool IsMapCache();
inline bool IsPrimitive();
inline bool IsGlobalObject();
inline bool IsJSGlobalObject();
inline bool IsJSBuiltinsObject();
Split window support from V8. Here is a description of the background and design of split window in Chrome and V8: https://docs.google.com/a/google.com/Doc?id=chhjkpg_47fwddxbfr This change list splits the window object into two parts: 1) an inner window object used as the global object of contexts; 2) an outer window object exposed to JavaScript and accessible by the name 'window'. Firefox did it awhile ago, here are some discussions: https://wiki.mozilla.org/Gecko:SplitWindow. One additional benefit of splitting window in Chrome is that accessing global variables don't need security checks anymore, it can improve applications that use many global variables. V8 support of split window: There are a small number of changes on V8 api to support split window: Security context is removed from V8, so does related API functions; A global object can be detached from its context and reused by a new context; Access checks on an object template can be turned on/off by default; An object can turn on its access checks later; V8 has a new object type, ApiGlobalObject, which is the outer window object type. The existing JSGlobalObject becomes the inner window object type. Security checks are moved from JSGlobalObject to ApiGlobalObject. ApiGlobalObject is the one exposed to JavaScript, it is accessible through Context::Global(). ApiGlobalObject's prototype is set to JSGlobalObject so that property lookups are forwarded to JSGlobalObject. ApiGlobalObject forwards all other property access requests to JSGlobalObject, such as SetProperty, DeleteProperty, etc. Security token is moved to a global context, and ApiGlobalObject has a reference to its global context. JSGlobalObject has a reference to its global context as well. When accessing properties on a global object in JavaScript, the domain security check is performed by comparing the security token of the lexical context (Top::global_context()) to the token of global object's context. The check is only needed when the receiver is a window object, such as 'window.document'. Accessing global variables, such as 'var foo = 3; foo' does not need checks because the receiver is the inner window object. When an outer window is detached from its global context (when a frame navigates away from a page), it is completely detached from the inner window. A new context is created for the new page, and the outer global object is reused. At this point, the access check on the DOMWindow wrapper of the old context is turned on. The code in old context is still able to access DOMWindow properties, but it has to go through domain security checks. It is debatable on how to implement the outer window object. Currently each property access function has to check if the receiver is ApiGlobalObject type. This approach might be error-prone that one may forget to check the receiver when adding new functions. It is unlikely a performance issue because accessing global variables are more common than 'window.foo' style coding. I am still working on the ARM port, and I'd like to hear comments and suggestions on the best way to support it in V8. Review URL: http://codereview.chromium.org/7366 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@540 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-10-21 19:07:58 +00:00
inline bool IsJSGlobalProxy();
inline bool IsUndetectableObject();
inline bool IsAccessCheckNeeded();
inline bool IsJSGlobalPropertyCell();
// Returns true if this object is an instance of the specified
// function template.
inline bool IsInstanceOf(FunctionTemplateInfo* type);
inline bool IsStruct();
#define DECLARE_STRUCT_PREDICATE(NAME, Name, name) inline bool Is##Name();
STRUCT_LIST(DECLARE_STRUCT_PREDICATE)
#undef DECLARE_STRUCT_PREDICATE
// Oddball testing.
INLINE(bool IsUndefined());
INLINE(bool IsTheHole());
INLINE(bool IsNull());
INLINE(bool IsTrue());
INLINE(bool IsFalse());
// Extract the number.
inline double Number();
inline bool HasSpecificClassOf(String* name);
Object* ToObject(); // ECMA-262 9.9.
Object* ToBoolean(); // ECMA-262 9.2.
// Convert to a JSObject if needed.
// global_context is used when creating wrapper object.
Object* ToObject(Context* global_context);
// Converts this to a Smi if possible.
// Failure is returned otherwise.
inline Object* ToSmi();
void Lookup(String* name, LookupResult* result);
// Property access.
inline Object* GetProperty(String* key);
inline Object* GetProperty(String* key, PropertyAttributes* attributes);
Object* GetPropertyWithReceiver(Object* receiver,
String* key,
PropertyAttributes* attributes);
Object* GetProperty(Object* receiver,
LookupResult* result,
String* key,
PropertyAttributes* attributes);
Object* GetPropertyWithCallback(Object* receiver,
Object* structure,
String* name,
Object* holder);
Object* GetPropertyWithDefinedGetter(Object* receiver,
JSFunction* getter);
inline Object* GetElement(uint32_t index);
Object* GetElementWithReceiver(Object* receiver, uint32_t index);
// Return the object's prototype (might be Heap::null_value()).
Object* GetPrototype();
// Returns true if this is a JSValue containing a string and the index is
// < the length of the string. Used to implement [] on strings.
inline bool IsStringObjectWithCharacterAt(uint32_t index);
#ifdef DEBUG
// Prints this object with details.
void Print();
void PrintLn();
// Verifies the object.
void Verify();
// Verify a pointer is a valid object pointer.
static void VerifyPointer(Object* p);
#endif
// Prints this object without details.
void ShortPrint();
// Prints this object without details to a message accumulator.
void ShortPrint(StringStream* accumulator);
// Casting: This cast is only needed to satisfy macros in objects-inl.h.
static Object* cast(Object* value) { return value; }
// Layout description.
static const int kHeaderSize = 0; // Object does not take up any space.
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(Object);
};
// Smi represents integer Numbers that can be stored in 31 bits.
// Smis are immediate which means they are NOT allocated in the heap.
// The this pointer has the following format: [31 bit signed int] 0
// For long smis it has the following format:
// [32 bit signed int] [31 bits zero padding] 0
// Smi stands for small integer.
class Smi: public Object {
public:
// Returns the integer value.
inline int value();
// Convert a value to a Smi object.
static inline Smi* FromInt(int value);
static inline Smi* FromIntptr(intptr_t value);
// Returns whether value can be represented in a Smi.
static inline bool IsValid(intptr_t value);
// Casting.
static inline Smi* cast(Object* object);
// Dispatched behavior.
void SmiPrint();
void SmiPrint(StringStream* accumulator);
#ifdef DEBUG
void SmiVerify();
#endif
static const int kMinValue = (-1 << (kSmiValueSize - 1));
static const int kMaxValue = -(kMinValue + 1);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(Smi);
};
// Failure is used for reporting out of memory situations and
// propagating exceptions through the runtime system. Failure objects
// are transient and cannot occur as part of the object graph.
//
// Failures are a single word, encoded as follows:
// +-------------------------+---+--+--+
// |...rrrrrrrrrrrrrrrrrrrrrr|sss|tt|11|
// +-------------------------+---+--+--+
// 7 6 4 32 10
//
//
// The low two bits, 0-1, are the failure tag, 11. The next two bits,
// 2-3, are a failure type tag 'tt' with possible values:
// 00 RETRY_AFTER_GC
// 01 EXCEPTION
// 10 INTERNAL_ERROR
// 11 OUT_OF_MEMORY_EXCEPTION
//
// The next three bits, 4-6, are an allocation space tag 'sss'. The
// allocation space tag is 000 for all failure types except
// RETRY_AFTER_GC. For RETRY_AFTER_GC, the possible values are the
// allocation spaces (the encoding is found in globals.h).
//
// The remaining bits is the size of the allocation request in units
// of the pointer size, and is zeroed except for RETRY_AFTER_GC
// failures. The 25 bits (on a 32 bit platform) gives a representable
// range of 2^27 bytes (128MB).
// Failure type tag info.
const int kFailureTypeTagSize = 2;
const int kFailureTypeTagMask = (1 << kFailureTypeTagSize) - 1;
class Failure: public Object {
public:
// RuntimeStubs assumes EXCEPTION = 1 in the compiler-generated code.
enum Type {
RETRY_AFTER_GC = 0,
EXCEPTION = 1, // Returning this marker tells the real exception
// is in Top::pending_exception.
INTERNAL_ERROR = 2,
OUT_OF_MEMORY_EXCEPTION = 3
};
inline Type type() const;
// Returns the space that needs to be collected for RetryAfterGC failures.
inline AllocationSpace allocation_space() const;
// Returns the number of bytes requested (up to the representable maximum)
// for RetryAfterGC failures.
inline int requested() const;
inline bool IsInternalError() const;
inline bool IsOutOfMemoryException() const;
static Failure* RetryAfterGC(int requested_bytes, AllocationSpace space);
static inline Failure* RetryAfterGC(int requested_bytes); // NEW_SPACE
static inline Failure* Exception();
static inline Failure* InternalError();
static inline Failure* OutOfMemoryException();
// Casting.
static inline Failure* cast(Object* object);
// Dispatched behavior.
void FailurePrint();
void FailurePrint(StringStream* accumulator);
#ifdef DEBUG
void FailureVerify();
#endif
private:
inline intptr_t value() const;
static inline Failure* Construct(Type type, intptr_t value = 0);
DISALLOW_IMPLICIT_CONSTRUCTORS(Failure);
};
// Heap objects typically have a map pointer in their first word. However,
// during GC other data (eg, mark bits, forwarding addresses) is sometimes
// encoded in the first word. The class MapWord is an abstraction of the
// value in a heap object's first word.
class MapWord BASE_EMBEDDED {
public:
// Normal state: the map word contains a map pointer.
// Create a map word from a map pointer.
static inline MapWord FromMap(Map* map);
// View this map word as a map pointer.
inline Map* ToMap();
// Scavenge collection: the map word of live objects in the from space
// contains a forwarding address (a heap object pointer in the to space).
// True if this map word is a forwarding address for a scavenge
// collection. Only valid during a scavenge collection (specifically,
// when all map words are heap object pointers, ie. not during a full GC).
inline bool IsForwardingAddress();
// Create a map word from a forwarding address.
static inline MapWord FromForwardingAddress(HeapObject* object);
// View this map word as a forwarding address.
inline HeapObject* ToForwardingAddress();
// True if this map word is a serialization address. This will only be the
// case during a destructive serialization of the heap.
inline bool IsSerializationAddress();
// Create a map word from a serialization address.
static inline MapWord FromSerializationAddress(int raw);
// View this map word as a serialization address.
inline int ToSerializationAddress();
// Marking phase of full collection: the map word of live objects is
// marked, and may be marked as overflowed (eg, the object is live, its
// children have not been visited, and it does not fit in the marking
// stack).
// True if this map word's mark bit is set.
inline bool IsMarked();
// Return this map word but with its mark bit set.
inline void SetMark();
// Return this map word but with its mark bit cleared.
inline void ClearMark();
// True if this map word's overflow bit is set.
inline bool IsOverflowed();
// Return this map word but with its overflow bit set.
inline void SetOverflow();
// Return this map word but with its overflow bit cleared.
inline void ClearOverflow();
// Compacting phase of a full compacting collection: the map word of live
// objects contains an encoding of the original map address along with the
// forwarding address (represented as an offset from the first live object
// in the same page as the (old) object address).
// Create a map word from a map address and a forwarding address offset.
static inline MapWord EncodeAddress(Address map_address, int offset);
// Return the map address encoded in this map word.
inline Address DecodeMapAddress(MapSpace* map_space);
// Return the forwarding offset encoded in this map word.
inline int DecodeOffset();
// During serialization: the map word is used to hold an encoded
// address, and possibly a mark bit (set and cleared with SetMark
// and ClearMark).
// Create a map word from an encoded address.
static inline MapWord FromEncodedAddress(Address address);
inline Address ToEncodedAddress();
// Bits used by the marking phase of the garbage collector.
//
// The first word of a heap object is normally a map pointer. The last two
// bits are tagged as '01' (kHeapObjectTag). We reuse the last two bits to
// mark an object as live and/or overflowed:
// last bit = 0, marked as alive
// second bit = 1, overflowed
// An object is only marked as overflowed when it is marked as live while
// the marking stack is overflowed.
static const int kMarkingBit = 0; // marking bit
static const int kMarkingMask = (1 << kMarkingBit); // marking mask
static const int kOverflowBit = 1; // overflow bit
static const int kOverflowMask = (1 << kOverflowBit); // overflow mask
// Forwarding pointers and map pointer encoding
// 31 21 20 10 9 0
// +-----------------+------------------+-----------------+
// |forwarding offset|page offset of map|page index of map|
// +-----------------+------------------+-----------------+
// 11 bits 11 bits 10 bits
static const int kMapPageIndexBits = 10;
static const int kMapPageOffsetBits = 11;
static const int kForwardingOffsetBits = 11;
static const int kMapPageIndexShift = 0;
static const int kMapPageOffsetShift =
kMapPageIndexShift + kMapPageIndexBits;
static const int kForwardingOffsetShift =
kMapPageOffsetShift + kMapPageOffsetBits;
// 0x000003FF
static const uint32_t kMapPageIndexMask =
(1 << kMapPageOffsetShift) - 1;
// 0x001FFC00
static const uint32_t kMapPageOffsetMask =
((1 << kForwardingOffsetShift) - 1) & ~kMapPageIndexMask;
// 0xFFE00000
static const uint32_t kForwardingOffsetMask =
~(kMapPageIndexMask | kMapPageOffsetMask);
private:
// HeapObject calls the private constructor and directly reads the value.
friend class HeapObject;
explicit MapWord(uintptr_t value) : value_(value) {}
uintptr_t value_;
};
// HeapObject is the superclass for all classes describing heap allocated
// objects.
class HeapObject: public Object {
public:
// [map]: Contains a map which contains the object's reflective
// information.
inline Map* map();
inline void set_map(Map* value);
// During garbage collection, the map word of a heap object does not
// necessarily contain a map pointer.
inline MapWord map_word();
inline void set_map_word(MapWord map_word);
// Converts an address to a HeapObject pointer.
static inline HeapObject* FromAddress(Address address);
// Returns the address of this HeapObject.
inline Address address();
// Iterates over pointers contained in the object (including the Map)
void Iterate(ObjectVisitor* v);
// Iterates over all pointers contained in the object except the
// first map pointer. The object type is given in the first
// parameter. This function does not access the map pointer in the
// object, and so is safe to call while the map pointer is modified.
void IterateBody(InstanceType type, int object_size, ObjectVisitor* v);
// This method only applies to struct objects. Iterates over all the fields
// of this struct.
void IterateStructBody(int object_size, ObjectVisitor* v);
// Returns the heap object's size in bytes
inline int Size();
// Given a heap object's map pointer, returns the heap size in bytes
// Useful when the map pointer field is used for other purposes.
// GC internal.
inline int SizeFromMap(Map* map);
// Support for the marking heap objects during the marking phase of GC.
// True if the object is marked live.
inline bool IsMarked();
// Mutate this object's map pointer to indicate that the object is live.
inline void SetMark();
// Mutate this object's map pointer to remove the indication that the
// object is live (ie, partially restore the map pointer).
inline void ClearMark();
// True if this object is marked as overflowed. Overflowed objects have
// been reached and marked during marking of the heap, but their children
// have not necessarily been marked and they have not been pushed on the
// marking stack.
inline bool IsOverflowed();
// Mutate this object's map pointer to indicate that the object is
// overflowed.
inline void SetOverflow();
// Mutate this object's map pointer to remove the indication that the
// object is overflowed (ie, partially restore the map pointer).
inline void ClearOverflow();
// Returns the field at offset in obj, as a read/write Object* reference.
// Does no checking, and is safe to use during GC, while maps are invalid.
// Does not update remembered sets, so should only be assigned to
// during marking GC.
static inline Object** RawField(HeapObject* obj, int offset);
// Casting.
static inline HeapObject* cast(Object* obj);
// Return the write barrier mode for this.
inline WriteBarrierMode GetWriteBarrierMode();
// Dispatched behavior.
void HeapObjectShortPrint(StringStream* accumulator);
#ifdef DEBUG
void HeapObjectPrint();
void HeapObjectVerify();
inline void VerifyObjectField(int offset);
void PrintHeader(const char* id);
// Verify a pointer is a valid HeapObject pointer that points to object
// areas in the heap.
static void VerifyHeapPointer(Object* p);
#endif
// Layout description.
// First field in a heap object is map.
static const int kMapOffset = Object::kHeaderSize;
static const int kHeaderSize = kMapOffset + kPointerSize;
STATIC_CHECK(kMapOffset == Internals::kHeapObjectMapOffset);
protected:
// helpers for calling an ObjectVisitor to iterate over pointers in the
// half-open range [start, end) specified as integer offsets
inline void IteratePointers(ObjectVisitor* v, int start, int end);
// as above, for the single element at "offset"
inline void IteratePointer(ObjectVisitor* v, int offset);
// Computes the object size from the map.
// Should only be used from SizeFromMap.
int SlowSizeFromMap(Map* map);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(HeapObject);
};
// The HeapNumber class describes heap allocated numbers that cannot be
// represented in a Smi (small integer)
class HeapNumber: public HeapObject {
public:
// [value]: number value.
inline double value();
inline void set_value(double value);
// Casting.
static inline HeapNumber* cast(Object* obj);
// Dispatched behavior.
Object* HeapNumberToBoolean();
void HeapNumberPrint();
void HeapNumberPrint(StringStream* accumulator);
#ifdef DEBUG
void HeapNumberVerify();
#endif
// Layout description.
static const int kValueOffset = HeapObject::kHeaderSize;
// IEEE doubles are two 32 bit words. The first is just mantissa, the second
// is a mixture of sign, exponent and mantissa. Our current platforms are all
// little endian apart from non-EABI arm which is little endian with big
// endian floating point word ordering!
#if !defined(V8_HOST_ARCH_ARM) || defined(USE_ARM_EABI)
static const int kMantissaOffset = kValueOffset;
static const int kExponentOffset = kValueOffset + 4;
#else
static const int kMantissaOffset = kValueOffset + 4;
static const int kExponentOffset = kValueOffset;
# define BIG_ENDIAN_FLOATING_POINT 1
#endif
static const int kSize = kValueOffset + kDoubleSize;
static const uint32_t kSignMask = 0x80000000u;
static const uint32_t kExponentMask = 0x7ff00000u;
static const uint32_t kMantissaMask = 0xfffffu;
static const int kExponentBias = 1023;
static const int kExponentShift = 20;
static const int kMantissaBitsInTopWord = 20;
static const int kNonMantissaBitsInTopWord = 12;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(HeapNumber);
};
// The JSObject describes real heap allocated JavaScript objects with
// properties.
// Note that the map of JSObject changes during execution to enable inline
// caching.
class JSObject: public HeapObject {
public:
enum DeleteMode { NORMAL_DELETION, FORCE_DELETION };
enum ElementsKind {
FAST_ELEMENTS,
DICTIONARY_ELEMENTS,
PIXEL_ELEMENTS,
EXTERNAL_BYTE_ELEMENTS,
EXTERNAL_UNSIGNED_BYTE_ELEMENTS,
EXTERNAL_SHORT_ELEMENTS,
EXTERNAL_UNSIGNED_SHORT_ELEMENTS,
EXTERNAL_INT_ELEMENTS,
EXTERNAL_UNSIGNED_INT_ELEMENTS,
EXTERNAL_FLOAT_ELEMENTS
};
// [properties]: Backing storage for properties.
// properties is a FixedArray in the fast case, and a Dictionary in the
// slow case.
DECL_ACCESSORS(properties, FixedArray) // Get and set fast properties.
inline void initialize_properties();
inline bool HasFastProperties();
inline StringDictionary* property_dictionary(); // Gets slow properties.
// [elements]: The elements (properties with names that are integers).
// elements is a FixedArray in the fast case, and a Dictionary in the slow
// case or a PixelArray in a special case.
DECL_ACCESSORS(elements, Array) // Get and set fast elements.
inline void initialize_elements();
inline ElementsKind GetElementsKind();
inline bool HasFastElements();
inline bool HasDictionaryElements();
inline bool HasPixelElements();
inline bool HasExternalArrayElements();
inline bool HasExternalByteElements();
inline bool HasExternalUnsignedByteElements();
inline bool HasExternalShortElements();
inline bool HasExternalUnsignedShortElements();
inline bool HasExternalIntElements();
inline bool HasExternalUnsignedIntElements();
inline bool HasExternalFloatElements();
inline NumberDictionary* element_dictionary(); // Gets slow elements.
// Collects elements starting at index 0.
// Undefined values are placed after non-undefined values.
// Returns the number of non-undefined values.
Object* PrepareElementsForSort(uint32_t limit);
// As PrepareElementsForSort, but only on objects where elements is
// a dictionary, and it will stay a dictionary.
Object* PrepareSlowElementsForSort(uint32_t limit);
Object* SetProperty(String* key,
Object* value,
PropertyAttributes attributes);
Object* SetProperty(LookupResult* result,
String* key,
Object* value,
PropertyAttributes attributes);
Object* SetPropertyWithFailedAccessCheck(LookupResult* result,
String* name,
Object* value);
Object* SetPropertyWithCallback(Object* structure,
String* name,
Object* value,
JSObject* holder);
Object* SetPropertyWithDefinedSetter(JSFunction* setter,
Object* value);
Object* SetPropertyWithInterceptor(String* name,
Object* value,
PropertyAttributes attributes);
Object* SetPropertyPostInterceptor(String* name,
Object* value,
PropertyAttributes attributes);
Object* IgnoreAttributesAndSetLocalProperty(String* key,
Object* value,
PropertyAttributes attributes);
// Retrieve a value in a normalized object given a lookup result.
// Handles the special representation of JS global objects.
Object* GetNormalizedProperty(LookupResult* result);
// Sets the property value in a normalized object given a lookup result.
// Handles the special representation of JS global objects.
Object* SetNormalizedProperty(LookupResult* result, Object* value);
// Sets the property value in a normalized object given (key, value, details).
// Handles the special representation of JS global objects.
Object* SetNormalizedProperty(String* name,
Object* value,
PropertyDetails details);
// Deletes the named property in a normalized object.
Object* DeleteNormalizedProperty(String* name, DeleteMode mode);
// Sets a property that currently has lazy loading.
Object* SetLazyProperty(LookupResult* result,
String* name,
Object* value,
PropertyAttributes attributes);
// Returns the class name ([[Class]] property in the specification).
String* class_name();
// Returns the constructor name (the name (possibly, inferred name) of the
// function that was used to instantiate the object).
String* constructor_name();
// Retrieve interceptors.
InterceptorInfo* GetNamedInterceptor();
InterceptorInfo* GetIndexedInterceptor();
inline PropertyAttributes GetPropertyAttribute(String* name);
PropertyAttributes GetPropertyAttributeWithReceiver(JSObject* receiver,
String* name);
PropertyAttributes GetLocalPropertyAttribute(String* name);
Object* DefineAccessor(String* name, bool is_getter, JSFunction* fun,
PropertyAttributes attributes);
Object* LookupAccessor(String* name, bool is_getter);
// Used from Object::GetProperty().
Object* GetPropertyWithFailedAccessCheck(Object* receiver,
LookupResult* result,
String* name,
PropertyAttributes* attributes);
Object* GetPropertyWithInterceptor(JSObject* receiver,
String* name,
PropertyAttributes* attributes);
Object* GetPropertyPostInterceptor(JSObject* receiver,
String* name,
PropertyAttributes* attributes);
Object* GetLazyProperty(Object* receiver,
LookupResult* result,
String* name,
PropertyAttributes* attributes);
// Tells whether this object needs to be loaded.
inline bool IsLoaded();
// Returns true if this is an instance of an api function and has
// been modified since it was created. May give false positives.
bool IsDirty();
bool HasProperty(String* name) {
return GetPropertyAttribute(name) != ABSENT;
}
// Can cause a GC if it hits an interceptor.
bool HasLocalProperty(String* name) {
return GetLocalPropertyAttribute(name) != ABSENT;
}
Object* DeleteProperty(String* name, DeleteMode mode);
Object* DeleteElement(uint32_t index, DeleteMode mode);
Object* DeleteLazyProperty(LookupResult* result,
String* name,
DeleteMode mode);
// Tests for the fast common case for property enumeration.
bool IsSimpleEnum();
// Do we want to keep the elements in fast case when increasing the
// capacity?
bool ShouldConvertToSlowElements(int new_capacity);
// Returns true if the backing storage for the slow-case elements of
// this object takes up nearly as much space as a fast-case backing
// storage would. In that case the JSObject should have fast
// elements.
bool ShouldConvertToFastElements();
// Return the object's prototype (might be Heap::null_value()).
inline Object* GetPrototype();
// Tells whether the index'th element is present.
inline bool HasElement(uint32_t index);
bool HasElementWithReceiver(JSObject* receiver, uint32_t index);
bool HasLocalElement(uint32_t index);
bool HasElementWithInterceptor(JSObject* receiver, uint32_t index);
bool HasElementPostInterceptor(JSObject* receiver, uint32_t index);
Object* SetFastElement(uint32_t index, Object* value);
// Set the index'th array element.
// A Failure object is returned if GC is needed.
Object* SetElement(uint32_t index, Object* value);
// Returns the index'th element.
// The undefined object if index is out of bounds.
Object* GetElementWithReceiver(JSObject* receiver, uint32_t index);
void SetFastElements(FixedArray* elements);
Object* SetSlowElements(Object* length);
// Lookup interceptors are used for handling properties controlled by host
// objects.
inline bool HasNamedInterceptor();
inline bool HasIndexedInterceptor();
// Support functions for v8 api (needed for correct interceptor behavior).
bool HasRealNamedProperty(String* key);
bool HasRealElementProperty(uint32_t index);
bool HasRealNamedCallbackProperty(String* key);
// Initializes the array to a certain length
Object* SetElementsLength(Object* length);
// Get the header size for a JSObject. Used to compute the index of
// internal fields as well as the number of internal fields.
inline int GetHeaderSize();
inline int GetInternalFieldCount();
inline Object* GetInternalField(int index);
inline void SetInternalField(int index, Object* value);
// Lookup a property. If found, the result is valid and has
// detailed information.
void LocalLookup(String* name, LookupResult* result);
void Lookup(String* name, LookupResult* result);
// The following lookup functions skip interceptors.
void LocalLookupRealNamedProperty(String* name, LookupResult* result);
void LookupRealNamedProperty(String* name, LookupResult* result);
void LookupRealNamedPropertyInPrototypes(String* name, LookupResult* result);
void LookupCallbackSetterInPrototypes(String* name, LookupResult* result);
Object* LookupCallbackSetterInPrototypes(uint32_t index);
void LookupCallback(String* name, LookupResult* result);
// Returns the number of properties on this object filtering out properties
// with the specified attributes (ignoring interceptors).
int NumberOfLocalProperties(PropertyAttributes filter);
// Returns the number of enumerable properties (ignoring interceptors).
int NumberOfEnumProperties();
// Fill in details for properties into storage starting at the specified
// index.
void GetLocalPropertyNames(FixedArray* storage, int index);
// Returns the number of properties on this object filtering out properties
// with the specified attributes (ignoring interceptors).
int NumberOfLocalElements(PropertyAttributes filter);
// Returns the number of enumerable elements (ignoring interceptors).
int NumberOfEnumElements();
// Returns the number of elements on this object filtering out elements
// with the specified attributes (ignoring interceptors).
int GetLocalElementKeys(FixedArray* storage, PropertyAttributes filter);
// Count and fill in the enumerable elements into storage.
// (storage->length() == NumberOfEnumElements()).
// If storage is NULL, will count the elements without adding
// them to any storage.
// Returns the number of enumerable elements.
int GetEnumElementKeys(FixedArray* storage);
// Add a property to a fast-case object using a map transition to
// new_map.
Object* AddFastPropertyUsingMap(Map* new_map,
String* name,
Object* value);
// Add a constant function property to a fast-case object.
// This leaves a CONSTANT_TRANSITION in the old map, and
// if it is called on a second object with this map, a
// normal property is added instead, with a map transition.
// This avoids the creation of many maps with the same constant
// function, all orphaned.
Object* AddConstantFunctionProperty(String* name,
JSFunction* function,
PropertyAttributes attributes);
Object* ReplaceSlowProperty(String* name,
Object* value,
PropertyAttributes attributes);
// Converts a descriptor of any other type to a real field,
// backed by the properties array. Descriptors of visible
// types, such as CONSTANT_FUNCTION, keep their enumeration order.
// Converts the descriptor on the original object's map to a
// map transition, and the the new field is on the object's new map.
Object* ConvertDescriptorToFieldAndMapTransition(
String* name,
Object* new_value,
PropertyAttributes attributes);
// Converts a descriptor of any other type to a real field,
// backed by the properties array. Descriptors of visible
// types, such as CONSTANT_FUNCTION, keep their enumeration order.
Object* ConvertDescriptorToField(String* name,
Object* new_value,
PropertyAttributes attributes);
// Add a property to a fast-case object.
Object* AddFastProperty(String* name,
Object* value,
PropertyAttributes attributes);
// Add a property to a slow-case object.
Object* AddSlowProperty(String* name,
Object* value,
PropertyAttributes attributes);
// Add a property to an object.
Object* AddProperty(String* name,
Object* value,
PropertyAttributes attributes);
// Convert the object to use the canonical dictionary
// representation. If the object is expected to have additional properties
// added this number can be indicated to have the backing store allocated to
// an initial capacity for holding these properties.
Object* NormalizeProperties(PropertyNormalizationMode mode,
int expected_additional_properties);
Object* NormalizeElements();
// Transform slow named properties to fast variants.
// Returns failure if allocation failed.
Object* TransformToFastProperties(int unused_property_fields);
// Access fast-case object properties at index.
inline Object* FastPropertyAt(int index);
inline Object* FastPropertyAtPut(int index, Object* value);
// Access to in object properties.
inline Object* InObjectPropertyAt(int index);
inline Object* InObjectPropertyAtPut(int index,
Object* value,
WriteBarrierMode mode
= UPDATE_WRITE_BARRIER);
// initializes the body after properties slot, properties slot is
// initialized by set_properties
// Note: this call does not update write barrier, it is caller's
// reponsibility to ensure that *v* can be collected without WB here.
inline void InitializeBody(int object_size);
// Check whether this object references another object
bool ReferencesObject(Object* obj);
// Casting.
static inline JSObject* cast(Object* obj);
// Dispatched behavior.
void JSObjectIterateBody(int object_size, ObjectVisitor* v);
void JSObjectShortPrint(StringStream* accumulator);
#ifdef DEBUG
void JSObjectPrint();
void JSObjectVerify();
void PrintProperties();
void PrintElements();
// Structure for collecting spill information about JSObjects.
class SpillInformation {
public:
void Clear();
void Print();
int number_of_objects_;
int number_of_objects_with_fast_properties_;
int number_of_objects_with_fast_elements_;
int number_of_fast_used_fields_;
int number_of_fast_unused_fields_;
int number_of_slow_used_properties_;
int number_of_slow_unused_properties_;
int number_of_fast_used_elements_;
int number_of_fast_unused_elements_;
int number_of_slow_used_elements_;
int number_of_slow_unused_elements_;
};
void IncrementSpillStatistics(SpillInformation* info);
#endif
Object* SlowReverseLookup(Object* value);
static const uint32_t kMaxGap = 1024;
static const int kMaxFastElementsLength = 5000;
static const int kInitialMaxFastElementArray = 100000;
static const int kMaxFastProperties = 8;
static const int kMaxInstanceSize = 255 * kPointerSize;
// When extending the backing storage for property values, we increase
// its size by more than the 1 entry necessary, so sequentially adding fields
// to the same object requires fewer allocations and copies.
static const int kFieldsAdded = 3;
// Layout description.
static const int kPropertiesOffset = HeapObject::kHeaderSize;
static const int kElementsOffset = kPropertiesOffset + kPointerSize;
static const int kHeaderSize = kElementsOffset + kPointerSize;
STATIC_CHECK(kHeaderSize == Internals::kJSObjectHeaderSize);
Object* GetElementWithInterceptor(JSObject* receiver, uint32_t index);
private:
Object* SetElementWithInterceptor(uint32_t index, Object* value);
Object* SetElementWithoutInterceptor(uint32_t index, Object* value);
Object* GetElementPostInterceptor(JSObject* receiver, uint32_t index);
Object* DeletePropertyPostInterceptor(String* name, DeleteMode mode);
Object* DeletePropertyWithInterceptor(String* name);
Object* DeleteElementPostInterceptor(uint32_t index, DeleteMode mode);
Object* DeleteElementWithInterceptor(uint32_t index);
PropertyAttributes GetPropertyAttributePostInterceptor(JSObject* receiver,
String* name,
bool continue_search);
PropertyAttributes GetPropertyAttributeWithInterceptor(JSObject* receiver,
String* name,
bool continue_search);
PropertyAttributes GetPropertyAttributeWithFailedAccessCheck(
Object* receiver,
LookupResult* result,
String* name,
bool continue_search);
PropertyAttributes GetPropertyAttribute(JSObject* receiver,
LookupResult* result,
String* name,
bool continue_search);
// Returns true if most of the elements backing storage is used.
bool HasDenseElements();
Object* DefineGetterSetter(String* name, PropertyAttributes attributes);
void LookupInDescriptor(String* name, LookupResult* result);
DISALLOW_IMPLICIT_CONSTRUCTORS(JSObject);
};
// Abstract super class arrays. It provides length behavior.
class Array: public HeapObject {
public:
// [length]: length of the array.
inline int length();
inline void set_length(int value);
// Convert an object to an array index.
// Returns true if the conversion succeeded.
static inline bool IndexFromObject(Object* object, uint32_t* index);
// Layout descriptor.
static const int kLengthOffset = HeapObject::kHeaderSize;
protected:
// No code should use the Array class directly, only its subclasses.
// Use the kHeaderSize of the appropriate subclass, which may be aligned.
static const int kHeaderSize = kLengthOffset + kIntSize;
static const int kAlignedSize = POINTER_SIZE_ALIGN(kHeaderSize);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(Array);
};
// FixedArray describes fixed sized arrays where element
// type is Object*.
class FixedArray: public Array {
public:
// Setter and getter for elements.
inline Object* get(int index);
// Setter that uses write barrier.
inline void set(int index, Object* value);
// Setter that doesn't need write barrier).
inline void set(int index, Smi* value);
// Setter with explicit barrier mode.
inline void set(int index, Object* value, WriteBarrierMode mode);
// Setters for frequently used oddballs located in old space.
inline void set_undefined(int index);
inline void set_null(int index);
inline void set_the_hole(int index);
// Copy operations.
inline Object* Copy();
Object* CopySize(int new_length);
// Add the elements of a JSArray to this FixedArray.
Object* AddKeysFromJSArray(JSArray* array);
// Compute the union of this and other.
Object* UnionOfKeys(FixedArray* other);
// Copy a sub array from the receiver to dest.
void CopyTo(int pos, FixedArray* dest, int dest_pos, int len);
// Garbage collection support.
static int SizeFor(int length) { return kHeaderSize + length * kPointerSize; }
// Code Generation support.
static int OffsetOfElementAt(int index) { return SizeFor(index); }
// Casting.
static inline FixedArray* cast(Object* obj);
// Align data at kPointerSize, even if Array.kHeaderSize isn't aligned.
static const int kHeaderSize = POINTER_SIZE_ALIGN(Array::kHeaderSize);
// Dispatched behavior.
int FixedArraySize() { return SizeFor(length()); }
void FixedArrayIterateBody(ObjectVisitor* v);
#ifdef DEBUG
void FixedArrayPrint();
void FixedArrayVerify();
// Checks if two FixedArrays have identical contents.
bool IsEqualTo(FixedArray* other);
#endif
// Swap two elements in a pair of arrays. If this array and the
// numbers array are the same object, the elements are only swapped
// once.
void SwapPairs(FixedArray* numbers, int i, int j);
// Sort prefix of this array and the numbers array as pairs wrt. the
// numbers. If the numbers array and the this array are the same
// object, the prefix of this array is sorted.
void SortPairs(FixedArray* numbers, uint32_t len);
protected:
// Set operation on FixedArray without using write barriers.
static inline void fast_set(FixedArray* array, int index, Object* value);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(FixedArray);
};
// DescriptorArrays are fixed arrays used to hold instance descriptors.
// The format of the these objects is:
// [0]: point to a fixed array with (value, detail) pairs.
// [1]: next enumeration index (Smi), or pointer to small fixed array:
// [0]: next enumeration index (Smi)
// [1]: pointer to fixed array with enum cache
// [2]: first key
// [length() - 1]: last key
//
class DescriptorArray: public FixedArray {
public:
// Is this the singleton empty_descriptor_array?
inline bool IsEmpty();
// Returns the number of descriptors in the array.
int number_of_descriptors() {
return IsEmpty() ? 0 : length() - kFirstIndex;
}
int NextEnumerationIndex() {
if (IsEmpty()) return PropertyDetails::kInitialIndex;
Object* obj = get(kEnumerationIndexIndex);
if (obj->IsSmi()) {
return Smi::cast(obj)->value();
} else {
Object* index = FixedArray::cast(obj)->get(kEnumCacheBridgeEnumIndex);
return Smi::cast(index)->value();
}
}
// Set next enumeration index and flush any enum cache.
void SetNextEnumerationIndex(int value) {
if (!IsEmpty()) {
fast_set(this, kEnumerationIndexIndex, Smi::FromInt(value));
}
}
bool HasEnumCache() {
return !IsEmpty() && !get(kEnumerationIndexIndex)->IsSmi();
}
Object* GetEnumCache() {
ASSERT(HasEnumCache());
FixedArray* bridge = FixedArray::cast(get(kEnumerationIndexIndex));
return bridge->get(kEnumCacheBridgeCacheIndex);
}
// Initialize or change the enum cache,
// using the supplied storage for the small "bridge".
void SetEnumCache(FixedArray* bridge_storage, FixedArray* new_cache);
// Accessors for fetching instance descriptor at descriptor number.
inline String* GetKey(int descriptor_number);
inline Object* GetValue(int descriptor_number);
inline Smi* GetDetails(int descriptor_number);
inline PropertyType GetType(int descriptor_number);
inline int GetFieldIndex(int descriptor_number);
inline JSFunction* GetConstantFunction(int descriptor_number);
inline Object* GetCallbacksObject(int descriptor_number);
inline AccessorDescriptor* GetCallbacks(int descriptor_number);
inline bool IsProperty(int descriptor_number);
inline bool IsTransition(int descriptor_number);
inline bool IsNullDescriptor(int descriptor_number);
inline bool IsDontEnum(int descriptor_number);
// Accessor for complete descriptor.
inline void Get(int descriptor_number, Descriptor* desc);
inline void Set(int descriptor_number, Descriptor* desc);
// Transfer complete descriptor from another descriptor array to
// this one.
inline void CopyFrom(int index, DescriptorArray* src, int src_index);
// Copy the descriptor array, insert a new descriptor and optionally
// remove map transitions. If the descriptor is already present, it is
// replaced. If a replaced descriptor is a real property (not a transition
// or null), its enumeration index is kept as is.
// If adding a real property, map transitions must be removed. If adding
// a transition, they must not be removed. All null descriptors are removed.
Object* CopyInsert(Descriptor* descriptor, TransitionFlag transition_flag);
// Remove all transitions. Return a copy of the array with all transitions
// removed, or a Failure object if the new array could not be allocated.
Object* RemoveTransitions();
// Sort the instance descriptors by the hash codes of their keys.
void Sort();
// Search the instance descriptors for given name.
inline int Search(String* name);
// Tells whether the name is present int the array.
bool Contains(String* name) { return kNotFound != Search(name); }
// Perform a binary search in the instance descriptors represented
// by this fixed array. low and high are descriptor indices. If there
// are three instance descriptors in this array it should be called
// with low=0 and high=2.
int BinarySearch(String* name, int low, int high);
// Perform a linear search in the instance descriptors represented
// by this fixed array. len is the number of descriptor indices that are
// valid. Does not require the descriptors to be sorted.
int LinearSearch(String* name, int len);
// Allocates a DescriptorArray, but returns the singleton
// empty descriptor array object if number_of_descriptors is 0.
static Object* Allocate(int number_of_descriptors);
// Casting.
static inline DescriptorArray* cast(Object* obj);
// Constant for denoting key was not found.
static const int kNotFound = -1;
static const int kContentArrayIndex = 0;
static const int kEnumerationIndexIndex = 1;
static const int kFirstIndex = 2;
// The length of the "bridge" to the enum cache.
static const int kEnumCacheBridgeLength = 2;
static const int kEnumCacheBridgeEnumIndex = 0;
static const int kEnumCacheBridgeCacheIndex = 1;
// Layout description.
static const int kContentArrayOffset = FixedArray::kHeaderSize;
static const int kEnumerationIndexOffset = kContentArrayOffset + kPointerSize;
static const int kFirstOffset = kEnumerationIndexOffset + kPointerSize;
// Layout description for the bridge array.
static const int kEnumCacheBridgeEnumOffset = FixedArray::kHeaderSize;
static const int kEnumCacheBridgeCacheOffset =
kEnumCacheBridgeEnumOffset + kPointerSize;
#ifdef DEBUG
// Print all the descriptors.
void PrintDescriptors();
// Is the descriptor array sorted and without duplicates?
bool IsSortedNoDuplicates();
// Are two DescriptorArrays equal?
bool IsEqualTo(DescriptorArray* other);
#endif
// The maximum number of descriptors we want in a descriptor array (should
// fit in a page).
static const int kMaxNumberOfDescriptors = 1024 + 512;
private:
// Conversion from descriptor number to array indices.
static int ToKeyIndex(int descriptor_number) {
return descriptor_number+kFirstIndex;
}
static int ToValueIndex(int descriptor_number) {
return descriptor_number << 1;
}
static int ToDetailsIndex(int descriptor_number) {
return( descriptor_number << 1) + 1;
}
bool is_null_descriptor(int descriptor_number) {
return PropertyDetails(GetDetails(descriptor_number)).type() ==
NULL_DESCRIPTOR;
}
// Swap operation on FixedArray without using write barriers.
static inline void fast_swap(FixedArray* array, int first, int second);
// Swap descriptor first and second.
inline void Swap(int first, int second);
FixedArray* GetContentArray() {
return FixedArray::cast(get(kContentArrayIndex));
}
DISALLOW_IMPLICIT_CONSTRUCTORS(DescriptorArray);
};
// HashTable is a subclass of FixedArray that implements a hash table
// that uses open addressing and quadratic probing.
//
// In order for the quadratic probing to work, elements that have not
// yet been used and elements that have been deleted are
// distinguished. Probing continues when deleted elements are
// encountered and stops when unused elements are encountered.
//
// - Elements with key == undefined have not been used yet.
// - Elements with key == null have been deleted.
//
// The hash table class is parameterized with a Shape and a Key.
// Shape must be a class with the following interface:
// class ExampleShape {
// public:
// // Tells whether key matches other.
// static bool IsMatch(Key key, Object* other);
// // Returns the hash value for key.
// static uint32_t Hash(Key key);
// // Returns the hash value for object.
// static uint32_t HashForObject(Key key, Object* object);
// // Convert key to an object.
// static inline Object* AsObject(Key key);
// // The prefix size indicates number of elements in the beginning
// // of the backing storage.
// static const int kPrefixSize = ..;
// // The Element size indicates number of elements per entry.
// static const int kEntrySize = ..;
// };
// The prefix size indicates an amount of memory in the
// beginning of the backing storage that can be used for non-element
// information by subclasses.
template<typename Shape, typename Key>
class HashTable: public FixedArray {
public:
// Returns the number of elements in the hash table.
int NumberOfElements() {
return Smi::cast(get(kNumberOfElementsIndex))->value();
}
// Returns the capacity of the hash table.
int Capacity() {
return Smi::cast(get(kCapacityIndex))->value();
}
// ElementAdded should be called whenever an element is added to a
// hash table.
void ElementAdded() { SetNumberOfElements(NumberOfElements() + 1); }
// ElementRemoved should be called whenever an element is removed from
// a hash table.
void ElementRemoved() { SetNumberOfElements(NumberOfElements() - 1); }
void ElementsRemoved(int n) { SetNumberOfElements(NumberOfElements() - n); }
// Returns a new HashTable object. Might return Failure.
static Object* Allocate(int at_least_space_for);
// Returns the key at entry.
Object* KeyAt(int entry) { return get(EntryToIndex(entry)); }
// Tells whether k is a real key. Null and undefined are not allowed
// as keys and can be used to indicate missing or deleted elements.
bool IsKey(Object* k) {
return !k->IsNull() && !k->IsUndefined();
}
// Garbage collection support.
void IteratePrefix(ObjectVisitor* visitor);
void IterateElements(ObjectVisitor* visitor);
// Casting.
static inline HashTable* cast(Object* obj);
// Compute the probe offset (quadratic probing).
INLINE(static uint32_t GetProbeOffset(uint32_t n)) {
return (n + n * n) >> 1;
}
static const int kNumberOfElementsIndex = 0;
static const int kCapacityIndex = 1;
static const int kPrefixStartIndex = 2;
static const int kElementsStartIndex =
kPrefixStartIndex + Shape::kPrefixSize;
static const int kEntrySize = Shape::kEntrySize;
static const int kElementsStartOffset =
kHeaderSize + kElementsStartIndex * kPointerSize;
// Constant used for denoting a absent entry.
static const int kNotFound = -1;
// Find entry for key otherwise return -1.
int FindEntry(Key key);
protected:
// Find the entry at which to insert element with the given key that
// has the given hash value.
uint32_t FindInsertionEntry(uint32_t hash);
// Returns the index for an entry (of the key)
static inline int EntryToIndex(int entry) {
return (entry * kEntrySize) + kElementsStartIndex;
}
// Update the number of elements in the hash table.
void SetNumberOfElements(int nof) {
fast_set(this, kNumberOfElementsIndex, Smi::FromInt(nof));
}
// Sets the capacity of the hash table.
void SetCapacity(int capacity) {
// To scale a computed hash code to fit within the hash table, we
// use bit-wise AND with a mask, so the capacity must be positive
// and non-zero.
ASSERT(capacity > 0);
fast_set(this, kCapacityIndex, Smi::FromInt(capacity));
}
// Returns probe entry.
static uint32_t GetProbe(uint32_t hash, uint32_t number, uint32_t size) {
ASSERT(IsPowerOf2(size));
return (hash + GetProbeOffset(number)) & (size - 1);
}
// Ensure enough space for n additional elements.
Object* EnsureCapacity(int n, Key key);
};
// HashTableKey is an abstract superclass for virtual key behavior.
class HashTableKey {
public:
// Returns whether the other object matches this key.
virtual bool IsMatch(Object* other) = 0;
// Returns the hash value for this key.
virtual uint32_t Hash() = 0;
// Returns the hash value for object.
virtual uint32_t HashForObject(Object* key) = 0;
// Returns the key object for storing into the hash table.
// If allocations fails a failure object is returned.
virtual Object* AsObject() = 0;
// Required.
virtual ~HashTableKey() {}
};
class SymbolTableShape {
public:
static bool IsMatch(HashTableKey* key, Object* value) {
return key->IsMatch(value);
}
static uint32_t Hash(HashTableKey* key) {
return key->Hash();
}
static uint32_t HashForObject(HashTableKey* key, Object* object) {
return key->HashForObject(object);
}
static Object* AsObject(HashTableKey* key) {
return key->AsObject();
}
static const int kPrefixSize = 0;
static const int kEntrySize = 1;
};
// SymbolTable.
//
// No special elements in the prefix and the element size is 1
// because only the symbol itself (the key) needs to be stored.
class SymbolTable: public HashTable<SymbolTableShape, HashTableKey*> {
public:
// Find symbol in the symbol table. If it is not there yet, it is
// added. The return value is the symbol table which might have
// been enlarged. If the return value is not a failure, the symbol
// pointer *s is set to the symbol found.
Object* LookupSymbol(Vector<const char> str, Object** s);
Object* LookupString(String* key, Object** s);
// Looks up a symbol that is equal to the given string and returns
// true if it is found, assigning the symbol to the given output
// parameter.
bool LookupSymbolIfExists(String* str, String** symbol);
// Casting.
static inline SymbolTable* cast(Object* obj);
private:
Object* LookupKey(HashTableKey* key, Object** s);
DISALLOW_IMPLICIT_CONSTRUCTORS(SymbolTable);
};
class MapCacheShape {
public:
static bool IsMatch(HashTableKey* key, Object* value) {
return key->IsMatch(value);
}
static uint32_t Hash(HashTableKey* key) {
return key->Hash();
}
static uint32_t HashForObject(HashTableKey* key, Object* object) {
return key->HashForObject(object);
}
static Object* AsObject(HashTableKey* key) {
return key->AsObject();
}
static const int kPrefixSize = 0;
static const int kEntrySize = 2;
};
// MapCache.
//
// Maps keys that are a fixed array of symbols to a map.
// Used for canonicalize maps for object literals.
class MapCache: public HashTable<MapCacheShape, HashTableKey*> {
public:
// Find cached value for a string key, otherwise return null.
Object* Lookup(FixedArray* key);
Object* Put(FixedArray* key, Map* value);
static inline MapCache* cast(Object* obj);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(MapCache);
};
template <typename Shape, typename Key>
class Dictionary: public HashTable<Shape, Key> {
public:
static inline Dictionary<Shape, Key>* cast(Object* obj) {
return reinterpret_cast<Dictionary<Shape, Key>*>(obj);
}
// Returns the value at entry.
Object* ValueAt(int entry) {
return get(HashTable<Shape, Key>::EntryToIndex(entry)+1);
}
// Set the value for entry.
void ValueAtPut(int entry, Object* value) {
set(HashTable<Shape, Key>::EntryToIndex(entry)+1, value);
}
// Returns the property details for the property at entry.
PropertyDetails DetailsAt(int entry) {
ASSERT(entry >= 0); // Not found is -1, which is not caught by get().
return PropertyDetails(
Smi::cast(get(HashTable<Shape, Key>::EntryToIndex(entry) + 2)));
}
// Set the details for entry.
void DetailsAtPut(int entry, PropertyDetails value) {
set(HashTable<Shape, Key>::EntryToIndex(entry) + 2, value.AsSmi());
}
// Sorting support
void CopyValuesTo(FixedArray* elements);
// Delete a property from the dictionary.
Object* DeleteProperty(int entry, JSObject::DeleteMode mode);
// Returns the number of elements in the dictionary filtering out properties
// with the specified attributes.
int NumberOfElementsFilterAttributes(PropertyAttributes filter);
// Returns the number of enumerable elements in the dictionary.
int NumberOfEnumElements();
// Copies keys to preallocated fixed array.
void CopyKeysTo(FixedArray* storage, PropertyAttributes filter);
// Fill in details for properties into storage.
void CopyKeysTo(FixedArray* storage);
// Accessors for next enumeration index.
void SetNextEnumerationIndex(int index) {
fast_set(this, kNextEnumerationIndexIndex, Smi::FromInt(index));
}
int NextEnumerationIndex() {
return Smi::cast(FixedArray::get(kNextEnumerationIndexIndex))->value();
}
// Returns a new array for dictionary usage. Might return Failure.
static Object* Allocate(int at_least_space_for);
// Ensure enough space for n additional elements.
Object* EnsureCapacity(int n, Key key);
#ifdef DEBUG
void Print();
#endif
// Returns the key (slow).
Object* SlowReverseLookup(Object* value);
// Sets the entry to (key, value) pair.
inline void SetEntry(int entry,
Object* key,
Object* value,
PropertyDetails details);
Object* Add(Key key, Object* value, PropertyDetails details);
protected:
// Generic at put operation.
Object* AtPut(Key key, Object* value);
// Add entry to dictionary.
Object* AddEntry(Key key,
Object* value,
PropertyDetails details,
uint32_t hash);
// Generate new enumeration indices to avoid enumeration index overflow.
Object* GenerateNewEnumerationIndices();
static const int kMaxNumberKeyIndex =
HashTable<Shape, Key>::kPrefixStartIndex;
static const int kNextEnumerationIndexIndex = kMaxNumberKeyIndex + 1;
};
class StringDictionaryShape {
public:
static inline bool IsMatch(String* key, Object* other);
static inline uint32_t Hash(String* key);
static inline uint32_t HashForObject(String* key, Object* object);
static inline Object* AsObject(String* key);
static const int kPrefixSize = 2;
static const int kEntrySize = 3;
static const bool kIsEnumerable = true;
};
class StringDictionary: public Dictionary<StringDictionaryShape, String*> {
public:
static inline StringDictionary* cast(Object* obj) {
ASSERT(obj->IsDictionary());
return reinterpret_cast<StringDictionary*>(obj);
}
// Copies enumerable keys to preallocated fixed array.
void CopyEnumKeysTo(FixedArray* storage, FixedArray* sort_array);
// For transforming properties of a JSObject.
Object* TransformPropertiesToFastFor(JSObject* obj,
int unused_property_fields);
};
class NumberDictionaryShape {
public:
static inline bool IsMatch(uint32_t key, Object* other);
static inline uint32_t Hash(uint32_t key);
static inline uint32_t HashForObject(uint32_t key, Object* object);
static inline Object* AsObject(uint32_t key);
static const int kPrefixSize = 2;
static const int kEntrySize = 3;
static const bool kIsEnumerable = false;
};
class NumberDictionary: public Dictionary<NumberDictionaryShape, uint32_t> {
public:
static NumberDictionary* cast(Object* obj) {
ASSERT(obj->IsDictionary());
return reinterpret_cast<NumberDictionary*>(obj);
}
// Type specific at put (default NONE attributes is used when adding).
Object* AtNumberPut(uint32_t key, Object* value);
Object* AddNumberEntry(uint32_t key,
Object* value,
PropertyDetails details);
// Set an existing entry or add a new one if needed.
Object* Set(uint32_t key, Object* value, PropertyDetails details);
void UpdateMaxNumberKey(uint32_t key);
// If slow elements are required we will never go back to fast-case
// for the elements kept in this dictionary. We require slow
// elements if an element has been added at an index larger than
// kRequiresSlowElementsLimit or set_requires_slow_elements() has been called
// when defining a getter or setter with a number key.
inline bool requires_slow_elements();
inline void set_requires_slow_elements();
// Get the value of the max number key that has been added to this
// dictionary. max_number_key can only be called if
// requires_slow_elements returns false.
inline uint32_t max_number_key();
// Remove all entries were key is a number and (from <= key && key < to).
void RemoveNumberEntries(uint32_t from, uint32_t to);
// Bit masks.
static const int kRequiresSlowElementsMask = 1;
static const int kRequiresSlowElementsTagSize = 1;
static const uint32_t kRequiresSlowElementsLimit = (1 << 29) - 1;
};
// ByteArray represents fixed sized byte arrays. Used by the outside world,
// such as PCRE, and also by the memory allocator and garbage collector to
// fill in free blocks in the heap.
class ByteArray: public Array {
public:
// Setter and getter.
inline byte get(int index);
inline void set(int index, byte value);
// Treat contents as an int array.
inline int get_int(int index);
static int SizeFor(int length) {
return OBJECT_SIZE_ALIGN(kHeaderSize + length);
}
// We use byte arrays for free blocks in the heap. Given a desired size in
// bytes that is a multiple of the word size and big enough to hold a byte
// array, this function returns the number of elements a byte array should
// have.
static int LengthFor(int size_in_bytes) {
ASSERT(IsAligned(size_in_bytes, kPointerSize));
ASSERT(size_in_bytes >= kHeaderSize);
return size_in_bytes - kHeaderSize;
}
// Returns data start address.
inline Address GetDataStartAddress();
// Returns a pointer to the ByteArray object for a given data start address.
static inline ByteArray* FromDataStartAddress(Address address);
// Casting.
static inline ByteArray* cast(Object* obj);
// Dispatched behavior.
int ByteArraySize() { return SizeFor(length()); }
#ifdef DEBUG
void ByteArrayPrint();
void ByteArrayVerify();
#endif
// ByteArray headers are not quadword aligned.
static const int kHeaderSize = Array::kHeaderSize;
static const int kAlignedSize = Array::kAlignedSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ByteArray);
};
// A PixelArray represents a fixed-size byte array with special semantics
// used for implementing the CanvasPixelArray object. Please see the
// specification at:
// http://www.whatwg.org/specs/web-apps/current-work/
// multipage/the-canvas-element.html#canvaspixelarray
// In particular, write access clamps the value written to 0 or 255 if the
// value written is outside this range.
class PixelArray: public Array {
public:
// [external_pointer]: The pointer to the external memory area backing this
// pixel array.
DECL_ACCESSORS(external_pointer, uint8_t) // Pointer to the data store.
// Setter and getter.
inline uint8_t get(int index);
inline void set(int index, uint8_t value);
// This accessor applies the correct conversion from Smi, HeapNumber and
// undefined and clamps the converted value between 0 and 255.
Object* SetValue(uint32_t index, Object* value);
// Casting.
static inline PixelArray* cast(Object* obj);
#ifdef DEBUG
void PixelArrayPrint();
void PixelArrayVerify();
#endif // DEBUG
// Maximal acceptable length for a pixel array.
static const int kMaxLength = 0x3fffffff;
// PixelArray headers are not quadword aligned.
static const int kExternalPointerOffset = Array::kAlignedSize;
static const int kHeaderSize = kExternalPointerOffset + kPointerSize;
static const int kAlignedSize = OBJECT_SIZE_ALIGN(kHeaderSize);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(PixelArray);
};
// An ExternalArray represents a fixed-size array of primitive values
// which live outside the JavaScript heap. Its subclasses are used to
// implement the CanvasArray types being defined in the WebGL
// specification. As of this writing the first public draft is not yet
// available, but Khronos members can access the draft at:
// https://cvs.khronos.org/svn/repos/3dweb/trunk/doc/spec/WebGL-spec.html
//
// The semantics of these arrays differ from CanvasPixelArray.
// Out-of-range values passed to the setter are converted via a C
// cast, not clamping. Out-of-range indices cause exceptions to be
// raised rather than being silently ignored.
class ExternalArray: public Array {
public:
// [external_pointer]: The pointer to the external memory area backing this
// external array.
DECL_ACCESSORS(external_pointer, void) // Pointer to the data store.
// Casting.
static inline ExternalArray* cast(Object* obj);
// Maximal acceptable length for an external array.
static const int kMaxLength = 0x3fffffff;
// ExternalArray headers are not quadword aligned.
static const int kExternalPointerOffset = Array::kAlignedSize;
static const int kHeaderSize = kExternalPointerOffset + kPointerSize;
static const int kAlignedSize = OBJECT_SIZE_ALIGN(kHeaderSize);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalArray);
};
class ExternalByteArray: public ExternalArray {
public:
// Setter and getter.
inline int8_t get(int index);
inline void set(int index, int8_t value);
// This accessor applies the correct conversion from Smi, HeapNumber
// and undefined.
Object* SetValue(uint32_t index, Object* value);
// Casting.
static inline ExternalByteArray* cast(Object* obj);
#ifdef DEBUG
void ExternalByteArrayPrint();
void ExternalByteArrayVerify();
#endif // DEBUG
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalByteArray);
};
class ExternalUnsignedByteArray: public ExternalArray {
public:
// Setter and getter.
inline uint8_t get(int index);
inline void set(int index, uint8_t value);
// This accessor applies the correct conversion from Smi, HeapNumber
// and undefined.
Object* SetValue(uint32_t index, Object* value);
// Casting.
static inline ExternalUnsignedByteArray* cast(Object* obj);
#ifdef DEBUG
void ExternalUnsignedByteArrayPrint();
void ExternalUnsignedByteArrayVerify();
#endif // DEBUG
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalUnsignedByteArray);
};
class ExternalShortArray: public ExternalArray {
public:
// Setter and getter.
inline int16_t get(int index);
inline void set(int index, int16_t value);
// This accessor applies the correct conversion from Smi, HeapNumber
// and undefined.
Object* SetValue(uint32_t index, Object* value);
// Casting.
static inline ExternalShortArray* cast(Object* obj);
#ifdef DEBUG
void ExternalShortArrayPrint();
void ExternalShortArrayVerify();
#endif // DEBUG
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalShortArray);
};
class ExternalUnsignedShortArray: public ExternalArray {
public:
// Setter and getter.
inline uint16_t get(int index);
inline void set(int index, uint16_t value);
// This accessor applies the correct conversion from Smi, HeapNumber
// and undefined.
Object* SetValue(uint32_t index, Object* value);
// Casting.
static inline ExternalUnsignedShortArray* cast(Object* obj);
#ifdef DEBUG
void ExternalUnsignedShortArrayPrint();
void ExternalUnsignedShortArrayVerify();
#endif // DEBUG
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalUnsignedShortArray);
};
class ExternalIntArray: public ExternalArray {
public:
// Setter and getter.
inline int32_t get(int index);
inline void set(int index, int32_t value);
// This accessor applies the correct conversion from Smi, HeapNumber
// and undefined.
Object* SetValue(uint32_t index, Object* value);
// Casting.
static inline ExternalIntArray* cast(Object* obj);
#ifdef DEBUG
void ExternalIntArrayPrint();
void ExternalIntArrayVerify();
#endif // DEBUG
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalIntArray);
};
class ExternalUnsignedIntArray: public ExternalArray {
public:
// Setter and getter.
inline uint32_t get(int index);
inline void set(int index, uint32_t value);
// This accessor applies the correct conversion from Smi, HeapNumber
// and undefined.
Object* SetValue(uint32_t index, Object* value);
// Casting.
static inline ExternalUnsignedIntArray* cast(Object* obj);
#ifdef DEBUG
void ExternalUnsignedIntArrayPrint();
void ExternalUnsignedIntArrayVerify();
#endif // DEBUG
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalUnsignedIntArray);
};
class ExternalFloatArray: public ExternalArray {
public:
// Setter and getter.
inline float get(int index);
inline void set(int index, float value);
// This accessor applies the correct conversion from Smi, HeapNumber
// and undefined.
Object* SetValue(uint32_t index, Object* value);
// Casting.
static inline ExternalFloatArray* cast(Object* obj);
#ifdef DEBUG
void ExternalFloatArrayPrint();
void ExternalFloatArrayVerify();
#endif // DEBUG
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalFloatArray);
};
// Code describes objects with on-the-fly generated machine code.
class Code: public HeapObject {
public:
// Opaque data type for encapsulating code flags like kind, inline
// cache state, and arguments count.
enum Flags { };
enum Kind {
FUNCTION,
STUB,
BUILTIN,
LOAD_IC,
KEYED_LOAD_IC,
CALL_IC,
STORE_IC,
KEYED_STORE_IC,
// No more than eight kinds. The value currently encoded in three bits in
// Flags.
// Pseudo-kinds.
REGEXP = BUILTIN,
FIRST_IC_KIND = LOAD_IC,
LAST_IC_KIND = KEYED_STORE_IC
};
enum {
NUMBER_OF_KINDS = KEYED_STORE_IC + 1
};
#ifdef ENABLE_DISASSEMBLER
// Printing
static const char* Kind2String(Kind kind);
static const char* ICState2String(InlineCacheState state);
static const char* PropertyType2String(PropertyType type);
void Disassemble(const char* name);
#endif // ENABLE_DISASSEMBLER
// [instruction_size]: Size of the native instructions
inline int instruction_size();
inline void set_instruction_size(int value);
// [relocation_size]: Size of relocation information.
inline int relocation_size();
inline void set_relocation_size(int value);
// [sinfo_size]: Size of scope information.
inline int sinfo_size();
inline void set_sinfo_size(int value);
// [flags]: Various code flags.
inline Flags flags();
inline void set_flags(Flags flags);
// [flags]: Access to specific code flags.
inline Kind kind();
inline InlineCacheState ic_state(); // Only valid for IC stubs.
inline InLoopFlag ic_in_loop(); // Only valid for IC stubs.
inline PropertyType type(); // Only valid for monomorphic IC stubs.
inline int arguments_count(); // Only valid for call IC stubs.
// Testers for IC stub kinds.
inline bool is_inline_cache_stub();
inline bool is_load_stub() { return kind() == LOAD_IC; }
inline bool is_keyed_load_stub() { return kind() == KEYED_LOAD_IC; }
inline bool is_store_stub() { return kind() == STORE_IC; }
inline bool is_keyed_store_stub() { return kind() == KEYED_STORE_IC; }
inline bool is_call_stub() { return kind() == CALL_IC; }
// [major_key]: For kind STUB, the major key.
inline CodeStub::Major major_key();
inline void set_major_key(CodeStub::Major major);
// Flags operations.
static inline Flags ComputeFlags(Kind kind,
InLoopFlag in_loop = NOT_IN_LOOP,
InlineCacheState ic_state = UNINITIALIZED,
PropertyType type = NORMAL,
int argc = -1);
static inline Flags ComputeMonomorphicFlags(
Kind kind,
PropertyType type,
InLoopFlag in_loop = NOT_IN_LOOP,
int argc = -1);
static inline Kind ExtractKindFromFlags(Flags flags);
static inline InlineCacheState ExtractICStateFromFlags(Flags flags);
static inline InLoopFlag ExtractICInLoopFromFlags(Flags flags);
static inline PropertyType ExtractTypeFromFlags(Flags flags);
static inline int ExtractArgumentsCountFromFlags(Flags flags);
static inline Flags RemoveTypeFromFlags(Flags flags);
// Convert a target address into a code object.
static inline Code* GetCodeFromTargetAddress(Address address);
// Returns the address of the first instruction.
inline byte* instruction_start();
// Returns the size of the instructions, padding, and relocation information.
inline int body_size();
// Returns the address of the first relocation info (read backwards!).
inline byte* relocation_start();
// Code entry point.
inline byte* entry();
// Returns true if pc is inside this object's instructions.
inline bool contains(byte* pc);
// Returns the address of the scope information.
inline byte* sinfo_start();
// Relocate the code by delta bytes. Called to signal that this code
// object has been moved by delta bytes.
void Relocate(int delta);
// Migrate code described by desc.
void CopyFrom(const CodeDesc& desc);
// Returns the object size for a given body and sinfo size (Used for
// allocation).
static int SizeFor(int body_size, int sinfo_size) {
ASSERT_SIZE_TAG_ALIGNED(body_size);
ASSERT_SIZE_TAG_ALIGNED(sinfo_size);
return RoundUp(kHeaderSize + body_size + sinfo_size, kCodeAlignment);
}
// Calculate the size of the code object to report for log events. This takes
// the layout of the code object into account.
int ExecutableSize() {
// Check that the assumptions about the layout of the code object holds.
ASSERT_EQ(static_cast<int>(instruction_start() - address()),
Code::kHeaderSize);
return instruction_size() + Code::kHeaderSize;
}
// Locating source position.
int SourcePosition(Address pc);
int SourceStatementPosition(Address pc);
// Casting.
static inline Code* cast(Object* obj);
// Dispatched behavior.
int CodeSize() { return SizeFor(body_size(), sinfo_size()); }
void CodeIterateBody(ObjectVisitor* v);
#ifdef DEBUG
void CodePrint();
void CodeVerify();
#endif
// Code entry points are aligned to 32 bytes.
static const int kCodeAlignmentBits = 5;
static const int kCodeAlignment = 1 << kCodeAlignmentBits;
static const int kCodeAlignmentMask = kCodeAlignment - 1;
// Layout description.
static const int kInstructionSizeOffset = HeapObject::kHeaderSize;
static const int kRelocationSizeOffset = kInstructionSizeOffset + kIntSize;
static const int kSInfoSizeOffset = kRelocationSizeOffset + kIntSize;
static const int kFlagsOffset = kSInfoSizeOffset + kIntSize;
static const int kKindSpecificFlagsOffset = kFlagsOffset + kIntSize;
// Add padding to align the instruction start following right after
// the Code object header.
static const int kHeaderSize =
(kKindSpecificFlagsOffset + kIntSize + kCodeAlignmentMask) &
~kCodeAlignmentMask;
// Byte offsets within kKindSpecificFlagsOffset.
static const int kStubMajorKeyOffset = kKindSpecificFlagsOffset + 1;
// Flags layout.
static const int kFlagsICStateShift = 0;
static const int kFlagsICInLoopShift = 3;
static const int kFlagsKindShift = 4;
static const int kFlagsTypeShift = 7;
static const int kFlagsArgumentsCountShift = 10;
static const int kFlagsICStateMask = 0x00000007; // 0000000111
static const int kFlagsICInLoopMask = 0x00000008; // 0000001000
static const int kFlagsKindMask = 0x00000070; // 0001110000
static const int kFlagsTypeMask = 0x00000380; // 1110000000
static const int kFlagsArgumentsCountMask = 0xFFFFFC00;
static const int kFlagsNotUsedInLookup =
(kFlagsICInLoopMask | kFlagsTypeMask);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(Code);
};
// All heap objects have a Map that describes their structure.
// A Map contains information about:
// - Size information about the object
// - How to iterate over an object (for garbage collection)
class Map: public HeapObject {
public:
// Instance size.
inline int instance_size();
inline void set_instance_size(int value);
// Count of properties allocated in the object.
inline int inobject_properties();
inline void set_inobject_properties(int value);
// Count of property fields pre-allocated in the object when first allocated.
inline int pre_allocated_property_fields();
inline void set_pre_allocated_property_fields(int value);
// Instance type.
inline InstanceType instance_type();
inline void set_instance_type(InstanceType value);
// Tells how many unused property fields are available in the
// instance (only used for JSObject in fast mode).
inline int unused_property_fields();
inline void set_unused_property_fields(int value);
// Bit field.
inline byte bit_field();
inline void set_bit_field(byte value);
// Bit field 2.
inline byte bit_field2();
inline void set_bit_field2(byte value);
// Tells whether the object in the prototype property will be used
// for instances created from this function. If the prototype
// property is set to a value that is not a JSObject, the prototype
// property will not be used to create instances of the function.
// See ECMA-262, 13.2.2.
inline void set_non_instance_prototype(bool value);
inline bool has_non_instance_prototype();
// Tells whether the instance with this map should be ignored by the
// __proto__ accessor.
inline void set_is_hidden_prototype() {
set_bit_field(bit_field() | (1 << kIsHiddenPrototype));
}
inline bool is_hidden_prototype() {
return ((1 << kIsHiddenPrototype) & bit_field()) != 0;
}
// Records and queries whether the instance has a named interceptor.
inline void set_has_named_interceptor() {
set_bit_field(bit_field() | (1 << kHasNamedInterceptor));
}
inline bool has_named_interceptor() {
return ((1 << kHasNamedInterceptor) & bit_field()) != 0;
}
// Records and queries whether the instance has an indexed interceptor.
inline void set_has_indexed_interceptor() {
set_bit_field(bit_field() | (1 << kHasIndexedInterceptor));
}
inline bool has_indexed_interceptor() {
return ((1 << kHasIndexedInterceptor) & bit_field()) != 0;
}
// Tells whether the instance is undetectable.
// An undetectable object is a special class of JSObject: 'typeof' operator
// returns undefined, ToBoolean returns false. Otherwise it behaves like
// a normal JS object. It is useful for implementing undetectable
// document.all in Firefox & Safari.
// See https://bugzilla.mozilla.org/show_bug.cgi?id=248549.
inline void set_is_undetectable() {
set_bit_field(bit_field() | (1 << kIsUndetectable));
}
inline bool is_undetectable() {
return ((1 << kIsUndetectable) & bit_field()) != 0;
}
inline void set_needs_loading(bool value) {
if (value) {
set_bit_field2(bit_field2() | (1 << kNeedsLoading));
} else {
set_bit_field2(bit_field2() & ~(1 << kNeedsLoading));
}
}
// Does this object or function require a lazily loaded script to be
// run before being used?
inline bool needs_loading() {
return ((1 << kNeedsLoading) & bit_field2()) != 0;
}
// Tells whether the instance has a call-as-function handler.
inline void set_has_instance_call_handler() {
set_bit_field(bit_field() | (1 << kHasInstanceCallHandler));
}
inline bool has_instance_call_handler() {
return ((1 << kHasInstanceCallHandler) & bit_field()) != 0;
}
// Tells whether the instance needs security checks when accessing its
// properties.
inline void set_is_access_check_needed(bool access_check_needed);
inline bool is_access_check_needed();
// [prototype]: implicit prototype object.
DECL_ACCESSORS(prototype, Object)
// [constructor]: points back to the function responsible for this map.
DECL_ACCESSORS(constructor, Object)
// [instance descriptors]: describes the object.
DECL_ACCESSORS(instance_descriptors, DescriptorArray)
// [stub cache]: contains stubs compiled for this map.
DECL_ACCESSORS(code_cache, FixedArray)
// Returns a copy of the map.
Object* CopyDropDescriptors();
// Returns a copy of the map, with all transitions dropped from the
// instance descriptors.
Object* CopyDropTransitions();
// Returns the property index for name (only valid for FAST MODE).
int PropertyIndexFor(String* name);
// Returns the next free property index (only valid for FAST MODE).
int NextFreePropertyIndex();
// Returns the number of properties described in instance_descriptors.
int NumberOfDescribedProperties();
// Casting.
static inline Map* cast(Object* obj);
// Locate an accessor in the instance descriptor.
AccessorDescriptor* FindAccessor(String* name);
// Code cache operations.
// Clears the code cache.
inline void ClearCodeCache();
// Update code cache.
Object* UpdateCodeCache(String* name, Code* code);
// Returns the found code or undefined if absent.
Object* FindInCodeCache(String* name, Code::Flags flags);
// Returns the non-negative index of the code object if it is in the
// cache and -1 otherwise.
int IndexInCodeCache(Code* code);
// Removes a code object from the code cache at the given index.
void RemoveFromCodeCache(int index);
// For every transition in this map, makes the transition's
// target's prototype pointer point back to this map.
// This is undone in MarkCompactCollector::ClearNonLiveTransitions().
void CreateBackPointers();
// Set all map transitions from this map to dead maps to null.
// Also, restore the original prototype on the targets of these
// transitions, so that we do not process this map again while
// following back pointers.
void ClearNonLiveTransitions(Object* real_prototype);
// Dispatched behavior.
void MapIterateBody(ObjectVisitor* v);
#ifdef DEBUG
void MapPrint();
void MapVerify();
#endif
static const int kMaxPreAllocatedPropertyFields = 255;
// Layout description.
static const int kInstanceSizesOffset = HeapObject::kHeaderSize;
static const int kInstanceAttributesOffset = kInstanceSizesOffset + kIntSize;
static const int kPrototypeOffset = kInstanceAttributesOffset + kIntSize;
static const int kConstructorOffset = kPrototypeOffset + kPointerSize;
static const int kInstanceDescriptorsOffset =
kConstructorOffset + kPointerSize;
static const int kCodeCacheOffset = kInstanceDescriptorsOffset + kPointerSize;
static const int kSize = kCodeCacheOffset + kPointerSize;
// Byte offsets within kInstanceSizesOffset.
static const int kInstanceSizeOffset = kInstanceSizesOffset + 0;
static const int kInObjectPropertiesByte = 1;
static const int kInObjectPropertiesOffset =
kInstanceSizesOffset + kInObjectPropertiesByte;
static const int kPreAllocatedPropertyFieldsByte = 2;
static const int kPreAllocatedPropertyFieldsOffset =
kInstanceSizesOffset + kPreAllocatedPropertyFieldsByte;
// The byte at position 3 is not in use at the moment.
// Byte offsets within kInstanceAttributesOffset attributes.
static const int kInstanceTypeOffset = kInstanceAttributesOffset + 0;
static const int kUnusedPropertyFieldsOffset = kInstanceAttributesOffset + 1;
static const int kBitFieldOffset = kInstanceAttributesOffset + 2;
static const int kBitField2Offset = kInstanceAttributesOffset + 3;
STATIC_CHECK(kInstanceTypeOffset == Internals::kMapInstanceTypeOffset);
// Bit positions for bit field.
static const int kUnused = 0; // To be used for marking recently used maps.
static const int kHasNonInstancePrototype = 1;
static const int kIsHiddenPrototype = 2;
static const int kHasNamedInterceptor = 3;
static const int kHasIndexedInterceptor = 4;
static const int kIsUndetectable = 5;
static const int kHasInstanceCallHandler = 6;
Split window support from V8. Here is a description of the background and design of split window in Chrome and V8: https://docs.google.com/a/google.com/Doc?id=chhjkpg_47fwddxbfr This change list splits the window object into two parts: 1) an inner window object used as the global object of contexts; 2) an outer window object exposed to JavaScript and accessible by the name 'window'. Firefox did it awhile ago, here are some discussions: https://wiki.mozilla.org/Gecko:SplitWindow. One additional benefit of splitting window in Chrome is that accessing global variables don't need security checks anymore, it can improve applications that use many global variables. V8 support of split window: There are a small number of changes on V8 api to support split window: Security context is removed from V8, so does related API functions; A global object can be detached from its context and reused by a new context; Access checks on an object template can be turned on/off by default; An object can turn on its access checks later; V8 has a new object type, ApiGlobalObject, which is the outer window object type. The existing JSGlobalObject becomes the inner window object type. Security checks are moved from JSGlobalObject to ApiGlobalObject. ApiGlobalObject is the one exposed to JavaScript, it is accessible through Context::Global(). ApiGlobalObject's prototype is set to JSGlobalObject so that property lookups are forwarded to JSGlobalObject. ApiGlobalObject forwards all other property access requests to JSGlobalObject, such as SetProperty, DeleteProperty, etc. Security token is moved to a global context, and ApiGlobalObject has a reference to its global context. JSGlobalObject has a reference to its global context as well. When accessing properties on a global object in JavaScript, the domain security check is performed by comparing the security token of the lexical context (Top::global_context()) to the token of global object's context. The check is only needed when the receiver is a window object, such as 'window.document'. Accessing global variables, such as 'var foo = 3; foo' does not need checks because the receiver is the inner window object. When an outer window is detached from its global context (when a frame navigates away from a page), it is completely detached from the inner window. A new context is created for the new page, and the outer global object is reused. At this point, the access check on the DOMWindow wrapper of the old context is turned on. The code in old context is still able to access DOMWindow properties, but it has to go through domain security checks. It is debatable on how to implement the outer window object. Currently each property access function has to check if the receiver is ApiGlobalObject type. This approach might be error-prone that one may forget to check the receiver when adding new functions. It is unlikely a performance issue because accessing global variables are more common than 'window.foo' style coding. I am still working on the ARM port, and I'd like to hear comments and suggestions on the best way to support it in V8. Review URL: http://codereview.chromium.org/7366 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@540 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-10-21 19:07:58 +00:00
static const int kIsAccessCheckNeeded = 7;
// Bit positions for bit field 2
static const int kNeedsLoading = 0;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(Map);
};
// An abstract superclass, a marker class really, for simple structure classes.
// It doesn't carry much functionality but allows struct classes to me
// identified in the type system.
class Struct: public HeapObject {
public:
inline void InitializeBody(int object_size);
static inline Struct* cast(Object* that);
};
// Script describes a script which has been added to the VM.
class Script: public Struct {
public:
// Script types.
enum Type {
TYPE_NATIVE = 0,
TYPE_EXTENSION = 1,
TYPE_NORMAL = 2
};
// Script compilation types.
enum CompilationType {
COMPILATION_TYPE_HOST = 0,
COMPILATION_TYPE_EVAL = 1,
COMPILATION_TYPE_JSON = 2
};
// [source]: the script source.
DECL_ACCESSORS(source, Object)
// [name]: the script name.
DECL_ACCESSORS(name, Object)
// [id]: the script id.
DECL_ACCESSORS(id, Object)
// [line_offset]: script line offset in resource from where it was extracted.
DECL_ACCESSORS(line_offset, Smi)
// [column_offset]: script column offset in resource from where it was
// extracted.
DECL_ACCESSORS(column_offset, Smi)
// [data]: additional data associated with this script.
DECL_ACCESSORS(data, Object)
// [context_data]: context data for the context this script was compiled in.
DECL_ACCESSORS(context_data, Object)
// [wrapper]: the wrapper cache.
DECL_ACCESSORS(wrapper, Proxy)
// [type]: the script type.
DECL_ACCESSORS(type, Smi)
// [compilation]: how the the script was compiled.
DECL_ACCESSORS(compilation_type, Smi)
// [line_ends]: array of line ends positions.
DECL_ACCESSORS(line_ends, Object)
// [eval_from_function]: for eval scripts the funcion from which eval was
// called.
DECL_ACCESSORS(eval_from_function, Object)
// [eval_from_instructions_offset]: the instruction offset in the code for the
// function from which eval was called where eval was called.
DECL_ACCESSORS(eval_from_instructions_offset, Smi)
static inline Script* cast(Object* obj);
// If script source is an external string, check that the underlying
// resource is accessible. Otherwise, always return true.
inline bool HasValidSource();
#ifdef DEBUG
void ScriptPrint();
void ScriptVerify();
#endif
static const int kSourceOffset = HeapObject::kHeaderSize;
static const int kNameOffset = kSourceOffset + kPointerSize;
static const int kLineOffsetOffset = kNameOffset + kPointerSize;
static const int kColumnOffsetOffset = kLineOffsetOffset + kPointerSize;
static const int kDataOffset = kColumnOffsetOffset + kPointerSize;
static const int kContextOffset = kDataOffset + kPointerSize;
static const int kWrapperOffset = kContextOffset + kPointerSize;
static const int kTypeOffset = kWrapperOffset + kPointerSize;
static const int kCompilationTypeOffset = kTypeOffset + kPointerSize;
static const int kLineEndsOffset = kCompilationTypeOffset + kPointerSize;
static const int kIdOffset = kLineEndsOffset + kPointerSize;
static const int kEvalFromFunctionOffset = kIdOffset + kPointerSize;
static const int kEvalFrominstructionsOffsetOffset =
kEvalFromFunctionOffset + kPointerSize;
static const int kSize = kEvalFrominstructionsOffsetOffset + kPointerSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(Script);
};
// SharedFunctionInfo describes the JSFunction information that can be
// shared by multiple instances of the function.
class SharedFunctionInfo: public HeapObject {
public:
// [name]: Function name.
DECL_ACCESSORS(name, Object)
// [code]: Function code.
DECL_ACCESSORS(code, Code)
// [construct stub]: Code stub for constructing instances of this function.
DECL_ACCESSORS(construct_stub, Code)
// Returns if this function has been compiled to native code yet.
inline bool is_compiled();
// [length]: The function length - usually the number of declared parameters.
// Use up to 2^30 parameters.
inline int length();
inline void set_length(int value);
// [formal parameter count]: The declared number of parameters.
inline int formal_parameter_count();
inline void set_formal_parameter_count(int value);
// Set the formal parameter count so the function code will be
// called without using argument adaptor frames.
inline void DontAdaptArguments();
// [expected_nof_properties]: Expected number of properties for the function.
inline int expected_nof_properties();
inline void set_expected_nof_properties(int value);
// [instance class name]: class name for instances.
DECL_ACCESSORS(instance_class_name, Object)
// [function data]: This field has been added for make benefit the API.
// In the long run we don't want all functions to have this field but
// we can fix that when we have a better model for storing hidden data
// on objects.
DECL_ACCESSORS(function_data, Object)
// [script info]: Script from which the function originates.
DECL_ACCESSORS(script, Object)
// [start_position_and_type]: Field used to store both the source code
// position, whether or not the function is a function expression,
// and whether or not the function is a toplevel function. The two
// least significants bit indicates whether the function is an
// expression and the rest contains the source code position.
inline int start_position_and_type();
inline void set_start_position_and_type(int value);
// [debug info]: Debug information.
DECL_ACCESSORS(debug_info, Object)
// [inferred name]: Name inferred from variable or property
// assignment of this function. Used to facilitate debugging and
// profiling of JavaScript code written in OO style, where almost
// all functions are anonymous but are assigned to object
// properties.
DECL_ACCESSORS(inferred_name, String)
// Position of the 'function' token in the script source.
inline int function_token_position();
inline void set_function_token_position(int function_token_position);
// Position of this function in the script source.
inline int start_position();
inline void set_start_position(int start_position);
// End position of this function in the script source.
inline int end_position();
inline void set_end_position(int end_position);
// Is this function a function expression in the source code.
inline bool is_expression();
inline void set_is_expression(bool value);
// Is this function a top-level function (scripts, evals).
inline bool is_toplevel();
inline void set_is_toplevel(bool value);
// Bit field containing various information collected by the compiler to
// drive optimization.
inline int compiler_hints();
inline void set_compiler_hints(int value);
// Add information on assignments of the form this.x = ...;
void SetThisPropertyAssignmentsInfo(
bool has_only_this_property_assignments,
bool has_only_simple_this_property_assignments,
FixedArray* this_property_assignments);
// Clear information on assignments of the form this.x = ...;
void ClearThisPropertyAssignmentsInfo();
// Indicate that this function only consists of assignments of the form
// this.x = ...;.
inline bool has_only_this_property_assignments();
// Indicate that this function only consists of assignments of the form
// this.x = y; where y is either a constant or refers to an argument.
inline bool has_only_simple_this_property_assignments();
inline bool try_fast_codegen();
inline void set_try_fast_codegen(bool flag);
// For functions which only contains this property assignments this provides
// access to the names for the properties assigned.
DECL_ACCESSORS(this_property_assignments, Object)
inline int this_property_assignments_count();
inline void set_this_property_assignments_count(int value);
String* GetThisPropertyAssignmentName(int index);
bool IsThisPropertyAssignmentArgument(int index);
int GetThisPropertyAssignmentArgument(int index);
Object* GetThisPropertyAssignmentConstant(int index);
// [source code]: Source code for the function.
bool HasSourceCode();
Object* GetSourceCode();
// Calculate the instance size.
int CalculateInstanceSize();
// Calculate the number of in-object properties.
int CalculateInObjectProperties();
// Dispatched behavior.
void SharedFunctionInfoIterateBody(ObjectVisitor* v);
// Set max_length to -1 for unlimited length.
void SourceCodePrint(StringStream* accumulator, int max_length);
#ifdef DEBUG
void SharedFunctionInfoPrint();
void SharedFunctionInfoVerify();
#endif
// Casting.
static inline SharedFunctionInfo* cast(Object* obj);
// Constants.
static const int kDontAdaptArgumentsSentinel = -1;
// Layout description.
// (An even number of integers has a size that is a multiple of a pointer.)
static const int kNameOffset = HeapObject::kHeaderSize;
static const int kCodeOffset = kNameOffset + kPointerSize;
static const int kConstructStubOffset = kCodeOffset + kPointerSize;
static const int kLengthOffset = kConstructStubOffset + kPointerSize;
static const int kFormalParameterCountOffset = kLengthOffset + kIntSize;
static const int kExpectedNofPropertiesOffset =
kFormalParameterCountOffset + kIntSize;
static const int kStartPositionAndTypeOffset =
kExpectedNofPropertiesOffset + kIntSize;
static const int kEndPositionOffset = kStartPositionAndTypeOffset + kIntSize;
static const int kFunctionTokenPositionOffset = kEndPositionOffset + kIntSize;
static const int kInstanceClassNameOffset =
kFunctionTokenPositionOffset + kIntSize;
static const int kExternalReferenceDataOffset =
kInstanceClassNameOffset + kPointerSize;
static const int kScriptOffset = kExternalReferenceDataOffset + kPointerSize;
static const int kDebugInfoOffset = kScriptOffset + kPointerSize;
static const int kInferredNameOffset = kDebugInfoOffset + kPointerSize;
static const int kCompilerHintsOffset = kInferredNameOffset + kPointerSize;
static const int kThisPropertyAssignmentsOffset =
kCompilerHintsOffset + kPointerSize;
static const int kThisPropertyAssignmentsCountOffset =
kThisPropertyAssignmentsOffset + kPointerSize;
static const int kSize = kThisPropertyAssignmentsCountOffset + kPointerSize;
private:
// Bit positions in length_and_flg.
// The least significant bit is used as the flag.
static const int kFlagBit = 0;
static const int kLengthShift = 1;
static const int kLengthMask = ~((1 << kLengthShift) - 1);
// Bit positions in start_position_and_type.
// The source code start position is in the 30 most significant bits of
// the start_position_and_type field.
static const int kIsExpressionBit = 0;
static const int kIsTopLevelBit = 1;
static const int kStartPositionShift = 2;
static const int kStartPositionMask = ~((1 << kStartPositionShift) - 1);
// Bit positions in compiler_hints.
static const int kHasOnlyThisPropertyAssignments = 0;
static const int kHasOnlySimpleThisPropertyAssignments = 1;
static const int kTryFastCodegen = 2;
DISALLOW_IMPLICIT_CONSTRUCTORS(SharedFunctionInfo);
};
// JSFunction describes JavaScript functions.
class JSFunction: public JSObject {
public:
// [prototype_or_initial_map]:
DECL_ACCESSORS(prototype_or_initial_map, Object)
// [shared_function_info]: The information about the function that
// can be shared by instances.
DECL_ACCESSORS(shared, SharedFunctionInfo)
// [context]: The context for this function.
inline Context* context();
inline Object* unchecked_context();
inline void set_context(Object* context);
// [code]: The generated code object for this function. Executed
// when the function is invoked, e.g. foo() or new foo(). See
// [[Call]] and [[Construct]] description in ECMA-262, section
// 8.6.2, page 27.
inline Code* code();
inline void set_code(Code* value);
// Tells whether this function is a context-independent boilerplate
// function.
inline bool IsBoilerplate();
// Tells whether this function is builtin.
inline bool IsBuiltin();
// [literals]: Fixed array holding the materialized literals.
//
// If the function contains object, regexp or array literals, the
// literals array prefix contains the object, regexp, and array
// function to be used when creating these literals. This is
// necessary so that we do not dynamically lookup the object, regexp
// or array functions. Performing a dynamic lookup, we might end up
// using the functions from a new context that we should not have
// access to.
DECL_ACCESSORS(literals, FixedArray)
// The initial map for an object created by this constructor.
inline Map* initial_map();
inline void set_initial_map(Map* value);
inline bool has_initial_map();
// Get and set the prototype property on a JSFunction. If the
// function has an initial map the prototype is set on the initial
// map. Otherwise, the prototype is put in the initial map field
// until an initial map is needed.
inline bool has_prototype();
inline bool has_instance_prototype();
inline Object* prototype();
inline Object* instance_prototype();
Object* SetInstancePrototype(Object* value);
Object* SetPrototype(Object* value);
// Accessor for this function's initial map's [[class]]
// property. This is primarily used by ECMA native functions. This
// method sets the class_name field of this function's initial map
// to a given value. It creates an initial map if this function does
// not have one. Note that this method does not copy the initial map
// if it has one already, but simply replaces it with the new value.
// Instances created afterwards will have a map whose [[class]] is
// set to 'value', but there is no guarantees on instances created
// before.
Object* SetInstanceClassName(String* name);
// Returns if this function has been compiled to native code yet.
inline bool is_compiled();
// Casting.
static inline JSFunction* cast(Object* obj);
// Dispatched behavior.
#ifdef DEBUG
void JSFunctionPrint();
void JSFunctionVerify();
#endif
// Returns the number of allocated literals.
inline int NumberOfLiterals();
// Retrieve the global context from a function's literal array.
static Context* GlobalContextFromLiterals(FixedArray* literals);
// Layout descriptors.
static const int kPrototypeOrInitialMapOffset = JSObject::kHeaderSize;
static const int kSharedFunctionInfoOffset =
kPrototypeOrInitialMapOffset + kPointerSize;
static const int kContextOffset = kSharedFunctionInfoOffset + kPointerSize;
static const int kLiteralsOffset = kContextOffset + kPointerSize;
static const int kSize = kLiteralsOffset + kPointerSize;
// Layout of the literals array.
static const int kLiteralsPrefixSize = 1;
static const int kLiteralGlobalContextIndex = 0;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(JSFunction);
};
Split window support from V8. Here is a description of the background and design of split window in Chrome and V8: https://docs.google.com/a/google.com/Doc?id=chhjkpg_47fwddxbfr This change list splits the window object into two parts: 1) an inner window object used as the global object of contexts; 2) an outer window object exposed to JavaScript and accessible by the name 'window'. Firefox did it awhile ago, here are some discussions: https://wiki.mozilla.org/Gecko:SplitWindow. One additional benefit of splitting window in Chrome is that accessing global variables don't need security checks anymore, it can improve applications that use many global variables. V8 support of split window: There are a small number of changes on V8 api to support split window: Security context is removed from V8, so does related API functions; A global object can be detached from its context and reused by a new context; Access checks on an object template can be turned on/off by default; An object can turn on its access checks later; V8 has a new object type, ApiGlobalObject, which is the outer window object type. The existing JSGlobalObject becomes the inner window object type. Security checks are moved from JSGlobalObject to ApiGlobalObject. ApiGlobalObject is the one exposed to JavaScript, it is accessible through Context::Global(). ApiGlobalObject's prototype is set to JSGlobalObject so that property lookups are forwarded to JSGlobalObject. ApiGlobalObject forwards all other property access requests to JSGlobalObject, such as SetProperty, DeleteProperty, etc. Security token is moved to a global context, and ApiGlobalObject has a reference to its global context. JSGlobalObject has a reference to its global context as well. When accessing properties on a global object in JavaScript, the domain security check is performed by comparing the security token of the lexical context (Top::global_context()) to the token of global object's context. The check is only needed when the receiver is a window object, such as 'window.document'. Accessing global variables, such as 'var foo = 3; foo' does not need checks because the receiver is the inner window object. When an outer window is detached from its global context (when a frame navigates away from a page), it is completely detached from the inner window. A new context is created for the new page, and the outer global object is reused. At this point, the access check on the DOMWindow wrapper of the old context is turned on. The code in old context is still able to access DOMWindow properties, but it has to go through domain security checks. It is debatable on how to implement the outer window object. Currently each property access function has to check if the receiver is ApiGlobalObject type. This approach might be error-prone that one may forget to check the receiver when adding new functions. It is unlikely a performance issue because accessing global variables are more common than 'window.foo' style coding. I am still working on the ARM port, and I'd like to hear comments and suggestions on the best way to support it in V8. Review URL: http://codereview.chromium.org/7366 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@540 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-10-21 19:07:58 +00:00
// JSGlobalProxy's prototype must be a JSGlobalObject or null,
// and the prototype is hidden. JSGlobalProxy always delegates
// property accesses to its prototype if the prototype is not null.
//
// A JSGlobalProxy can be reinitialized which will preserve its identity.
//
// Accessing a JSGlobalProxy requires security check.
class JSGlobalProxy : public JSObject {
public:
// [context]: the owner global context of this proxy object.
// It is null value if this object is not used by any context.
DECL_ACCESSORS(context, Object)
// Casting.
static inline JSGlobalProxy* cast(Object* obj);
// Dispatched behavior.
#ifdef DEBUG
void JSGlobalProxyPrint();
void JSGlobalProxyVerify();
#endif
// Layout description.
static const int kContextOffset = JSObject::kHeaderSize;
static const int kSize = kContextOffset + kPointerSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(JSGlobalProxy);
};
// Forward declaration.
class JSBuiltinsObject;
// Common super class for JavaScript global objects and the special
// builtins global objects.
class GlobalObject: public JSObject {
public:
// [builtins]: the object holding the runtime routines written in JS.
DECL_ACCESSORS(builtins, JSBuiltinsObject)
// [global context]: the global context corresponding to this global object.
DECL_ACCESSORS(global_context, Context)
Split window support from V8. Here is a description of the background and design of split window in Chrome and V8: https://docs.google.com/a/google.com/Doc?id=chhjkpg_47fwddxbfr This change list splits the window object into two parts: 1) an inner window object used as the global object of contexts; 2) an outer window object exposed to JavaScript and accessible by the name 'window'. Firefox did it awhile ago, here are some discussions: https://wiki.mozilla.org/Gecko:SplitWindow. One additional benefit of splitting window in Chrome is that accessing global variables don't need security checks anymore, it can improve applications that use many global variables. V8 support of split window: There are a small number of changes on V8 api to support split window: Security context is removed from V8, so does related API functions; A global object can be detached from its context and reused by a new context; Access checks on an object template can be turned on/off by default; An object can turn on its access checks later; V8 has a new object type, ApiGlobalObject, which is the outer window object type. The existing JSGlobalObject becomes the inner window object type. Security checks are moved from JSGlobalObject to ApiGlobalObject. ApiGlobalObject is the one exposed to JavaScript, it is accessible through Context::Global(). ApiGlobalObject's prototype is set to JSGlobalObject so that property lookups are forwarded to JSGlobalObject. ApiGlobalObject forwards all other property access requests to JSGlobalObject, such as SetProperty, DeleteProperty, etc. Security token is moved to a global context, and ApiGlobalObject has a reference to its global context. JSGlobalObject has a reference to its global context as well. When accessing properties on a global object in JavaScript, the domain security check is performed by comparing the security token of the lexical context (Top::global_context()) to the token of global object's context. The check is only needed when the receiver is a window object, such as 'window.document'. Accessing global variables, such as 'var foo = 3; foo' does not need checks because the receiver is the inner window object. When an outer window is detached from its global context (when a frame navigates away from a page), it is completely detached from the inner window. A new context is created for the new page, and the outer global object is reused. At this point, the access check on the DOMWindow wrapper of the old context is turned on. The code in old context is still able to access DOMWindow properties, but it has to go through domain security checks. It is debatable on how to implement the outer window object. Currently each property access function has to check if the receiver is ApiGlobalObject type. This approach might be error-prone that one may forget to check the receiver when adding new functions. It is unlikely a performance issue because accessing global variables are more common than 'window.foo' style coding. I am still working on the ARM port, and I'd like to hear comments and suggestions on the best way to support it in V8. Review URL: http://codereview.chromium.org/7366 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@540 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-10-21 19:07:58 +00:00
// [global receiver]: the global receiver object of the context
DECL_ACCESSORS(global_receiver, JSObject)
// Retrieve the property cell used to store a property.
Object* GetPropertyCell(LookupResult* result);
// Ensure that the global object has a cell for the given property name.
Object* EnsurePropertyCell(String* name);
// Casting.
static inline GlobalObject* cast(Object* obj);
// Layout description.
static const int kBuiltinsOffset = JSObject::kHeaderSize;
static const int kGlobalContextOffset = kBuiltinsOffset + kPointerSize;
Split window support from V8. Here is a description of the background and design of split window in Chrome and V8: https://docs.google.com/a/google.com/Doc?id=chhjkpg_47fwddxbfr This change list splits the window object into two parts: 1) an inner window object used as the global object of contexts; 2) an outer window object exposed to JavaScript and accessible by the name 'window'. Firefox did it awhile ago, here are some discussions: https://wiki.mozilla.org/Gecko:SplitWindow. One additional benefit of splitting window in Chrome is that accessing global variables don't need security checks anymore, it can improve applications that use many global variables. V8 support of split window: There are a small number of changes on V8 api to support split window: Security context is removed from V8, so does related API functions; A global object can be detached from its context and reused by a new context; Access checks on an object template can be turned on/off by default; An object can turn on its access checks later; V8 has a new object type, ApiGlobalObject, which is the outer window object type. The existing JSGlobalObject becomes the inner window object type. Security checks are moved from JSGlobalObject to ApiGlobalObject. ApiGlobalObject is the one exposed to JavaScript, it is accessible through Context::Global(). ApiGlobalObject's prototype is set to JSGlobalObject so that property lookups are forwarded to JSGlobalObject. ApiGlobalObject forwards all other property access requests to JSGlobalObject, such as SetProperty, DeleteProperty, etc. Security token is moved to a global context, and ApiGlobalObject has a reference to its global context. JSGlobalObject has a reference to its global context as well. When accessing properties on a global object in JavaScript, the domain security check is performed by comparing the security token of the lexical context (Top::global_context()) to the token of global object's context. The check is only needed when the receiver is a window object, such as 'window.document'. Accessing global variables, such as 'var foo = 3; foo' does not need checks because the receiver is the inner window object. When an outer window is detached from its global context (when a frame navigates away from a page), it is completely detached from the inner window. A new context is created for the new page, and the outer global object is reused. At this point, the access check on the DOMWindow wrapper of the old context is turned on. The code in old context is still able to access DOMWindow properties, but it has to go through domain security checks. It is debatable on how to implement the outer window object. Currently each property access function has to check if the receiver is ApiGlobalObject type. This approach might be error-prone that one may forget to check the receiver when adding new functions. It is unlikely a performance issue because accessing global variables are more common than 'window.foo' style coding. I am still working on the ARM port, and I'd like to hear comments and suggestions on the best way to support it in V8. Review URL: http://codereview.chromium.org/7366 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@540 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-10-21 19:07:58 +00:00
static const int kGlobalReceiverOffset = kGlobalContextOffset + kPointerSize;
static const int kHeaderSize = kGlobalReceiverOffset + kPointerSize;
private:
friend class AGCCVersionRequiresThisClassToHaveAFriendSoHereItIs;
DISALLOW_IMPLICIT_CONSTRUCTORS(GlobalObject);
};
// JavaScript global object.
class JSGlobalObject: public GlobalObject {
public:
// Casting.
static inline JSGlobalObject* cast(Object* obj);
// Dispatched behavior.
#ifdef DEBUG
void JSGlobalObjectPrint();
void JSGlobalObjectVerify();
#endif
// Layout description.
Split window support from V8. Here is a description of the background and design of split window in Chrome and V8: https://docs.google.com/a/google.com/Doc?id=chhjkpg_47fwddxbfr This change list splits the window object into two parts: 1) an inner window object used as the global object of contexts; 2) an outer window object exposed to JavaScript and accessible by the name 'window'. Firefox did it awhile ago, here are some discussions: https://wiki.mozilla.org/Gecko:SplitWindow. One additional benefit of splitting window in Chrome is that accessing global variables don't need security checks anymore, it can improve applications that use many global variables. V8 support of split window: There are a small number of changes on V8 api to support split window: Security context is removed from V8, so does related API functions; A global object can be detached from its context and reused by a new context; Access checks on an object template can be turned on/off by default; An object can turn on its access checks later; V8 has a new object type, ApiGlobalObject, which is the outer window object type. The existing JSGlobalObject becomes the inner window object type. Security checks are moved from JSGlobalObject to ApiGlobalObject. ApiGlobalObject is the one exposed to JavaScript, it is accessible through Context::Global(). ApiGlobalObject's prototype is set to JSGlobalObject so that property lookups are forwarded to JSGlobalObject. ApiGlobalObject forwards all other property access requests to JSGlobalObject, such as SetProperty, DeleteProperty, etc. Security token is moved to a global context, and ApiGlobalObject has a reference to its global context. JSGlobalObject has a reference to its global context as well. When accessing properties on a global object in JavaScript, the domain security check is performed by comparing the security token of the lexical context (Top::global_context()) to the token of global object's context. The check is only needed when the receiver is a window object, such as 'window.document'. Accessing global variables, such as 'var foo = 3; foo' does not need checks because the receiver is the inner window object. When an outer window is detached from its global context (when a frame navigates away from a page), it is completely detached from the inner window. A new context is created for the new page, and the outer global object is reused. At this point, the access check on the DOMWindow wrapper of the old context is turned on. The code in old context is still able to access DOMWindow properties, but it has to go through domain security checks. It is debatable on how to implement the outer window object. Currently each property access function has to check if the receiver is ApiGlobalObject type. This approach might be error-prone that one may forget to check the receiver when adding new functions. It is unlikely a performance issue because accessing global variables are more common than 'window.foo' style coding. I am still working on the ARM port, and I'd like to hear comments and suggestions on the best way to support it in V8. Review URL: http://codereview.chromium.org/7366 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@540 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-10-21 19:07:58 +00:00
static const int kSize = GlobalObject::kHeaderSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(JSGlobalObject);
};
// Builtins global object which holds the runtime routines written in
// JavaScript.
class JSBuiltinsObject: public GlobalObject {
public:
// Accessors for the runtime routines written in JavaScript.
inline Object* javascript_builtin(Builtins::JavaScript id);
inline void set_javascript_builtin(Builtins::JavaScript id, Object* value);
// Casting.
static inline JSBuiltinsObject* cast(Object* obj);
// Dispatched behavior.
#ifdef DEBUG
void JSBuiltinsObjectPrint();
void JSBuiltinsObjectVerify();
#endif
// Layout description. The size of the builtins object includes
// room for one pointer per runtime routine written in javascript.
static const int kJSBuiltinsCount = Builtins::id_count;
static const int kJSBuiltinsOffset = GlobalObject::kHeaderSize;
static const int kSize =
kJSBuiltinsOffset + (kJSBuiltinsCount * kPointerSize);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(JSBuiltinsObject);
};
// Representation for JS Wrapper objects, String, Number, Boolean, Date, etc.
class JSValue: public JSObject {
public:
// [value]: the object being wrapped.
DECL_ACCESSORS(value, Object)
// Casting.
static inline JSValue* cast(Object* obj);
// Dispatched behavior.
#ifdef DEBUG
void JSValuePrint();
void JSValueVerify();
#endif
// Layout description.
static const int kValueOffset = JSObject::kHeaderSize;
static const int kSize = kValueOffset + kPointerSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(JSValue);
};
// Regular expressions
// The regular expression holds a single reference to a FixedArray in
// the kDataOffset field.
// The FixedArray contains the following data:
// - tag : type of regexp implementation (not compiled yet, atom or irregexp)
// - reference to the original source string
// - reference to the original flag string
// If it is an atom regexp
// - a reference to a literal string to search for
// If it is an irregexp regexp:
// - a reference to code for ASCII inputs (bytecode or compiled).
// - a reference to code for UC16 inputs (bytecode or compiled).
// - max number of registers used by irregexp implementations.
// - number of capture registers (output values) of the regexp.
class JSRegExp: public JSObject {
public:
// Meaning of Type:
// NOT_COMPILED: Initial value. No data has been stored in the JSRegExp yet.
// ATOM: A simple string to match against using an indexOf operation.
// IRREGEXP: Compiled with Irregexp.
// IRREGEXP_NATIVE: Compiled to native code with Irregexp.
enum Type { NOT_COMPILED, ATOM, IRREGEXP };
enum Flag { NONE = 0, GLOBAL = 1, IGNORE_CASE = 2, MULTILINE = 4 };
class Flags {
public:
explicit Flags(uint32_t value) : value_(value) { }
bool is_global() { return (value_ & GLOBAL) != 0; }
bool is_ignore_case() { return (value_ & IGNORE_CASE) != 0; }
bool is_multiline() { return (value_ & MULTILINE) != 0; }
uint32_t value() { return value_; }
private:
uint32_t value_;
};
DECL_ACCESSORS(data, Object)
inline Type TypeTag();
inline int CaptureCount();
inline Flags GetFlags();
inline String* Pattern();
inline Object* DataAt(int index);
// Set implementation data after the object has been prepared.
inline void SetDataAt(int index, Object* value);
static int code_index(bool is_ascii) {
if (is_ascii) {
return kIrregexpASCIICodeIndex;
} else {
return kIrregexpUC16CodeIndex;
}
}
static inline JSRegExp* cast(Object* obj);
// Dispatched behavior.
#ifdef DEBUG
void JSRegExpVerify();
#endif
static const int kDataOffset = JSObject::kHeaderSize;
static const int kSize = kDataOffset + kPointerSize;
// Indices in the data array.
static const int kTagIndex = 0;
static const int kSourceIndex = kTagIndex + 1;
static const int kFlagsIndex = kSourceIndex + 1;
static const int kDataIndex = kFlagsIndex + 1;
// The data fields are used in different ways depending on the
// value of the tag.
// Atom regexps (literal strings).
static const int kAtomPatternIndex = kDataIndex;
static const int kAtomDataSize = kAtomPatternIndex + 1;
// Irregexp compiled code or bytecode for ASCII. If compilation
// fails, this fields hold an exception object that should be
// thrown if the regexp is used again.
static const int kIrregexpASCIICodeIndex = kDataIndex;
// Irregexp compiled code or bytecode for UC16. If compilation
// fails, this fields hold an exception object that should be
// thrown if the regexp is used again.
static const int kIrregexpUC16CodeIndex = kDataIndex + 1;
// Maximal number of registers used by either ASCII or UC16.
// Only used to check that there is enough stack space
static const int kIrregexpMaxRegisterCountIndex = kDataIndex + 2;
// Number of captures in the compiled regexp.
static const int kIrregexpCaptureCountIndex = kDataIndex + 3;
static const int kIrregexpDataSize = kIrregexpCaptureCountIndex + 1;
};
class CompilationCacheShape {
public:
static inline bool IsMatch(HashTableKey* key, Object* value) {
return key->IsMatch(value);
}
static inline uint32_t Hash(HashTableKey* key) {
return key->Hash();
}
static inline uint32_t HashForObject(HashTableKey* key, Object* object) {
return key->HashForObject(object);
}
static Object* AsObject(HashTableKey* key) {
return key->AsObject();
}
static const int kPrefixSize = 0;
static const int kEntrySize = 2;
};
class CompilationCacheTable: public HashTable<CompilationCacheShape,
HashTableKey*> {
public:
// Find cached value for a string key, otherwise return null.
Object* Lookup(String* src);
Object* LookupEval(String* src, Context* context);
Object* LookupRegExp(String* source, JSRegExp::Flags flags);
Object* Put(String* src, Object* value);
Object* PutEval(String* src, Context* context, Object* value);
Object* PutRegExp(String* src, JSRegExp::Flags flags, FixedArray* value);
static inline CompilationCacheTable* cast(Object* obj);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(CompilationCacheTable);
};
enum AllowNullsFlag {ALLOW_NULLS, DISALLOW_NULLS};
enum RobustnessFlag {ROBUST_STRING_TRAVERSAL, FAST_STRING_TRAVERSAL};
class StringHasher {
public:
inline StringHasher(int length);
// Returns true if the hash of this string can be computed without
// looking at the contents.
inline bool has_trivial_hash();
// Add a character to the hash and update the array index calculation.
inline void AddCharacter(uc32 c);
// Adds a character to the hash but does not update the array index
// calculation. This can only be called when it has been verified
// that the input is not an array index.
inline void AddCharacterNoIndex(uc32 c);
// Returns the value to store in the hash field of a string with
// the given length and contents.
uint32_t GetHashField();
// Returns true if the characters seen so far make up a legal array
// index.
bool is_array_index() { return is_array_index_; }
bool is_valid() { return is_valid_; }
void invalidate() { is_valid_ = false; }
private:
uint32_t array_index() {
ASSERT(is_array_index());
return array_index_;
}
inline uint32_t GetHash();
int length_;
uint32_t raw_running_hash_;
uint32_t array_index_;
bool is_array_index_;
bool is_first_char_;
bool is_valid_;
};
// The characteristics of a string are stored in its map. Retrieving these
// few bits of information is moderately expensive, involving two memory
// loads where the second is dependent on the first. To improve efficiency
// the shape of the string is given its own class so that it can be retrieved
// once and used for several string operations. A StringShape is small enough
// to be passed by value and is immutable, but be aware that flattening a
// string can potentially alter its shape. Also be aware that a GC caused by
// something else can alter the shape of a string due to ConsString
// shortcutting. Keeping these restrictions in mind has proven to be error-
// prone and so we no longer put StringShapes in variables unless there is a
// concrete performance benefit at that particular point in the code.
class StringShape BASE_EMBEDDED {
public:
inline explicit StringShape(String* s);
inline explicit StringShape(Map* s);
inline explicit StringShape(InstanceType t);
inline bool IsSequential();
inline bool IsExternal();
inline bool IsCons();
inline bool IsExternalAscii();
inline bool IsExternalTwoByte();
inline bool IsSequentialAscii();
inline bool IsSequentialTwoByte();
inline bool IsSymbol();
inline StringRepresentationTag representation_tag();
inline uint32_t full_representation_tag();
inline uint32_t size_tag();
#ifdef DEBUG
inline uint32_t type() { return type_; }
inline void invalidate() { valid_ = false; }
inline bool valid() { return valid_; }
#else
inline void invalidate() { }
#endif
private:
uint32_t type_;
#ifdef DEBUG
inline void set_valid() { valid_ = true; }
bool valid_;
#else
inline void set_valid() { }
#endif
};
// The String abstract class captures JavaScript string values:
//
// Ecma-262:
// 4.3.16 String Value
// A string value is a member of the type String and is a finite
// ordered sequence of zero or more 16-bit unsigned integer values.
//
// All string values have a length field.
class String: public HeapObject {
public:
// Get and set the length of the string.
inline int length();
inline void set_length(int value);
// Get and set the uninterpreted length field of the string. Notice
// that the length field is also used to cache the hash value of
// strings. In order to get or set the actual length of the string
// use the length() and set_length methods.
inline uint32_t length_field();
inline void set_length_field(uint32_t value);
inline bool IsAsciiRepresentation();
inline bool IsTwoByteRepresentation();
// Get and set individual two byte chars in the string.
inline void Set(int index, uint16_t value);
// Get individual two byte char in the string. Repeated calls
// to this method are not efficient unless the string is flat.
inline uint16_t Get(int index);
// Try to flatten the top level ConsString that is hiding behind this
// string. This is a no-op unless the string is a ConsString. Flatten
// mutates the ConsString and might return a failure.
Object* TryFlatten();
// Try to flatten the string. Checks first inline to see if it is necessary.
// Do not handle allocation failures. After calling TryFlattenIfNotFlat, the
// string could still be a ConsString, in which case a failure is returned.
// Use FlattenString from Handles.cc to be sure to flatten.
inline Object* TryFlattenIfNotFlat();
Vector<const char> ToAsciiVector();
Vector<const uc16> ToUC16Vector();
// Mark the string as an undetectable object. It only applies to
// ascii and two byte string types.
bool MarkAsUndetectable();
// Return a substring.
Object* SubString(int from, int to);
// String equality operations.
inline bool Equals(String* other);
bool IsEqualTo(Vector<const char> str);
// Return a UTF8 representation of the string. The string is null
// terminated but may optionally contain nulls. Length is returned
// in length_output if length_output is not a null pointer The string
// should be nearly flat, otherwise the performance of this method may
// be very slow (quadratic in the length). Setting robustness_flag to
// ROBUST_STRING_TRAVERSAL invokes behaviour that is robust This means it
// handles unexpected data without causing assert failures and it does not
// do any heap allocations. This is useful when printing stack traces.
SmartPointer<char> ToCString(AllowNullsFlag allow_nulls,
RobustnessFlag robustness_flag,
int offset,
int length,
int* length_output = 0);
SmartPointer<char> ToCString(
AllowNullsFlag allow_nulls = DISALLOW_NULLS,
RobustnessFlag robustness_flag = FAST_STRING_TRAVERSAL,
int* length_output = 0);
int Utf8Length();
// Return a 16 bit Unicode representation of the string.
// The string should be nearly flat, otherwise the performance of
// of this method may be very bad. Setting robustness_flag to
// ROBUST_STRING_TRAVERSAL invokes behaviour that is robust This means it
// handles unexpected data without causing assert failures and it does not
// do any heap allocations. This is useful when printing stack traces.
SmartPointer<uc16> ToWideCString(
RobustnessFlag robustness_flag = FAST_STRING_TRAVERSAL);
// Tells whether the hash code has been computed.
inline bool HasHashCode();
// Returns a hash value used for the property table
inline uint32_t Hash();
static uint32_t ComputeLengthAndHashField(unibrow::CharacterStream* buffer,
int length);
static bool ComputeArrayIndex(unibrow::CharacterStream* buffer,
uint32_t* index,
int length);
// Externalization.
bool MakeExternal(v8::String::ExternalStringResource* resource);
bool MakeExternal(v8::String::ExternalAsciiStringResource* resource);
// Conversion.
inline bool AsArrayIndex(uint32_t* index);
// Casting.
static inline String* cast(Object* obj);
void PrintOn(FILE* out);
// For use during stack traces. Performs rudimentary sanity check.
bool LooksValid();
// Dispatched behavior.
void StringShortPrint(StringStream* accumulator);
#ifdef DEBUG
void StringPrint();
void StringVerify();
#endif
inline bool IsFlat();
// Layout description.
static const int kLengthOffset = HeapObject::kHeaderSize;
static const int kSize = kLengthOffset + kIntSize;
// Notice: kSize is not pointer-size aligned if pointers are 64-bit.
// Maximum number of characters to consider when trying to convert a string
// value into an array index.
static const int kMaxArrayIndexSize = 10;
// Max ascii char code.
Split window support from V8. Here is a description of the background and design of split window in Chrome and V8: https://docs.google.com/a/google.com/Doc?id=chhjkpg_47fwddxbfr This change list splits the window object into two parts: 1) an inner window object used as the global object of contexts; 2) an outer window object exposed to JavaScript and accessible by the name 'window'. Firefox did it awhile ago, here are some discussions: https://wiki.mozilla.org/Gecko:SplitWindow. One additional benefit of splitting window in Chrome is that accessing global variables don't need security checks anymore, it can improve applications that use many global variables. V8 support of split window: There are a small number of changes on V8 api to support split window: Security context is removed from V8, so does related API functions; A global object can be detached from its context and reused by a new context; Access checks on an object template can be turned on/off by default; An object can turn on its access checks later; V8 has a new object type, ApiGlobalObject, which is the outer window object type. The existing JSGlobalObject becomes the inner window object type. Security checks are moved from JSGlobalObject to ApiGlobalObject. ApiGlobalObject is the one exposed to JavaScript, it is accessible through Context::Global(). ApiGlobalObject's prototype is set to JSGlobalObject so that property lookups are forwarded to JSGlobalObject. ApiGlobalObject forwards all other property access requests to JSGlobalObject, such as SetProperty, DeleteProperty, etc. Security token is moved to a global context, and ApiGlobalObject has a reference to its global context. JSGlobalObject has a reference to its global context as well. When accessing properties on a global object in JavaScript, the domain security check is performed by comparing the security token of the lexical context (Top::global_context()) to the token of global object's context. The check is only needed when the receiver is a window object, such as 'window.document'. Accessing global variables, such as 'var foo = 3; foo' does not need checks because the receiver is the inner window object. When an outer window is detached from its global context (when a frame navigates away from a page), it is completely detached from the inner window. A new context is created for the new page, and the outer global object is reused. At this point, the access check on the DOMWindow wrapper of the old context is turned on. The code in old context is still able to access DOMWindow properties, but it has to go through domain security checks. It is debatable on how to implement the outer window object. Currently each property access function has to check if the receiver is ApiGlobalObject type. This approach might be error-prone that one may forget to check the receiver when adding new functions. It is unlikely a performance issue because accessing global variables are more common than 'window.foo' style coding. I am still working on the ARM port, and I'd like to hear comments and suggestions on the best way to support it in V8. Review URL: http://codereview.chromium.org/7366 git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@540 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2008-10-21 19:07:58 +00:00
static const int kMaxAsciiCharCode = unibrow::Utf8::kMaxOneByteChar;
static const unsigned kMaxAsciiCharCodeU = unibrow::Utf8::kMaxOneByteChar;
static const int kMaxUC16CharCode = 0xffff;
// Minimum length for a cons string.
static const int kMinNonFlatLength = 13;
// Mask constant for checking if a string has a computed hash code
// and if it is an array index. The least significant bit indicates
// whether a hash code has been computed. If the hash code has been
// computed the 2nd bit tells whether the string can be used as an
// array index.
static const int kHashComputedMask = 1;
static const int kIsArrayIndexMask = 1 << 1;
static const int kNofLengthBitFields = 2;
// Array index strings this short can keep their index in the hash
// field.
static const int kMaxCachedArrayIndexLength = 7;
// Shift constants for retrieving length and hash code from
// length/hash field.
static const int kHashShift = kNofLengthBitFields;
static const int kShortLengthShift = kHashShift + kShortStringTag;
static const int kMediumLengthShift = kHashShift + kMediumStringTag;
static const int kLongLengthShift = kHashShift + kLongStringTag;
// Maximal string length that can be stored in the hash/length field for
// different types of strings.
static const int kMaxShortSize = (1 << (32 - kShortLengthShift)) - 1;
static const int kMaxMediumSize = (1 << (32 - kMediumLengthShift)) - 1;
static const int kMaxLength = (1 << (32 - kLongLengthShift)) - 1;
// Limit for truncation in short printing.
static const int kMaxShortPrintLength = 1024;
// Support for regular expressions.
const uc16* GetTwoByteData();
const uc16* GetTwoByteData(unsigned start);
// Support for StringInputBuffer
static const unibrow::byte* ReadBlock(String* input,
unibrow::byte* util_buffer,
unsigned capacity,
unsigned* remaining,
unsigned* offset);
static const unibrow::byte* ReadBlock(String** input,
unibrow::byte* util_buffer,
unsigned capacity,
unsigned* remaining,
unsigned* offset);
// Helper function for flattening strings.
template <typename sinkchar>
static void WriteToFlat(String* source,
sinkchar* sink,
int from,
int to);
protected:
class ReadBlockBuffer {
public:
ReadBlockBuffer(unibrow::byte* util_buffer_,
unsigned cursor_,
unsigned capacity_,
unsigned remaining_) :
util_buffer(util_buffer_),
cursor(cursor_),
capacity(capacity_),
remaining(remaining_) {
}
unibrow::byte* util_buffer;
unsigned cursor;
unsigned capacity;
unsigned remaining;
};
static inline const unibrow::byte* ReadBlock(String* input,
ReadBlockBuffer* buffer,
unsigned* offset,
unsigned max_chars);
static void ReadBlockIntoBuffer(String* input,
ReadBlockBuffer* buffer,
unsigned* offset_ptr,
unsigned max_chars);
private:
// Slow case of String::Equals. This implementation works on any strings
// but it is most efficient on strings that are almost flat.
bool SlowEquals(String* other);
// Slow case of AsArrayIndex.
bool SlowAsArrayIndex(uint32_t* index);
// Compute and set the hash code.
uint32_t ComputeAndSetHash();
DISALLOW_IMPLICIT_CONSTRUCTORS(String);
};
// The SeqString abstract class captures sequential string values.
class SeqString: public String {
public:
// Casting.
static inline SeqString* cast(Object* obj);
// Dispatched behaviour.
// For regexp code.
uint16_t* SeqStringGetTwoByteAddress();
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(SeqString);
};
// The AsciiString class captures sequential ascii string objects.
// Each character in the AsciiString is an ascii character.
class SeqAsciiString: public SeqString {
public:
// Dispatched behavior.
inline uint16_t SeqAsciiStringGet(int index);
inline void SeqAsciiStringSet(int index, uint16_t value);
// Get the address of the characters in this string.
inline Address GetCharsAddress();
inline char* GetChars();
// Casting
static inline SeqAsciiString* cast(Object* obj);
// Garbage collection support. This method is called by the
// garbage collector to compute the actual size of an AsciiString
// instance.
inline int SeqAsciiStringSize(InstanceType instance_type);
// Computes the size for an AsciiString instance of a given length.
static int SizeFor(int length) {
return OBJECT_SIZE_ALIGN(kHeaderSize + length * kCharSize);
}
// Layout description.
static const int kHeaderSize = String::kSize;
static const int kAlignedSize = POINTER_SIZE_ALIGN(kHeaderSize);
// Support for StringInputBuffer.
inline void SeqAsciiStringReadBlockIntoBuffer(ReadBlockBuffer* buffer,
unsigned* offset,
unsigned chars);
inline const unibrow::byte* SeqAsciiStringReadBlock(unsigned* remaining,
unsigned* offset,
unsigned chars);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(SeqAsciiString);
};
// The TwoByteString class captures sequential unicode string objects.
// Each character in the TwoByteString is a two-byte uint16_t.
class SeqTwoByteString: public SeqString {
public:
// Dispatched behavior.
inline uint16_t SeqTwoByteStringGet(int index);
inline void SeqTwoByteStringSet(int index, uint16_t value);
// Get the address of the characters in this string.
inline Address GetCharsAddress();
inline uc16* GetChars();
// For regexp code.
const uint16_t* SeqTwoByteStringGetData(unsigned start);
// Casting
static inline SeqTwoByteString* cast(Object* obj);
// Garbage collection support. This method is called by the
// garbage collector to compute the actual size of a TwoByteString
// instance.
inline int SeqTwoByteStringSize(InstanceType instance_type);
// Computes the size for a TwoByteString instance of a given length.
static int SizeFor(int length) {
return OBJECT_SIZE_ALIGN(kHeaderSize + length * kShortSize);
}
// Layout description.
static const int kHeaderSize = String::kSize;
static const int kAlignedSize = POINTER_SIZE_ALIGN(kHeaderSize);
// Support for StringInputBuffer.
inline void SeqTwoByteStringReadBlockIntoBuffer(ReadBlockBuffer* buffer,
unsigned* offset_ptr,
unsigned chars);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(SeqTwoByteString);
};
// The ConsString class describes string values built by using the
// addition operator on strings. A ConsString is a pair where the
// first and second components are pointers to other string values.
// One or both components of a ConsString can be pointers to other
// ConsStrings, creating a binary tree of ConsStrings where the leaves
// are non-ConsString string values. The string value represented by
// a ConsString can be obtained by concatenating the leaf string
// values in a left-to-right depth-first traversal of the tree.
class ConsString: public String {
public:
// First string of the cons cell.
inline String* first();
// Doesn't check that the result is a string, even in debug mode. This is
// useful during GC where the mark bits confuse the checks.
inline Object* unchecked_first();
inline void set_first(String* first,
WriteBarrierMode mode = UPDATE_WRITE_BARRIER);
// Second string of the cons cell.
inline String* second();
// Doesn't check that the result is a string, even in debug mode. This is
// useful during GC where the mark bits confuse the checks.
inline Object* unchecked_second();
inline void set_second(String* second,
WriteBarrierMode mode = UPDATE_WRITE_BARRIER);
// Dispatched behavior.
uint16_t ConsStringGet(int index);
// Casting.
static inline ConsString* cast(Object* obj);
// Garbage collection support. This method is called during garbage
// collection to iterate through the heap pointers in the body of
// the ConsString.
void ConsStringIterateBody(ObjectVisitor* v);
// Layout description.
static const int kFirstOffset = POINTER_SIZE_ALIGN(String::kSize);
static const int kSecondOffset = kFirstOffset + kPointerSize;
static const int kSize = kSecondOffset + kPointerSize;
// Support for StringInputBuffer.
inline const unibrow::byte* ConsStringReadBlock(ReadBlockBuffer* buffer,
unsigned* offset_ptr,
unsigned chars);
inline void ConsStringReadBlockIntoBuffer(ReadBlockBuffer* buffer,
unsigned* offset_ptr,
unsigned chars);
// Minimum length for a cons string.
static const int kMinLength = 13;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ConsString);
};
// The ExternalString class describes string values that are backed by
// a string resource that lies outside the V8 heap. ExternalStrings
// consist of the length field common to all strings, a pointer to the
// external resource. It is important to ensure (externally) that the
// resource is not deallocated while the ExternalString is live in the
// V8 heap.
//
// The API expects that all ExternalStrings are created through the
// API. Therefore, ExternalStrings should not be used internally.
class ExternalString: public String {
public:
// Casting
static inline ExternalString* cast(Object* obj);
// Layout description.
static const int kResourceOffset = POINTER_SIZE_ALIGN(String::kSize);
static const int kSize = kResourceOffset + kPointerSize;
STATIC_CHECK(kResourceOffset == Internals::kStringResourceOffset);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalString);
};
// The ExternalAsciiString class is an external string backed by an
// ASCII string.
class ExternalAsciiString: public ExternalString {
public:
typedef v8::String::ExternalAsciiStringResource Resource;
// The underlying resource.
inline Resource* resource();
inline void set_resource(Resource* buffer);
// Dispatched behavior.
uint16_t ExternalAsciiStringGet(int index);
// Casting.
static inline ExternalAsciiString* cast(Object* obj);
// Garbage collection support.
void ExternalAsciiStringIterateBody(ObjectVisitor* v);
// Support for StringInputBuffer.
const unibrow::byte* ExternalAsciiStringReadBlock(unsigned* remaining,
unsigned* offset,
unsigned chars);
inline void ExternalAsciiStringReadBlockIntoBuffer(ReadBlockBuffer* buffer,
unsigned* offset,
unsigned chars);
// Identify the map for the external string/symbol with a particular length.
static inline Map* StringMap(int length);
static inline Map* SymbolMap(int length);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalAsciiString);
};
// The ExternalTwoByteString class is an external string backed by a UTF-16
// encoded string.
class ExternalTwoByteString: public ExternalString {
public:
typedef v8::String::ExternalStringResource Resource;
// The underlying string resource.
inline Resource* resource();
inline void set_resource(Resource* buffer);
// Dispatched behavior.
uint16_t ExternalTwoByteStringGet(int index);
// For regexp code.
const uint16_t* ExternalTwoByteStringGetData(unsigned start);
// Casting.
static inline ExternalTwoByteString* cast(Object* obj);
// Garbage collection support.
void ExternalTwoByteStringIterateBody(ObjectVisitor* v);
// Support for StringInputBuffer.
void ExternalTwoByteStringReadBlockIntoBuffer(ReadBlockBuffer* buffer,
unsigned* offset_ptr,
unsigned chars);
// Identify the map for the external string/symbol with a particular length.
static inline Map* StringMap(int length);
static inline Map* SymbolMap(int length);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(ExternalTwoByteString);
};
// Utility superclass for stack-allocated objects that must be updated
// on gc. It provides two ways for the gc to update instances, either
// iterating or updating after gc.
class Relocatable BASE_EMBEDDED {
public:
inline Relocatable() : prev_(top_) { top_ = this; }
virtual ~Relocatable() {
ASSERT_EQ(top_, this);
top_ = prev_;
}
virtual void IterateInstance(ObjectVisitor* v) { }
virtual void PostGarbageCollection() { }
static void PostGarbageCollectionProcessing();
static int ArchiveSpacePerThread();
static char* ArchiveState(char* to);
static char* RestoreState(char* from);
static void Iterate(ObjectVisitor* v);
static void Iterate(ObjectVisitor* v, Relocatable* top);
static char* Iterate(ObjectVisitor* v, char* t);
private:
static Relocatable* top_;
Relocatable* prev_;
};
// A flat string reader provides random access to the contents of a
// string independent of the character width of the string. The handle
// must be valid as long as the reader is being used.
class FlatStringReader : public Relocatable {
public:
explicit FlatStringReader(Handle<String> str);
explicit FlatStringReader(Vector<const char> input);
void PostGarbageCollection();
inline uc32 Get(int index);
int length() { return length_; }
private:
String** str_;
bool is_ascii_;
int length_;
const void* start_;
};
// Note that StringInputBuffers are not valid across a GC! To fix this
// it would have to store a String Handle instead of a String* and
// AsciiStringReadBlock would have to be modified to use memcpy.
//
// StringInputBuffer is able to traverse any string regardless of how
// deeply nested a sequence of ConsStrings it is made of. However,
// performance will be better if deep strings are flattened before they
// are traversed. Since flattening requires memory allocation this is
// not always desirable, however (esp. in debugging situations).
class StringInputBuffer: public unibrow::InputBuffer<String, String*, 1024> {
public:
virtual void Seek(unsigned pos);
inline StringInputBuffer(): unibrow::InputBuffer<String, String*, 1024>() {}
inline StringInputBuffer(String* backing):
unibrow::InputBuffer<String, String*, 1024>(backing) {}
};
class SafeStringInputBuffer
: public unibrow::InputBuffer<String, String**, 256> {
public:
virtual void Seek(unsigned pos);
inline SafeStringInputBuffer()
: unibrow::InputBuffer<String, String**, 256>() {}
inline SafeStringInputBuffer(String** backing)
: unibrow::InputBuffer<String, String**, 256>(backing) {}
};
template <typename T>
class VectorIterator {
public:
VectorIterator(T* d, int l) : data_(Vector<const T>(d, l)), index_(0) { }
explicit VectorIterator(Vector<const T> data) : data_(data), index_(0) { }
T GetNext() { return data_[index_++]; }
bool has_more() { return index_ < data_.length(); }
private:
Vector<const T> data_;
int index_;
};
// The Oddball describes objects null, undefined, true, and false.
class Oddball: public HeapObject {
public:
// [to_string]: Cached to_string computed at startup.
DECL_ACCESSORS(to_string, String)
// [to_number]: Cached to_number computed at startup.
DECL_ACCESSORS(to_number, Object)
// Casting.
static inline Oddball* cast(Object* obj);
// Dispatched behavior.
void OddballIterateBody(ObjectVisitor* v);
#ifdef DEBUG
void OddballVerify();
#endif
// Initialize the fields.
Object* Initialize(const char* to_string, Object* to_number);
// Layout description.
static const int kToStringOffset = HeapObject::kHeaderSize;
static const int kToNumberOffset = kToStringOffset + kPointerSize;
static const int kSize = kToNumberOffset + kPointerSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(Oddball);
};
class JSGlobalPropertyCell: public HeapObject {
public:
// [value]: value of the global property.
DECL_ACCESSORS(value, Object)
// Casting.
static inline JSGlobalPropertyCell* cast(Object* obj);
// Dispatched behavior.
void JSGlobalPropertyCellIterateBody(ObjectVisitor* v);
#ifdef DEBUG
void JSGlobalPropertyCellVerify();
void JSGlobalPropertyCellPrint();
#endif
// Layout description.
static const int kValueOffset = HeapObject::kHeaderSize;
static const int kSize = kValueOffset + kPointerSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(JSGlobalPropertyCell);
};
// Proxy describes objects pointing from JavaScript to C structures.
// Since they cannot contain references to JS HeapObjects they can be
// placed in old_data_space.
class Proxy: public HeapObject {
public:
// [proxy]: field containing the address.
inline Address proxy();
inline void set_proxy(Address value);
// Casting.
static inline Proxy* cast(Object* obj);
// Dispatched behavior.
inline void ProxyIterateBody(ObjectVisitor* v);
#ifdef DEBUG
void ProxyPrint();
void ProxyVerify();
#endif
// Layout description.
static const int kProxyOffset = HeapObject::kHeaderSize;
static const int kSize = kProxyOffset + kPointerSize;
STATIC_CHECK(kProxyOffset == Internals::kProxyProxyOffset);
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(Proxy);
};
// The JSArray describes JavaScript Arrays
// Such an array can be in one of two modes:
// - fast, backing storage is a FixedArray and length <= elements.length();
// Please note: push and pop can be used to grow and shrink the array.
// - slow, backing storage is a HashTable with numbers as keys.
class JSArray: public JSObject {
public:
// [length]: The length property.
DECL_ACCESSORS(length, Object)
Object* JSArrayUpdateLengthFromIndex(uint32_t index, Object* value);
// Initialize the array with the given capacity. The function may
// fail due to out-of-memory situations, but only if the requested
// capacity is non-zero.
Object* Initialize(int capacity);
// Set the content of the array to the content of storage.
inline void SetContent(FixedArray* storage);
// Casting.
static inline JSArray* cast(Object* obj);
// Uses handles. Ensures that the fixed array backing the JSArray has at
// least the stated size.
inline void EnsureSize(int minimum_size_of_backing_fixed_array);
// Dispatched behavior.
#ifdef DEBUG
void JSArrayPrint();
void JSArrayVerify();
#endif
// Number of element slots to pre-allocate for an empty array.
static const int kPreallocatedArrayElements = 4;
// Layout description.
static const int kLengthOffset = JSObject::kHeaderSize;
static const int kSize = kLengthOffset + kPointerSize;
private:
// Expand the fixed array backing of a fast-case JSArray to at least
// the requested size.
void Expand(int minimum_size_of_backing_fixed_array);
DISALLOW_IMPLICIT_CONSTRUCTORS(JSArray);
};
// An accessor must have a getter, but can have no setter.
//
// When setting a property, V8 searches accessors in prototypes.
// If an accessor was found and it does not have a setter,
// the request is ignored.
//
// If the accessor in the prototype has the READ_ONLY property attribute, then
// a new value is added to the local object when the property is set.
// This shadows the accessor in the prototype.
class AccessorInfo: public Struct {
public:
DECL_ACCESSORS(getter, Object)
DECL_ACCESSORS(setter, Object)
DECL_ACCESSORS(data, Object)
DECL_ACCESSORS(name, Object)
DECL_ACCESSORS(flag, Smi)
DECL_ACCESSORS(load_stub_cache, Object)
inline bool all_can_read();
inline void set_all_can_read(bool value);
inline bool all_can_write();
inline void set_all_can_write(bool value);
inline bool prohibits_overwriting();
inline void set_prohibits_overwriting(bool value);
inline PropertyAttributes property_attributes();
inline void set_property_attributes(PropertyAttributes attributes);
static inline AccessorInfo* cast(Object* obj);
#ifdef DEBUG
void AccessorInfoPrint();
void AccessorInfoVerify();
#endif
static const int kGetterOffset = HeapObject::kHeaderSize;
static const int kSetterOffset = kGetterOffset + kPointerSize;
static const int kDataOffset = kSetterOffset + kPointerSize;
static const int kNameOffset = kDataOffset + kPointerSize;
static const int kFlagOffset = kNameOffset + kPointerSize;
static const int kLoadStubCacheOffset = kFlagOffset + kPointerSize;
static const int kSize = kLoadStubCacheOffset + kPointerSize;
private:
// Bit positions in flag.
static const int kAllCanReadBit = 0;
static const int kAllCanWriteBit = 1;
static const int kProhibitsOverwritingBit = 2;
class AttributesField: public BitField<PropertyAttributes, 3, 3> {};
DISALLOW_IMPLICIT_CONSTRUCTORS(AccessorInfo);
};
class AccessCheckInfo: public Struct {
public:
DECL_ACCESSORS(named_callback, Object)
DECL_ACCESSORS(indexed_callback, Object)
DECL_ACCESSORS(data, Object)
static inline AccessCheckInfo* cast(Object* obj);
#ifdef DEBUG
void AccessCheckInfoPrint();
void AccessCheckInfoVerify();
#endif
static const int kNamedCallbackOffset = HeapObject::kHeaderSize;
static const int kIndexedCallbackOffset = kNamedCallbackOffset + kPointerSize;
static const int kDataOffset = kIndexedCallbackOffset + kPointerSize;
static const int kSize = kDataOffset + kPointerSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(AccessCheckInfo);
};
class InterceptorInfo: public Struct {
public:
DECL_ACCESSORS(getter, Object)
DECL_ACCESSORS(setter, Object)
DECL_ACCESSORS(query, Object)
DECL_ACCESSORS(deleter, Object)
DECL_ACCESSORS(enumerator, Object)
DECL_ACCESSORS(data, Object)
static inline InterceptorInfo* cast(Object* obj);
#ifdef DEBUG
void InterceptorInfoPrint();
void InterceptorInfoVerify();
#endif
static const int kGetterOffset = HeapObject::kHeaderSize;
static const int kSetterOffset = kGetterOffset + kPointerSize;
static const int kQueryOffset = kSetterOffset + kPointerSize;
static const int kDeleterOffset = kQueryOffset + kPointerSize;
static const int kEnumeratorOffset = kDeleterOffset + kPointerSize;
static const int kDataOffset = kEnumeratorOffset + kPointerSize;
static const int kSize = kDataOffset + kPointerSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(InterceptorInfo);
};
class CallHandlerInfo: public Struct {
public:
DECL_ACCESSORS(callback, Object)
DECL_ACCESSORS(data, Object)
static inline CallHandlerInfo* cast(Object* obj);
#ifdef DEBUG
void CallHandlerInfoPrint();
void CallHandlerInfoVerify();
#endif
static const int kCallbackOffset = HeapObject::kHeaderSize;
static const int kDataOffset = kCallbackOffset + kPointerSize;
static const int kSize = kDataOffset + kPointerSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(CallHandlerInfo);
};
class TemplateInfo: public Struct {
public:
DECL_ACCESSORS(tag, Object)
DECL_ACCESSORS(property_list, Object)
#ifdef DEBUG
void TemplateInfoVerify();
#endif
static const int kTagOffset = HeapObject::kHeaderSize;
static const int kPropertyListOffset = kTagOffset + kPointerSize;
static const int kHeaderSize = kPropertyListOffset + kPointerSize;
protected:
friend class AGCCVersionRequiresThisClassToHaveAFriendSoHereItIs;
DISALLOW_IMPLICIT_CONSTRUCTORS(TemplateInfo);
};
class FunctionTemplateInfo: public TemplateInfo {
public:
DECL_ACCESSORS(serial_number, Object)
DECL_ACCESSORS(call_code, Object)
DECL_ACCESSORS(property_accessors, Object)
DECL_ACCESSORS(prototype_template, Object)
DECL_ACCESSORS(parent_template, Object)
DECL_ACCESSORS(named_property_handler, Object)
DECL_ACCESSORS(indexed_property_handler, Object)
DECL_ACCESSORS(instance_template, Object)
DECL_ACCESSORS(class_name, Object)
DECL_ACCESSORS(signature, Object)
DECL_ACCESSORS(instance_call_handler, Object)
DECL_ACCESSORS(access_check_info, Object)
DECL_ACCESSORS(flag, Smi)
// Following properties use flag bits.
DECL_BOOLEAN_ACCESSORS(hidden_prototype)
DECL_BOOLEAN_ACCESSORS(undetectable)
// If the bit is set, object instances created by this function
// requires access check.
DECL_BOOLEAN_ACCESSORS(needs_access_check)
static inline FunctionTemplateInfo* cast(Object* obj);
#ifdef DEBUG
void FunctionTemplateInfoPrint();
void FunctionTemplateInfoVerify();
#endif
static const int kSerialNumberOffset = TemplateInfo::kHeaderSize;
static const int kCallCodeOffset = kSerialNumberOffset + kPointerSize;
static const int kPropertyAccessorsOffset = kCallCodeOffset + kPointerSize;
static const int kPrototypeTemplateOffset =
kPropertyAccessorsOffset + kPointerSize;
static const int kParentTemplateOffset =
kPrototypeTemplateOffset + kPointerSize;
static const int kNamedPropertyHandlerOffset =
kParentTemplateOffset + kPointerSize;
static const int kIndexedPropertyHandlerOffset =
kNamedPropertyHandlerOffset + kPointerSize;
static const int kInstanceTemplateOffset =
kIndexedPropertyHandlerOffset + kPointerSize;
static const int kClassNameOffset = kInstanceTemplateOffset + kPointerSize;
static const int kSignatureOffset = kClassNameOffset + kPointerSize;
static const int kInstanceCallHandlerOffset = kSignatureOffset + kPointerSize;
static const int kAccessCheckInfoOffset =
kInstanceCallHandlerOffset + kPointerSize;
static const int kFlagOffset = kAccessCheckInfoOffset + kPointerSize;
static const int kSize = kFlagOffset + kPointerSize;
private:
// Bit position in the flag, from least significant bit position.
static const int kHiddenPrototypeBit = 0;
static const int kUndetectableBit = 1;
static const int kNeedsAccessCheckBit = 2;
DISALLOW_IMPLICIT_CONSTRUCTORS(FunctionTemplateInfo);
};
class ObjectTemplateInfo: public TemplateInfo {
public:
DECL_ACCESSORS(constructor, Object)
DECL_ACCESSORS(internal_field_count, Object)
static inline ObjectTemplateInfo* cast(Object* obj);
#ifdef DEBUG
void ObjectTemplateInfoPrint();
void ObjectTemplateInfoVerify();
#endif
static const int kConstructorOffset = TemplateInfo::kHeaderSize;
static const int kInternalFieldCountOffset =
kConstructorOffset + kPointerSize;
static const int kSize = kInternalFieldCountOffset + kPointerSize;
};
class SignatureInfo: public Struct {
public:
DECL_ACCESSORS(receiver, Object)
DECL_ACCESSORS(args, Object)
static inline SignatureInfo* cast(Object* obj);
#ifdef DEBUG
void SignatureInfoPrint();
void SignatureInfoVerify();
#endif
static const int kReceiverOffset = Struct::kHeaderSize;
static const int kArgsOffset = kReceiverOffset + kPointerSize;
static const int kSize = kArgsOffset + kPointerSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(SignatureInfo);
};
class TypeSwitchInfo: public Struct {
public:
DECL_ACCESSORS(types, Object)
static inline TypeSwitchInfo* cast(Object* obj);
#ifdef DEBUG
void TypeSwitchInfoPrint();
void TypeSwitchInfoVerify();
#endif
static const int kTypesOffset = Struct::kHeaderSize;
static const int kSize = kTypesOffset + kPointerSize;
};
#ifdef ENABLE_DEBUGGER_SUPPORT
// The DebugInfo class holds additional information for a function being
// debugged.
class DebugInfo: public Struct {
public:
// The shared function info for the source being debugged.
DECL_ACCESSORS(shared, SharedFunctionInfo)
// Code object for the original code.
DECL_ACCESSORS(original_code, Code)
// Code object for the patched code. This code object is the code object
// currently active for the function.
DECL_ACCESSORS(code, Code)
// Fixed array holding status information for each active break point.
DECL_ACCESSORS(break_points, FixedArray)
// Check if there is a break point at a code position.
bool HasBreakPoint(int code_position);
// Get the break point info object for a code position.
Object* GetBreakPointInfo(int code_position);
// Clear a break point.
static void ClearBreakPoint(Handle<DebugInfo> debug_info,
int code_position,
Handle<Object> break_point_object);
// Set a break point.
static void SetBreakPoint(Handle<DebugInfo> debug_info, int code_position,
int source_position, int statement_position,
Handle<Object> break_point_object);
// Get the break point objects for a code position.
Object* GetBreakPointObjects(int code_position);
// Find the break point info holding this break point object.
static Object* FindBreakPointInfo(Handle<DebugInfo> debug_info,
Handle<Object> break_point_object);
// Get the number of break points for this function.
int GetBreakPointCount();
static inline DebugInfo* cast(Object* obj);
#ifdef DEBUG
void DebugInfoPrint();
void DebugInfoVerify();
#endif
static const int kSharedFunctionInfoIndex = Struct::kHeaderSize;
static const int kOriginalCodeIndex = kSharedFunctionInfoIndex + kPointerSize;
static const int kPatchedCodeIndex = kOriginalCodeIndex + kPointerSize;
static const int kActiveBreakPointsCountIndex =
kPatchedCodeIndex + kPointerSize;
static const int kBreakPointsStateIndex =
kActiveBreakPointsCountIndex + kPointerSize;
static const int kSize = kBreakPointsStateIndex + kPointerSize;
private:
static const int kNoBreakPointInfo = -1;
// Lookup the index in the break_points array for a code position.
int GetBreakPointInfoIndex(int code_position);
DISALLOW_IMPLICIT_CONSTRUCTORS(DebugInfo);
};
// The BreakPointInfo class holds information for break points set in a
// function. The DebugInfo object holds a BreakPointInfo object for each code
// position with one or more break points.
class BreakPointInfo: public Struct {
public:
// The position in the code for the break point.
DECL_ACCESSORS(code_position, Smi)
// The position in the source for the break position.
DECL_ACCESSORS(source_position, Smi)
// The position in the source for the last statement before this break
// position.
DECL_ACCESSORS(statement_position, Smi)
// List of related JavaScript break points.
DECL_ACCESSORS(break_point_objects, Object)
// Removes a break point.
static void ClearBreakPoint(Handle<BreakPointInfo> info,
Handle<Object> break_point_object);
// Set a break point.
static void SetBreakPoint(Handle<BreakPointInfo> info,
Handle<Object> break_point_object);
// Check if break point info has this break point object.
static bool HasBreakPointObject(Handle<BreakPointInfo> info,
Handle<Object> break_point_object);
// Get the number of break points for this code position.
int GetBreakPointCount();
static inline BreakPointInfo* cast(Object* obj);
#ifdef DEBUG
void BreakPointInfoPrint();
void BreakPointInfoVerify();
#endif
static const int kCodePositionIndex = Struct::kHeaderSize;
static const int kSourcePositionIndex = kCodePositionIndex + kPointerSize;
static const int kStatementPositionIndex =
kSourcePositionIndex + kPointerSize;
static const int kBreakPointObjectsIndex =
kStatementPositionIndex + kPointerSize;
static const int kSize = kBreakPointObjectsIndex + kPointerSize;
private:
DISALLOW_IMPLICIT_CONSTRUCTORS(BreakPointInfo);
};
#endif // ENABLE_DEBUGGER_SUPPORT
#undef DECL_BOOLEAN_ACCESSORS
#undef DECL_ACCESSORS
// Abstract base class for visiting, and optionally modifying, the
// pointers contained in Objects. Used in GC and serialization/deserialization.
class ObjectVisitor BASE_EMBEDDED {
public:
virtual ~ObjectVisitor() {}
// Visits a contiguous arrays of pointers in the half-open range
// [start, end). Any or all of the values may be modified on return.
virtual void VisitPointers(Object** start, Object** end) = 0;
// To allow lazy clearing of inline caches the visitor has
// a rich interface for iterating over Code objects..
// Visits a code target in the instruction stream.
virtual void VisitCodeTarget(RelocInfo* rinfo);
// Visits a runtime entry in the instruction stream.
virtual void VisitRuntimeEntry(RelocInfo* rinfo) {}
// Visits the resource of an ASCII or two-byte string.
virtual void VisitExternalAsciiString(
v8::String::ExternalAsciiStringResource** resource) {}
virtual void VisitExternalTwoByteString(
v8::String::ExternalStringResource** resource) {}
// Visits a debug call target in the instruction stream.
virtual void VisitDebugTarget(RelocInfo* rinfo);
// Handy shorthand for visiting a single pointer.
virtual void VisitPointer(Object** p) { VisitPointers(p, p + 1); }
// Visits a contiguous arrays of external references (references to the C++
// heap) in the half-open range [start, end). Any or all of the values
// may be modified on return.
virtual void VisitExternalReferences(Address* start, Address* end) {}
inline void VisitExternalReference(Address* p) {
VisitExternalReferences(p, p + 1);
}
#ifdef DEBUG
// Intended for serialization/deserialization checking: insert, or
// check for the presence of, a tag at this position in the stream.
virtual void Synchronize(const char* tag) {}
#else
inline void Synchronize(const char* tag) {}
#endif
};
// BooleanBit is a helper class for setting and getting a bit in an
// integer or Smi.
class BooleanBit : public AllStatic {
public:
static inline bool get(Smi* smi, int bit_position) {
return get(smi->value(), bit_position);
}
static inline bool get(int value, int bit_position) {
return (value & (1 << bit_position)) != 0;
}
static inline Smi* set(Smi* smi, int bit_position, bool v) {
return Smi::FromInt(set(smi->value(), bit_position, v));
}
static inline int set(int value, int bit_position, bool v) {
if (v) {
value |= (1 << bit_position);
} else {
value &= ~(1 << bit_position);
}
return value;
}
};
} } // namespace v8::internal
#endif // V8_OBJECTS_H_