[MERGE #4077 @MikeHolman] Optimize JSON

Merge pull request #4077 from MikeHolman:lazyjson This change represents a major overhaul to JSON.stringify. The new architecture works as follows: We first calculate the length of the string we need and create a side data structure with the values we will need, but do not actually create any concat strings, etc. like we used to do. If/when the string is needed, we create a buffer using the length and metadata collected in the first pass, filling it element by element. If the string is passed to JSON.parse, we use our metadata to construct an object without needing to parse the string. By avoiding creation of all the intermediate concat strings, this makes JSON.stringify much more efficient. In micro-benchmarks I'm seeing JSON.stringify take around 40% less time than today. And in case the string is actually sent to JSON.parse, that now takes 70% less time. This improves Speedometer 2 geomean by about 3%. Also improves AcmeAir by about 3%.
chakra-core · Nov 3, 2017 · 0cfaf74 · 0cfaf74
2 parents 36c1f14 + 80f5a23
commit 0cfaf74
Show file tree

Hide file tree

Showing 36 changed files with 1,718 additions and 1,108 deletions.
diff --git a/lib/Backend/JITTimeFunctionBody.cpp b/lib/Backend/JITTimeFunctionBody.cpp
@@ -957,17 +957,6 @@ JITTimeFunctionBody::GetConstTable() const
     return m_bodyData.constTable;
 }
 
-bool
-JITTimeFunctionBody::IsConstRegPropertyString(Js::RegSlot reg, ScriptContextInfo * context) const
-{
-    RecyclableObjectIDL * content = m_bodyData.constTableContent->content[reg - Js::FunctionBody::FirstRegSlot];
-    if (content != nullptr && content->vtbl == context->GetVTableAddress(VtablePropertyString))
-    {
-        return true;
-    }
-    return false;
-}
-
 intptr_t
 JITTimeFunctionBody::GetRootObject() const
 {

diff --git a/lib/Backend/JITTimeFunctionBody.h b/lib/Backend/JITTimeFunctionBody.h
@@ -153,7 +153,6 @@ class JITTimeFunctionBody
     intptr_t GetIsInstInlineCache(uint index) const;
     Js::TypeId GetConstantType(Js::RegSlot location) const;
     void * GetConstTable() const;
-    bool IsConstRegPropertyString(Js::RegSlot reg, ScriptContextInfo * context) const;
 
     intptr_t GetRootObject() const;
     intptr_t GetLoopHeaderAddr(uint loopNum) const;

diff --git a/lib/Backend/JnHelperMethodList.h b/lib/Backend/JnHelperMethodList.h
@@ -35,7 +35,7 @@ HELPERCALL(ScrObj_LdStrictInnerFrameDisplayNoParent, Js::JavascriptOperators::OP
 HELPERCALL(ScrObj_OP_IsInst, Js::JavascriptOperators::OP_IsInst, AttrCanThrow)
 
 HELPERCALL(Op_IsIn, Js::JavascriptOperators::IsIn, AttrCanThrow)
-HELPERCALL(Op_IsObject, Js::JavascriptOperators::IsObject, AttrCanThrow)
+HELPERCALL(Op_IsObject, (BOOL (*) (Js::Var))Js::JavascriptOperators::IsObject, AttrCanThrow)
 HELPERCALL(Op_IsClassConstructor, Js::JavascriptOperators::IsClassConstructor, AttrCanThrow)
 HELPERCALL(Op_IsBaseConstructorKind, Js::JavascriptOperators::IsBaseConstructorKind, AttrCanThrow)
 HELPERCALL(Op_LoadHeapArguments, Js::JavascriptOperators::LoadHeapArguments, 0)

diff --git a/lib/Common/BackendApi.h b/lib/Common/BackendApi.h
@@ -235,6 +235,7 @@ enum VTableValue {
     VtableDynamicObject,
     VtableInvalid,
     VtablePropertyString,
+    VtableLazyJSONString,
     VtableLiteralStringWithPropertyStringPtr,
     VtableJavascriptBoolean,
     VtableJavascriptArray,

diff --git a/lib/JITIDL/JITTypes.h b/lib/JITIDL/JITTypes.h
@@ -77,7 +77,7 @@ typedef unsigned char boolean;
 #define __JITTypes_h__
 
 // TODO: OOP JIT, how do we make this better?
-const int VTABLE_COUNT = 48;
+const int VTABLE_COUNT = 49;
 const int EQUIVALENT_TYPE_CACHE_SIZE = 8;
 
 typedef IDL_DEF([context_handle]) void * PTHREADCONTEXT_HANDLE;

diff --git a/lib/Runtime/Base/Constants.h b/lib/Runtime/Base/Constants.h
@@ -155,14 +155,17 @@ namespace Js
         static const  char16 StringReplace[];
         static const  char16 StringMatch[];
 
-        static const size_t AnonymousFunctionLength = _countof(_u("Anonymous function")) - 1;
-        static const size_t AnonymousLength         = _countof(_u("anonymous")) - 1;
-        static const size_t AnonymousClassLength    = _countof(_u("Anonymous class")) - 1;
-        static const size_t FunctionCodeLength      = _countof(_u("Function code")) - 1;
-        static const size_t GlobalFunctionLength    = _countof(_u("glo")) - 1;
-        static const size_t GlobalCodeLength        = _countof(_u("Global code")) - 1;
-        static const size_t EvalCodeLength          = _countof(_u("eval code")) - 1;
-        static const size_t UnknownScriptCodeLength = _countof(_u("Unknown script code")) - 1;
+        static const charcount_t AnonymousFunctionLength = _countof(_u("Anonymous function")) - 1;
+        static const charcount_t AnonymousLength         = _countof(_u("anonymous")) - 1;
+        static const charcount_t AnonymousClassLength    = _countof(_u("Anonymous class")) - 1;
+        static const charcount_t FunctionCodeLength      = _countof(_u("Function code")) - 1;
+        static const charcount_t GlobalFunctionLength    = _countof(_u("glo")) - 1;
+        static const charcount_t GlobalCodeLength        = _countof(_u("Global code")) - 1;
+        static const charcount_t EvalCodeLength          = _countof(_u("eval code")) - 1;
+        static const charcount_t UnknownScriptCodeLength = _countof(_u("Unknown script code")) - 1;
+        static const charcount_t NullStringLength = _countof(_u("Null")) - 1;
+        static const charcount_t TrueStringLength = _countof(_u("True")) - 1;
+        static const charcount_t FalseStringLength = _countof(_u("False")) - 1;
     };
 
     extern const FrameDisplay NullFrameDisplay;

diff --git a/lib/Runtime/Language/JavascriptConversion.cpp b/lib/Runtime/Language/JavascriptConversion.cpp
@@ -538,27 +538,25 @@ namespace Js
         }
 
         // If exoticToPrim is not undefined, then
-        if (nullptr != exoticToPrim)
+        Assert(nullptr != exoticToPrim);
+        ThreadContext * threadContext = requestContext->GetThreadContext();
+        result = threadContext->ExecuteImplicitCall(exoticToPrim, ImplicitCall_ToPrimitive, [=]()->Js::Var
         {
-            ThreadContext * threadContext = requestContext->GetThreadContext();
-            result = threadContext->ExecuteImplicitCall(exoticToPrim, ImplicitCall_ToPrimitive, [=]()->Js::Var
-            {
-                // Stack object should have a pre-op bail on implicit call.  We shouldn't see them here.
-                Assert(!ThreadContext::IsOnStack(recyclableObject));
-
-                // Let result be the result of calling the[[Call]] internal method of exoticToPrim, with input as thisArgument and(hint) as argumentsList.
-                return CALL_FUNCTION(threadContext, exoticToPrim, CallInfo(CallFlags_Value, 2), recyclableObject, hintString);
-            });
+            // Stack object should have a pre-op bail on implicit call.  We shouldn't see them here.
+            Assert(!ThreadContext::IsOnStack(recyclableObject));
 
-            if (!result)
-            {
-                // There was an implicit call and implicit calls are disabled. This would typically cause a bailout.
-                Assert(threadContext->IsDisableImplicitCall());
-                return requestContext->GetLibrary()->GetNull();
-            }
+            // Let result be the result of calling the[[Call]] internal method of exoticToPrim, with input as thisArgument and(hint) as argumentsList.
+            return CALL_FUNCTION(threadContext, exoticToPrim, CallInfo(CallFlags_Value, 2), recyclableObject, hintString);
+        });
 
-            Assert(!CrossSite::NeedMarshalVar(result, requestContext));
+        if (!result)
+        {
+            // There was an implicit call and implicit calls are disabled. This would typically cause a bailout.
+            Assert(threadContext->IsDisableImplicitCall());
+            return requestContext->GetLibrary()->GetNull();
         }
+
+        Assert(!CrossSite::NeedMarshalVar(result, requestContext));
         // If result is an ECMAScript language value and Type(result) is not Object, then return result.
         if (TaggedInt::Is(result) || !JavascriptOperators::IsObjectType(JavascriptOperators::GetTypeId(result)))
         {

diff --git a/lib/Runtime/Language/JavascriptOperators.cpp b/lib/Runtime/Language/JavascriptOperators.cpp
@@ -1325,32 +1325,36 @@ namespace Js
         return type->GetPrototype();
     }
 
-    BOOL JavascriptOperators::IsArray(Var instanceVar)
+    BOOL JavascriptOperators::IsArray(_In_ RecyclableObject* instance)
     {
-        if (!RecyclableObject::Is(instanceVar))
-        {
-            return FALSE;
-        }
-        RecyclableObject* instance = RecyclableObject::UnsafeFromVar(instanceVar);
         if (DynamicObject::IsAnyArray(instance))
         {
             return TRUE;
         }
 
-        JavascriptProxy* proxy = JavascriptOperators::TryFromVar<JavascriptProxy>(instanceVar);
+        JavascriptProxy* proxy = JavascriptOperators::TryFromVar<JavascriptProxy>(instance);
         if (proxy)
         {
             return IsArray(proxy->GetTarget());
         }
         TypeId remoteTypeId = TypeIds_Limit;
-        if (JavascriptOperators::GetRemoteTypeId(instanceVar, &remoteTypeId) &&
+        if (JavascriptOperators::GetRemoteTypeId(instance, &remoteTypeId) &&
             DynamicObject::IsAnyArrayTypeId(remoteTypeId))
         {
             return TRUE;
         }
         return FALSE;
     }
 
+    BOOL JavascriptOperators::IsArray(_In_ Var instanceVar)
+    {
+        if (!RecyclableObject::Is(instanceVar))
+        {
+            return FALSE;
+        }
+        return IsArray(RecyclableObject::FromVar(instanceVar));
+    }
+
     BOOL JavascriptOperators::IsConstructor(Var instanceVar)
     {
         if (!RecyclableObject::Is(instanceVar))
@@ -10045,9 +10049,14 @@ namespace Js
         return JavascriptNumber::ToVarNoCheck(JavascriptConversion::ToNumber_Full(aRight, scriptContext), scriptContext);
     }
 
-    BOOL JavascriptOperators::IsObject(Var aValue)
+    BOOL JavascriptOperators::IsObject(_In_ RecyclableObject* instance)
+    {
+        return GetTypeId(instance) > TypeIds_LastJavascriptPrimitiveType;
+    }
+
+    BOOL JavascriptOperators::IsObject(_In_ Var instance)
     {
-        return GetTypeId(aValue) > TypeIds_LastJavascriptPrimitiveType;
+        return GetTypeId(instance) > TypeIds_LastJavascriptPrimitiveType;
     }
 
     BOOL JavascriptOperators::IsObjectType(TypeId typeId)
@@ -10066,6 +10075,11 @@ namespace Js
         return IsObjectType(typeId) || typeId == TypeIds_Null;
     }
 
+    BOOL JavascriptOperators::IsUndefined(_In_ RecyclableObject* instance)
+    {
+        return JavascriptOperators::GetTypeId(instance) == TypeIds_Undefined;
+    }
+
     BOOL JavascriptOperators::IsUndefined(Var instance)
     {
         return JavascriptOperators::GetTypeId(instance) == TypeIds_Undefined;
@@ -10408,6 +10422,11 @@ namespace Js
 
     BOOL JavascriptOperators::GetItem(RecyclableObject* instance, uint64 index, Var* value, ScriptContext* requestContext)
     {
+        if (index < JavascriptArray::InvalidIndex)
+        {
+            // In case index fits in uint32, we can avoid the (slower) big-index path
+            return GetItem(instance, static_cast<uint32>(index), value, requestContext);
+        }
         PropertyRecord const * propertyRecord = nullptr;
         JavascriptOperators::GetPropertyIdForInt(index, requestContext, &propertyRecord);
         return JavascriptOperators::GetProperty(instance, propertyRecord->GetPropertyId(), value, requestContext);

diff --git a/lib/Runtime/Language/JavascriptOperators.h b/lib/Runtime/Language/JavascriptOperators.h
@@ -102,7 +102,8 @@ namespace Js
     {
     // Methods
     public:
-        static BOOL IsArray(Var instanceVar);
+        static BOOL IsArray(_In_ RecyclableObject* instanceObj);
+        static BOOL IsArray(_In_ Var instanceVar);
         static BOOL IsConstructor(Var instanceVar);
         static BOOL IsConcatSpreadable(Var instanceVar);
         static RecyclableObject* ToObject(Var aRight,ScriptContext* scriptContext);
@@ -207,17 +208,20 @@ namespace Js
         static BOOL DeletePropertyUnscopables(RecyclableObject* instance, PropertyId propertyId, PropertyOperationFlags propertyOperationFlags = PropertyOperation_None);
         template<bool unscopables>
         static BOOL DeleteProperty_Impl(RecyclableObject* instance, PropertyId propertyId, PropertyOperationFlags propertyOperationFlags = PropertyOperation_None);
-        static TypeId GetTypeId(Var instance);
+        static TypeId GetTypeId(_In_ const Var instance);
+        static TypeId GetTypeId(_In_ RecyclableObject* instance);
         static TypeId GetTypeIdNoCheck(Var instance);
         template <typename T>
         __forceinline static T* TryFromVar(Var value)
         {
             return T::Is(value) ? T::UnsafeFromVar(value) : nullptr;
         }
-        static BOOL IsObject(Var instance);
+        static BOOL IsObject(_In_ Var instance);
+        static BOOL IsObject(_In_ RecyclableObject* instance);
         static BOOL IsExposedType(TypeId typeId);
         static BOOL IsObjectType(TypeId typeId);
         static BOOL IsObjectOrNull(Var instance);
+        static BOOL IsUndefined(_In_ RecyclableObject* instance);
         static BOOL IsUndefined(Var instance);
         static BOOL IsUndefinedObject(RecyclableObject* instance);
         static BOOL IsUndefinedOrNullType(TypeId);

diff --git a/lib/Runtime/Language/JavascriptOperators.inl b/lib/Runtime/Language/JavascriptOperators.inl
@@ -6,7 +6,19 @@
 
 namespace Js
 {
-    __forceinline TypeId JavascriptOperators::GetTypeId(const Var aValue)
+    __forceinline TypeId JavascriptOperators::GetTypeId(_In_ RecyclableObject* obj)
+    {
+        AssertMsg(obj != nullptr, "GetTypeId aValue is null");
+
+        auto typeId = obj->GetTypeId();
+#if DBG
+        auto isExternal = obj->CanHaveInterceptors();
+        AssertMsg(typeId < TypeIds_Limit || isExternal, "GetTypeId aValue has invalid TypeId");
+#endif
+        return typeId;
+    }
+
+    __forceinline TypeId JavascriptOperators::GetTypeId(_In_ const Var aValue)
     {
         AssertMsg(aValue != nullptr, "GetTypeId aValue is null");
 
@@ -22,12 +34,7 @@ namespace Js
 #endif
         else
         {
-            auto typeId = RecyclableObject::UnsafeFromVar(aValue)->GetTypeId();
-#if DBG
-            auto isExternal = RecyclableObject::FromVar(aValue)->CanHaveInterceptors();
-            AssertMsg(typeId < TypeIds_Limit || isExternal, "GetTypeId aValue has invalid TypeId");
-#endif
-            return typeId;
+            return JavascriptOperators::GetTypeId(RecyclableObject::UnsafeFromVar(aValue));
         }
     }
 

diff --git a/lib/Runtime/Library/CMakeLists.txt b/lib/Runtime/Library/CMakeLists.txt
@@ -45,6 +45,8 @@ set(CRLIB_SOURCE_CODES
     JSONScanner.cpp
     JSONStack.cpp
     JSONString.cpp
+    JSONStringBuilder.cpp
+    JSONStringifier.cpp
     JavascriptArray.cpp
     JavascriptArrayIndexEnumerator.cpp
     JavascriptArrayIndexEnumeratorBase.cpp
@@ -101,6 +103,7 @@ set(CRLIB_SOURCE_CODES
     JavascriptVariantDate.cpp
     JavascriptWeakMap.cpp
     JavascriptWeakSet.cpp
+    LazyJSONString.cpp
     LiteralString.cpp
     MathLibrary.cpp
     ModuleRoot.cpp

diff --git a/lib/Runtime/Library/Chakra.Runtime.Library.vcxproj b/lib/Runtime/Library/Chakra.Runtime.Library.vcxproj
@@ -168,6 +168,9 @@
     <ClCompile Include="$(MSBuildThisFileDirectory)CustomExternalIterator.cpp" />
     <ClCompile Include="$(MSBuildThisFileDirectory)JavascriptExceptionMetadata.cpp" />
     <ClCompile Include="$(MSBuildThisFileDirectory)VerifyMarkFalseReference.cpp" />
+    <ClCompile Include="JSONStringBuilder.cpp" />
+    <ClCompile Include="JSONStringifier.cpp" />
+    <ClCompile Include="LazyJSONString.cpp" />
   </ItemGroup>
   <ItemGroup>
     <ClInclude Include="..\DetachedStateBase.h" />
@@ -193,6 +196,9 @@
     <ClInclude Include="JavascriptSimdBool32x4.h" />
     <ClInclude Include="JavascriptSimdBool16x8.h" />
     <ClInclude Include="JavascriptSimdBool8x16.h" />
+    <ClInclude Include="JSONStringBuilder.h" />
+    <ClInclude Include="JSONStringifier.h" />
+    <ClInclude Include="LazyJSONString.h" />
     <ClInclude Include="SharedArrayBuffer.h" />
     <ClInclude Include="SimdFloat32x4Lib.h" />
     <ClInclude Include="SimdFloat64x2Lib.h" />

diff --git a/lib/Runtime/Library/Chakra.Runtime.Library.vcxproj.filters b/lib/Runtime/Library/Chakra.Runtime.Library.vcxproj.filters
@@ -121,6 +121,9 @@
     <ClCompile Include="$(MSBuildThisFileDirectory)WabtInterface.cpp" />
     <ClCompile Include="$(MSBuildThisFileDirectory)CustomExternalIterator.cpp" />
     <ClCompile Include="$(MSBuildThisFileDirectory)VerifyMarkFalseReference.cpp" />
+    <ClCompile Include="LazyJSONString.cpp" />
+    <ClCompile Include="JSONStringifier.cpp" />
+    <ClCompile Include="JSONStringBuilder.cpp" />
   </ItemGroup>
   <ItemGroup>
     <ClInclude Include="..\InternalPropertyList.h" />
@@ -259,6 +262,9 @@
     <ClInclude Include="CustomExternalIterator.h" />
     <ClInclude Include="JavascriptExceptionMetadata.h" />
     <ClInclude Include="..\DetachedStateBase.h" />
+    <ClInclude Include="LazyJSONString.h" />
+    <ClInclude Include="JSONStringifier.h" />
+    <ClInclude Include="JSONStringBuilder.h" />
   </ItemGroup>
   <ItemGroup>
     <None Include="ConcatString.inl" />