Attachment 371138 Details for Bug 198399 – WIP

[patch] WIP

bug-198399-20190601233812.patch (text/plain), 130.99 KB, created by Myles C. Maxfield on 2019-06-01 23:38:13 PDT

(hide)

Description:

Filename:

MIME Type:

Creator: Myles C. Maxfield

Created: 2019-06-01 23:38:13 PDT

Size: 130.99 KB

patch

obsolete

>Subversion Revision: 246012
>diff --git a/Source/WebCore/ChangeLog b/Source/WebCore/ChangeLog
>index 2458cfb74339378bebf54199c9dd048de52a7fdf..a6dfb2c75229abe521148483765ea0dded2fec26 100644
>--- a/Source/WebCore/ChangeLog
>+++ b/Source/WebCore/ChangeLog
>@@ -1,3 +1,28 @@
>+2019-05-31  Myles C. Maxfield  <mmaxfield@apple.com>
>+
>+        [WHLSL] Educate the property resolver about IndexExpressions
>+        https://bugs.webkit.org/show_bug.cgi?id=198399
>+
>+        Reviewed by NOBODY (OOPS!).
>+
>+        No new tests (OOPS!).
>+
>+        * Modules/webgpu/WHLSL/AST/WHLSLIndexExpression.h:
>+        (WebCore::WHLSL::AST::IndexExpression::takeIndex):
>+        * Modules/webgpu/WHLSL/AST/WHLSLReferenceType.h:
>+        * Modules/webgpu/WHLSL/WHLSLChecker.cpp:
>+        (WebCore::WHLSL::Checker::finishVisiting):
>+        (WebCore::WHLSL::Checker::visit):
>+        * Modules/webgpu/WHLSL/WHLSLInferTypes.h:
>+        * Modules/webgpu/WHLSL/WHLSLPropertyResolver.cpp:
>+        (WebCore::WHLSL::PropertyResolver::visit):
>+        (WebCore::WHLSL::setterCall):
>+        (WebCore::WHLSL::getterCall):
>+        (WebCore::WHLSL::modify):
>+        (WebCore::WHLSL::PropertyResolver::simplifyRightValue):
>+        (WebCore::WHLSL::LeftValueSimplifier::finishVisiting):
>+        (WebCore::WHLSL::LeftValueSimplifier::visit):
>+
> 2019-05-31  Youenn Fablet  <youenn@apple.com>
> 
>         Add an option to mute audio capture automatically when page is not visible
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLAddressSpace.h b/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLAddressSpace.h
>index 425d134667851dab85c827b250cb57afdefc2dde..aba33d5d07834176360a8b3bf8173dbae300281f 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLAddressSpace.h
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLAddressSpace.h
>@@ -43,7 +43,7 @@ enum class AddressSpace : uint8_t {
>     Thread
> };
> 
>-static ALWAYS_INLINE String toString(AddressSpace addressSpace)
>+ALWAYS_INLINE String toString(AddressSpace addressSpace)
> {
>     switch (addressSpace) {
>     case AddressSpace::Constant:
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLEntryPointType.h b/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLEntryPointType.h
>index 65f6873b46909dead4cb20262b7536e117e57d22..b45424fd3fd387ed0a3ffb1d103a765560088dbc 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLEntryPointType.h
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLEntryPointType.h
>@@ -41,7 +41,7 @@ enum class EntryPointType : uint8_t {
>     Compute,
> };
> 
>-static ALWAYS_INLINE String toString(EntryPointType type)
>+ALWAYS_INLINE String toString(EntryPointType type)
> {
>     switch (type) {
>     case EntryPointType::Vertex:
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLIndexExpression.h b/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLIndexExpression.h
>index aa197b5af43bd60388c7185ad146e8ec03d32e8c..5e49116ab84f67fc026e4572cbe8f5844c813364 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLIndexExpression.h
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLIndexExpression.h
>@@ -59,15 +59,16 @@ public:
> 
>     String setterFunctionName() const override
>     {
>-        return "operator&[]"_str;
>+        return "operator[]="_str;
>     }
> 
>     String anderFunctionName() const override
>     {
>-        return "operator[]="_str;
>+        return "operator&[]"_str;
>     }
> 
>     Expression& indexExpression() { return m_index; }
>+    UniqueRef<Expression> takeIndex() { return WTFMove(m_index); }
> 
> private:
>     UniqueRef<Expression> m_index;
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLReferenceType.h b/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLReferenceType.h
>index 8811ac9ad427dd02e7f141b06432bc5033869653..1e6ce164c6cbdb1938c92b594e0060e091295944 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLReferenceType.h
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/AST/WHLSLReferenceType.h
>@@ -53,7 +53,7 @@ public:
>     ReferenceType(const ReferenceType&) = delete;
>     ReferenceType(ReferenceType&&) = default;
> 
>-    bool isReferenceType() const override { return false; }
>+    bool isReferenceType() const override { return true; }
> 
>     AddressSpace addressSpace() const { return m_addressSpace; }
>     const UnnamedType& elementType() const { return m_elementType; }
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/Metal/WHLSLNativeFunctionWriter.cpp b/Source/WebCore/Modules/webgpu/WHLSL/Metal/WHLSLNativeFunctionWriter.cpp
>index 38e7a71876ee850b0d758945ad1971fec91fd6e0..69dffdfe5a5c413e118753c38d1d39c014c8082f 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/Metal/WHLSLNativeFunctionWriter.cpp
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/Metal/WHLSLNativeFunctionWriter.cpp
>@@ -63,21 +63,6 @@ static String mapFunctionName(String& functionName)
>     return functionName;
> }
> 
>-static String convertAddressSpace(AST::AddressSpace addressSpace)
>-{
>-    switch (addressSpace) {
>-    case AST::AddressSpace::Constant:
>-        return "constant"_str;
>-    case AST::AddressSpace::Device:
>-        return "device"_str;
>-    case AST::AddressSpace::Threadgroup:
>-        return "threadgroup"_str;
>-    default:
>-        ASSERT(addressSpace == AST::AddressSpace::Thread);
>-        return "thread"_str;
>-    }
>-}
>-
> static String atomicName(String input)
> {
>     if (input == "Add")
>@@ -375,7 +360,7 @@ String writeNativeFunction(AST::NativeFunctionDeclaration& nativeFunctionDeclara
>             auto& fourthArgumentPointer = downcast<AST::PointerType>(*nativeFunctionDeclaration.parameters()[3]->type());
>             auto fourthArgumentAddressSpace = fourthArgumentPointer.addressSpace();
>             auto fourthArgumentPointee = typeNamer.mangledNameForType(fourthArgumentPointer.elementType());
>-            stringBuilder.append(makeString("void ", outputFunctionName, '(', convertAddressSpace(firstArgumentAddressSpace), ' ', firstArgumentPointee, "* object, ", secondArgument, " compare, ", thirdArgument, " desired, ", convertAddressSpace(fourthArgumentAddressSpace), ' ', fourthArgumentPointee, "* out) {\n"));
>+            stringBuilder.append(makeString("void ", outputFunctionName, '(', toString(firstArgumentAddressSpace), ' ', firstArgumentPointee, "* object, ", secondArgument, " compare, ", thirdArgument, " desired, ", toString(fourthArgumentAddressSpace), ' ', fourthArgumentPointee, "* out) {\n"));
>             stringBuilder.append("    atomic_compare_exchange_weak_explicit(object, &compare, desired, memory_order_relaxed);\n");
>             stringBuilder.append("    *out = compare;\n");
>             stringBuilder.append("}\n");
>@@ -393,7 +378,7 @@ String writeNativeFunction(AST::NativeFunctionDeclaration& nativeFunctionDeclara
>         auto thirdArgumentAddressSpace = thirdArgumentPointer.addressSpace();
>         auto thirdArgumentPointee = typeNamer.mangledNameForType(thirdArgumentPointer.elementType());
>         auto name = atomicName(nativeFunctionDeclaration.name().substring("Interlocked"_str.length()));
>-        stringBuilder.append(makeString("void ", outputFunctionName, '(', convertAddressSpace(firstArgumentAddressSpace), ' ', firstArgumentPointee, "* object, ", secondArgument, " operand, ", convertAddressSpace(thirdArgumentAddressSpace), ' ', thirdArgumentPointee, "* out) {\n"));
>+        stringBuilder.append(makeString("void ", outputFunctionName, '(', toString(firstArgumentAddressSpace), ' ', firstArgumentPointee, "* object, ", secondArgument, " operand, ", toString(thirdArgumentAddressSpace), ' ', thirdArgumentPointee, "* out) {\n"));
>         stringBuilder.append(makeString("    *out = atomic_fetch_", name, "_explicit(object, operand, memory_order_relaxed);\n"));
>         stringBuilder.append("}\n");
>         return stringBuilder.toString();
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLASTDumper.cpp b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLASTDumper.cpp
>index b7a99336406d5bacc64ecb03fa9de3911a35d1ef..302020a414be3425b55909530eec9318b7aa9f03 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLASTDumper.cpp
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLASTDumper.cpp
>@@ -584,11 +584,14 @@ void ASTDumper::visit(AST::CommaExpression& commaExpression)
> {
>     m_out.print("(");
>     bool once = false;
>+    int i = 0;
>     for (auto& expression : commaExpression.list()) {
>+        m_out.print(i);
>         if (once)
>             m_out.print(", ");
>         once = true;
>         visit(expression);
>+        ++i;
>     }
>     m_out.print(")");
> }
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLChecker.cpp b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLChecker.cpp
>index 75578a49b29a894ddf70211d6e5e1ce9775ae7ac..d9c37ce27bb8b954329d7b2a96500171390948c4 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLChecker.cpp
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLChecker.cpp
>@@ -36,6 +36,7 @@
> #include "WHLSLDereferenceExpression.h"
> #include "WHLSLDoWhileLoop.h"
> #include "WHLSLDotExpression.h"
>+#include "WHLSLEntryPointType.h"
> #include "WHLSLForLoop.h"
> #include "WHLSLGatherEntryPointItems.h"
> #include "WHLSLIfStatement.h"
>@@ -354,7 +355,7 @@ static bool checkOperatorOverload(const AST::FunctionDefinition& functionDefinit
>         Vector<ResolvingType> argumentTypes;
>         Vector<std::reference_wrapper<ResolvingType>> argumentTypeReferences;
>         for (size_t i = 0; i < numExpectedParameters - 1; ++i)
>-            argumentTypes.append((*functionDefinition.parameters()[0]->type())->clone());
>+            argumentTypes.append((*functionDefinition.parameters()[i]->type())->clone());
>         for (auto& argumentType : argumentTypes)
>             argumentTypeReferences.append(argumentType);
>         auto* overload = resolveFunctionOverloadImpl(*getterFuncs, argumentTypeReferences, nullptr);
>@@ -491,6 +492,8 @@ private:
>     void visit(AST::TernaryExpression&) override;
>     void visit(AST::CallExpression&) override;
> 
>+    void finishVisiting(AST::PropertyAccessExpression&, ResolvingType* additionalArgumentType = nullptr);
>+
>     HashMap<AST::Expression*, ResolvingType> m_typeMap;
>     HashMap<AST::Expression*, AST::TypeAnnotation> m_typeAnnotations;
>     HashSet<String> m_vertexEntryPoints;
>@@ -966,9 +969,9 @@ void Checker::visit(AST::MakeArrayReferenceExpression& makeArrayReferenceExpress
>     assignType(makeArrayReferenceExpression, makeUniqueRef<AST::ArrayReferenceType>(Lexer::Token(makeArrayReferenceExpression.origin()), *leftAddressSpace, leftValueType->clone()));
> }
> 
>-void Checker::visit(AST::DotExpression& dotExpression)
>+void Checker::finishVisiting(AST::PropertyAccessExpression& propertyAccessExpression, ResolvingType* additionalArgumentType)
> {
>-    auto baseInfo = recurseAndGetInfo(dotExpression.base());
>+    auto baseInfo = recurseAndGetInfo(propertyAccessExpression.base());
>     if (!baseInfo)
>         return;
>     auto baseUnnamedType = commit(baseInfo->resolvingType);
>@@ -979,65 +982,89 @@ void Checker::visit(AST::DotExpression& dotExpression)
>     AST::UnnamedType* getterReturnType = nullptr;
>     {
>         Vector<std::reference_wrapper<ResolvingType>> getterArgumentTypes { baseInfo->resolvingType };
>-        getterFunction = resolveFunctionOverloadImpl(dotExpression.possibleGetterOverloads(), getterArgumentTypes, nullptr);
>-        if (getterFunction)
>+        if (additionalArgumentType)
>+            getterArgumentTypes.append(*additionalArgumentType);
>+        if ((getterFunction = resolveFunctionOverloadImpl(propertyAccessExpression.possibleGetterOverloads(), getterArgumentTypes, nullptr)))
>             getterReturnType = &getterFunction->type();
>     }
> 
>     AST::FunctionDeclaration* anderFunction = nullptr;
>     AST::UnnamedType* anderReturnType = nullptr;
>-    if (auto leftAddressSpace = baseInfo->typeAnnotation.leftAddressSpace()) {
>-        auto argumentType = makeUniqueRef<AST::PointerType>(Lexer::Token(dotExpression.origin()), *leftAddressSpace, baseUnnamedType->get().clone());
>-        Vector<std::reference_wrapper<ResolvingType>> anderArgumentTypes { baseInfo->resolvingType };
>-        anderFunction = resolveFunctionOverloadImpl(dotExpression.possibleAnderOverloads(), anderArgumentTypes, nullptr);
>-        if (anderFunction)
>+    auto leftAddressSpace = baseInfo->typeAnnotation.leftAddressSpace();
>+    if (leftAddressSpace) {
>+        ResolvingType argumentType = { makeUniqueRef<AST::PointerType>(Lexer::Token(propertyAccessExpression.origin()), *leftAddressSpace, baseUnnamedType->get().clone()) };
>+        Vector<std::reference_wrapper<ResolvingType>> anderArgumentTypes { argumentType };
>+        if (additionalArgumentType)
>+            anderArgumentTypes.append(*additionalArgumentType);
>+        if ((anderFunction = resolveFunctionOverloadImpl(propertyAccessExpression.possibleAnderOverloads(), anderArgumentTypes, nullptr)))
>             anderReturnType = &downcast<AST::PointerType>(anderFunction->type()).elementType(); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198164 Enforce the return of anders will always be a pointer
>     }
> 
>     AST::FunctionDeclaration* threadAnderFunction = nullptr;
>     AST::UnnamedType* threadAnderReturnType = nullptr;
>     {
>-        auto argumentType = makeUniqueRef<AST::PointerType>(Lexer::Token(dotExpression.origin()), AST::AddressSpace::Thread, baseUnnamedType->get().clone());
>-        Vector<std::reference_wrapper<ResolvingType>> threadAnderArgumentTypes { baseInfo->resolvingType };
>-        threadAnderFunction = resolveFunctionOverloadImpl(dotExpression.possibleAnderOverloads(), threadAnderArgumentTypes, nullptr);
>-        if (threadAnderFunction)
>+        ResolvingType argumentType = { makeUniqueRef<AST::PointerType>(Lexer::Token(propertyAccessExpression.origin()), AST::AddressSpace::Thread, baseUnnamedType->get().clone()) };
>+        Vector<std::reference_wrapper<ResolvingType>> threadAnderArgumentTypes { argumentType };
>+        if (additionalArgumentType)
>+            threadAnderArgumentTypes.append(*additionalArgumentType);
>+        if ((threadAnderFunction = resolveFunctionOverloadImpl(propertyAccessExpression.possibleAnderOverloads(), threadAnderArgumentTypes, nullptr)))
>             threadAnderReturnType = &downcast<AST::PointerType>(threadAnderFunction->type()).elementType(); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198164 Enforce the return of anders will always be a pointer
>     }
> 
>-    if (!getterFunction && !anderFunction) {
>+    if (leftAddressSpace && !anderFunction && !getterFunction) {
>+        setError();
>+        return;
>+    }
>+
>+    if (!leftAddressSpace && !threadAnderFunction && !getterFunction) {
>         setError();
>         return;
>     }
>-    if (getterFunction && anderFunction) {
>+
>+    if (threadAnderFunction && getterFunction) {
>         setError();
>         return;
>     }
>+
>     if (anderFunction && threadAnderFunction && !matches(*anderReturnType, *threadAnderReturnType)) {
>         setError();
>         return;
>     }
> 
>-    AST::UnnamedType* fieldType = getterReturnType ? getterReturnType : anderReturnType;
>+    if (getterFunction && anderFunction && !matches(*getterReturnType, *anderReturnType)) {
>+        setError();
>+        return;
>+    }
>+
>+    if (getterFunction && threadAnderFunction && !matches(*getterReturnType, *threadAnderReturnType)) {
>+        setError();
>+        return;
>+    }
>+
>+    AST::UnnamedType* fieldType = getterReturnType ? getterReturnType : anderReturnType ? anderReturnType : threadAnderReturnType;
> 
>     AST::FunctionDeclaration* setterFunction = nullptr;
>     AST::UnnamedType* setterReturnType = nullptr;
>     {
>         ResolvingType fieldResolvingType(fieldType->clone());
>-        Vector<std::reference_wrapper<ResolvingType>> setterArgumentTypes { baseInfo->resolvingType, fieldResolvingType };
>-        setterFunction = resolveFunctionOverloadImpl(dotExpression.possibleSetterOverloads(), setterArgumentTypes, nullptr);
>+        Vector<std::reference_wrapper<ResolvingType>> setterArgumentTypes { baseInfo->resolvingType };
>+        if (additionalArgumentType)
>+            setterArgumentTypes.append(*additionalArgumentType);
>+        setterArgumentTypes.append(fieldResolvingType);
>+        setterFunction = resolveFunctionOverloadImpl(propertyAccessExpression.possibleSetterOverloads(), setterArgumentTypes, nullptr);
>         if (setterFunction)
>             setterReturnType = &setterFunction->type();
>     }
> 
>-    if (setterFunction && anderFunction) {
>+    if (setterFunction && !getterFunction) {
>         setError();
>         return;
>     }
> 
>-    dotExpression.setGetterFunction(getterFunction);
>-    dotExpression.setAnderFunction(anderFunction);
>-    dotExpression.setThreadAnderFunction(threadAnderFunction);
>-    dotExpression.setSetterFunction(setterFunction);
>+    propertyAccessExpression.setGetterFunction(getterFunction);
>+    propertyAccessExpression.setAnderFunction(anderFunction);
>+    propertyAccessExpression.setThreadAnderFunction(threadAnderFunction);
>+    propertyAccessExpression.setSetterFunction(setterFunction);
> 
>     AST::TypeAnnotation typeAnnotation = AST::RightValue();
>     if (auto leftAddressSpace = baseInfo->typeAnnotation.leftAddressSpace()) {
>@@ -1045,14 +1072,22 @@ void Checker::visit(AST::DotExpression& dotExpression)
>             typeAnnotation = AST::LeftValue { *leftAddressSpace };
>         else if (setterFunction)
>             typeAnnotation = AST::AbstractLeftValue();
>-    } else if (!baseInfo->typeAnnotation.isRightValue() && (setterFunction || anderFunction))
>+    } else if (!baseInfo->typeAnnotation.isRightValue() && (setterFunction || threadAnderFunction))
>         typeAnnotation = AST::AbstractLeftValue();
>-    assignType(dotExpression, fieldType->clone(), WTFMove(typeAnnotation));
>+    assignType(propertyAccessExpression, fieldType->clone(), WTFMove(typeAnnotation));
> }
> 
>-void Checker::visit(AST::IndexExpression&)
>+void Checker::visit(AST::DotExpression& dotExpression)
>+{
>+    finishVisiting(dotExpression);
>+}
>+
>+void Checker::visit(AST::IndexExpression& indexExpression)
> {
>-    // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198163 Implement this.
>+    auto baseInfo = recurseAndGetInfo(indexExpression.indexExpression());
>+    if (!baseInfo)
>+        return;
>+    finishVisiting(indexExpression, &baseInfo->resolvingType);
> }
> 
> void Checker::visit(AST::VariableReference& variableReference)
>@@ -1362,7 +1397,7 @@ void Checker::visit(AST::CommaExpression& commaExpression)
>     if (error())
>         return;
>     auto lastInfo = getInfo(commaExpression.list().last());
>-    forwardType(commaExpression, lastInfo->resolvingType, lastInfo->typeAnnotation);
>+    forwardType(commaExpression, lastInfo->resolvingType);
> }
> 
> void Checker::visit(AST::TernaryExpression& ternaryExpression)
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLInferTypes.h b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLInferTypes.h
>index 173d3014497d2a3742e80670f80bc96ff0f79c4d..19fb1ffcf814414cd6a227dea8943055028eb0ac 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLInferTypes.h
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLInferTypes.h
>@@ -47,7 +47,6 @@ class UnnamedType;
> bool matches(const AST::UnnamedType&, const AST::UnnamedType&);
> bool matches(const AST::NamedType&, const AST::NamedType&);
> bool matches(const AST::UnnamedType&, const AST::NamedType&);
>-// FIXME: Is anyone actually using the return type here?
> Optional<UniqueRef<AST::UnnamedType>> matchAndCommit(AST::UnnamedType&, AST::ResolvableType&);
> Optional<UniqueRef<AST::UnnamedType>> matchAndCommit(AST::NamedType&, AST::ResolvableType&);
> Optional<UniqueRef<AST::UnnamedType>> matchAndCommit(AST::ResolvableType&, AST::ResolvableType&);
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPrepare.cpp b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPrepare.cpp
>index e3535a4c4d5d8d02385cf38ccfe4a3d048cc1348..39e193bfe1fa6a702b05067f417cd23a3f93c304 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPrepare.cpp
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPrepare.cpp
>@@ -55,7 +55,7 @@ namespace WebCore {
> 
> namespace WHLSL {
> 
>-static constexpr bool dumpASTBeforeEachPass = false;
>+static constexpr bool dumpASTBeforeEachPass = true;
> static constexpr bool dumpASTAfterParsing = false;
> static constexpr bool dumpASTAtEnd = true;
> static constexpr bool alwaysDumpPassFailures = false;
>@@ -124,7 +124,9 @@ static Optional<Program> prepareShared(String& whlslSource)
>     RUN_PASS(check, program);
> 
>     checkLiteralTypes(program);
>+    dumpASTBetweenEachPassIfNeeded(program, "AST before resolveProperties");
>     resolveProperties(program);
>+    dumpASTBetweenEachPassIfNeeded(program, "AST before findHighZombies");
>     findHighZombies(program);
>     RUN_PASS(checkStatementBehavior, program);
>     RUN_PASS(checkRecursion, program);
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPropertyResolver.cpp b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPropertyResolver.cpp
>index fa069cabce204d379ef6568e7d926073792f85e7..fd55934ac71713d59ff9f8b662b0cb1bb38074c4 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPropertyResolver.cpp
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLPropertyResolver.cpp
>@@ -51,10 +51,11 @@ public:
> private:
>     void visit(AST::FunctionDefinition&) override;
>     void visit(AST::DotExpression&) override;
>+    void visit(AST::IndexExpression&) override;
>     void visit(AST::AssignmentExpression&) override;
>     void visit(AST::ReadModifyWriteExpression&) override;
> 
>-    bool simplifyRightValue(AST::DotExpression&);
>+    bool simplifyRightValue(AST::PropertyAccessExpression&);
>     bool simplifyAbstractLeftValue(AST::AssignmentExpression&, AST::DotExpression&, UniqueRef<AST::Expression>&& right);
>     void simplifyLeftValue(AST::Expression&);
> 
>@@ -68,6 +69,14 @@ void PropertyResolver::visit(AST::DotExpression& dotExpression)
>         setError();
> }
> 
>+void PropertyResolver::visit(AST::IndexExpression& indexExpression)
>+{
>+    checkErrorAndVisit(indexExpression.indexExpression());
>+    // Unless we're inside an AssignmentExpression or a ReadModifyWriteExpression, we're a right value.
>+    if (!simplifyRightValue(indexExpression))
>+        setError();
>+}
>+
> void PropertyResolver::visit(AST::FunctionDefinition& functionDefinition)
> {
>     Visitor::visit(functionDefinition);
>@@ -75,79 +84,105 @@ void PropertyResolver::visit(AST::FunctionDefinition& functionDefinition)
>         functionDefinition.block().statements().insert(0, makeUniqueRef<AST::VariableDeclarationsStatement>(Lexer::Token(m_variableDeclarations[0]->origin()), WTFMove(m_variableDeclarations)));
> }
> 
>-static Optional<UniqueRef<AST::Expression>> setterCall(AST::DotExpression& dotExpression, UniqueRef<AST::Expression>&& newValue, const std::function<UniqueRef<AST::Expression>()>& leftValueFactory, const std::function<UniqueRef<AST::Expression>()>& pointerToLeftValueFactory)
>+static Optional<UniqueRef<AST::Expression>> setterCall(AST::PropertyAccessExpression& propertyAccessExpression, AST::FunctionDeclaration* relevantAnder, UniqueRef<AST::Expression>&& newValue, const std::function<UniqueRef<AST::Expression>()>& leftValueFactory, const std::function<UniqueRef<AST::Expression>()>& pointerToLeftValueFactory, AST::VariableDeclaration* indexVariable)
> {
>-    if (dotExpression.anderFunction()) {
>+    if (relevantAnder) {
>         // *operator&.foo(&v) = newValue
>-        if (!dotExpression.threadAnderFunction())
>-            return WTF::nullopt;
>-        
>         Vector<UniqueRef<AST::Expression>> arguments;
>         arguments.append(pointerToLeftValueFactory());
>-        auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(dotExpression.origin()), String(dotExpression.threadAnderFunction()->name()), WTFMove(arguments));
>-        callExpression->setType(dotExpression.threadAnderFunction()->type().clone());
>+        if (indexVariable) {
>+            auto variableReference = makeUniqueRef<AST::VariableReference>(AST::VariableReference::wrap(*indexVariable));
>+            ASSERT(indexVariable->type());
>+            variableReference->setType(indexVariable->type()->clone());
>+            variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
>+            arguments.append(WTFMove(variableReference));
>+        }
>+
>+        auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(propertyAccessExpression.origin()), String(relevantAnder->name()), WTFMove(arguments));
>+        callExpression->setType(relevantAnder->type().clone());
>         callExpression->setTypeAnnotation(AST::RightValue());
>-        callExpression->setFunction(*dotExpression.threadAnderFunction());
>+        callExpression->setFunction(*relevantAnder);
> 
>-        auto dereferenceExpression = makeUniqueRef<AST::DereferenceExpression>(Lexer::Token(dotExpression.origin()), WTFMove(callExpression));
>-        dereferenceExpression->setType(downcast<AST::PointerType>(dotExpression.threadAnderFunction()->type()).elementType().clone());
>+        auto dereferenceExpression = makeUniqueRef<AST::DereferenceExpression>(Lexer::Token(propertyAccessExpression.origin()), WTFMove(callExpression));
>+        dereferenceExpression->setType(downcast<AST::PointerType>(relevantAnder->type()).elementType().clone());
>         dereferenceExpression->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread });
> 
>-        auto assignmentExpression = makeUniqueRef<AST::AssignmentExpression>(Lexer::Token(dotExpression.origin()), WTFMove(dereferenceExpression), WTFMove(newValue));
>-        assignmentExpression->setType(downcast<AST::PointerType>(dotExpression.threadAnderFunction()->type()).elementType().clone());
>+        auto assignmentExpression = makeUniqueRef<AST::AssignmentExpression>(Lexer::Token(propertyAccessExpression.origin()), WTFMove(dereferenceExpression), WTFMove(newValue));
>+        assignmentExpression->setType(downcast<AST::PointerType>(relevantAnder->type()).elementType().clone());
>         assignmentExpression->setTypeAnnotation(AST::RightValue());
> 
>         return UniqueRef<AST::Expression>(WTFMove(assignmentExpression));
>     }
> 
>     // v = operator.foo=(v, newValue)
>-    ASSERT(dotExpression.setterFunction());
>+    ASSERT(propertyAccessExpression.setterFunction());
> 
>     Vector<UniqueRef<AST::Expression>> arguments;
>     arguments.append(leftValueFactory());
>+    if (indexVariable) {
>+        auto variableReference = makeUniqueRef<AST::VariableReference>(AST::VariableReference::wrap(*indexVariable));
>+        ASSERT(indexVariable->type());
>+        variableReference->setType(indexVariable->type()->clone());
>+        variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
>+        arguments.append(WTFMove(variableReference));
>+    }
>     arguments.append(WTFMove(newValue));
>-    auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(dotExpression.origin()), String(dotExpression.setterFunction()->name()), WTFMove(arguments));
>-    callExpression->setType(dotExpression.setterFunction()->type().clone());
>+
>+    auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(propertyAccessExpression.origin()), String(propertyAccessExpression.setterFunction()->name()), WTFMove(arguments));
>+    callExpression->setType(propertyAccessExpression.setterFunction()->type().clone());
>     callExpression->setTypeAnnotation(AST::RightValue());
>-    callExpression->setFunction(*dotExpression.setterFunction());
>+    callExpression->setFunction(*propertyAccessExpression.setterFunction());
> 
>-    auto assignmentExpression = makeUniqueRef<AST::AssignmentExpression>(Lexer::Token(dotExpression.origin()), leftValueFactory(), WTFMove(callExpression));
>-    assignmentExpression->setType(dotExpression.setterFunction()->type().clone());
>+    auto assignmentExpression = makeUniqueRef<AST::AssignmentExpression>(Lexer::Token(propertyAccessExpression.origin()), leftValueFactory(), WTFMove(callExpression));
>+    assignmentExpression->setType(propertyAccessExpression.setterFunction()->type().clone());
>     assignmentExpression->setTypeAnnotation(AST::RightValue());
> 
>     return UniqueRef<AST::Expression>(WTFMove(assignmentExpression));
> }
> 
>-static Optional<UniqueRef<AST::Expression>> getterCall(AST::DotExpression& dotExpression, const std::function<UniqueRef<AST::Expression>()>& leftValueFactory, const std::function<UniqueRef<AST::Expression>()>& pointerToLeftValueFactory)
>+static Optional<UniqueRef<AST::Expression>> getterCall(AST::PropertyAccessExpression& propertyAccessExpression, AST::FunctionDeclaration* relevantAnder, const std::function<UniqueRef<AST::Expression>()>& leftValueFactory, const std::function<UniqueRef<AST::Expression>()>& pointerToLeftValueFactory, AST::VariableDeclaration* indexVariable)
> {
>-    if (dotExpression.anderFunction()) {
>+    if (relevantAnder) {
>         // *operator&.foo(&v)
>-        if (!dotExpression.threadAnderFunction())
>-            return WTF::nullopt;
>-        
>         Vector<UniqueRef<AST::Expression>> arguments;
>         arguments.append(pointerToLeftValueFactory());
>-        auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(dotExpression.origin()), String(dotExpression.threadAnderFunction()->name()), WTFMove(arguments));
>-        callExpression->setType(dotExpression.threadAnderFunction()->type().clone());
>+        if (indexVariable) {
>+            auto variableReference = makeUniqueRef<AST::VariableReference>(AST::VariableReference::wrap(*indexVariable));
>+            ASSERT(indexVariable->type());
>+            variableReference->setType(indexVariable->type()->clone());
>+            variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
>+            arguments.append(WTFMove(variableReference));
>+        }
>+
>+        auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(propertyAccessExpression.origin()), String(relevantAnder->name()), WTFMove(arguments));
>+        callExpression->setType(relevantAnder->type().clone());
>         callExpression->setTypeAnnotation(AST::RightValue());
>-        callExpression->setFunction(*dotExpression.threadAnderFunction());
>+        callExpression->setFunction(*relevantAnder);
> 
>-        auto dereferenceExpression = makeUniqueRef<AST::DereferenceExpression>(Lexer::Token(dotExpression.origin()), WTFMove(callExpression));
>-        dereferenceExpression->setType(downcast<AST::PointerType>(dotExpression.threadAnderFunction()->type()).elementType().clone());
>+        auto dereferenceExpression = makeUniqueRef<AST::DereferenceExpression>(Lexer::Token(propertyAccessExpression.origin()), WTFMove(callExpression));
>+        dereferenceExpression->setType(downcast<AST::PointerType>(relevantAnder->type()).elementType().clone());
>         dereferenceExpression->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread });
> 
>         return UniqueRef<AST::Expression>(WTFMove(dereferenceExpression));
>     }
> 
>     // operator.foo(v)
>-    ASSERT(dotExpression.getterFunction());
>+    ASSERT(propertyAccessExpression.getterFunction());
>     
>     Vector<UniqueRef<AST::Expression>> arguments;
>     arguments.append(leftValueFactory());
>-    auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(dotExpression.origin()), String(dotExpression.getterFunction()->name()), WTFMove(arguments));
>-    callExpression->setType(dotExpression.getterFunction()->type().clone());
>+    if (indexVariable) {
>+        auto variableReference = makeUniqueRef<AST::VariableReference>(AST::VariableReference::wrap(*indexVariable));
>+        ASSERT(indexVariable->type());
>+        variableReference->setType(indexVariable->type()->clone());
>+        variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
>+        arguments.append(WTFMove(variableReference));
>+    }
>+
>+    auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(propertyAccessExpression.origin()), String(propertyAccessExpression.getterFunction()->name()), WTFMove(arguments));
>+    callExpression->setType(propertyAccessExpression.getterFunction()->type().clone());
>     callExpression->setTypeAnnotation(AST::RightValue());
>-    callExpression->setFunction(*dotExpression.getterFunction());
>+    callExpression->setFunction(*propertyAccessExpression.getterFunction());
> 
>     return UniqueRef<AST::Expression>(WTFMove(callExpression));
> }
>@@ -161,7 +196,7 @@ struct ModificationResult {
>     Vector<UniqueRef<AST::Expression>> expressions;
>     UniqueRef<AST::Expression> result;
> };
>-static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::function<Optional<ModificationResult>(Optional<UniqueRef<AST::Expression>>&&)> modification)
>+static Optional<ModifyResult> modify(AST::PropertyAccessExpression& propertyAccessExpression, std::function<Optional<ModificationResult>(Optional<UniqueRef<AST::Expression>>&&)> modification)
> {
>     // Consider a.b.c.d++;
>     // This would get transformed into:
>@@ -181,22 +216,22 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
>     // r = operator.d=(r, newValue);
>     // q = operator.c=(q, r);
>     //
>-    // Step 4:
>+    // Step 5:
>     // *p = operator.b=(*p, q);
> 
>     // If the expression is a.b.c.d = e, Step 3 disappears and "newValue" in step 4 becomes "e".
>     
> 
>     // Find the ".b" ".c" and ".d" expressions. They end up in the order [".d", ".c", ".b"].
>-    Vector<std::reference_wrapper<AST::DotExpression>> chain;
>-    AST::DotExpression* iterator = &dotExpression;
>+    Vector<std::reference_wrapper<AST::PropertyAccessExpression>> chain;
>+    AST::PropertyAccessExpression* iterator = &propertyAccessExpression;
>     while (true) {
>         chain.append(*iterator);
>         if (iterator->base().typeAnnotation().leftAddressSpace())
>             break;
>         ASSERT(!iterator->base().typeAnnotation().isRightValue());
>-        ASSERT(is<AST::DotExpression>(iterator->base())); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198163 Make this work with index expressions
>-        iterator = &downcast<AST::DotExpression>(iterator->base());
>+        ASSERT(is<AST::PropertyAccessExpression>(iterator->base()));
>+        iterator = &downcast<AST::PropertyAccessExpression>(iterator->base());
>     }
>     auto leftExpression = iterator->takeBase();
>     AST::Expression& innerLeftExpression = leftExpression;
>@@ -208,8 +243,29 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
>     Vector<UniqueRef<AST::VariableDeclaration>> intermediateVariables;
>     intermediateVariables.reserveInitialCapacity(chain.size() - 1);
>     for (size_t i = 1; i < chain.size(); ++i) {
>-        auto& dotExpression = static_cast<AST::DotExpression&>(chain[i]);
>-        intermediateVariables.uncheckedAppend(makeUniqueRef<AST::VariableDeclaration>(Lexer::Token(dotExpression.origin()), AST::Qualifiers(), dotExpression.resolvedType().clone(), String(), WTF::nullopt, WTF::nullopt));
>+        auto& propertyAccessExpression = static_cast<AST::PropertyAccessExpression&>(chain[i]);
>+        intermediateVariables.uncheckedAppend(makeUniqueRef<AST::VariableDeclaration>(Lexer::Token(propertyAccessExpression.origin()), AST::Qualifiers(), propertyAccessExpression.resolvedType().clone(), String(), WTF::nullopt, WTF::nullopt));
>+    }
>+
>+    // Consider a[foo()][b] = c;
>+    // Naively, This would get expanded to:
>+    //
>+    // temp = operator[](a, foo());
>+    // temp = operator[]=(temp, b, c);
>+    // a = operator[]=(a, foo(), temp);
>+    //
>+    // However, if we did this, we would have to run foo() twice, which would be incorrect.
>+    // Instead, we need to save foo() and b into more temporary variables.
>+    // These temporary variables are parallel to "chain" above, with nullopt referring to a DotExpression (which doesn't have an index value to save to a variable).
>+    Vector<Optional<UniqueRef<AST::VariableDeclaration>>> indexVariables;
>+    indexVariables.reserveInitialCapacity(chain.size());
>+    for (AST::PropertyAccessExpression& propertyAccessExpression : chain) {
>+        if (!is<AST::IndexExpression>(propertyAccessExpression)) {
>+            indexVariables.append(WTF::nullopt);
>+            continue;
>+        }
>+        auto& indexExpression = downcast<AST::IndexExpression>(propertyAccessExpression);
>+        indexVariables.uncheckedAppend(makeUniqueRef<AST::VariableDeclaration>(Lexer::Token(propertyAccessExpression.origin()), AST::Qualifiers(), indexExpression.indexExpression().resolvedType().clone(), String(), WTF::nullopt, WTF::nullopt));
>     }
> 
>     Vector<UniqueRef<AST::Expression>> expressions;
>@@ -248,7 +304,7 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
>         variableReference->setType(pointerVariable->type()->clone());
>         variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
> 
>-        auto dereferenceExpression = makeUniqueRef<AST::DereferenceExpression>(Lexer::Token(dotExpression.origin()), WTFMove(variableReference));
>+        auto dereferenceExpression = makeUniqueRef<AST::DereferenceExpression>(Lexer::Token(propertyAccessExpression.origin()), WTFMove(variableReference));
>         ASSERT(pointerVariable->type());
>         dereferenceExpression->setType(downcast<AST::PointerType>(*pointerVariable->type()).elementType().clone());
>         dereferenceExpression->setTypeAnnotation(AST::LeftValue { downcast<AST::PointerType>(*pointerVariable->type()).addressSpace() });
>@@ -261,9 +317,9 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
>             variableReference->setType(previous->type()->clone());
>             variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
> 
>-            auto makePointerExpression = makeUniqueRef<AST::MakePointerExpression>(Lexer::Token(dotExpression.origin()), WTFMove(variableReference));
>+            auto makePointerExpression = makeUniqueRef<AST::MakePointerExpression>(Lexer::Token(propertyAccessExpression.origin()), WTFMove(variableReference));
>             ASSERT(previous->type());
>-            makePointerExpression->setType(makeUniqueRef<AST::PointerType>(Lexer::Token(dotExpression.origin()), AST::AddressSpace::Thread, previous->type()->clone()));
>+            makePointerExpression->setType(makeUniqueRef<AST::PointerType>(Lexer::Token(propertyAccessExpression.origin()), AST::AddressSpace::Thread, previous->type()->clone()));
>             makePointerExpression->setTypeAnnotation(AST::RightValue());
>             return makePointerExpression;
>         }
>@@ -274,11 +330,32 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
>         variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
>         return variableReference;
>     };
>+    auto appendIndexAssignment = [&](AST::PropertyAccessExpression& propertyAccessExpression, Optional<UniqueRef<AST::VariableDeclaration>>& indexVariable) {
>+        if (!indexVariable)
>+            return;
>+
>+        auto& indexExpression = downcast<AST::IndexExpression>(propertyAccessExpression);
>+
>+        auto variableReference = makeUniqueRef<AST::VariableReference>(AST::VariableReference::wrap(*indexVariable));
>+        ASSERT(indexVariable->get().type());
>+        variableReference->setType(indexVariable->get().type()->clone());
>+        variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
>+
>+        auto assignmentExpression = makeUniqueRef<AST::AssignmentExpression>(Lexer::Token(propertyAccessExpression.origin()), WTFMove(variableReference), indexExpression.takeIndex());
>+        assignmentExpression->setType(indexVariable->get().type()->clone());
>+        assignmentExpression->setTypeAnnotation(AST::RightValue());
>+
>+        expressions.append(WTFMove(assignmentExpression));
>+    };
>     for (size_t i = chain.size(); --i; ) {
>-        AST::DotExpression& dotExpression = chain[i];
>+        AST::PropertyAccessExpression& propertyAccessExpression = chain[i];
>         AST::VariableDeclaration& variableDeclaration = intermediateVariables[i - 1];
>+        Optional<UniqueRef<AST::VariableDeclaration>>& indexVariable = indexVariables[i];
>+
>+        appendIndexAssignment(propertyAccessExpression, indexVariable);
> 
>-        auto callExpression = getterCall(dotExpression, previousLeftValue, pointerToPreviousLeftValue);
>+        AST::FunctionDeclaration* relevantAnder = i == chain.size() - 1 ? propertyAccessExpression.anderFunction() : propertyAccessExpression.threadAnderFunction();
>+        auto callExpression = getterCall(propertyAccessExpression, relevantAnder, previousLeftValue, pointerToPreviousLeftValue, indexVariable ? &*indexVariable : nullptr);
> 
>         if (!callExpression)
>             return WTF::nullopt;
>@@ -288,7 +365,7 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
>         variableReference->setType(variableDeclaration.type()->clone());
>         variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
> 
>-        auto assignmentExpression = makeUniqueRef<AST::AssignmentExpression>(Lexer::Token(dotExpression.origin()), WTFMove(variableReference), WTFMove(*callExpression));
>+        auto assignmentExpression = makeUniqueRef<AST::AssignmentExpression>(Lexer::Token(propertyAccessExpression.origin()), WTFMove(variableReference), WTFMove(*callExpression));
>         assignmentExpression->setType(variableDeclaration.type()->clone());
>         assignmentExpression->setTypeAnnotation(AST::RightValue());
> 
>@@ -296,7 +373,9 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
>         
>         previous = &variableDeclaration;
>     }
>-    auto lastGetterCallExpression = getterCall(chain[0], previousLeftValue, pointerToPreviousLeftValue);
>+    appendIndexAssignment(chain[0], indexVariables[0]);
>+    AST::FunctionDeclaration* relevantAnder = chain.size() == 1 ? propertyAccessExpression.anderFunction() : propertyAccessExpression.threadAnderFunction();
>+    auto lastGetterCallExpression = getterCall(chain[0], relevantAnder, previousLeftValue, pointerToPreviousLeftValue, indexVariables[0] ? &*(indexVariables[0]) : nullptr);
> 
>     // Step 3:
>     auto modificationResult = modification(WTFMove(lastGetterCallExpression));
>@@ -309,10 +388,11 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
>     UniqueRef<AST::Expression> rightValue = WTFMove(modificationResult->result);
>     auto expressionType = rightValue->resolvedType().clone();
>     for (size_t i = 0; i < chain.size() - 1; ++i) {
>-        AST::DotExpression& dotExpression = chain[i];
>+        AST::PropertyAccessExpression& propertyAccessExpression = chain[i];
>         AST::VariableDeclaration& variableDeclaration = intermediateVariables[i];
>+        Optional<UniqueRef<AST::VariableDeclaration>>& indexVariable = indexVariables[i];
> 
>-        auto assignmentExpression = setterCall(dotExpression, WTFMove(rightValue), [&]() -> UniqueRef<AST::Expression> {
>+        auto assignmentExpression = setterCall(propertyAccessExpression, propertyAccessExpression.threadAnderFunction(), WTFMove(rightValue), [&]() -> UniqueRef<AST::Expression> {
>             auto variableReference = makeUniqueRef<AST::VariableReference>(AST::VariableReference::wrap(variableDeclaration));
>             ASSERT(variableDeclaration.type());
>             variableReference->setType(variableDeclaration.type()->clone());
>@@ -324,12 +404,12 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
>             variableReference->setType(variableDeclaration.type()->clone());
>             variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
> 
>-            auto makePointerExpression = makeUniqueRef<AST::MakePointerExpression>(Lexer::Token(dotExpression.origin()), WTFMove(variableReference));
>+            auto makePointerExpression = makeUniqueRef<AST::MakePointerExpression>(Lexer::Token(propertyAccessExpression.origin()), WTFMove(variableReference));
>             ASSERT(variableDeclaration.type());
>-            makePointerExpression->setType(makeUniqueRef<AST::PointerType>(Lexer::Token(dotExpression.origin()), AST::AddressSpace::Thread, variableDeclaration.type()->clone()));
>+            makePointerExpression->setType(makeUniqueRef<AST::PointerType>(Lexer::Token(propertyAccessExpression.origin()), AST::AddressSpace::Thread, variableDeclaration.type()->clone()));
>             makePointerExpression->setTypeAnnotation(AST::RightValue());
>             return makePointerExpression;
>-        });
>+        }, indexVariable ? &*indexVariable : nullptr);
> 
>         if (!assignmentExpression)
>             return WTF::nullopt;
>@@ -344,13 +424,14 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
> 
>     // Step 5:
>     {
>-        auto assignmentExpression = setterCall(chain[chain.size() - 1], WTFMove(rightValue), [&]() -> UniqueRef<AST::Expression> {
>+        AST::PropertyAccessExpression& propertyAccessExpression = chain[chain.size() - 1];
>+        auto assignmentExpression = setterCall(propertyAccessExpression, propertyAccessExpression.anderFunction(), WTFMove(rightValue), [&]() -> UniqueRef<AST::Expression> {
>             auto variableReference = makeUniqueRef<AST::VariableReference>(AST::VariableReference::wrap(pointerVariable));
>             ASSERT(pointerVariable->type());
>             variableReference->setType(pointerVariable->type()->clone());
>             variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
> 
>-            auto dereferenceExpression = makeUniqueRef<AST::DereferenceExpression>(Lexer::Token(dotExpression.origin()), WTFMove(variableReference));
>+            auto dereferenceExpression = makeUniqueRef<AST::DereferenceExpression>(Lexer::Token(propertyAccessExpression.origin()), WTFMove(variableReference));
>             ASSERT(pointerVariable->type());
>             dereferenceExpression->setType(downcast<AST::PointerType>(*pointerVariable->type()).elementType().clone());
>             dereferenceExpression->setTypeAnnotation(AST::LeftValue { downcast<AST::PointerType>(*pointerVariable->type()).addressSpace() });
>@@ -361,7 +442,7 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
>             variableReference->setType(pointerVariable->type()->clone());
>             variableReference->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread }); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198169 Is this right?
>             return variableReference;
>-        });
>+        }, indexVariables[indexVariables.size() - 1] ? &*(indexVariables[indexVariables.size() - 1]) : nullptr);
> 
>         if (!assignmentExpression)
>             return WTF::nullopt;
>@@ -373,6 +454,10 @@ static Optional<ModifyResult> modify(AST::DotExpression& dotExpression, std::fun
>     variableDeclarations.append(WTFMove(pointerVariable));
>     for (auto& intermediateVariable : intermediateVariables)
>         variableDeclarations.append(WTFMove(intermediateVariable));
>+    for (auto& indexVariable : indexVariables) {
>+        if (indexVariable)
>+            variableDeclarations.append(WTFMove(*indexVariable));
>+    }
> 
>     return {{ innerLeftExpression, WTFMove(expressions), WTFMove(variableDeclarations) }};
> }
>@@ -385,16 +470,13 @@ void PropertyResolver::visit(AST::AssignmentExpression& assignmentExpression)
>         return;
>     }
>     ASSERT(!assignmentExpression.left().typeAnnotation().isRightValue());
>-    if (!is<AST::DotExpression>(assignmentExpression.left())) {
>-        setError(); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198163 Make this work with index expressions.
>-        return;
>-    }
>+    ASSERT(is<AST::PropertyAccessExpression>(assignmentExpression.left()));
> 
>     auto type = assignmentExpression.right().resolvedType().clone();
> 
>     checkErrorAndVisit(assignmentExpression.right());
> 
>-    auto modifyResult = modify(downcast<AST::DotExpression>(assignmentExpression.left()), [&](Optional<UniqueRef<AST::Expression>>&&) -> Optional<ModificationResult> {
>+    auto modifyResult = modify(downcast<AST::PropertyAccessExpression>(assignmentExpression.left()), [&](Optional<UniqueRef<AST::Expression>>&&) -> Optional<ModificationResult> {
>         return {{ Vector<UniqueRef<AST::Expression>>(), assignmentExpression.takeRight() }};
>     });
> 
>@@ -424,6 +506,8 @@ void PropertyResolver::visit(AST::ReadModifyWriteExpression& readModifyWriteExpr
>         // newValue = ...;
>         // *p = newValue;
> 
>+        simplifyLeftValue(readModifyWriteExpression.leftValue());
>+
>         auto baseType = readModifyWriteExpression.leftValue().resolvedType().clone();
>         auto pointerType = makeUniqueRef<AST::PointerType>(Lexer::Token(readModifyWriteExpression.leftValue().origin()), *readModifyWriteExpression.leftValue().typeAnnotation().leftAddressSpace(), baseType->clone());
> 
>@@ -520,11 +604,11 @@ void PropertyResolver::visit(AST::ReadModifyWriteExpression& readModifyWriteExpr
>     }
> 
>     ASSERT(!readModifyWriteExpression.leftValue().typeAnnotation().isRightValue());
>-    if (!is<AST::DotExpression>(readModifyWriteExpression.leftValue())) {
>-        setError(); // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198163 Make this work with index expressions.
>+    if (!is<AST::PropertyAccessExpression>(readModifyWriteExpression.leftValue())) {
>+        setError();
>         return;
>     }
>-    auto modifyResult = modify(downcast<AST::DotExpression>(readModifyWriteExpression.leftValue()), [&](Optional<UniqueRef<AST::Expression>>&& lastGetterCallExpression) -> Optional<ModificationResult> {
>+    auto modifyResult = modify(downcast<AST::PropertyAccessExpression>(readModifyWriteExpression.leftValue()), [&](Optional<UniqueRef<AST::Expression>>&& lastGetterCallExpression) -> Optional<ModificationResult> {
>         Vector<UniqueRef<AST::Expression>> expressions;
>         if (!lastGetterCallExpression)
>             return WTF::nullopt;
>@@ -581,128 +665,150 @@ void PropertyResolver::visit(AST::ReadModifyWriteExpression& readModifyWriteExpr
>     m_variableDeclarations.append(WTFMove(newVariableDeclaration));
> }
> 
>-bool PropertyResolver::simplifyRightValue(AST::DotExpression& dotExpression)
>+bool PropertyResolver::simplifyRightValue(AST::PropertyAccessExpression& propertyAccessExpression)
> {
>-    Lexer::Token origin = dotExpression.origin();
>+    Lexer::Token origin = propertyAccessExpression.origin();
> 
>-    checkErrorAndVisit(dotExpression.base());
>+    checkErrorAndVisit(propertyAccessExpression.base());
> 
>-    if (auto* anderFunction = dotExpression.anderFunction()) {
>-        auto& base = dotExpression.base();
>-        if (auto leftAddressSpace = base.typeAnnotation().leftAddressSpace()) {
>-            auto makePointerExpression = makeUniqueRef<AST::MakePointerExpression>(Lexer::Token(origin), dotExpression.takeBase());
>+    auto& base = propertyAccessExpression.base();
>+    if (auto leftAddressSpace = base.typeAnnotation().leftAddressSpace()) {
>+        if (auto* anderFunction = propertyAccessExpression.anderFunction()) {
>+            auto makePointerExpression = makeUniqueRef<AST::MakePointerExpression>(Lexer::Token(origin), propertyAccessExpression.takeBase());
>             makePointerExpression->setType(makeUniqueRef<AST::PointerType>(Lexer::Token(origin), *leftAddressSpace, base.resolvedType().clone()));
>             makePointerExpression->setTypeAnnotation(AST::RightValue());
> 
>             Vector<UniqueRef<AST::Expression>> arguments;
>             arguments.append(WTFMove(makePointerExpression));
>+            if (is<AST::IndexExpression>(propertyAccessExpression))
>+                arguments.append(downcast<AST::IndexExpression>(propertyAccessExpression).takeIndex());
>             auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(origin), String(anderFunction->name()), WTFMove(arguments));
>             callExpression->setType(anderFunction->type().clone());
>             callExpression->setTypeAnnotation(AST::RightValue());
>             callExpression->setFunction(*anderFunction);
> 
>-            auto* dereferenceExpression = AST::replaceWith<AST::DereferenceExpression>(dotExpression, WTFMove(origin), WTFMove(callExpression));
>+            auto* dereferenceExpression = AST::replaceWith<AST::DereferenceExpression>(propertyAccessExpression, WTFMove(origin), WTFMove(callExpression));
>             dereferenceExpression->setType(downcast<AST::PointerType>(anderFunction->type()).elementType().clone());
>             dereferenceExpression->setTypeAnnotation(AST::LeftValue { downcast<AST::PointerType>(anderFunction->type()).addressSpace() });
>             return true;
>         }
>+    }
> 
>-        // We have an ander, but no left value to call it on. Let's save the value into a temporary variable to create a left value.
>-        // This is effectively inlining the functions the spec says are generated.
>-        if (!dotExpression.threadAnderFunction())
>-            return false;
>+    if (propertyAccessExpression.getterFunction()) {
>+        auto& getterFunction = *propertyAccessExpression.getterFunction();
>+        Vector<UniqueRef<AST::Expression>> arguments;
>+        arguments.append(propertyAccessExpression.takeBase());
>+        if (is<AST::IndexExpression>(propertyAccessExpression))
>+            arguments.append(downcast<AST::IndexExpression>(propertyAccessExpression).takeIndex());
>+        auto* callExpression = AST::replaceWith<AST::CallExpression>(propertyAccessExpression, WTFMove(origin), String(getterFunction.name()), WTFMove(arguments));
>+        callExpression->setFunction(getterFunction);
>+        callExpression->setType(getterFunction.type().clone());
>+        callExpression->setTypeAnnotation(AST::RightValue());
>+        return true;
>+    }
> 
>-        auto variableDeclaration = makeUniqueRef<AST::VariableDeclaration>(Lexer::Token(origin), AST::Qualifiers(), base.resolvedType().clone(), String(), WTF::nullopt, WTF::nullopt);
>+    // We have an ander, but no left value to call it on. Let's save the value into a temporary variable to create a left value.
>+    // This is effectively inlining the functions the spec says are generated.
>+    ASSERT(propertyAccessExpression.threadAnderFunction());
>+    auto* threadAnderFunction = propertyAccessExpression.threadAnderFunction();
> 
>-        auto variableReference1 = makeUniqueRef<AST::VariableReference>(AST::VariableReference::wrap(variableDeclaration));
>-        variableReference1->setType(base.resolvedType().clone());
>-        variableReference1->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread });
>+    auto variableDeclaration = makeUniqueRef<AST::VariableDeclaration>(Lexer::Token(origin), AST::Qualifiers(), base.resolvedType().clone(), String(), WTF::nullopt, WTF::nullopt);
> 
>-        auto assignmentExpression = makeUniqueRef<AST::AssignmentExpression>(Lexer::Token(origin), WTFMove(variableReference1), dotExpression.takeBase());
>-        assignmentExpression->setType(base.resolvedType().clone());
>-        assignmentExpression->setTypeAnnotation(AST::RightValue());
>+    auto variableReference1 = makeUniqueRef<AST::VariableReference>(AST::VariableReference::wrap(variableDeclaration));
>+    variableReference1->setType(base.resolvedType().clone());
>+    variableReference1->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread });
> 
>-        auto variableReference2 = makeUniqueRef<AST::VariableReference>(AST::VariableReference::wrap(variableDeclaration));
>-        variableReference2->setType(base.resolvedType().clone());
>-        variableReference2->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread });
>+    auto assignmentExpression = makeUniqueRef<AST::AssignmentExpression>(Lexer::Token(origin), WTFMove(variableReference1), propertyAccessExpression.takeBase());
>+    assignmentExpression->setType(base.resolvedType().clone());
>+    assignmentExpression->setTypeAnnotation(AST::RightValue());
> 
>-        auto makePointerExpression = makeUniqueRef<AST::MakePointerExpression>(Lexer::Token(origin), WTFMove(variableReference2));
>-        makePointerExpression->setType(makeUniqueRef<AST::PointerType>(Lexer::Token(origin), AST::AddressSpace::Thread, base.resolvedType().clone()));
>-        makePointerExpression->setTypeAnnotation(AST::RightValue());
>+    auto variableReference2 = makeUniqueRef<AST::VariableReference>(AST::VariableReference::wrap(variableDeclaration));
>+    variableReference2->setType(base.resolvedType().clone());
>+    variableReference2->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread });
> 
>-        Vector<UniqueRef<AST::Expression>> arguments;
>-        arguments.append(WTFMove(makePointerExpression));
>-        auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(origin), String(anderFunction->name()), WTFMove(arguments));
>-        callExpression->setType(anderFunction->type().clone());
>-        callExpression->setTypeAnnotation(AST::RightValue());
>-        callExpression->setFunction(*anderFunction);
>+    auto makePointerExpression = makeUniqueRef<AST::MakePointerExpression>(Lexer::Token(origin), WTFMove(variableReference2));
>+    makePointerExpression->setType(makeUniqueRef<AST::PointerType>(Lexer::Token(origin), AST::AddressSpace::Thread, base.resolvedType().clone()));
>+    makePointerExpression->setTypeAnnotation(AST::RightValue());
> 
>-        auto dereferenceExpression = makeUniqueRef<AST::DereferenceExpression>(WTFMove(origin), WTFMove(callExpression));
>-        dereferenceExpression->setType(downcast<AST::PointerType>(anderFunction->type()).elementType().clone());
>-        dereferenceExpression->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread });
>+    Vector<UniqueRef<AST::Expression>> arguments;
>+    arguments.append(WTFMove(makePointerExpression));
>+    if (is<AST::IndexExpression>(propertyAccessExpression))
>+        arguments.append(downcast<AST::IndexExpression>(propertyAccessExpression).takeIndex());
>+    auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(origin), String(threadAnderFunction->name()), WTFMove(arguments));
>+    callExpression->setType(threadAnderFunction->type().clone());
>+    callExpression->setTypeAnnotation(AST::RightValue());
>+    callExpression->setFunction(*threadAnderFunction);
> 
>-        Vector<UniqueRef<AST::Expression>> expressions;
>-        expressions.append(WTFMove(assignmentExpression));
>-        expressions.append(WTFMove(dereferenceExpression));
>-        auto* commaExpression = AST::replaceWith<AST::CommaExpression>(dotExpression, WTFMove(origin), WTFMove(expressions));
>-        commaExpression->setType(downcast<AST::PointerType>(anderFunction->type()).elementType().clone());
>-        commaExpression->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread });
>+    auto dereferenceExpression = makeUniqueRef<AST::DereferenceExpression>(WTFMove(origin), WTFMove(callExpression));
>+    dereferenceExpression->setType(downcast<AST::PointerType>(threadAnderFunction->type()).elementType().clone());
>+    dereferenceExpression->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread });
> 
>-        m_variableDeclarations.append(WTFMove(variableDeclaration));
>-        return true;
>-    }
>+    Vector<UniqueRef<AST::Expression>> expressions;
>+    expressions.append(WTFMove(assignmentExpression));
>+    expressions.append(WTFMove(dereferenceExpression));
>+    auto* commaExpression = AST::replaceWith<AST::CommaExpression>(propertyAccessExpression, WTFMove(origin), WTFMove(expressions));
>+    commaExpression->setType(downcast<AST::PointerType>(threadAnderFunction->type()).elementType().clone());
>+    commaExpression->setTypeAnnotation(AST::LeftValue { AST::AddressSpace::Thread });
> 
>-    ASSERT(dotExpression.getterFunction());
>-    auto& getterFunction = *dotExpression.getterFunction();
>-    Vector<UniqueRef<AST::Expression>> arguments;
>-    arguments.append(dotExpression.takeBase());
>-    auto* callExpression = AST::replaceWith<AST::CallExpression>(dotExpression, WTFMove(origin), String(getterFunction.name()), WTFMove(arguments));
>-    callExpression->setFunction(getterFunction);
>-    callExpression->setType(getterFunction.type().clone());
>-    callExpression->setTypeAnnotation(AST::RightValue());
>+    m_variableDeclarations.append(WTFMove(variableDeclaration));
>     return true;
>+
> }
> 
> class LeftValueSimplifier : public Visitor {
>-public:
>+private:
>     void visit(AST::DotExpression&) override;
>+    void visit(AST::IndexExpression&) override;
>     void visit(AST::DereferenceExpression&) override;
> 
>-private:
>+    void finishVisiting(AST::PropertyAccessExpression&);
> };
> 
>-void LeftValueSimplifier::visit(AST::DotExpression& dotExpression)
>+void LeftValueSimplifier::finishVisiting(AST::PropertyAccessExpression& propertyAccessExpression)
> {
>-    Visitor::visit(dotExpression);
>-    ASSERT(dotExpression.base().typeAnnotation().leftAddressSpace());
>-    ASSERT(dotExpression.anderFunction());
>-
>-    Lexer::Token origin = dotExpression.origin();
>-    auto* anderFunction = dotExpression.anderFunction();
>-    auto& base = dotExpression.base();
>-    auto leftAddressSpace = *dotExpression.base().typeAnnotation().leftAddressSpace();
>-    auto makePointerExpression = makeUniqueRef<AST::MakePointerExpression>(Lexer::Token(origin), dotExpression.takeBase());
>+    ASSERT(propertyAccessExpression.base().typeAnnotation().leftAddressSpace());
>+    ASSERT(propertyAccessExpression.anderFunction());
>+
>+    Lexer::Token origin = propertyAccessExpression.origin();
>+    auto* anderFunction = propertyAccessExpression.anderFunction();
>+    auto& base = propertyAccessExpression.base();
>+    auto leftAddressSpace = *propertyAccessExpression.base().typeAnnotation().leftAddressSpace();
>+    auto makePointerExpression = makeUniqueRef<AST::MakePointerExpression>(Lexer::Token(origin), propertyAccessExpression.takeBase());
>     makePointerExpression->setType(makeUniqueRef<AST::PointerType>(Lexer::Token(origin), leftAddressSpace, base.resolvedType().clone()));
>     makePointerExpression->setTypeAnnotation(AST::RightValue());
> 
>     Vector<UniqueRef<AST::Expression>> arguments;
>     arguments.append(WTFMove(makePointerExpression));
>+    if (is<AST::IndexExpression>(propertyAccessExpression))
>+        arguments.append(downcast<AST::IndexExpression>(propertyAccessExpression).takeIndex());
>     auto callExpression = makeUniqueRef<AST::CallExpression>(Lexer::Token(origin), String(anderFunction->name()), WTFMove(arguments));
>     callExpression->setType(anderFunction->type().clone());
>     callExpression->setTypeAnnotation(AST::RightValue());
>     callExpression->setFunction(*anderFunction);
> 
>-    auto* dereferenceExpression = AST::replaceWith<AST::DereferenceExpression>(dotExpression, WTFMove(origin), WTFMove(callExpression));
>+    auto* dereferenceExpression = AST::replaceWith<AST::DereferenceExpression>(propertyAccessExpression, WTFMove(origin), WTFMove(callExpression));
>     dereferenceExpression->setType(downcast<AST::PointerType>(anderFunction->type()).elementType().clone());
>     dereferenceExpression->setTypeAnnotation(AST::LeftValue { downcast<AST::PointerType>(anderFunction->type()).addressSpace() });
> }
> 
>+void LeftValueSimplifier::visit(AST::DotExpression& dotExpression)
>+{
>+    Visitor::visit(dotExpression);
>+    finishVisiting(dotExpression);
>+}
>+
>+void LeftValueSimplifier::visit(AST::IndexExpression& indexExpression)
>+{
>+    Visitor::visit(indexExpression);
>+    PropertyResolver().Visitor::visit(indexExpression.indexExpression());
>+    finishVisiting(indexExpression);
>+}
>+
> void LeftValueSimplifier::visit(AST::DereferenceExpression& dereferenceExpression)
> {
>     // Dereference expressions are the only expressions where the children might be more-right than we are.
>     // For example, a dereference expression may be a left value but its child may be a call expression which is a right value.
>     // LeftValueSimplifier doesn't handle right values, so we instead need to use PropertyResolver.
>-    // FIXME: https://bugs.webkit.org/show_bug.cgi?id=198170 What about function call arguments?
>     PropertyResolver().Visitor::visit(dereferenceExpression);
> }
> 
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLStandardLibrary.txt b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLStandardLibrary.txt
>index 0fb5d3932d45fb0afde0959615ebd81feaa91c53..b8c63621540528f1eee58095b08961031be6862c 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLStandardLibrary.txt
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLStandardLibrary.txt
>@@ -375,6 +375,63 @@ native typedef TextureDepth2DArray<float>;
> native typedef RWTextureDepth2DArray<float>;
> native typedef TextureDepthCube<float>;
> 
>+native operator uchar(ushort);
>+native operator uchar(uint);
>+native operator uchar(char);
>+native operator uchar(short);
>+native operator uchar(int);
>+native operator uchar(half);
>+native operator uchar(float);
>+native operator ushort(uchar);
>+native operator ushort(uint);
>+native operator ushort(char);
>+native operator ushort(short);
>+native operator ushort(int);
>+native operator ushort(half);
>+native operator ushort(float);
>+native operator uint(uchar);
>+native operator uint(ushort);
>+native operator uint(char);
>+native operator uint(short);
>+native operator uint(int);
>+native operator uint(half);
>+native operator uint(float);
>+native operator char(uchar);
>+native operator char(ushort);
>+native operator char(uint);
>+native operator char(short);
>+native operator char(int);
>+native operator char(half);
>+native operator char(float);
>+native operator short(uchar);
>+native operator short(ushort);
>+native operator short(uint);
>+native operator short(char);
>+native operator short(int);
>+native operator short(half);
>+native operator short(float);
>+native operator int(uchar);
>+native operator int(ushort);
>+native operator int(uint);
>+native operator int(char);
>+native operator int(short);
>+native operator int(half);
>+native operator int(float);
>+native operator half(uchar);
>+native operator half(ushort);
>+native operator half(uint);
>+native operator half(char);
>+native operator half(short);
>+native operator half(int);
>+native operator half(float);
>+native operator float(uchar);
>+native operator float(ushort);
>+native operator float(uint);
>+native operator float(char);
>+native operator float(short);
>+native operator float(int);
>+native operator float(half);
>+
> native float operator.x(float4);
> native float operator.y(float4);
> native float operator.z(float4);
>diff --git a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLSynthesizeStructureAccessors.cpp b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLSynthesizeStructureAccessors.cpp
>index 945d0ceefaa39ce5074607d2c4493cf397e983de..f89787d17804c7b4f80dc919c70da3f7a75e2b98 100644
>--- a/Source/WebCore/Modules/webgpu/WHLSL/WHLSLSynthesizeStructureAccessors.cpp
>+++ b/Source/WebCore/Modules/webgpu/WHLSL/WHLSLSynthesizeStructureAccessors.cpp
>@@ -44,7 +44,7 @@ bool synthesizeStructureAccessors(Program& program)
>     bool isOperator = true;
>     for (auto& structureDefinition : program.structureDefinitions()) {
>         for (auto& structureElement : structureDefinition->structureElements()) {
>-            {
>+            /*{
>                 // The getter: operator.field
>                 auto variableDeclaration = makeUniqueRef<AST::VariableDeclaration>(Lexer::Token(structureElement.origin()), AST::Qualifiers(), UniqueRef<AST::UnnamedType>(AST::TypeReference::wrap(Lexer::Token(structureElement.origin()), structureDefinition)), String(), WTF::nullopt, WTF::nullopt);
>                 AST::VariableDeclarations parameters;
>@@ -64,7 +64,7 @@ bool synthesizeStructureAccessors(Program& program)
>                 AST::NativeFunctionDeclaration nativeFunctionDeclaration(AST::FunctionDeclaration(Lexer::Token(structureElement.origin()), AST::AttributeBlock(), WTF::nullopt, AST::TypeReference::wrap(Lexer::Token(structureElement.origin()), structureDefinition), makeString("operator.", structureElement.name(), '='), WTFMove(parameters), WTF::nullopt, isOperator));
>                 if (!program.append(WTFMove(nativeFunctionDeclaration)))
>                     return false;
>-            }
>+            }*/
> 
>             // The ander: operator&.field
>             auto createAnder = [&](AST::AddressSpace addressSpace) -> AST::NativeFunctionDeclaration {
>diff --git a/LayoutTests/webgpu/propertyresolver/ander-abstract-lvalue.html b/LayoutTests/webgpu/propertyresolver/ander-abstract-lvalue.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..ce14a68f14f23e8e96cc67e5706166c2f0870d5d
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/ander-abstract-lvalue.html
>@@ -0,0 +1,115 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+thread float* operator&.y(thread Foo* f) {
>+    return &f->x;
>+}
>+
>+struct Bar {
>+    Foo z;
>+}
>+
>+Foo operator.w(Bar b) {
>+    return b.z;
>+}
>+
>+Bar operator.w=(Bar b, Foo a) {
>+    b.z = a;
>+    return b;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Bar b;
>+    b.w.y = 1.0;
>+    return float4(b.w.y, b.w.y, b.w.y, 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/ander-lvalue-3-levels.html b/LayoutTests/webgpu/propertyresolver/ander-lvalue-3-levels.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..b3446091d0be8b605be3de3f252f365e75a9ce02
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/ander-lvalue-3-levels.html
>@@ -0,0 +1,118 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+thread float* operator&.y(thread Foo* f) {
>+    return &f->x;
>+}
>+
>+struct Bar {
>+    Foo z;
>+}
>+
>+thread Foo* operator&.w(thread Bar* b) {
>+    return &b->z;
>+}
>+
>+struct Baz {
>+    Bar p;
>+}
>+
>+thread Bar* operator&.q(thread Baz* b) {
>+    return &b->p;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Baz b;
>+    b.q.w.y = 1.0;
>+    return float4(b.q.w.y, b.q.w.y, b.q.w.y, 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/ander-lvalue.html b/LayoutTests/webgpu/propertyresolver/ander-lvalue.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..31f23f923133f934509b884e2aa2f6af0c657c81
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/ander-lvalue.html
>@@ -0,0 +1,110 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+thread float* operator&.y(thread Foo* f) {
>+    return &f->x;
>+}
>+
>+struct Bar {
>+    Foo z;
>+}
>+
>+thread Foo* operator&.w(thread Bar* b) {
>+    return &b->z;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Bar b;
>+    b.w.y = 1.0;
>+    return float4(b.w.y, b.w.y, b.w.y, 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/ander.html b/LayoutTests/webgpu/propertyresolver/ander.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..f7b60e6fd13c2bc5165cc4948d3884173ec1e1ff
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/ander.html
>@@ -0,0 +1,106 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+thread float* operator&.y(thread Foo* f) {
>+    return &f->x;
>+}
>+
>+Foo generateFoo() {
>+    Foo f;
>+    f.x = 1.0;
>+    return f;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    return float4(generateFoo().y, generateFoo().y, generateFoo().y, 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/getter.html b/LayoutTests/webgpu/propertyresolver/getter.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..cc5cba018e5045afefdf51aa7d274a61ebc394f3
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/getter.html
>@@ -0,0 +1,102 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+float operator.y(Foo f) {
>+    return f.x;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Foo f;
>+    f.x = 1.0;
>+    return float4(f.y, f.y, f.y, 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/indexer-ander-abstract-lvalue.html b/LayoutTests/webgpu/propertyresolver/indexer-ander-abstract-lvalue.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..afce1afdfcccbb2049db3f9d055a4349817b6c6b
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/indexer-ander-abstract-lvalue.html
>@@ -0,0 +1,115 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+thread float* operator&[](thread Foo* f, uint index) {
>+    return &f->x;
>+}
>+
>+struct Bar {
>+    Foo z;
>+}
>+
>+Foo operator[](Bar b, uint index) {
>+    return b.z;
>+}
>+
>+Bar operator[]=(Bar b, uint index, Foo a) {
>+    b.z = a;
>+    return b;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Bar b;
>+    b[2][3] = 1.0;
>+    return float4(b[2][3], b[2][3], b[2][3], 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/indexer-ander-lvalue-3-levels.html b/LayoutTests/webgpu/propertyresolver/indexer-ander-lvalue-3-levels.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..c3fec6528c30a2fd36a5b05ccee90f14b122cd03
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/indexer-ander-lvalue-3-levels.html
>@@ -0,0 +1,118 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+thread float* operator&[](thread Foo* f, uint index) {
>+    return &f->x;
>+}
>+
>+struct Bar {
>+    Foo z;
>+}
>+
>+thread Foo* operator&[](thread Bar* b, uint index) {
>+    return &b->z;
>+}
>+
>+struct Baz {
>+    Bar p;
>+}
>+
>+thread Bar* operator&[](thread Baz* b, uint index) {
>+    return &b->p;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Baz b;
>+    b[1][2][3] = 1.0;
>+    return float4(b[1][2][3], b[1][2][3], b[1][2][3], 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/indexer-ander-lvalue.html b/LayoutTests/webgpu/propertyresolver/indexer-ander-lvalue.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..354ba5ef06cf970e1ab09e03e36cfef3ac3fa595
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/indexer-ander-lvalue.html
>@@ -0,0 +1,110 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+thread float* operator&[](thread Foo* f, uint index) {
>+    return &f->x;
>+}
>+
>+struct Bar {
>+    Foo z;
>+}
>+
>+thread Foo* operator&[](thread Bar* b, uint index) {
>+    return &b->z;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Bar b;
>+    b[1][2] = 1.0;
>+    return float4(b[1][2], b[1][2], b[1][2], 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/indexer-ander.html b/LayoutTests/webgpu/propertyresolver/indexer-ander.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..c66111446f4dd260cdfd8c6a88a490a2bb46cf9c
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/indexer-ander.html
>@@ -0,0 +1,106 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+thread float* operator&[](thread Foo* f, uint index) {
>+    return &f->x;
>+}
>+
>+Foo generateFoo() {
>+    Foo f;
>+    f.x = 1.0;
>+    return f;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    return float4(generateFoo()[0], generateFoo()[1], generateFoo()[2], 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/indexer-getter.html b/LayoutTests/webgpu/propertyresolver/indexer-getter.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..df16838963a2c8a69d13c4bb9978b1fdd160a871
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/indexer-getter.html
>@@ -0,0 +1,101 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+float operator[](Foo f, uint index) {
>+    return float(index);
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Foo f;
>+    return float4(f[1], f[1], f[1], 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/indexer-setter.html b/LayoutTests/webgpu/propertyresolver/indexer-setter.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..61da193d342ea67f966e3086c8a83a222d596000
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/indexer-setter.html
>@@ -0,0 +1,118 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+thread float* operator&[](thread Foo* f, uint index) {
>+    return &f->x;
>+}
>+
>+struct Bar {
>+    Foo z;
>+}
>+
>+Foo operator[](Bar b, uint index) {
>+    return b.z;
>+}
>+
>+Bar operator[]=(Bar b, uint index, Foo a) {
>+    b.z = a;
>+    return b;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Bar b;
>+    Foo foo;
>+    foo.x = 1.0;
>+    b[1] = foo;
>+    foo = b[1];
>+    return float4(foo.x, foo.x, foo.x, 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/setter-abstract-lvalue-3-levels.html b/LayoutTests/webgpu/propertyresolver/setter-abstract-lvalue-3-levels.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..6e912f516d7bdf672112b356f24b427814268659
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/setter-abstract-lvalue-3-levels.html
>@@ -0,0 +1,133 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+float operator.y(Foo f) {
>+    return f.x;
>+}
>+
>+Foo operator.y=(Foo f, float x) {
>+    f.x = x;
>+    return f;
>+}
>+
>+struct Bar {
>+    Foo z;
>+}
>+
>+Foo operator.w(Bar b) {
>+    return b.z;
>+}
>+
>+Bar operator.w=(Bar b, Foo a) {
>+    b.z = a;
>+    return b;
>+}
>+
>+struct Baz {
>+    Bar p;
>+}
>+
>+Bar operator.q(Baz b) {
>+    return b.p;
>+}
>+
>+Baz operator.q=(Baz b, Bar a) {
>+    b.p = a;
>+    return b;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Baz b;
>+    b.q.w.y = 1.0;
>+    return float4(b.q.w.y, b.q.w.y, b.q.w.y, 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/setter-abstract-lvalue.html b/LayoutTests/webgpu/propertyresolver/setter-abstract-lvalue.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..0ddc7b459c17fda29a1b1f758b79460008811d91
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/setter-abstract-lvalue.html
>@@ -0,0 +1,120 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+float operator.y(Foo f) {
>+    return f.x;
>+}
>+
>+Foo operator.y=(Foo f, float x) {
>+    f.x = x;
>+    return f;
>+}
>+
>+struct Bar {
>+    Foo z;
>+}
>+
>+Foo operator.w(Bar b) {
>+    return b.z;
>+}
>+
>+Bar operator.w=(Bar b, Foo a) {
>+    b.z = a;
>+    return b;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Bar b;
>+    b.w.y = 1.0;
>+    return float4(b.w.y, b.w.y, b.w.y, 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>
>diff --git a/LayoutTests/webgpu/propertyresolver/setter-lvalue.html b/LayoutTests/webgpu/propertyresolver/setter-lvalue.html
>new file mode 100644
>index 0000000000000000000000000000000000000000..9a9b4f29555bde124d99182beb3ab27da7be9c58
>--- /dev/null
>+++ b/LayoutTests/webgpu/propertyresolver/setter-lvalue.html
>@@ -0,0 +1,115 @@
>+<!DOCTYPE html>
>+<html>
>+<head>
>+</head>
>+<body>
>+<canvas id="canvas" width="400" height="400"></canvas>
>+<script>
>+const shaderSource = `
>+vertex float4 vertexShader(float4 position : attribute(0)) : SV_Position {
>+    return position;
>+}
>+
>+struct Foo {
>+    float x;
>+}
>+
>+float operator.y(Foo f) {
>+    return f.x;
>+}
>+
>+Foo operator.y=(Foo f, float x) {
>+    f.x = x;
>+    return f;
>+}
>+
>+struct Bar {
>+    Foo z;
>+}
>+
>+thread Foo* operator&.w(thread Bar* b) {
>+    return &b->z;
>+}
>+
>+fragment float4 fragmentShader() : SV_Target 0 {
>+    Bar b;
>+    b.w.y = 1.0;
>+    return float4(b.w.y, b.w.y, b.w.y, 1.0);
>+}
>+`;
>+async function start() {
>+    const adapter = await navigator.gpu.requestAdapter();
>+    const device = await adapter.requestDevice();
>+
>+    const shaderModule = device.createShaderModule({code: shaderSource, isWHLSL: true});
>+    const vertexStage = {module: shaderModule, entryPoint: "vertexShader"};
>+    const fragmentStage = {module: shaderModule, entryPoint: "fragmentShader"};
>+    const primitiveTopology = "triangle-strip";
>+    const rasterizationState = {frontFace: "cw", cullMode: "none"};
>+    const alphaBlend = {};
>+    const colorBlend = {};
>+    const colorStates = [{format: "rgba8unorm", alphaBlend, colorBlend, writeMask: 15}]; // GPUColorWriteBits.ALL
>+    const depthStencilState = null;
>+    
>+    const attribute0 = {shaderLocation: 0, format: "float4"};
>+    const input0 = {stride: 16, attributeSet: [attribute0]};
>+    const inputs = [input0];
>+    const vertexInput = {vertexBuffers: inputs};
>+
>+    const pipelineLayoutDescriptor = {bindGroupLayouts: []};
>+    const pipelineLayout = device.createPipelineLayout(pipelineLayoutDescriptor);
>+
>+    const renderPipelineDescriptor = {vertexStage, fragmentStage, primitiveTopology, rasterizationState, colorStates, depthStencilState, vertexInput, sampleCount: 1, layout: pipelineLayout};
>+    const renderPipeline = device.createRenderPipeline(renderPipelineDescriptor);
>+
>+    const vertexBufferDescriptor = {size: Float32Array.BYTES_PER_ELEMENT * 4 * 4, usage: GPUBufferUsage.VERTEX | GPUBufferUsage.MAP_WRITE};
>+    const vertexBuffer = device.createBuffer(vertexBufferDescriptor);
>+    const vertexBufferArrayBuffer = await vertexBuffer.mapWriteAsync();
>+    const vertexBufferFloat32Array = new Float32Array(vertexBufferArrayBuffer);
>+    vertexBufferFloat32Array[0] = -0.5;
>+    vertexBufferFloat32Array[1] = -0.5;
>+    vertexBufferFloat32Array[2] = 1.0;
>+    vertexBufferFloat32Array[3] = 1;
>+    vertexBufferFloat32Array[4] = -0.5;
>+    vertexBufferFloat32Array[5] = 0.5;
>+    vertexBufferFloat32Array[6] = 1.0;
>+    vertexBufferFloat32Array[7] = 1;
>+    vertexBufferFloat32Array[8] = 0.5;
>+    vertexBufferFloat32Array[9] = -0.5;
>+    vertexBufferFloat32Array[10] = 1.0;
>+    vertexBufferFloat32Array[11] = 1;
>+    vertexBufferFloat32Array[12] = 0.5;
>+    vertexBufferFloat32Array[13] = 0.5;
>+    vertexBufferFloat32Array[14] = 1.0;
>+    vertexBufferFloat32Array[15] = 1;
>+    vertexBuffer.unmap();
>+
>+    const canvas = document.getElementById("canvas");
>+    const context = canvas.getContext("gpu");
>+    const swapChainDescriptor = {device, format: "bgra8unorm"};
>+    const swapChain = context.configureSwapChain(swapChainDescriptor);
>+    const outputTexture = swapChain.getCurrentTexture();
>+    const outputTextureView = outputTexture.createDefaultView();
>+
>+    const commandEncoder = device.createCommandEncoder(); // {}
>+    const red = {r: 0, g: 0, b: 1, a: 1};
>+    const colorAttachments = [{attachment: outputTextureView, resolveTarget: null, loadOp: "clear", storeOp: "store", clearColor: red}];
>+    const depthStencilAttachment = null;
>+    const renderPassDescriptor = {colorAttachments, depthStencilAttachment};
>+    const renderPassEncoder = commandEncoder.beginRenderPass(renderPassDescriptor);
>+    renderPassEncoder.setPipeline(renderPipeline);
>+    renderPassEncoder.setVertexBuffers(0, [vertexBuffer], [0]);
>+    renderPassEncoder.draw(4, 1, 0, 0);
>+    renderPassEncoder.endPass();
>+    const commandBuffer = commandEncoder.finish();
>+    device.getQueue().submit([commandBuffer]);
>+
>+    if (window.testRunner)
>+        testRunner.notifyDone();
>+}
>+if (window.testRunner)
>+    testRunner.waitUntilDone();
>+window.addEventListener("load", start);
>+</script>
>+</body>
>+</html>

Actions: View | Formatted Diff | Diff

Attachments on bug 198399: 371002 | 371003 | 371042 | 371101 | 371110 | 371138 | 371139 | 371857 | 371872