[EH] Fuzz throws from JS (WebAssembly#7027)

kripken · web-flow · commit dcc70bbfb16c · 2024-10-23T10:17:13.000-07:00
We already generated (throw ..) instructions in wasm, but it makes sense to model
throws from outside as well, as they cross the module boundary. This adds a new fuzzer
import to the generated modules, "throw", that just does a throw from JS etc.

Also be more precise about handling fuzzing-support imports in fuzz-exec: we now
check that logging functions start with "log*" and error otherwise (this check is
now needed given we have "throw", which is not logging). Also fix a minor issue
with name conflicts for logging functions by using getValidFunctionName for them,
both for logging and for throw.
diff --git a/scripts/fuzz_shell.js b/scripts/fuzz_shell.js
@@ -138,6 +138,7 @@ function logValue(x, y) {
 var tempRet0;
 var imports = {
   'fuzzing-support': {
+    // Logging.
     'log-i32': logValue,
     'log-i64': logValue,
     'log-f32': logValue,
@@ -147,7 +148,13 @@ var imports = {
     // we could avoid running JS on code with SIMD in it, but it is useful to
     // fuzz such code as much as we can.)
     'log-v128': logValue,
+
+    // Throw an exception from JS.
+    'throw': () => {
+      throw 'some JS error';
+    }
   },
+  // Emscripten support.
   'env': {
     'setTempRet0': function(x) { tempRet0 = x },
     'getTempRet0': function() { return tempRet0 },
diff --git a/src/tools/execution-results.h b/src/tools/execution-results.h
@@ -27,6 +27,7 @@ using Loggings = std::vector<Literal>;
 
 // Logs every relevant import call parameter.
 struct LoggingExternalInterface : public ShellExternalInterface {
+private:
   Loggings& loggings;
 
   struct State {
@@ -37,30 +38,40 @@ struct LoggingExternalInterface : public ShellExternalInterface {
     uint32_t tempRet0 = 0;
   } state;
 
+public:
   LoggingExternalInterface(Loggings& loggings) : loggings(loggings) {}
 
   Literals callImport(Function* import, const Literals& arguments) override {
     if (import->module == "fuzzing-support") {
-      std::cout << "[LoggingExternalInterface logging";
-      loggings.push_back(Literal()); // buffer with a None between calls
-      for (auto argument : arguments) {
-        if (argument.type == Type::i64) {
-          // To avoid JS legalization changing logging results, treat a logging
-          // of an i64 as two i32s (which is what legalization would turn us
-          // into).
-          auto low = Literal(int32_t(argument.getInteger()));
-          auto high = Literal(int32_t(argument.getInteger() >> int32_t(32)));
-          std::cout << ' ' << low;
-          loggings.push_back(low);
-          std::cout << ' ' << high;
-          loggings.push_back(high);
-        } else {
-          std::cout << ' ' << argument;
-          loggings.push_back(argument);
+      if (import->base.startsWith("log")) {
+        // This is a logging function like log-i32 or log-f64
+        std::cout << "[LoggingExternalInterface logging";
+        loggings.push_back(Literal()); // buffer with a None between calls
+        for (auto argument : arguments) {
+          if (argument.type == Type::i64) {
+            // To avoid JS legalization changing logging results, treat a
+            // logging of an i64 as two i32s (which is what legalization would
+            // turn us into).
+            auto low = Literal(int32_t(argument.getInteger()));
+            auto high = Literal(int32_t(argument.getInteger() >> int32_t(32)));
+            std::cout << ' ' << low;
+            loggings.push_back(low);
+            std::cout << ' ' << high;
+            loggings.push_back(high);
+          } else {
+            std::cout << ' ' << argument;
+            loggings.push_back(argument);
+          }
         }
+        std::cout << "]\n";
+        return {};
+      } else if (import->base == "throw") {
+        // Throw something. We use a (hopefully) private name here.
+        auto payload = std::make_shared<ExnData>("__private", Literals{});
+        throwException(WasmException{Literal(payload)});
+      } else {
+        WASM_UNREACHABLE("unknown fuzzer import");
       }
-      std::cout << "]\n";
-      return {};
     } else if (import->module == ENV) {
       if (import->base == "log_execution") {
         std::cout << "[LoggingExternalInterface log-execution";
diff --git a/src/tools/fuzzing.h b/src/tools/fuzzing.h
@@ -104,6 +104,10 @@ class TranslateToFuzzReader {
 
   Name funcrefTableName;
 
+  std::unordered_map<Type, Name> logImportNames;
+
+  Name throwImportName;
+
   std::unordered_map<Type, std::vector<Name>> globalsByType;
   std::unordered_map<Type, std::vector<Name>> mutableGlobalsByType;
   std::unordered_map<Type, std::vector<Name>> immutableGlobalsByType;
@@ -220,12 +224,16 @@ class TranslateToFuzzReader {
   void finalizeTable();
   void prepareHangLimitSupport();
   void addHangLimitSupport();
+  // Imports that we call to log out values.
   void addImportLoggingSupport();
+  // An import that we call to throw an exception from outside.
+  void addImportThrowingSupport();
   void addHashMemorySupport();
 
   // Special expression makers
   Expression* makeHangLimitCheck();
-  Expression* makeLogging();
+  Expression* makeImportLogging();
+  Expression* makeImportThrowing(Type type);
   Expression* makeMemoryHashLogging();
 
   // Function creation
diff --git a/src/tools/fuzzing/fuzzing.cpp b/src/tools/fuzzing/fuzzing.cpp
@@ -176,9 +176,10 @@ void TranslateToFuzzReader::build() {
   setupGlobals();
   if (wasm.features.hasExceptionHandling()) {
     setupTags();
+    addImportThrowingSupport();
   }
-  modifyInitialFunctions();
   addImportLoggingSupport();
+  modifyInitialFunctions();
   // keep adding functions until we run out of input
   while (!random.finished()) {
     auto* func = addFunction();
@@ -583,16 +584,31 @@ void TranslateToFuzzReader::addHangLimitSupport() {
 
 void TranslateToFuzzReader::addImportLoggingSupport() {
   for (auto type : loggableTypes) {
-    auto* func = new Function;
-    Name name = std::string("log-") + type.toString();
-    func->name = name;
+    auto func = std::make_unique<Function>();
+    Name baseName = std::string("log-") + type.toString();
+    func->name = Names::getValidFunctionName(wasm, baseName);
+    logImportNames[type] = func->name;
     func->module = "fuzzing-support";
-    func->base = name;
+    func->base = baseName;
     func->type = Signature(type, Type::none);
-    wasm.addFunction(func);
+    wasm.addFunction(std::move(func));
   }
 }
 
+void TranslateToFuzzReader::addImportThrowingSupport() {
+  // Throw some kind of exception from JS.
+  // TODO: Send an index, which is which exported wasm Tag we should throw, or
+  //       something not exported if out of bounds. First we must also export
+  //       tags sometimes.
+  throwImportName = Names::getValidFunctionName(wasm, "throw");
+  auto func = std::make_unique<Function>();
+  func->name = throwImportName;
+  func->module = "fuzzing-support";
+  func->base = "throw";
+  func->type = Signature(Type::none, Type::none);
+  wasm.addFunction(std::move(func));
+}
+
 void TranslateToFuzzReader::addHashMemorySupport() {
   // Add memory hasher helper (for the hash, see hash.h). The function looks
   // like:
@@ -692,21 +708,30 @@ Expression* TranslateToFuzzReader::makeHangLimitCheck() {
                          builder.makeConst(int32_t(1)))));
 }
 
-Expression* TranslateToFuzzReader::makeLogging() {
+Expression* TranslateToFuzzReader::makeImportLogging() {
   auto type = getLoggableType();
-  return builder.makeCall(
-    std::string("log-") + type.toString(), {make(type)}, Type::none);
+  return builder.makeCall(logImportNames[type], {make(type)}, Type::none);
+}
+
+Expression* TranslateToFuzzReader::makeImportThrowing(Type type) {
+  // We throw from the import, so this call appears to be none and not
+  // unreachable.
+  assert(type == Type::none);
+
+  // TODO: This and makeThrow should probably be rare, as they halt the program.
+  return builder.makeCall(throwImportName, {}, Type::none);
 }
 
 Expression* TranslateToFuzzReader::makeMemoryHashLogging() {
   auto* hash = builder.makeCall(std::string("hashMemory"), {}, Type::i32);
-  return builder.makeCall(std::string("log-i32"), {hash}, Type::none);
+  return builder.makeCall(logImportNames[Type::i32], {hash}, Type::none);
 }
 
 // TODO: return std::unique_ptr<Function>
 Function* TranslateToFuzzReader::addFunction() {
   LOGGING_PERCENT = upToSquared(100);
-  auto* func = new Function;
+  auto allocation = std::make_unique<Function>();
+  auto* func = allocation.get();
   func->name = Names::getValidFunctionName(wasm, "func");
   FunctionCreationContext context(*this, func);
   assert(funcContext->typeLocals.empty());
@@ -765,7 +790,7 @@ Function* TranslateToFuzzReader::addFunction() {
   }
 
   // Add hang limit checks after all other operations on the function body.
-  wasm.addFunction(func);
+  wasm.addFunction(std::move(allocation));
   // Export some functions, but not all (to allow inlining etc.). Try to export
   // at least one, though, to keep each testcase interesting. Only functions
   // with valid params and returns can be exported because the trap fuzzer
@@ -1215,10 +1240,13 @@ void TranslateToFuzzReader::modifyInitialFunctions() {
   // the end (currently that is not needed atm, but it might in the future).
   for (Index i = 0; i < wasm.functions.size(); i++) {
     auto* func = wasm.functions[i].get();
+    // We can't allow extra imports, as the fuzzing infrastructure wouldn't
+    // know what to provide. Keep only our own fuzzer imports.
+    if (func->imported() && func->module == "fuzzing-support") {
+      continue;
+    }
     FunctionCreationContext context(*this, func);
     if (func->imported()) {
-      // We can't allow extra imports, as the fuzzing infrastructure wouldn't
-      // know what to provide.
       func->module = func->base = Name();
       func->body = make(func->getResults());
     }
@@ -1261,10 +1289,9 @@ void TranslateToFuzzReader::dropToLog(Function* func) {
 
     void visitDrop(Drop* curr) {
       if (parent.isLoggableType(curr->value->type) && parent.oneIn(2)) {
-        replaceCurrent(parent.builder.makeCall(std::string("log-") +
-                                                 curr->value->type.toString(),
-                                               {curr->value},
-                                               Type::none));
+        auto target = parent.logImportNames[curr->value->type];
+        replaceCurrent(
+          parent.builder.makeCall(target, {curr->value}, Type::none));
       }
     }
   };
@@ -1430,7 +1457,7 @@ Expression* TranslateToFuzzReader::_makenone() {
   auto choice = upTo(100);
   if (choice < LOGGING_PERCENT) {
     if (choice < LOGGING_PERCENT / 2) {
-      return makeLogging();
+      return makeImportLogging();
     } else {
       return makeMemoryHashLogging();
     }
@@ -1455,6 +1482,7 @@ Expression* TranslateToFuzzReader::_makenone() {
     .add(FeatureSet::Atomics, &Self::makeAtomic)
     .add(FeatureSet::ExceptionHandling, &Self::makeTry)
     .add(FeatureSet::ExceptionHandling, &Self::makeTryTable)
+    .add(FeatureSet::ExceptionHandling, &Self::makeImportThrowing)
     .add(FeatureSet::ReferenceTypes | FeatureSet::GC, &Self::makeCallRef)
     .add(FeatureSet::ReferenceTypes | FeatureSet::GC, &Self::makeStructSet)
     .add(FeatureSet::ReferenceTypes | FeatureSet::GC, &Self::makeArraySet)
diff --git a/test/lit/exec/fuzzing-api.wast b/test/lit/exec/fuzzing-api.wast
@@ -0,0 +1,39 @@
+;; NOTE: Assertions have been generated by update_lit_checks.py --output=fuzz-exec and should not be edited.
+
+;; RUN: wasm-opt %s -all --fuzz-exec -o /dev/null 2>&1 | filecheck %s
+
+;; Test the fuzzing-support module imports.
+
+(module
+ (import "fuzzing-support" "log-i32" (func $log-i32 (param i32)))
+ (import "fuzzing-support" "log-f64" (func $log-f64 (param f64)))
+
+ (import "fuzzing-support" "throw" (func $throw))
+
+ ;; CHECK:      [fuzz-exec] calling logging
+ ;; CHECK-NEXT: [LoggingExternalInterface logging 42]
+ ;; CHECK-NEXT: [LoggingExternalInterface logging 3.14159]
+ (func $logging (export "logging")
+  (call $log-i32
+   (i32.const 42)
+  )
+  (call $log-f64
+   (f64.const 3.14159)
+  )
+ )
+
+ ;; CHECK:      [fuzz-exec] calling throwing
+ ;; CHECK-NEXT: [exception thrown: __private ()]
+ ;; CHECK-NEXT: warning: no passes specified, not doing any work
+ (func $throwing (export "throwing")
+  (call $throw)
+ )
+)
+;; CHECK:      [fuzz-exec] calling logging
+;; CHECK-NEXT: [LoggingExternalInterface logging 42]
+;; CHECK-NEXT: [LoggingExternalInterface logging 3.14159]
+
+;; CHECK:      [fuzz-exec] calling throwing
+;; CHECK-NEXT: [exception thrown: __private ()]
+;; CHECK-NEXT: [fuzz-exec] comparing logging
+;; CHECK-NEXT: [fuzz-exec] comparing throwing
diff --git a/test/lit/exec/host-limit.wast b/test/lit/exec/host-limit.wast
@@ -9,7 +9,7 @@
 (module
  (type $type$0 (array i8))
 
- (import "fuzzing-support" "log" (func $log (param i32)))
+ (import "fuzzing-support" "log-i32" (func $log (param i32)))
 
  (global $global (mut (ref $type$0)) (array.new_default $type$0
   (i32.const -1)
diff --git a/test/passes/fuzz_metrics_noprint.bin.txt b/test/passes/fuzz_metrics_noprint.bin.txt
@@ -1,35 +1,35 @@
 Metrics
 total
- [exports]      : 23      
- [funcs]        : 34      
+ [exports]      : 50      
+ [funcs]        : 72      
  [globals]      : 9       
  [imports]      : 4       
  [memories]     : 1       
  [memory-data]  : 2       
- [table-data]   : 6       
+ [table-data]   : 25      
  [tables]       : 1       
  [tags]         : 0       
- [total]        : 4303    
- [vars]         : 100     
- Binary         : 355     
- Block          : 684     
- Break          : 149     
- Call           : 219     
+ [total]        : 4381    
+ [vars]         : 218     
+ Binary         : 335     
+ Block          : 725     
+ Break          : 120     
+ Call           : 210     
  CallIndirect   : 23      
- Const          : 643     
- Drop           : 50      
- GlobalGet      : 367     
- GlobalSet      : 258     
- If             : 206     
- Load           : 78      
- LocalGet       : 339     
- LocalSet       : 236     
- Loop           : 93      
- Nop            : 41      
- RefFunc        : 6       
- Return         : 45      
- Select         : 41      
- Store          : 36      
- Switch         : 1       
- Unary          : 304     
- Unreachable    : 129     
+ Const          : 692     
+ Drop           : 64      
+ GlobalGet      : 391     
+ GlobalSet      : 298     
+ If             : 236     
+ Load           : 71      
+ LocalGet       : 285     
+ LocalSet       : 209     
+ Loop           : 76      
+ Nop            : 63      
+ RefFunc        : 25      
+ Return         : 60      
+ Select         : 23      
+ Store          : 29      
+ Switch         : 2       
+ Unary          : 293     
+ Unreachable    : 151     
diff --git a/test/passes/simplify-locals_all-features.txt b/test/passes/simplify-locals_all-features.txt
@@ -1220,9 +1220,9 @@
  (type $9 (func (result f64)))
  (type $10 (func (param i32 i32) (result f64)))
  (type $11 (func (param i32 i32) (result i32)))
- (import "fuzzing-support" "log1" (func $fimport$0 (type $FUNCSIG$i) (result i32)))
- (import "fuzzing-support" "log2" (func $fimport$1 (type $4) (param i32)))
- (import "fuzzing-support" "log3" (func $fimport$2 (type $7) (param f32)))
+ (import "env" "get1" (func $fimport$0 (type $FUNCSIG$i) (result i32)))
+ (import "fuzzing-support" "log-i32" (func $fimport$1 (type $4) (param i32)))
+ (import "fuzzing-support" "log-f32" (func $fimport$2 (type $7) (param f32)))
  (global $global$0 (mut i32) (i32.const 10))
  (memory $0 256 256 shared)
  (func $nonatomics (type $FUNCSIG$i) (result i32)
diff --git a/test/passes/simplify-locals_all-features.wast b/test/passes/simplify-locals_all-features.wast
@@ -1194,9 +1194,9 @@
   (type $4 (func (param i32)))
   (type $5 (func (param i32) (result i32)))
   (type $6 (func (param i32 i32 i32 i32 i32 i32)))
-  (import "fuzzing-support" "log1" (func $fimport$0 (result i32)))
-  (import "fuzzing-support" "log2" (func $fimport$1 (param i32)))
-  (import "fuzzing-support" "log3" (func $fimport$2 (param f32)))
+  (import "env" "get1" (func $fimport$0 (result i32)))
+  (import "fuzzing-support" "log-i32" (func $fimport$1 (param i32)))
+  (import "fuzzing-support" "log-f32" (func $fimport$2 (param f32)))
   (memory 256 256 shared)
   (global $global$0 (mut i32) (i32.const 10))
   (func $nonatomics (result i32) ;; loads are reordered
diff --git a/test/passes/simplify-locals_all-features_disable-exception-handling.txt b/test/passes/simplify-locals_all-features_disable-exception-handling.txt
diff --git a/test/passes/simplify-locals_all-features_disable-exception-handling.wast b/test/passes/simplify-locals_all-features_disable-exception-handling.wast
diff --git a/test/passes/translate-to-fuzz_all-features_metrics_noprint.txt b/test/passes/translate-to-fuzz_all-features_metrics_noprint.txt
