[Dependency Scanning] Parallelize Clang module queries

This change refactors the top-level dependency scanning flow to follow the following procedure:

Scan():
1. From the source target under scan, query all imported module identifiers for a *Swift* module. Leave unresolved identifiers unresolved. Proceed transitively to build a *Swift* module dependency graph.
2. Take every unresolved import identifier in the graph from (1) and, assuming that it must be a Clang module, dispatch all of them to be queried in-parallel by the scanner's worker pool.
3. Resolve bridging header Clang module dpendencies
4. Resolve all Swift overlay dependencies, relying on all Clang modules collected in (2) and (3)
5. For the source target under scan, use all of the above discovered module dependencies to resolve all cross-import overlay dependencies

Author: artemcm

include/swift/AST/ModuleDependencies.h

@@ -180,7 +180,7 @@ class ModuleDependencyInfoStorageBase {
                                   ArrayRef<LinkLibrary> linkLibraries,
                                   StringRef moduleCacheKey = "")
       : dependencyKind(dependencyKind), linkLibraries(linkLibraries),
-        moduleCacheKey(moduleCacheKey.str()), resolved(false),
+        moduleCacheKey(moduleCacheKey.str()),
         finalized(false) {}
 
   ModuleDependencyInfoStorageBase(
@@ -191,7 +191,7 @@ class ModuleDependencyInfoStorageBase {
       : dependencyKind(dependencyKind), moduleImports(moduleImports),
         optionalModuleImports(optionalModuleImports),
         linkLibraries(linkLibraries), moduleCacheKey(moduleCacheKey.str()),
-        resolved(false), finalized(false) {}
+        finalized(false) {}
 
   virtual ModuleDependencyInfoStorageBase *clone() const = 0;
 
@@ -209,11 +209,13 @@ class ModuleDependencyInfoStorageBase {
   /// use this module.
   std::vector<LinkLibrary> linkLibraries;
 
-  /// The set of modules on which this module depends, resolved
-  /// to Module IDs, qualified by module kind: Swift, Clang, etc.
-  std::vector<ModuleDependencyID> resolvedDirectModuleDependencies;
-
-  /// Dependencies comprised of Swift overlay modules of direct and
+  /// All directly-imported Swift module dependencies.
+  std::vector<ModuleDependencyID> importedSwiftModules;
+  /// All directly-imported Clang module dependencies.
+  std::vector<ModuleDependencyID> importedClangModules;
+  /// All cross-import overlay module dependencies.
+  std::vector<ModuleDependencyID> crossImportOverlayModules;
+  /// All dependencies comprised of Swift overlay modules of direct and
   /// transitive Clang dependencies.
   std::vector<ModuleDependencyID> swiftOverlayDependencies;
 
@@ -227,8 +229,6 @@ class ModuleDependencyInfoStorageBase {
   /// The macro dependencies.
   std::map<std::string, MacroPluginDependency> macroDependencies;
 
-  /// The direct dependency of the module is resolved by scanner.
-  bool resolved;
   /// ModuleDependencyInfo is finalized (with all transitive dependencies
   /// and inputs).
   bool finalized;
@@ -239,6 +239,8 @@ struct CommonSwiftTextualModuleDependencyDetails {
       ArrayRef<StringRef> extraPCMArgs, ArrayRef<StringRef> buildCommandLine,
       StringRef CASFileSystemRootID)
       : extraPCMArgs(extraPCMArgs.begin(), extraPCMArgs.end()),
+        bridgingHeaderFile(std::nullopt),
+        bridgingSourceFiles(), bridgingModuleDependencies(),
         buildCommandLine(buildCommandLine.begin(), buildCommandLine.end()),
         CASFileSystemRootID(CASFileSystemRootID) {}
 
@@ -254,7 +256,7 @@ struct CommonSwiftTextualModuleDependencyDetails {
   std::vector<std::string> bridgingSourceFiles;
 
   /// (Clang) modules on which the bridging header depends.
-  std::vector<std::string> bridgingModuleDependencies;
+  std::vector<ModuleDependencyID> bridgingModuleDependencies;
 
   /// The Swift frontend invocation arguments to build the Swift module from the
   /// interface.
@@ -426,7 +428,7 @@ class SwiftBinaryModuleDependencyStorage
   std::vector<std::string> headerSourceFiles;
 
   /// (Clang) modules on which the header inputs depend.
-  std::vector<std::string> headerModuleDependencies;
+  std::vector<ModuleDependencyID> headerModuleDependencies;
 
   /// A flag that indicates this dependency is a framework
   const bool isFramework;
@@ -660,42 +662,105 @@ class ModuleDependencyInfo {
     return storage->optionalModuleImports;
   }
 
-  /// Retreive the module-level dependencies.
-  const ArrayRef<ModuleDependencyID> getDirectModuleDependencies() const {
-    assert(storage->resolved);
-    return storage->resolvedDirectModuleDependencies;
-  }
-
   std::string getModuleCacheKey() const {
-    assert(storage->resolved);
     return storage->moduleCacheKey;
   }
 
   void updateModuleCacheKey(const std::string &key) {
     storage->moduleCacheKey = key;
   }
 
-  /// Resolve a dependency's set of `imports` with qualified Module IDs
   void
-  resolveDirectDependencies(const ArrayRef<ModuleDependencyID> dependencyIDs) {
-    assert(!storage->resolved && "Resolving an already-resolved dependency");
-    storage->resolved = true;
-    storage->resolvedDirectModuleDependencies.assign(dependencyIDs.begin(),
-                                                     dependencyIDs.end());
+  setImportedSwiftDependencies(const ArrayRef<ModuleDependencyID> dependencyIDs) {
+    assert(isSwiftModule());
+    storage->importedSwiftModules.assign(dependencyIDs.begin(),
+                                         dependencyIDs.end());
+  }
+  const ArrayRef<ModuleDependencyID> getImportedSwiftDependencies() const {
+    return storage->importedSwiftModules;
+  }
+
+  void
+  setImportedClangDependencies(const ArrayRef<ModuleDependencyID> dependencyIDs) {
+    storage->importedClangModules.assign(dependencyIDs.begin(),
+                                         dependencyIDs.end());
+  }
+  const ArrayRef<ModuleDependencyID> getImportedClangDependencies() const {
+    return storage->importedClangModules;
+  }
+
+  void
+  setHeaderClangDependencies(const ArrayRef<ModuleDependencyID> dependencyIDs) {
+    assert(isSwiftModule());
+    switch (getKind()) {
+    case swift::ModuleDependencyKind::SwiftInterface: {
+      auto swiftInterfaceStorage =
+          cast<SwiftInterfaceModuleDependenciesStorage>(storage.get());
+      swiftInterfaceStorage->textualModuleDetails.bridgingModuleDependencies.assign(dependencyIDs.begin(),
+                                                                                    dependencyIDs.end());
+      break;
+    }
+    case swift::ModuleDependencyKind::SwiftSource: {
+      auto swiftSourceStorage =
+          cast<SwiftSourceModuleDependenciesStorage>(storage.get());
+      swiftSourceStorage->textualModuleDetails.bridgingModuleDependencies.assign(dependencyIDs.begin(),
+                                                                                 dependencyIDs.end());
+      break;
+    }
+    case swift::ModuleDependencyKind::SwiftBinary: {
+      auto swiftBinaryStorage =
+          cast<SwiftBinaryModuleDependencyStorage>(storage.get());
+      swiftBinaryStorage->headerModuleDependencies.assign(dependencyIDs.begin(),
+                                                          dependencyIDs.end());
+      break;
+    }
+    default: {
+      llvm_unreachable("Unexpected dependency kind");
+    }
+    }
+  }
+  const ArrayRef<ModuleDependencyID> getHeaderClangDependencies() const {
+    switch (getKind()) {
+    case swift::ModuleDependencyKind::SwiftInterface: {
+      auto swiftInterfaceStorage =
+          cast<SwiftInterfaceModuleDependenciesStorage>(storage.get());
+      return swiftInterfaceStorage->textualModuleDetails.bridgingModuleDependencies;
+    }
+    case swift::ModuleDependencyKind::SwiftSource: {
+      auto swiftSourceStorage =
+          cast<SwiftSourceModuleDependenciesStorage>(storage.get());
+      return swiftSourceStorage->textualModuleDetails.bridgingModuleDependencies;
+    }
+    case swift::ModuleDependencyKind::SwiftBinary: {
+      auto swiftBinaryStorage =
+          cast<SwiftBinaryModuleDependencyStorage>(storage.get());
+      return swiftBinaryStorage->headerModuleDependencies;
+    }
+    default:
+      return {};
+    }
   }
 
-  /// Set this module's set of Swift Overlay dependencies
   void
-  setOverlayDependencies(const ArrayRef<ModuleDependencyID> dependencyIDs) {
+  setSwiftOverlayDependencies(const ArrayRef<ModuleDependencyID> dependencyIDs) {
     assert(isSwiftModule());
     storage->swiftOverlayDependencies.assign(dependencyIDs.begin(),
                                              dependencyIDs.end());
   }
-
   const ArrayRef<ModuleDependencyID> getSwiftOverlayDependencies() const {
     return storage->swiftOverlayDependencies;
   }
 
+  void
+  setCrossImportOverlayDependencies(const ArrayRef<ModuleDependencyID> dependencyIDs) {
+    assert(isSwiftModule());
+    storage->crossImportOverlayModules.assign(dependencyIDs.begin(),
+                                              dependencyIDs.end());
+  }
+  const ArrayRef<ModuleDependencyID> getCrossImportOverlayDependencies() const {
+    return storage->crossImportOverlayModules;
+  }
+
   const ArrayRef<LinkLibrary> getLinkLibraries() const {
     return storage->linkLibraries;
   }
@@ -723,16 +788,6 @@ class ModuleDependencyInfo {
     return {};
   }
 
-  const ArrayRef<std::string> getHeaderDependencies() const {
-    if (auto *detail = getAsSwiftInterfaceModule())
-      return detail->textualModuleDetails.bridgingModuleDependencies;
-    if (auto *detail = getAsSwiftSourceModule())
-      return detail->textualModuleDetails.bridgingModuleDependencies;
-    if (auto *detail = getAsSwiftBinaryModule())
-      return detail->headerModuleDependencies;
-    return {};
-  }
-
   std::vector<std::string> getCommandline() const {
     if (auto *detail = getAsClangModule())
       return detail->buildCommandLine;
@@ -798,9 +853,8 @@ class ModuleDependencyInfo {
       llvm_unreachable("Unexpected type");
   }
 
-  bool isResolved() const { return storage->resolved; }
-  void setIsResolved(bool isResolved) { storage->resolved = isResolved; }
-
+  /// Whether explicit input paths of all the module dependencies
+  /// have been specified on the command-line recipe for this module.
   bool isFinalized() const { return storage->finalized; }
   void setIsFinalized(bool isFinalized) { storage->finalized = isFinalized; }
 
@@ -876,10 +930,6 @@ class ModuleDependencyInfo {
                        const SourceManager *sourceManager = nullptr,
                        SourceLoc sourceLocation = SourceLoc());
 
-  /// Add a kind-qualified module dependency ID to the set of
-  /// module dependencies.
-  void addModuleDependency(ModuleDependencyID dependencyID);
-
   /// Get the bridging header.
   std::optional<std::string> getBridgingHeader() const;
 
@@ -904,10 +954,6 @@ class ModuleDependencyInfo {
   /// Add source files that the header input depends on.
   void addHeaderSourceFile(StringRef bridgingSourceFile);
 
-  /// Add (Clang) modules on which a non-bridging header input depends.
-  void addHeaderInputModuleDependency(StringRef module,
-                                      llvm::StringSet<> &alreadyAddedModules);
-
   /// Add bridging header include tree.
   void addBridgingHeaderIncludeTree(StringRef ID);
 
@@ -1175,6 +1221,8 @@ class ModuleDependenciesCache {
                      std::optional<ModuleDependencyKind> kind) const;
   /// Whether we have cached dependency information for the given module Name.
   bool hasDependency(StringRef moduleName) const;
+  /// Whether we have cached dependency information for the given Swift module.
+  bool hasSwiftDependency(StringRef moduleName) const;
 
   SwiftDependencyScanningService &getScanService() {
     return globalScanningService;
@@ -1191,17 +1239,29 @@ class ModuleDependenciesCache {
   }
   std::string getModuleOutputPath() const { return moduleOutputPath; }
 
-  /// Query all dependencies, direct and Swift overlay.
-  std::vector<ModuleDependencyID>
+  /// Query all dependencies
+  ModuleDependencyIDSetVector
   getAllDependencies(const ModuleDependencyID &moduleID) const;
-
-  /// Query only direct import dependencies
+  
+  /// Query all Clang module dependencies.
+  ModuleDependencyIDSetVector
+  getClangDependencies(const ModuleDependencyID &moduleID) const;
+  
+  /// Query all directly-imported Swift dependencies
   llvm::ArrayRef<ModuleDependencyID>
-  getOnlyDirectDependencies(const ModuleDependencyID &moduleID) const;
-
-  /// Query only Swift overlay dependencies
+  getImportedSwiftDependencies(const ModuleDependencyID &moduleID) const;
+  /// Query all directly-imported Clang dependencies
+  llvm::ArrayRef<ModuleDependencyID>
+  getImportedClangDependencies(const ModuleDependencyID &moduleID) const;
+  /// Query all Clang module dependencies of this module's imported (bridging) header
   llvm::ArrayRef<ModuleDependencyID>
-  getOnlyOverlayDependencies(const ModuleDependencyID &moduleID) const;
+  getHeaderClangDependencies(const ModuleDependencyID &moduleID) const;
+  /// Query Swift overlay dependencies
+  llvm::ArrayRef<ModuleDependencyID>
+  getSwiftOverlayDependencies(const ModuleDependencyID &moduleID) const;
+  /// Query all cross-import overlay dependencies
+  llvm::ArrayRef<ModuleDependencyID>
+  getCrossImportOverlayDependencies(const ModuleDependencyID &moduleID) const;
 
   /// Look for module dependencies for a module with the given ID
   ///
@@ -1222,6 +1282,12 @@ class ModuleDependenciesCache {
   std::optional<const ModuleDependencyInfo *>
   findDependency(StringRef moduleName) const;
 
+  /// Look for Swift module dependencies for a module with the given name
+  ///
+  /// \returns the cached result, or \c None if there is no cached entry.
+  std::optional<const ModuleDependencyInfo *>
+  findSwiftDependency(StringRef moduleName) const;
+
   /// Look for known existing dependencies.
   ///
   /// \returns the cached result.
@@ -1237,19 +1303,32 @@ class ModuleDependenciesCache {
 
   /// Update stored dependencies for the given module.
   void updateDependency(ModuleDependencyID moduleID,
-                        ModuleDependencyInfo dependencies);
+                        ModuleDependencyInfo dependencyInfo);
 
-  /// Resolve a dependency module's set of imports
-  /// to a kind-qualified set of module IDs.
+  /// Resolve this module's set of directly-imported Swift module
+  /// dependencies
   void
-  resolveDependencyImports(ModuleDependencyID moduleID,
-                           const ArrayRef<ModuleDependencyID> dependencyIDs);
-
-  /// Resolve a dependency module's set of Swift module dependencies
+  setImportedSwiftDependencies(ModuleDependencyID moduleID,
+                               const ArrayRef<ModuleDependencyID> dependencyIDs);
+  /// Resolve this module's set of directly-imported Clang module
+  /// dependencies
+  void
+  setImportedClangDependencies(ModuleDependencyID moduleID,
+                               const ArrayRef<ModuleDependencyID> dependencyIDs);
+  /// Resolve this module's set of Swift module dependencies
   /// that are Swift overlays of Clang module dependencies.
   void
   setSwiftOverlayDependencies(ModuleDependencyID moduleID,
                               const ArrayRef<ModuleDependencyID> dependencyIDs);
+  /// Resolve this Swift module's imported (bridging) header's
+  /// Clang module dependencies
+  void
+  setHeaderClangDependencies(ModuleDependencyID moduleID,
+                             const ArrayRef<ModuleDependencyID> dependencyIDs);
+  /// Resolve this module's cross-import overlay dependencies
+  void
+  setCrossImportOverlayDependencies(ModuleDependencyID moduleID,
+                                    const ArrayRef<ModuleDependencyID> dependencyIDs);
 
   StringRef getMainModuleName() const { return mainScanModuleName; }
 };

include/swift/ClangImporter/ClangImporter.h

@@ -90,6 +90,9 @@ class SwiftLookupTable;
 class TypeDecl;
 class ValueDecl;
 class VisibleDeclConsumer;
+using ModuleDependencyIDSetVector =
+    llvm::SetVector<ModuleDependencyID, std::vector<ModuleDependencyID>,
+                    std::set<ModuleDependencyID>>;
 enum class SelectorSplitKind;
 
 /// Kinds of optional types.
@@ -492,7 +495,11 @@ class ClangImporter final : public ClangModuleLoader {
       ModuleDependencyInfo &MDI,
       const clang::tooling::dependencies::TranslationUnitDeps &deps);
 
-  /// Add dependency information for header dependencies
+  void getBridgingHeaderOptions(
+      const clang::tooling::dependencies::TranslationUnitDeps &deps,
+      std::vector<std::string> &swiftArgs);
+
+  /// Query dependency information for header dependencies
   /// of a binary Swift module.
   ///
   /// \param moduleID the name of the Swift module whose dependency
@@ -505,10 +512,14 @@ class ClangImporter final : public ClangModuleLoader {
   /// about new Clang modules discovered along the way.
   ///
   /// \returns \c true if an error occurred, \c false otherwise
-  bool addHeaderDependencies(
+  bool getHeaderDependencies(
       ModuleDependencyID moduleID,
       clang::tooling::dependencies::DependencyScanningTool &clangScanningTool,
-      ModuleDependenciesCache &cache);
+      ModuleDependenciesCache &cache,
+      ModuleDependencyIDSetVector &headerClangModuleDependencies,
+      std::vector<std::string> &headerFileInputs,
+      std::vector<std::string> &bridgingHeaderCommandLine,
+      std::optional<std::string> &includeTreeID);
 
   clang::TargetInfo &getModuleAvailabilityTarget() const override;
   clang::ASTContext &getClangASTContext() const override;