[clang][modules] Delay creating IdentifierInfo for names of explicit modules

When using explicit Clang modules, some declarations might unexpectedly become invisible.

This is caused by the mechanism that loads PCM files passed via `-fmodule-file=<path>` and creates an `IdentifierInfo` for the module name. The `IdentifierInfo` creation takes place when the `ASTReader` is in a weird state, with modules that are loaded but not yet set up properly. This patch delays the creation of `IdentifierInfo` until the `ASTReader` is done with reading the PCM.

Note that the `-fmodule-file=<name>=<path>` form of the argument doesn't suffer from this issue, since it doesn't create `IdentifierInfo` for the module name.

Reviewed By: dexonsmith

Differential Revision: https://reviews.llvm.org/D111543

Author: jansvoboda11

clang/lib/Frontend/CompilerInstance.cpp

@@ -566,25 +566,28 @@ namespace {
 // the files we were handed.
 struct ReadModuleNames : ASTReaderListener {
   Preprocessor &PP;
-  llvm::SmallVector<IdentifierInfo*, 8> LoadedModules;
+  llvm::SmallVector<std::string, 8> LoadedModules;
 
   ReadModuleNames(Preprocessor &PP) : PP(PP) {}
 
   void ReadModuleName(StringRef ModuleName) override {
-    LoadedModules.push_back(PP.getIdentifierInfo(ModuleName));
+    // Keep the module name as a string for now. It's not safe to create a new
+    // IdentifierInfo from an ASTReader callback.
+    LoadedModules.push_back(ModuleName.str());
   }
 
   void registerAll() {
     ModuleMap &MM = PP.getHeaderSearchInfo().getModuleMap();
-    for (auto *II : LoadedModules)
-      MM.cacheModuleLoad(*II, MM.findModule(II->getName()));
+    for (const std::string &LoadedModule : LoadedModules)
+      MM.cacheModuleLoad(*PP.getIdentifierInfo(LoadedModule),
+                         MM.findModule(LoadedModule));
     LoadedModules.clear();
   }
 
   void markAllUnavailable() {
-    for (auto *II : LoadedModules) {
+    for (const std::string &LoadedModule : LoadedModules) {
       if (Module *M = PP.getHeaderSearchInfo().getModuleMap().findModule(
-              II->getName())) {
+              LoadedModule)) {
         M->HasIncompatibleModuleFile = true;
 
         // Mark module as available if the only reason it was unavailable

clang/test/Modules/Inputs/module-name-used-by-objc-bridge/Interface.h

@@ -0,0 +1,2 @@
+@interface Interface
+@end

clang/test/Modules/Inputs/module-name-used-by-objc-bridge/InterfaceBridge.h

@@ -0,0 +1 @@
+typedef struct __attribute__((objc_bridge(Interface))) Foo *Bar;

clang/test/Modules/Inputs/module-name-used-by-objc-bridge/module.modulemap

@@ -0,0 +1,7 @@
+module InterfaceBridge {
+  header "InterfaceBridge.h"
+}
+
+module Interface {
+  header "Interface.h"
+}

clang/test/Modules/module-name-used-by-objc-bridge.m

@@ -0,0 +1,25 @@
+// RUN: rm -rf %t && mkdir %t
+
+// RUN: %clang_cc1 -fmodules -x objective-c -emit-module -fmodule-name=InterfaceBridge \
+// RUN:   %S/Inputs/module-name-used-by-objc-bridge/module.modulemap -o %t/InterfaceBridge.pcm
+
+// RUN: %clang_cc1 -fmodules -x objective-c -emit-module -fmodule-name=Interface \
+// RUN:   %S/Inputs/module-name-used-by-objc-bridge/module.modulemap -o %t/Interface.pcm
+
+// Check that the `-fmodule-file=<name>=<path>` form succeeds:
+// RUN: %clang_cc1 -fmodules -fsyntax-only %s -I %S/Inputs/module-name-used-by-objc-bridge \
+// RUN:   -fmodule-file=InterfaceBridge=%t/InterfaceBridge.pcm -fmodule-file=Interface=%t/Interface.pcm \
+// RUN:   -fmodule-map-file=%S/Inputs/module-name-used-by-objc-bridge/module.modulemap -verify
+
+// Check that the `-fmodule-file=<path>` form succeeds:
+// RUN: %clang_cc1 -fmodules -fsyntax-only %s -I %S/Inputs/module-name-used-by-objc-bridge \
+// RUN:   -fmodule-file=%t/InterfaceBridge.pcm -fmodule-file=%t/Interface.pcm \
+// RUN:   -fmodule-map-file=%S/Inputs/module-name-used-by-objc-bridge/module.modulemap -verify
+
+#import "InterfaceBridge.h"
+#import "Interface.h"
+
+@interface Interface (User)
+@end
+
+// expected-no-diagnostics