[clang] Support '-fmodule-file-home-is-cwd' for C++ modules. (PR #135147)

[Michael Park via cfe-commits]

https://github.com/mpark updated https://github.com/llvm/llvm-project/pull/135147

>From 37a5c579d37d4314f79bcf77b7421bfbf61243aa Mon Sep 17 00:00:00 2001
From: Michael Park <mcypark at gmail.com>
Date: Thu, 10 Apr 2025 02:01:07 -0700
Subject: [PATCH] Support '-fmodule-file-home-is-cwd' for C++ modules.

'-fmodule-file-home-is-cwd' was added in https://github.com/llvm/llvm-project/commit/646e502de0d854cb3ecaca90ab52bebfe59a40cd
to support relocatable PCMs for Clang modules. This PR extends the
functionality for standard C++ modules.
---
 clang/lib/Serialization/ASTWriter.cpp      | 69 +++++++++++-----------
 clang/test/Modules/relocatable-modules.cpp | 54 +++++++++++++++++
 2 files changed, 90 insertions(+), 33 deletions(-)
 create mode 100644 clang/test/Modules/relocatable-modules.cpp

diff --git a/clang/lib/Serialization/ASTWriter.cpp b/clang/lib/Serialization/ASTWriter.cpp
index a48c05061626a..95b5718f1d140 100644
--- a/clang/lib/Serialization/ASTWriter.cpp
+++ b/clang/lib/Serialization/ASTWriter.cpp
@@ -1493,42 +1493,45 @@ void ASTWriter::WriteControlBlock(Preprocessor &PP, StringRef isysroot) {
     unsigned AbbrevCode = Stream.EmitAbbrev(std::move(Abbrev));
     RecordData::value_type Record[] = {MODULE_NAME};
     Stream.EmitRecordWithBlob(AbbrevCode, Record, WritingModule->Name);
-  }
 
-  if (WritingModule && WritingModule->Directory) {
-    SmallString<128> BaseDir;
-    if (PP.getHeaderSearchInfo().getHeaderSearchOpts().ModuleFileHomeIsCwd) {
-      // Use the current working directory as the base path for all inputs.
-      auto CWD = FileMgr.getOptionalDirectoryRef(".");
-      BaseDir.assign(CWD->getName());
-    } else {
-      BaseDir.assign(WritingModule->Directory->getName());
-    }
-    cleanPathForOutput(FileMgr, BaseDir);
-
-    // If the home of the module is the current working directory, then we
-    // want to pick up the cwd of the build process loading the module, not
-    // our cwd, when we load this module.
-    if (!PP.getHeaderSearchInfo().getHeaderSearchOpts().ModuleFileHomeIsCwd &&
-        (!PP.getHeaderSearchInfo()
-              .getHeaderSearchOpts()
-              .ModuleMapFileHomeIsCwd ||
-         WritingModule->Directory->getName() != ".")) {
-      // Module directory.
-      auto Abbrev = std::make_shared<BitCodeAbbrev>();
-      Abbrev->Add(BitCodeAbbrevOp(MODULE_DIRECTORY));
-      Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Blob)); // Directory
-      unsigned AbbrevCode = Stream.EmitAbbrev(std::move(Abbrev));
+    auto BaseDir = [&]() -> std::optional<SmallString<128>> {
+      if (PP.getHeaderSearchInfo().getHeaderSearchOpts().ModuleFileHomeIsCwd) {
+        // Use the current working directory as the base path for all inputs.
+        auto CWD = FileMgr.getOptionalDirectoryRef(".");
+        return CWD->getName();
+      }
+      if (WritingModule->Directory) {
+        return WritingModule->Directory->getName();
+      }
+      return std::nullopt;
+    }();
+    if (BaseDir) {
+      cleanPathForOutput(FileMgr, *BaseDir);
+
+      // If the home of the module is the current working directory, then we
+      // want to pick up the cwd of the build process loading the module, not
+      // our cwd, when we load this module.
+      if (!PP.getHeaderSearchInfo().getHeaderSearchOpts().ModuleFileHomeIsCwd &&
+          (!PP.getHeaderSearchInfo()
+                .getHeaderSearchOpts()
+                .ModuleMapFileHomeIsCwd ||
+           WritingModule->Directory->getName() != ".")) {
+        // Module directory.
+        auto Abbrev = std::make_shared<BitCodeAbbrev>();
+        Abbrev->Add(BitCodeAbbrevOp(MODULE_DIRECTORY));
+        Abbrev->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Blob)); // Directory
+        unsigned AbbrevCode = Stream.EmitAbbrev(std::move(Abbrev));
+
+        RecordData::value_type Record[] = {MODULE_DIRECTORY};
+        Stream.EmitRecordWithBlob(AbbrevCode, Record, *BaseDir);
+      }
 
-      RecordData::value_type Record[] = {MODULE_DIRECTORY};
-      Stream.EmitRecordWithBlob(AbbrevCode, Record, BaseDir);
+      // Write out all other paths relative to the base directory if possible.
+      BaseDirectory.assign(BaseDir->begin(), BaseDir->end());
+    } else if (!isysroot.empty()) {
+      // Write out paths relative to the sysroot if possible.
+      BaseDirectory = std::string(isysroot);
     }
-
-    // Write out all other paths relative to the base directory if possible.
-    BaseDirectory.assign(BaseDir.begin(), BaseDir.end());
-  } else if (!isysroot.empty()) {
-    // Write out paths relative to the sysroot if possible.
-    BaseDirectory = std::string(isysroot);
   }
 
   // Module map file
diff --git a/clang/test/Modules/relocatable-modules.cpp b/clang/test/Modules/relocatable-modules.cpp
new file mode 100644
index 0000000000000..c8d1e6d455666
--- /dev/null
+++ b/clang/test/Modules/relocatable-modules.cpp
@@ -0,0 +1,54 @@
+// RUN: rm -rf %t
+// RUN: mkdir -p %t
+// RUN: split-file %s %t
+// RUN: cd %t
+
+// RUN: %clang_cc1 -std=c++20 -emit-header-unit -xc++-user-header hu-01.h \
+// RUN:   -fmodule-name=hu-01 -o hu-01.pcm
+
+// RUN: %clang_cc1 -std=c++20 -emit-header-unit -xc++-user-header hu-02.h \
+// RUN:  -Wno-experimental-header-units -fmodule-file=hu-01.pcm -o hu-02-abs.pcm
+
+// RUN: %clang_cc1 -std=c++20 -emit-header-unit -xc++-user-header hu-02.h \
+// RUN:  -Wno-experimental-header-units -fmodule-file=hu-01.pcm -o hu-02-rel.pcm \
+// RUN:  -fmodule-file-home-is-cwd
+
+// RUN: %clang -module-file-info hu-02-abs.pcm | FileCheck %s --check-prefix=IMPORT-ABS -DPREFIX=%t
+// IMPORT-ABS: Imports module 'hu-01': [[PREFIX]]{{/|\\}}hu-01.pcm
+
+// RUN: %clang -module-file-info hu-02-rel.pcm | FileCheck %s --check-prefix=IMPORT-REL
+// IMPORT-REL: Imports module 'hu-01': hu-01.pcm
+
+// RUN: llvm-bcanalyzer --dump --disable-histogram %t/hu-02-abs.pcm \
+// RUN:   | FileCheck %s --check-prefix=INPUT-ABS -DPREFIX=%t
+// INPUT-ABS: <INPUT_FILE {{.*}}/> blob data = '[[PREFIX]]{{/|\\}}hu-02.h'
+
+// RUN: llvm-bcanalyzer --dump --disable-histogram %t/hu-02-rel.pcm \
+// RUN:   | FileCheck %s --check-prefix=INPUT-REL
+// INPUT-REL: <INPUT_FILE {{.*}}/> blob data = 'hu-02.h'
+
+//--- hu-01.h
+inline void f() {}
+
+//--- hu-02.h
+import "hu-01.h";
+
+inline void g() {
+  f();
+}
+
+// RUN: %clang_cc1 -std=c++20 -emit-module-interface %t/a.cppm -o %t/a-abs.pcm
+
+// RUN: %clang_cc1 -std=c++20 -emit-module-interface %t/a.cppm -o %t/a-rel.pcm \
+// RUN:   -fmodule-file-home-is-cwd
+
+// RUN: llvm-bcanalyzer --dump --disable-histogram %t/a-abs.pcm \
+// RUN:   | FileCheck %s --check-prefix=M-INPUT-ABS -DPREFIX=%t
+// M-INPUT-ABS: <INPUT_FILE {{.*}}/> blob data = '[[PREFIX]]{{/|\\}}a.cppm'
+
+// RUN: llvm-bcanalyzer --dump --disable-histogram %t/a-rel.pcm \
+// RUN:   | FileCheck %s --check-prefix=M-INPUT-REL
+// M-INPUT-REL: <INPUT_FILE {{.*}}/> blob data = 'a.cppm'
+
+//--- a.cppm
+export module a;