[clang] [flang] [flang] New -fdebug-unparse-with-modules option (PR #91660)

via cfe-commits cfe-commits at lists.llvm.org
Thu May 9 14:34:08 PDT 2024


llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-clang

Author: Peter Klausler (klausler)

<details>
<summary>Changes</summary>

This option is a compilation action that parses a source file and performs semantic analysis on it, like the existing -fdebug-unparse option does.  Its output, however, is preceded by the effective contents of all of the non-intrinsic modules on which it depends but does not define, transitively preceded by the closure of all of those modules' dependencies.

The output from this option is therefore the analyzed parse tree for a source file encapsulated with all of its non-intrinsic module dependencies.  This output may be useful for extracting code from large applications for use as an attachment to a bug report, or as input to a test case reduction tool for problem isolation.

---
Full diff: https://github.com/llvm/llvm-project/pull/91660.diff


11 Files Affected:

- (modified) clang/include/clang/Driver/Options.td (+3-1) 
- (modified) flang/include/flang/Frontend/FrontendActions.h (+4) 
- (modified) flang/include/flang/Frontend/FrontendOptions.h (+4) 
- (modified) flang/include/flang/Semantics/unparse-with-symbols.h (+4) 
- (modified) flang/lib/Frontend/CompilerInvocation.cpp (+3) 
- (modified) flang/lib/Frontend/FrontendActions.cpp (+9) 
- (modified) flang/lib/FrontendTool/ExecuteCompilerInvocation.cpp (+2) 
- (modified) flang/lib/Semantics/mod-file.cpp (+18-2) 
- (modified) flang/lib/Semantics/mod-file.h (+3) 
- (modified) flang/lib/Semantics/unparse-with-symbols.cpp (+38) 
- (added) flang/test/Driver/unparse-with-modules.f90 (+34) 


``````````diff
diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td
index 73a2518480e9b..2e6db8d5ebd56 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -6644,7 +6644,9 @@ def fdebug_unparse : Flag<["-"], "fdebug-unparse">, Group<Action_Group>,
   DocBrief<[{Run the parser and the semantic checks. Then unparse the
 parse-tree and output the generated Fortran source file.}]>;
 def fdebug_unparse_with_symbols : Flag<["-"], "fdebug-unparse-with-symbols">, Group<Action_Group>,
-  HelpText<"Unparse and stop.">;
+  HelpText<"Unparse with symbols and stop.">;
+def fdebug_unparse_with_modules : Flag<["-"], "fdebug-unparse-with-modules">, Group<Action_Group>,
+  HelpText<"Unparse with dependent modules and stop.">;
 def fdebug_dump_symbols : Flag<["-"], "fdebug-dump-symbols">, Group<Action_Group>,
   HelpText<"Dump symbols after the semantic analysis">;
 def fdebug_dump_parse_tree : Flag<["-"], "fdebug-dump-parse-tree">, Group<Action_Group>,
diff --git a/flang/include/flang/Frontend/FrontendActions.h b/flang/include/flang/Frontend/FrontendActions.h
index e2e859f3a81bd..7823565eb815f 100644
--- a/flang/include/flang/Frontend/FrontendActions.h
+++ b/flang/include/flang/Frontend/FrontendActions.h
@@ -108,6 +108,10 @@ class DebugUnparseWithSymbolsAction : public PrescanAndSemaAction {
   void executeAction() override;
 };
 
+class DebugUnparseWithModulesAction : public PrescanAndSemaAction {
+  void executeAction() override;
+};
+
 class DebugUnparseAction : public PrescanAndSemaAction {
   void executeAction() override;
 };
diff --git a/flang/include/flang/Frontend/FrontendOptions.h b/flang/include/flang/Frontend/FrontendOptions.h
index 06b1318f243b0..82ca99672ec61 100644
--- a/flang/include/flang/Frontend/FrontendOptions.h
+++ b/flang/include/flang/Frontend/FrontendOptions.h
@@ -63,6 +63,10 @@ enum ActionKind {
   /// Fortran source file
   DebugUnparseWithSymbols,
 
+  /// Parse, run semantics, and output a Fortran source file preceded
+  /// by all the necessary modules (transitively)
+  DebugUnparseWithModules,
+
   /// Parse, run semantics and then output symbols from semantics
   DebugDumpSymbols,
 
diff --git a/flang/include/flang/Semantics/unparse-with-symbols.h b/flang/include/flang/Semantics/unparse-with-symbols.h
index d70110245e2b2..5e18b3fc3063d 100644
--- a/flang/include/flang/Semantics/unparse-with-symbols.h
+++ b/flang/include/flang/Semantics/unparse-with-symbols.h
@@ -21,8 +21,12 @@ struct Program;
 }
 
 namespace Fortran::semantics {
+class SemanticsContext;
 void UnparseWithSymbols(llvm::raw_ostream &, const parser::Program &,
     parser::Encoding encoding = parser::Encoding::UTF_8);
+void UnparseWithModules(llvm::raw_ostream &, SemanticsContext &,
+    const parser::Program &,
+    parser::Encoding encoding = parser::Encoding::UTF_8);
 }
 
 #endif // FORTRAN_SEMANTICS_UNPARSE_WITH_SYMBOLS_H_
diff --git a/flang/lib/Frontend/CompilerInvocation.cpp b/flang/lib/Frontend/CompilerInvocation.cpp
index 4318286e74152..e0d5043823a6c 100644
--- a/flang/lib/Frontend/CompilerInvocation.cpp
+++ b/flang/lib/Frontend/CompilerInvocation.cpp
@@ -488,6 +488,9 @@ static bool parseFrontendArgs(FrontendOptions &opts, llvm::opt::ArgList &args,
     case clang::driver::options::OPT_fdebug_unparse_with_symbols:
       opts.programAction = DebugUnparseWithSymbols;
       break;
+    case clang::driver::options::OPT_fdebug_unparse_with_modules:
+      opts.programAction = DebugUnparseWithModules;
+      break;
     case clang::driver::options::OPT_fdebug_dump_symbols:
       opts.programAction = DebugDumpSymbols;
       break;
diff --git a/flang/lib/Frontend/FrontendActions.cpp b/flang/lib/Frontend/FrontendActions.cpp
index 2f65ab6102f4d..4341c104a69df 100644
--- a/flang/lib/Frontend/FrontendActions.cpp
+++ b/flang/lib/Frontend/FrontendActions.cpp
@@ -477,6 +477,15 @@ void DebugUnparseWithSymbolsAction::executeAction() {
   reportFatalSemanticErrors();
 }
 
+void DebugUnparseWithModulesAction::executeAction() {
+  auto &parseTree{*getInstance().getParsing().parseTree()};
+  CompilerInstance &ci{getInstance()};
+  Fortran::semantics::UnparseWithModules(
+      llvm::outs(), ci.getSemantics().context(), parseTree,
+      /*encoding=*/Fortran::parser::Encoding::UTF_8);
+  reportFatalSemanticErrors();
+}
+
 void DebugDumpSymbolsAction::executeAction() {
   CompilerInstance &ci = this->getInstance();
 
diff --git a/flang/lib/FrontendTool/ExecuteCompilerInvocation.cpp b/flang/lib/FrontendTool/ExecuteCompilerInvocation.cpp
index 4cad640562c61..e2cbd5112d6ea 100644
--- a/flang/lib/FrontendTool/ExecuteCompilerInvocation.cpp
+++ b/flang/lib/FrontendTool/ExecuteCompilerInvocation.cpp
@@ -59,6 +59,8 @@ createFrontendAction(CompilerInstance &ci) {
     return std::make_unique<DebugUnparseNoSemaAction>();
   case DebugUnparseWithSymbols:
     return std::make_unique<DebugUnparseWithSymbolsAction>();
+  case DebugUnparseWithModules:
+    return std::make_unique<DebugUnparseWithModulesAction>();
   case DebugDumpSymbols:
     return std::make_unique<DebugDumpSymbolsAction>();
   case DebugDumpParseTree:
diff --git a/flang/lib/Semantics/mod-file.cpp b/flang/lib/Semantics/mod-file.cpp
index e9aebe5b08f2b..bb8c6c7567b8d 100644
--- a/flang/lib/Semantics/mod-file.cpp
+++ b/flang/lib/Semantics/mod-file.cpp
@@ -132,11 +132,11 @@ static std::string ModFileName(const SourceName &name,
 
 // Write the module file for symbol, which must be a module or submodule.
 void ModFileWriter::Write(const Symbol &symbol) {
-  auto &module{symbol.get<ModuleDetails>()};
+  const auto &module{symbol.get<ModuleDetails>()};
   if (module.moduleFileHash()) {
     return; // already written
   }
-  auto *ancestor{module.ancestor()};
+  const auto *ancestor{module.ancestor()};
   isSubmodule_ = ancestor != nullptr;
   auto ancestorName{ancestor ? ancestor->GetName().value().ToString() : ""s};
   auto path{context_.moduleDirectory() + '/' +
@@ -151,6 +151,21 @@ void ModFileWriter::Write(const Symbol &symbol) {
   const_cast<ModuleDetails &>(module).set_moduleFileHash(checkSum);
 }
 
+void ModFileWriter::WriteClosure(llvm::raw_ostream &out, const Symbol &symbol,
+    UnorderedSymbolSet &nonIntrinsicModulesWritten) {
+  if (!symbol.has<ModuleDetails>() || symbol.owner().IsIntrinsicModules() ||
+      !nonIntrinsicModulesWritten.insert(symbol).second) {
+    return;
+  }
+  PutSymbols(DEREF(symbol.scope()));
+  needsBuf_.clear(); // omit module checksums
+  auto str{GetAsString(symbol)};
+  for (auto depRef : std::move(usedNonIntrinsicModules_)) {
+    WriteClosure(out, *depRef, nonIntrinsicModulesWritten);
+  }
+  out << std::move(str);
+}
+
 // Return the entire body of the module file
 // and clear saved uses, decls, and contains.
 std::string ModFileWriter::GetAsString(const Symbol &symbol) {
@@ -710,6 +725,7 @@ void ModFileWriter::PutUse(const Symbol &symbol) {
     uses_ << "use,intrinsic::";
   } else {
     uses_ << "use ";
+    usedNonIntrinsicModules_.insert(module);
   }
   uses_ << module.name() << ",only:";
   PutGenericName(uses_, symbol);
diff --git a/flang/lib/Semantics/mod-file.h b/flang/lib/Semantics/mod-file.h
index b4ece4018c054..739add32c2e0e 100644
--- a/flang/lib/Semantics/mod-file.h
+++ b/flang/lib/Semantics/mod-file.h
@@ -35,6 +35,8 @@ class ModFileWriter {
 public:
   explicit ModFileWriter(SemanticsContext &context) : context_{context} {}
   bool WriteAll();
+  void WriteClosure(llvm::raw_ostream &, const Symbol &,
+      UnorderedSymbolSet &nonIntrinsicModulesWritten);
 
 private:
   SemanticsContext &context_;
@@ -46,6 +48,7 @@ class ModFileWriter {
   std::string containsBuf_;
   // Tracks nested DEC structures and fields of that type
   UnorderedSymbolSet emittedDECStructures_, emittedDECFields_;
+  UnorderedSymbolSet usedNonIntrinsicModules_;
 
   llvm::raw_string_ostream needs_{needsBuf_};
   llvm::raw_string_ostream uses_{usesBuf_};
diff --git a/flang/lib/Semantics/unparse-with-symbols.cpp b/flang/lib/Semantics/unparse-with-symbols.cpp
index 67016e85777c7..c451f885c0627 100644
--- a/flang/lib/Semantics/unparse-with-symbols.cpp
+++ b/flang/lib/Semantics/unparse-with-symbols.cpp
@@ -7,6 +7,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "flang/Semantics/unparse-with-symbols.h"
+#include "mod-file.h"
 #include "flang/Parser/parse-tree-visitor.h"
 #include "flang/Parser/parse-tree.h"
 #include "flang/Parser/unparse.h"
@@ -98,4 +99,41 @@ void UnparseWithSymbols(llvm::raw_ostream &out, const parser::Program &program,
           int indent) { visitor.PrintSymbols(location, out, indent); }};
   parser::Unparse(out, program, encoding, false, true, &preStatement);
 }
+
+// UnparseWithModules()
+
+class UsedModuleVisitor {
+public:
+  UnorderedSymbolSet &modulesUsed() { return modulesUsed_; }
+  UnorderedSymbolSet &modulesDefined() { return modulesDefined_; }
+  template <typename T> bool Pre(const T &) { return true; }
+  template <typename T> void Post(const T &) {}
+  void Post(const parser::ModuleStmt &module) {
+    if (module.v.symbol) {
+      modulesDefined_.insert(*module.v.symbol);
+    }
+  }
+  void Post(const parser::UseStmt &use) {
+    if (use.moduleName.symbol) {
+      modulesUsed_.insert(*use.moduleName.symbol);
+    }
+  }
+
+private:
+  UnorderedSymbolSet modulesUsed_;
+  UnorderedSymbolSet modulesDefined_;
+};
+
+void UnparseWithModules(llvm::raw_ostream &out, SemanticsContext &context,
+    const parser::Program &program, parser::Encoding encoding) {
+  UsedModuleVisitor visitor;
+  parser::Walk(program, visitor);
+  UnorderedSymbolSet nonIntrinsicModulesWritten{
+      std::move(visitor.modulesDefined())};
+  ModFileWriter writer{context};
+  for (SymbolRef moduleRef : visitor.modulesUsed()) {
+    writer.WriteClosure(out, *moduleRef, nonIntrinsicModulesWritten);
+  }
+  parser::Unparse(out, program, encoding, false, true);
+}
 } // namespace Fortran::semantics
diff --git a/flang/test/Driver/unparse-with-modules.f90 b/flang/test/Driver/unparse-with-modules.f90
new file mode 100644
index 0000000000000..53997f7804efa
--- /dev/null
+++ b/flang/test/Driver/unparse-with-modules.f90
@@ -0,0 +1,34 @@
+! RUN: %flang_fc1 -I %S/Inputs/module-dir -fdebug-unparse-with-modules %s | FileCheck %s
+module m1
+  use iso_fortran_env
+  use BasicTestModuleTwo
+  implicit none
+  type(t2) y
+  real(real32) x
+end
+
+program test
+  use m1
+  use BasicTestModuleTwo
+  implicit none
+  x = 123.
+  y = t2()
+end
+
+!CHECK-NOT: module iso_fortran_env
+!CHECK: module basictestmoduletwo
+!CHECK: type::t2
+!CHECK: end type
+!CHECK: end
+!CHECK: module m1
+!CHECK:  use :: iso_fortran_env
+!CHECK:  implicit none
+!CHECK:  real(kind=real32) x
+!CHECK: end module
+!CHECK: program test
+!CHECK:  use :: m1
+!CHECK:  use :: basictestmoduletwo
+!CHECK:  implicit none
+!CHECK:  x = 123.
+!CHECK:  y = t2()
+!CHECK: end program

``````````

</details>


https://github.com/llvm/llvm-project/pull/91660


More information about the cfe-commits mailing list