[clang][modules] Reset codegen options. (#74006)

CodeGen options do not affect the AST, so they usually can be ignored.
The only exception to the rule is when a PCM is created with
`-gmodules`.
In that case the Clang module format is switched to object file
container and contains also serialized debug information that can be
affected by debug options. There the following approach was choosen:

1.) Split out all the debug options into a separate `DebugOptions.def`
    file. The file is included by `CodeGenOptions.def`, so the change is
    transparent to all existing users of `CodeGenOptions.def`.
2.) Reset all CodeGen options, but excluding affecting debug options.
3.) Conditionally reset debug options that can affect the PCM.

This fixes rdar://113135909.

Author: ributzka

clang/include/clang/Basic/CodeGenOptions.def

@@ -28,12 +28,8 @@ CODEGENOPT(Name, Bits, Default)
 #endif
 
 CODEGENOPT(DisableIntegratedAS, 1, 0) ///< -no-integrated-as
-ENUM_CODEGENOPT(CompressDebugSections, llvm::DebugCompressionType, 2,
-                llvm::DebugCompressionType::None)
 CODEGENOPT(RelaxELFRelocations, 1, 1) ///< -Wa,-mrelax-relocations={yes,no}
 CODEGENOPT(AsmVerbose        , 1, 0) ///< -dA, -fverbose-asm.
-CODEGENOPT(Dwarf64           , 1, 0) ///< -gdwarf64.
-CODEGENOPT(Dwarf32           , 1, 1) ///< -gdwarf32.
 CODEGENOPT(PreserveAsmComments, 1, 1) ///< -dA, -fno-preserve-as-comments.
 CODEGENOPT(AssumeSaneOperatorNew , 1, 1) ///< implicit __attribute__((malloc)) operator new
 CODEGENOPT(AssumeUniqueVTables , 1, 1) ///< Assume a class has only one vtable.
@@ -73,10 +69,6 @@ CODEGENOPT(DebugPassManager, 1, 0) ///< Prints debug information for the new
 CODEGENOPT(DisableRedZone    , 1, 0) ///< Set when -mno-red-zone is enabled.
 CODEGENOPT(EmitCallSiteInfo, 1, 0) ///< Emit call site info only in the case of
                                    ///< '-g' + 'O>0' level.
-CODEGENOPT(EnableDIPreservationVerify, 1, 0) ///< Enable di preservation verify
-                                             ///< each (it means check
-                                             ///< the original debug info
-                                             ///< metadata preservation).
 CODEGENOPT(IndirectTlsSegRefs, 1, 0) ///< Set when -mno-tls-direct-seg-refs
                                      ///< is specified.
 CODEGENOPT(DisableTailCalls  , 1, 0) ///< Do not emit tail calls.
@@ -113,16 +105,10 @@ CODEGENOPT(IndirectBranchCSPrefix, 1, 0) ///< if -mindirect-branch-cs-prefix
 CODEGENOPT(XRayInstrumentFunctions , 1, 0) ///< Set when -fxray-instrument is
                                            ///< enabled.
 CODEGENOPT(StackSizeSection  , 1, 0) ///< Set when -fstack-size-section is enabled.
-CODEGENOPT(ForceDwarfFrameSection , 1, 0) ///< Set when -fforce-dwarf-frame is
-                                          ///< enabled.
 
 ///< Set when -femit-compact-unwind-non-canonical is enabled.
 CODEGENOPT(EmitCompactUnwindNonCanonical, 1, 0)
 
-///< Set when -femit-dwarf-unwind is passed.
-ENUM_CODEGENOPT(EmitDwarfUnwind, llvm::EmitDwarfUnwindType, 2,
-                llvm::EmitDwarfUnwindType::Default)
-
 ///< Set when -fxray-always-emit-customevents is enabled.
 CODEGENOPT(XRayAlwaysEmitCustomEvents , 1, 0)
 
@@ -178,17 +164,13 @@ CODEGENOPT(IncrementalLinkerCompatible, 1, 0) ///< Emit an object file which can
 CODEGENOPT(MergeAllConstants , 1, 1) ///< Merge identical constants.
 CODEGENOPT(MergeFunctions    , 1, 0) ///< Set when -fmerge-functions is enabled.
 CODEGENOPT(NoCommon          , 1, 0) ///< Set when -fno-common or C++ is enabled.
-CODEGENOPT(NoDwarfDirectoryAsm , 1, 0) ///< Set when -fno-dwarf-directory-asm is
-                                       ///< enabled.
 CODEGENOPT(NoExecStack       , 1, 0) ///< Set when -Wa,--noexecstack is enabled.
 CODEGENOPT(FatalWarnings     , 1, 0) ///< Set when -Wa,--fatal-warnings is
                                      ///< enabled.
 CODEGENOPT(NoWarn            , 1, 0) ///< Set when -Wa,--no-warn is enabled.
 CODEGENOPT(NoTypeCheck       , 1, 0) ///< Set when -Wa,--no-type-check is enabled.
 CODEGENOPT(MisExpect         , 1, 0) ///< Set when -Wmisexpect is enabled
 CODEGENOPT(EnableSegmentedStacks , 1, 0) ///< Set when -fsplit-stack is enabled.
-CODEGENOPT(NoInlineLineTables, 1, 0) ///< Whether debug info should contain
-                                     ///< inline line tables.
 CODEGENOPT(StackClashProtector, 1, 0) ///< Set when -fstack-clash-protection is enabled.
 CODEGENOPT(NoImplicitFloat   , 1, 0) ///< Set when -mno-implicit-float is enabled.
 CODEGENOPT(NullPointerIsValid , 1, 0) ///< Assume Null pointer deference is defined.
@@ -341,37 +323,6 @@ VALUE_CODEGENOPT(StackProbeSize    , 32, 4096) ///< Overrides default stack
                                                ///< probe size, even if 0.
 VALUE_CODEGENOPT(WarnStackSize     , 32, UINT_MAX) ///< Set via -fwarn-stack-size.
 CODEGENOPT(NoStackArgProbe, 1, 0) ///< Set when -mno-stack-arg-probe is used
-CODEGENOPT(DebugStrictDwarf, 1, 1) ///< Whether or not to use strict DWARF info.
-
-/// Control the Assignment Tracking debug info feature.
-ENUM_CODEGENOPT(AssignmentTrackingMode, AssignmentTrackingOpts, 2, AssignmentTrackingOpts::Disabled)
-
-CODEGENOPT(DebugColumnInfo, 1, 0) ///< Whether or not to use column information
-                                  ///< in debug info.
-
-CODEGENOPT(DebugTypeExtRefs, 1, 0) ///< Whether or not debug info should contain
-                                   ///< external references to a PCH or module.
-
-CODEGENOPT(DebugExplicitImport, 1, 0)  ///< Whether or not debug info should
-                                       ///< contain explicit imports for
-                                       ///< anonymous namespaces
-
-/// Set debug info source file hashing algorithm.
-ENUM_CODEGENOPT(DebugSrcHash, DebugSrcHashKind, 2, DSH_MD5)
-
-CODEGENOPT(SplitDwarfInlining, 1, 1) ///< Whether to include inlining info in the
-                                     ///< skeleton CU to allow for symbolication
-                                     ///< of inline stack frames without .dwo files.
-CODEGENOPT(DebugFwdTemplateParams, 1, 0) ///< Whether to emit complete
-                                         ///< template parameter descriptions in
-                                         ///< forward declarations (versus just
-                                         ///< including them in the name).
-ENUM_CODEGENOPT(DebugSimpleTemplateNames, llvm::codegenoptions::DebugTemplateNamesKind, 2, llvm::codegenoptions::DebugTemplateNamesKind::Full) ///< Whether to emit template parameters
-                                           ///< in the textual names of template
-					   ///< specializations.
-					   ///< Implies DebugFwdTemplateNames to
-					   ///< allow decorated names to be
-					   ///< reconstructed when needed.
 CODEGENOPT(EmitLLVMUseLists, 1, 0) ///< Control whether to serialize use-lists.
 
 CODEGENOPT(WholeProgramVTables, 1, 0) ///< Whether to apply whole-program
@@ -395,30 +346,6 @@ VALUE_CODEGENOPT(SmallDataLimit, 32, 0)
 /// The lower bound for a buffer to be considered for stack protection.
 VALUE_CODEGENOPT(SSPBufferSize, 32, 0)
 
-/// The kind of generated debug info.
-ENUM_CODEGENOPT(DebugInfo, llvm::codegenoptions::DebugInfoKind, 4, llvm::codegenoptions::NoDebugInfo)
-
-/// Whether to generate macro debug info.
-CODEGENOPT(MacroDebugInfo, 1, 0)
-
-/// Tune the debug info for this debugger.
-ENUM_CODEGENOPT(DebuggerTuning, llvm::DebuggerKind, 3,
-                llvm::DebuggerKind::Default)
-
-/// Dwarf version. Version zero indicates to LLVM that no DWARF should be
-/// emitted.
-VALUE_CODEGENOPT(DwarfVersion, 3, 0)
-
-/// Whether we should emit CodeView debug information. It's possible to emit
-/// CodeView and DWARF into the same object.
-CODEGENOPT(EmitCodeView, 1, 0)
-
-/// Whether to emit the .debug$H section containing hashes of CodeView types.
-CODEGENOPT(CodeViewGHash, 1, 0)
-
-/// Whether to emit the compiler path and command line into the CodeView debug information.
-CODEGENOPT(CodeViewCommandLine, 1, 0)
-
 /// The kind of inlining to perform.
 ENUM_CODEGENOPT(Inlining, InliningMethod, 2, NormalInlining)
 
@@ -451,26 +378,14 @@ CODEGENOPT(DirectAccessExternalData, 1, 0)
 /// paths that reach the end of a function without executing a required return.
 CODEGENOPT(StrictReturn, 1, 1)
 
-/// Whether emit extra debug info for sample pgo profile collection.
-CODEGENOPT(DebugInfoForProfiling, 1, 0)
-
 /// Whether emit pseudo probes for sample pgo profile collection.
 CODEGENOPT(PseudoProbeForProfiling, 1, 0)
 
 /// Whether 3-component vector type is preserved.
 CODEGENOPT(PreserveVec3Type, 1, 0)
 
-/// Whether to emit .debug_gnu_pubnames section instead of .debug_pubnames.
-CODEGENOPT(DebugNameTable, 2, 0)
-
-/// Whether to use DWARF base address specifiers in .debug_ranges.
-CODEGENOPT(DebugRangesBaseAddress, 1, 0)
-
 CODEGENOPT(NoPLT, 1, 0)
 
-/// Whether to embed source in DWARF debug line section.
-CODEGENOPT(EmbedSource, 1, 0)
-
 /// Whether to emit all vtables
 CODEGENOPT(ForceEmitVTables, 1, 0)
 
@@ -515,6 +430,9 @@ ENUM_CODEGENOPT(ZeroCallUsedRegs, llvm::ZeroCallUsedRegs::ZeroCallUsedRegsKind,
 /// non-deleting destructors. (No effect on Microsoft ABI.)
 CODEGENOPT(CtorDtorReturnThis, 1, 0)
 
+/// FIXME: Make DebugOptions its own top-level .def file.
+#include "DebugOptions.def"
+
 #undef CODEGENOPT
 #undef ENUM_CODEGENOPT
 #undef VALUE_CODEGENOPT

clang/include/clang/Basic/CodeGenOptions.h

@@ -530,6 +530,10 @@ class CodeGenOptions : public CodeGenOptionsBase {
     return SanitizeBinaryMetadataCovered || SanitizeBinaryMetadataAtomics ||
            SanitizeBinaryMetadataUAR;
   }
+
+  /// Reset all of the options that are not considered when building a
+  /// module.
+  void resetNonModularOptions(StringRef ModuleFormat);
 };
 
 }  // end namespace clang

clang/include/clang/Basic/DebugOptions.def

@@ -0,0 +1,146 @@
+//===--- DebugOptions.def - Debug option database ----------------- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines debug-specific codegen options. Users of this file
+// must define the CODEGENOPT macro to make use of this information.
+// Optionally, the user may also define DEBUGOPT (for flags), ENUM_DEBUGOPT (for
+// options that have enumeration type), and VALUE_DEBUGOPT (is a debug option
+// that describes a value rather than a flag).
+//
+// BENIGN_ variants of the macros are used to describe options that do not
+// affect the generated PCM.
+//
+//===----------------------------------------------------------------------===//
+#ifndef DEBUGOPT
+#define DEBUGOPT(Name, Bits, Default) \
+CODEGENOPT(Name, Bits, Default)
+#endif
+
+#ifndef VALUE_DEBUGOPT
+#  define VALUE_DEBUGOPT(Name, Bits, Default) \
+VALUE_CODEGENOPT(Name, Bits, Default)
+#endif
+
+#ifndef ENUM_DEBUGOPT
+#  define ENUM_DEBUGOPT(Name, Type, Bits, Default) \
+ENUM_CODEGENOPT(Name, Type, Bits, Default)
+#endif
+
+#ifndef BENIGN_DEBUGOPT
+#define BENIGN_DEBUGOPT(Name, Bits, Default) \
+DEBUGOPT(Name, Bits, Default)
+#endif
+
+#ifndef BENIGN_VALUE_DEBUGOPT
+#  define BENIGN_VALUE_DEBUGOPT(Name, Bits, Default) \
+VALUE_DEBUGOPT(Name, Bits, Default)
+#endif
+
+#ifndef BENIGN_ENUM_DEBUGOPT
+#  define BENIGN_ENUM_DEBUGOPT(Name, Type, Bits, Default) \
+ENUM_DEBUGOPT(Name, Type, Bits, Default)
+#endif
+
+BENIGN_ENUM_DEBUGOPT(CompressDebugSections, llvm::DebugCompressionType, 2,
+                     llvm::DebugCompressionType::None)
+DEBUGOPT(Dwarf64, 1, 0) ///< -gdwarf64.
+BENIGN_DEBUGOPT(EnableDIPreservationVerify, 1, 0) ///< Enable di preservation
+                                                  ///< verify each (it means
+                                                  ///< check the original debug
+                                                  ///< info metadata
+                                                  ///< preservation).
+BENIGN_DEBUGOPT(ForceDwarfFrameSection , 1, 0) ///< Set when -fforce-dwarf-frame
+                                               ///< is enabled.
+
+///< Set when -femit-dwarf-unwind is passed.
+BENIGN_ENUM_DEBUGOPT(EmitDwarfUnwind, llvm::EmitDwarfUnwindType, 2,
+                     llvm::EmitDwarfUnwindType::Default)
+
+BENIGN_DEBUGOPT(NoDwarfDirectoryAsm , 1, 0) ///< Set when -fno-dwarf-directory-asm
+                                            ///< is enabled.
+
+BENIGN_DEBUGOPT(NoInlineLineTables, 1, 0) ///< Whether debug info should contain
+                                          ///< inline line tables.
+
+DEBUGOPT(DebugStrictDwarf, 1, 1) ///< Whether or not to use strict DWARF info.
+
+/// Control the Assignment Tracking debug info feature.
+BENIGN_ENUM_DEBUGOPT(AssignmentTrackingMode, AssignmentTrackingOpts, 2,
+                     AssignmentTrackingOpts::Disabled)
+
+DEBUGOPT(DebugColumnInfo, 1, 0) ///< Whether or not to use column information
+                                ///< in debug info.
+
+DEBUGOPT(DebugTypeExtRefs, 1, 0) ///< Whether or not debug info should contain
+                                 ///< external references to a PCH or module.
+
+DEBUGOPT(DebugExplicitImport, 1, 0)  ///< Whether or not debug info should
+                                     ///< contain explicit imports for
+                                     ///< anonymous namespaces
+
+/// Set debug info source file hashing algorithm.
+ENUM_DEBUGOPT(DebugSrcHash, DebugSrcHashKind, 2, DSH_MD5)
+
+DEBUGOPT(SplitDwarfInlining, 1, 1) ///< Whether to include inlining info in the
+                                   ///< skeleton CU to allow for symbolication
+                                   ///< of inline stack frames without .dwo files.
+DEBUGOPT(DebugFwdTemplateParams, 1, 0) ///< Whether to emit complete
+                                       ///< template parameter descriptions in
+                                       ///< forward declarations (versus just
+                                       ///< including them in the name).
+ENUM_DEBUGOPT(DebugSimpleTemplateNames,
+              llvm::codegenoptions::DebugTemplateNamesKind, 2,
+              llvm::codegenoptions::DebugTemplateNamesKind::Full)
+              ///< Whether to emit template parameters in the textual names of
+              ///< template specializations.
+			  ///< Implies DebugFwdTemplateNames to allow decorated names to be
+			  ///< reconstructed when needed.
+
+/// The kind of generated debug info.
+ENUM_DEBUGOPT(DebugInfo, llvm::codegenoptions::DebugInfoKind, 4,
+              llvm::codegenoptions::NoDebugInfo)
+
+/// Whether to generate macro debug info.
+DEBUGOPT(MacroDebugInfo, 1, 0)
+
+/// Tune the debug info for this debugger.
+ENUM_DEBUGOPT(DebuggerTuning, llvm::DebuggerKind, 3,
+                llvm::DebuggerKind::Default)
+
+/// Dwarf version. Version zero indicates to LLVM that no DWARF should be
+/// emitted.
+VALUE_DEBUGOPT(DwarfVersion, 3, 0)
+
+/// Whether we should emit CodeView debug information. It's possible to emit
+/// CodeView and DWARF into the same object.
+DEBUGOPT(EmitCodeView, 1, 0)
+
+/// Whether to emit the .debug$H section containing hashes of CodeView types.
+DEBUGOPT(CodeViewGHash, 1, 0)
+
+/// Whether to emit the compiler path and command line into the CodeView debug information.
+DEBUGOPT(CodeViewCommandLine, 1, 0)
+
+/// Whether emit extra debug info for sample pgo profile collection.
+DEBUGOPT(DebugInfoForProfiling, 1, 0)
+
+/// Whether to emit .debug_gnu_pubnames section instead of .debug_pubnames.
+DEBUGOPT(DebugNameTable, 2, 0)
+
+/// Whether to use DWARF base address specifiers in .debug_ranges.
+DEBUGOPT(DebugRangesBaseAddress, 1, 0)
+
+/// Whether to embed source in DWARF debug line section.
+DEBUGOPT(EmbedSource, 1, 0)
+
+#undef DEBUGOPT
+#undef ENUM_DEBUGOPT
+#undef VALUE_DEBUGOPT
+#undef BENIGN_DEBUGOPT
+#undef BENIGN_ENUM_DEBUGOPT
+#undef BENIGN_VALUE_DEBUGOPT

clang/include/module.modulemap

@@ -68,6 +68,7 @@ module Clang_Basic {
   textual header "clang/Basic/BuiltinsX86_64.def"
   textual header "clang/Basic/BuiltinsXCore.def"
   textual header "clang/Basic/CodeGenOptions.def"
+  textual header "clang/Basic/DebugOptions.def"
   textual header "clang/Basic/DiagnosticOptions.def"
   textual header "clang/Basic/Features.def"
   textual header "clang/Basic/FPOptions.def"

clang/lib/Basic/CodeGenOptions.cpp

@@ -20,4 +20,39 @@ CodeGenOptions::CodeGenOptions() {
   memcpy(CoverageVersion, "408*", 4);
 }
 
+void CodeGenOptions::resetNonModularOptions(StringRef ModuleFormat) {
+  // First reset all CodeGen options only. The Debug options are handled later.
+#define DEBUGOPT(Name, Bits, Default)
+#define VALUE_DEBUGOPT(Name, Bits, Default)
+#define ENUM_DEBUGOPT(Name, Type, Bits, Default)
+#define CODEGENOPT(Name, Bits, Default) Name = Default;
+#define ENUM_CODEGENOPT(Name, Type, Bits, Default) set##Name(Default);
+#include "clang/Basic/CodeGenOptions.def"
+
+  // Next reset all debug options that can always be reset, because they never
+  // affect the PCM.
+#define DEBUGOPT(Name, Bits, Default)
+#define VALUE_DEBUGOPT(Name, Bits, Default)
+#define ENUM_DEBUGOPT(Name, Type, Bits, Default)
+#define BENIGN_DEBUGOPT(Name, Bits, Default) Name = Default;
+#define BENIGN_VALUE_DEBUGOPT(Name, Bits, Default) Name = Default;
+#define BENIGN_ENUM_DEBUGOPT(Name, Type, Bits, Default) set##Name(Default);
+#include "clang/Basic/DebugOptions.def"
+
+  // Conditionally reset debug options that only matter when the debug info is
+  // emitted into the PCM (-gmodules).
+  if (ModuleFormat == "raw") {
+#define DEBUGOPT(Name, Bits, Default) Name = Default;
+#define VALUE_DEBUGOPT(Name, Bits, Default) Name = Default;
+#define ENUM_DEBUGOPT(Name, Type, Bits, Default) set##Name(Default);
+#define BENIGN_DEBUGOPT(Name, Bits, Default)
+#define BENIGN_VALUE_DEBUGOPT(Name, Bits, Default)
+#define BENIGN_ENUM_DEBUGOPT(Name, Type, Bits, Default)
+#include "clang/Basic/DebugOptions.def"
+  }
+
+  RelocationModel = llvm::Reloc::PIC_;
+  memcpy(CoverageVersion, "408*", 4);
+}
+
 }  // end namespace clang

clang/lib/Frontend/CompilerInvocation.cpp

@@ -4787,6 +4787,18 @@ std::string CompilerInvocation::getModuleHash() const {
   if (getCodeGenOpts().DebugTypeExtRefs)
     HBuilder.addRange(getCodeGenOpts().DebugPrefixMap);
 
+  // Extend the signature with the affecting debug options.
+  if (getHeaderSearchOpts().ModuleFormat == "obj") {
+#define DEBUGOPT(Name, Bits, Default) HBuilder.add(CodeGenOpts->Name);
+#define VALUE_DEBUGOPT(Name, Bits, Default) HBuilder.add(CodeGenOpts->Name);
+#define ENUM_DEBUGOPT(Name, Type, Bits, Default)                               \
+  HBuilder.add(static_cast<unsigned>(CodeGenOpts->get##Name()));
+#define BENIGN_DEBUGOPT(Name, Bits, Default)
+#define BENIGN_VALUE_DEBUGOPT(Name, Bits, Default)
+#define BENIGN_ENUM_DEBUGOPT(Name, Type, Bits, Default)
+#include "clang/Basic/DebugOptions.def"
+  }
+
   // Extend the signature with the enabled sanitizers, if at least one is
   // enabled. Sanitizers which cannot affect AST generation aren't hashed.
   SanitizerSet SanHash = getLangOpts().Sanitize;
@@ -4833,6 +4845,7 @@ std::vector<std::string> CompilerInvocationBase::getCC1CommandLine() const {
 void CompilerInvocation::resetNonModularOptions() {
   getLangOpts().resetNonModularOptions();
   getPreprocessorOpts().resetNonModularOptions();
+  getCodeGenOpts().resetNonModularOptions(getHeaderSearchOpts().ModuleFormat);
 }
 
 void CompilerInvocation::clearImplicitModuleBuildOptions() {