diff options
Diffstat (limited to 'compiler/rustc_llvm/llvm-wrapper')
| -rw-r--r-- | compiler/rustc_llvm/llvm-wrapper/.editorconfig | 6 | ||||
| -rw-r--r-- | compiler/rustc_llvm/llvm-wrapper/ArchiveWrapper.cpp | 212 | ||||
| -rw-r--r-- | compiler/rustc_llvm/llvm-wrapper/CoverageMappingWrapper.cpp | 292 | ||||
| -rw-r--r-- | compiler/rustc_llvm/llvm-wrapper/LLVMWrapper.h | 123 | ||||
| -rw-r--r-- | compiler/rustc_llvm/llvm-wrapper/Linker.cpp | 41 | ||||
| -rw-r--r-- | compiler/rustc_llvm/llvm-wrapper/PassWrapper.cpp | 1626 | ||||
| -rw-r--r-- | compiler/rustc_llvm/llvm-wrapper/README | 16 | ||||
| -rw-r--r-- | compiler/rustc_llvm/llvm-wrapper/RustWrapper.cpp | 2193 | ||||
| -rw-r--r-- | compiler/rustc_llvm/llvm-wrapper/SuppressLLVMWarnings.h | 17 | ||||
| -rw-r--r-- | compiler/rustc_llvm/llvm-wrapper/SymbolWrapper.cpp | 99 |
10 files changed, 4625 insertions, 0 deletions
diff --git a/compiler/rustc_llvm/llvm-wrapper/.editorconfig b/compiler/rustc_llvm/llvm-wrapper/.editorconfig new file mode 100644 index 00000000000..865cd45f708 --- /dev/null +++ b/compiler/rustc_llvm/llvm-wrapper/.editorconfig @@ -0,0 +1,6 @@ +[*.{h,cpp}] +end_of_line = lf +insert_final_newline = true +charset = utf-8 +indent_style = space +indent_size = 2 diff --git a/compiler/rustc_llvm/llvm-wrapper/ArchiveWrapper.cpp b/compiler/rustc_llvm/llvm-wrapper/ArchiveWrapper.cpp new file mode 100644 index 00000000000..a8c278741a7 --- /dev/null +++ b/compiler/rustc_llvm/llvm-wrapper/ArchiveWrapper.cpp @@ -0,0 +1,212 @@ +#include "LLVMWrapper.h" + +#include "llvm/Object/Archive.h" +#include "llvm/Object/ArchiveWriter.h" +#include "llvm/Support/Path.h" + +using namespace llvm; +using namespace llvm::object; + +struct RustArchiveMember { + const char *Filename; + const char *Name; + Archive::Child Child; + + RustArchiveMember() + : Filename(nullptr), Name(nullptr), Child(nullptr, nullptr, nullptr) {} + ~RustArchiveMember() {} +}; + +struct RustArchiveIterator { + bool First; + Archive::child_iterator Cur; + Archive::child_iterator End; + std::unique_ptr<Error> Err; + + RustArchiveIterator(Archive::child_iterator Cur, Archive::child_iterator End, + std::unique_ptr<Error> Err) + : First(true), Cur(Cur), End(End), Err(std::move(Err)) {} +}; + +enum class LLVMRustArchiveKind { + GNU, + BSD, + DARWIN, + COFF, + AIX_BIG, +}; + +static Archive::Kind fromRust(LLVMRustArchiveKind Kind) { + switch (Kind) { + case LLVMRustArchiveKind::GNU: + return Archive::K_GNU; + case LLVMRustArchiveKind::BSD: + return Archive::K_BSD; + case LLVMRustArchiveKind::DARWIN: + return Archive::K_DARWIN; + case LLVMRustArchiveKind::COFF: + return Archive::K_COFF; + case LLVMRustArchiveKind::AIX_BIG: + return Archive::K_AIXBIG; + default: + report_fatal_error("Bad ArchiveKind."); + } +} + +typedef OwningBinary<Archive> *LLVMRustArchiveRef; +typedef RustArchiveMember *LLVMRustArchiveMemberRef; +typedef Archive::Child *LLVMRustArchiveChildRef; +typedef Archive::Child const *LLVMRustArchiveChildConstRef; +typedef RustArchiveIterator *LLVMRustArchiveIteratorRef; + +extern "C" LLVMRustArchiveRef LLVMRustOpenArchive(char *Path) { + ErrorOr<std::unique_ptr<MemoryBuffer>> BufOr = MemoryBuffer::getFile( + Path, /*IsText*/ false, /*RequiresNullTerminator=*/false); + if (!BufOr) { + LLVMRustSetLastError(BufOr.getError().message().c_str()); + return nullptr; + } + + Expected<std::unique_ptr<Archive>> ArchiveOr = + Archive::create(BufOr.get()->getMemBufferRef()); + + if (!ArchiveOr) { + LLVMRustSetLastError(toString(ArchiveOr.takeError()).c_str()); + return nullptr; + } + + OwningBinary<Archive> *Ret = new OwningBinary<Archive>( + std::move(ArchiveOr.get()), std::move(BufOr.get())); + + return Ret; +} + +extern "C" void LLVMRustDestroyArchive(LLVMRustArchiveRef RustArchive) { + delete RustArchive; +} + +extern "C" LLVMRustArchiveIteratorRef +LLVMRustArchiveIteratorNew(LLVMRustArchiveRef RustArchive) { + Archive *Archive = RustArchive->getBinary(); + std::unique_ptr<Error> Err = std::make_unique<Error>(Error::success()); + auto Cur = Archive->child_begin(*Err); + if (*Err) { + LLVMRustSetLastError(toString(std::move(*Err)).c_str()); + return nullptr; + } + auto End = Archive->child_end(); + return new RustArchiveIterator(Cur, End, std::move(Err)); +} + +extern "C" LLVMRustArchiveChildConstRef +LLVMRustArchiveIteratorNext(LLVMRustArchiveIteratorRef RAI) { + if (RAI->Cur == RAI->End) + return nullptr; + + // Advancing the iterator validates the next child, and this can + // uncover an error. LLVM requires that we check all Errors, + // so we only advance the iterator if we actually need to fetch + // the next child. + // This means we must not advance the iterator in the *first* call, + // but instead advance it *before* fetching the child in all later calls. + if (!RAI->First) { + ++RAI->Cur; + if (*RAI->Err) { + LLVMRustSetLastError(toString(std::move(*RAI->Err)).c_str()); + return nullptr; + } + } else { + RAI->First = false; + } + + if (RAI->Cur == RAI->End) + return nullptr; + + const Archive::Child &Child = *RAI->Cur.operator->(); + Archive::Child *Ret = new Archive::Child(Child); + + return Ret; +} + +extern "C" void LLVMRustArchiveChildFree(LLVMRustArchiveChildRef Child) { + delete Child; +} + +extern "C" void LLVMRustArchiveIteratorFree(LLVMRustArchiveIteratorRef RAI) { + delete RAI; +} + +extern "C" const char * +LLVMRustArchiveChildName(LLVMRustArchiveChildConstRef Child, size_t *Size) { + Expected<StringRef> NameOrErr = Child->getName(); + if (!NameOrErr) { + // rustc_codegen_llvm currently doesn't use this error string, but it might + // be useful in the future, and in the meantime this tells LLVM that the + // error was not ignored and that it shouldn't abort the process. + LLVMRustSetLastError(toString(NameOrErr.takeError()).c_str()); + return nullptr; + } + StringRef Name = NameOrErr.get(); + *Size = Name.size(); + return Name.data(); +} + +extern "C" LLVMRustArchiveMemberRef +LLVMRustArchiveMemberNew(char *Filename, char *Name, + LLVMRustArchiveChildRef Child) { + RustArchiveMember *Member = new RustArchiveMember; + Member->Filename = Filename; + Member->Name = Name; + if (Child) + Member->Child = *Child; + return Member; +} + +extern "C" void LLVMRustArchiveMemberFree(LLVMRustArchiveMemberRef Member) { + delete Member; +} + +extern "C" LLVMRustResult LLVMRustWriteArchive( + char *Dst, size_t NumMembers, const LLVMRustArchiveMemberRef *NewMembers, + bool WriteSymbtab, LLVMRustArchiveKind RustKind, bool isEC) { + + std::vector<NewArchiveMember> Members; + auto Kind = fromRust(RustKind); + + for (size_t I = 0; I < NumMembers; I++) { + auto Member = NewMembers[I]; + assert(Member->Name); + if (Member->Filename) { + Expected<NewArchiveMember> MOrErr = + NewArchiveMember::getFile(Member->Filename, true); + if (!MOrErr) { + LLVMRustSetLastError(toString(MOrErr.takeError()).c_str()); + return LLVMRustResult::Failure; + } + MOrErr->MemberName = sys::path::filename(MOrErr->MemberName); + Members.push_back(std::move(*MOrErr)); + } else { + Expected<NewArchiveMember> MOrErr = + NewArchiveMember::getOldMember(Member->Child, true); + if (!MOrErr) { + LLVMRustSetLastError(toString(MOrErr.takeError()).c_str()); + return LLVMRustResult::Failure; + } + Members.push_back(std::move(*MOrErr)); + } + } + +#if LLVM_VERSION_LT(18, 0) + auto Result = writeArchive(Dst, Members, WriteSymbtab, Kind, true, false); +#else + auto SymtabMode = WriteSymbtab ? SymtabWritingMode::NormalSymtab + : SymtabWritingMode::NoSymtab; + auto Result = + writeArchive(Dst, Members, SymtabMode, Kind, true, false, nullptr, isEC); +#endif + if (!Result) + return LLVMRustResult::Success; + LLVMRustSetLastError(toString(std::move(Result)).c_str()); + + return LLVMRustResult::Failure; +} diff --git a/compiler/rustc_llvm/llvm-wrapper/CoverageMappingWrapper.cpp b/compiler/rustc_llvm/llvm-wrapper/CoverageMappingWrapper.cpp new file mode 100644 index 00000000000..e842f47f48c --- /dev/null +++ b/compiler/rustc_llvm/llvm-wrapper/CoverageMappingWrapper.cpp @@ -0,0 +1,292 @@ +#include "LLVMWrapper.h" +#include "llvm/ADT/ArrayRef.h" +#include "llvm/ProfileData/Coverage/CoverageMapping.h" +#include "llvm/ProfileData/Coverage/CoverageMappingWriter.h" +#include "llvm/ProfileData/InstrProf.h" + +using namespace llvm; + +// FFI equivalent of enum `llvm::coverage::Counter::CounterKind` +// https://github.com/rust-lang/llvm-project/blob/ea6fa9c2/llvm/include/llvm/ProfileData/Coverage/CoverageMapping.h#L97-L99 +enum class LLVMRustCounterKind { + Zero = 0, + CounterValueReference = 1, + Expression = 2, +}; + +// FFI equivalent of struct `llvm::coverage::Counter` +// https://github.com/rust-lang/llvm-project/blob/ea6fa9c2/llvm/include/llvm/ProfileData/Coverage/CoverageMapping.h#L94-L149 +struct LLVMRustCounter { + LLVMRustCounterKind CounterKind; + uint32_t ID; +}; + +static coverage::Counter fromRust(LLVMRustCounter Counter) { + switch (Counter.CounterKind) { + case LLVMRustCounterKind::Zero: + return coverage::Counter::getZero(); + case LLVMRustCounterKind::CounterValueReference: + return coverage::Counter::getCounter(Counter.ID); + case LLVMRustCounterKind::Expression: + return coverage::Counter::getExpression(Counter.ID); + } + report_fatal_error("Bad LLVMRustCounterKind!"); +} + +// FFI equivalent of enum `llvm::coverage::CounterMappingRegion::RegionKind` +// https://github.com/rust-lang/llvm-project/blob/ea6fa9c2/llvm/include/llvm/ProfileData/Coverage/CoverageMapping.h#L213-L234 +enum class LLVMRustCounterMappingRegionKind { + CodeRegion = 0, + ExpansionRegion = 1, + SkippedRegion = 2, + GapRegion = 3, + BranchRegion = 4, + MCDCDecisionRegion = 5, + MCDCBranchRegion = 6 +}; + +static coverage::CounterMappingRegion::RegionKind +fromRust(LLVMRustCounterMappingRegionKind Kind) { + switch (Kind) { + case LLVMRustCounterMappingRegionKind::CodeRegion: + return coverage::CounterMappingRegion::CodeRegion; + case LLVMRustCounterMappingRegionKind::ExpansionRegion: + return coverage::CounterMappingRegion::ExpansionRegion; + case LLVMRustCounterMappingRegionKind::SkippedRegion: + return coverage::CounterMappingRegion::SkippedRegion; + case LLVMRustCounterMappingRegionKind::GapRegion: + return coverage::CounterMappingRegion::GapRegion; + case LLVMRustCounterMappingRegionKind::BranchRegion: + return coverage::CounterMappingRegion::BranchRegion; +#if LLVM_VERSION_GE(18, 0) + case LLVMRustCounterMappingRegionKind::MCDCDecisionRegion: + return coverage::CounterMappingRegion::MCDCDecisionRegion; + case LLVMRustCounterMappingRegionKind::MCDCBranchRegion: + return coverage::CounterMappingRegion::MCDCBranchRegion; +#else + case LLVMRustCounterMappingRegionKind::MCDCDecisionRegion: + break; + case LLVMRustCounterMappingRegionKind::MCDCBranchRegion: + break; +#endif + } + report_fatal_error("Bad LLVMRustCounterMappingRegionKind!"); +} + +enum LLVMRustMCDCParametersTag { + None = 0, + Decision = 1, + Branch = 2, +}; + +struct LLVMRustMCDCDecisionParameters { + uint32_t BitmapIdx; + uint16_t NumConditions; +}; + +struct LLVMRustMCDCBranchParameters { + int16_t ConditionID; + int16_t ConditionIDs[2]; +}; + +struct LLVMRustMCDCParameters { + LLVMRustMCDCParametersTag Tag; + LLVMRustMCDCDecisionParameters DecisionParameters; + LLVMRustMCDCBranchParameters BranchParameters; +}; + +// LLVM representations for `MCDCParameters` evolved from LLVM 18 to 19. +// Look at representations in 18 +// https://github.com/rust-lang/llvm-project/blob/66a2881a/llvm/include/llvm/ProfileData/Coverage/CoverageMapping.h#L253-L263 +// and representations in 19 +// https://github.com/llvm/llvm-project/blob/843cc474faefad1d639f4c44c1cf3ad7dbda76c8/llvm/include/llvm/ProfileData/Coverage/MCDCTypes.h +#if LLVM_VERSION_GE(18, 0) && LLVM_VERSION_LT(19, 0) +static coverage::CounterMappingRegion::MCDCParameters +fromRust(LLVMRustMCDCParameters Params) { + auto parameter = coverage::CounterMappingRegion::MCDCParameters{}; + switch (Params.Tag) { + case LLVMRustMCDCParametersTag::None: + return parameter; + case LLVMRustMCDCParametersTag::Decision: + parameter.BitmapIdx = + static_cast<unsigned>(Params.DecisionParameters.BitmapIdx), + parameter.NumConditions = + static_cast<unsigned>(Params.DecisionParameters.NumConditions); + return parameter; + case LLVMRustMCDCParametersTag::Branch: + parameter.ID = static_cast<coverage::CounterMappingRegion::MCDCConditionID>( + Params.BranchParameters.ConditionID), + parameter.FalseID = + static_cast<coverage::CounterMappingRegion::MCDCConditionID>( + Params.BranchParameters.ConditionIDs[0]), + parameter.TrueID = + static_cast<coverage::CounterMappingRegion::MCDCConditionID>( + Params.BranchParameters.ConditionIDs[1]); + return parameter; + } + report_fatal_error("Bad LLVMRustMCDCParametersTag!"); +} +#elif LLVM_VERSION_GE(19, 0) +static coverage::mcdc::Parameters fromRust(LLVMRustMCDCParameters Params) { + switch (Params.Tag) { + case LLVMRustMCDCParametersTag::None: + return std::monostate(); + case LLVMRustMCDCParametersTag::Decision: + return coverage::mcdc::DecisionParameters( + Params.DecisionParameters.BitmapIdx, + Params.DecisionParameters.NumConditions); + case LLVMRustMCDCParametersTag::Branch: + return coverage::mcdc::BranchParameters( + static_cast<coverage::mcdc::ConditionID>( + Params.BranchParameters.ConditionID), + {static_cast<coverage::mcdc::ConditionID>( + Params.BranchParameters.ConditionIDs[0]), + static_cast<coverage::mcdc::ConditionID>( + Params.BranchParameters.ConditionIDs[1])}); + } + report_fatal_error("Bad LLVMRustMCDCParametersTag!"); +} +#endif + +// FFI equivalent of struct `llvm::coverage::CounterMappingRegion` +// https://github.com/rust-lang/llvm-project/blob/ea6fa9c2/llvm/include/llvm/ProfileData/Coverage/CoverageMapping.h#L211-L304 +struct LLVMRustCounterMappingRegion { + LLVMRustCounter Count; + LLVMRustCounter FalseCount; + LLVMRustMCDCParameters MCDCParameters; + uint32_t FileID; + uint32_t ExpandedFileID; + uint32_t LineStart; + uint32_t ColumnStart; + uint32_t LineEnd; + uint32_t ColumnEnd; + LLVMRustCounterMappingRegionKind Kind; +}; + +// FFI equivalent of enum `llvm::coverage::CounterExpression::ExprKind` +// https://github.com/rust-lang/llvm-project/blob/ea6fa9c2/llvm/include/llvm/ProfileData/Coverage/CoverageMapping.h#L154 +enum class LLVMRustCounterExprKind { + Subtract = 0, + Add = 1, +}; + +// FFI equivalent of struct `llvm::coverage::CounterExpression` +// https://github.com/rust-lang/llvm-project/blob/ea6fa9c2/llvm/include/llvm/ProfileData/Coverage/CoverageMapping.h#L151-L160 +struct LLVMRustCounterExpression { + LLVMRustCounterExprKind Kind; + LLVMRustCounter LHS; + LLVMRustCounter RHS; +}; + +static coverage::CounterExpression::ExprKind +fromRust(LLVMRustCounterExprKind Kind) { + switch (Kind) { + case LLVMRustCounterExprKind::Subtract: + return coverage::CounterExpression::Subtract; + case LLVMRustCounterExprKind::Add: + return coverage::CounterExpression::Add; + } + report_fatal_error("Bad LLVMRustCounterExprKind!"); +} + +extern "C" void LLVMRustCoverageWriteFilenamesSectionToBuffer( + const char *const Filenames[], size_t FilenamesLen, // String start pointers + const size_t *const Lengths, size_t LengthsLen, // Corresponding lengths + RustStringRef BufferOut) { + if (FilenamesLen != LengthsLen) { + report_fatal_error( + "Mismatched lengths in LLVMRustCoverageWriteFilenamesSectionToBuffer"); + } + + SmallVector<std::string, 32> FilenameRefs; + FilenameRefs.reserve(FilenamesLen); + for (size_t i = 0; i < FilenamesLen; i++) { + FilenameRefs.emplace_back(Filenames[i], Lengths[i]); + } + auto FilenamesWriter = coverage::CoverageFilenamesSectionWriter( + ArrayRef<std::string>(FilenameRefs)); + auto OS = RawRustStringOstream(BufferOut); + FilenamesWriter.write(OS); +} + +extern "C" void LLVMRustCoverageWriteMappingToBuffer( + const unsigned *VirtualFileMappingIDs, unsigned NumVirtualFileMappingIDs, + const LLVMRustCounterExpression *RustExpressions, unsigned NumExpressions, + const LLVMRustCounterMappingRegion *RustMappingRegions, + unsigned NumMappingRegions, RustStringRef BufferOut) { + // Convert from FFI representation to LLVM representation. + SmallVector<coverage::CounterMappingRegion, 0> MappingRegions; + MappingRegions.reserve(NumMappingRegions); + for (const auto &Region : ArrayRef<LLVMRustCounterMappingRegion>( + RustMappingRegions, NumMappingRegions)) { + MappingRegions.emplace_back( + fromRust(Region.Count), fromRust(Region.FalseCount), +#if LLVM_VERSION_GE(18, 0) && LLVM_VERSION_LT(19, 0) + // LLVM 19 may move this argument to last. + fromRust(Region.MCDCParameters), +#endif + Region.FileID, Region.ExpandedFileID, // File IDs, then region info. + Region.LineStart, Region.ColumnStart, Region.LineEnd, Region.ColumnEnd, + fromRust(Region.Kind)); + } + + std::vector<coverage::CounterExpression> Expressions; + Expressions.reserve(NumExpressions); + for (const auto &Expression : + ArrayRef<LLVMRustCounterExpression>(RustExpressions, NumExpressions)) { + Expressions.emplace_back(fromRust(Expression.Kind), + fromRust(Expression.LHS), + fromRust(Expression.RHS)); + } + + auto CoverageMappingWriter = coverage::CoverageMappingWriter( + ArrayRef<unsigned>(VirtualFileMappingIDs, NumVirtualFileMappingIDs), + Expressions, MappingRegions); + auto OS = RawRustStringOstream(BufferOut); + CoverageMappingWriter.write(OS); +} + +extern "C" LLVMValueRef +LLVMRustCoverageCreatePGOFuncNameVar(LLVMValueRef F, const char *FuncName, + size_t FuncNameLen) { + auto FuncNameRef = StringRef(FuncName, FuncNameLen); + return wrap(createPGOFuncNameVar(*cast<Function>(unwrap(F)), FuncNameRef)); +} + +extern "C" uint64_t LLVMRustCoverageHashByteArray(const char *Bytes, + size_t NumBytes) { + auto StrRef = StringRef(Bytes, NumBytes); + return IndexedInstrProf::ComputeHash(StrRef); +} + +static void WriteSectionNameToString(LLVMModuleRef M, InstrProfSectKind SK, + RustStringRef Str) { + auto TargetTriple = Triple(unwrap(M)->getTargetTriple()); + auto name = getInstrProfSectionName(SK, TargetTriple.getObjectFormat()); + auto OS = RawRustStringOstream(Str); + OS << name; +} + +extern "C" void LLVMRustCoverageWriteMapSectionNameToString(LLVMModuleRef M, + RustStringRef Str) { + WriteSectionNameToString(M, IPSK_covmap, Str); +} + +extern "C" void +LLVMRustCoverageWriteFuncSectionNameToString(LLVMModuleRef M, + RustStringRef Str) { + WriteSectionNameToString(M, IPSK_covfun, Str); +} + +extern "C" void LLVMRustCoverageWriteMappingVarNameToString(RustStringRef Str) { + auto name = getCoverageMappingVarName(); + auto OS = RawRustStringOstream(Str); + OS << name; +} + +extern "C" uint32_t LLVMRustCoverageMappingVersion() { + // This should always be `CurrentVersion`, because that's the version LLVM + // will use when encoding the data we give it. If for some reason we ever + // want to override the version number we _emit_, do it on the Rust side. + return coverage::CovMapVersion::CurrentVersion; +} diff --git a/compiler/rustc_llvm/llvm-wrapper/LLVMWrapper.h b/compiler/rustc_llvm/llvm-wrapper/LLVMWrapper.h new file mode 100644 index 00000000000..a493abbbc7e --- /dev/null +++ b/compiler/rustc_llvm/llvm-wrapper/LLVMWrapper.h @@ -0,0 +1,123 @@ +#include "SuppressLLVMWarnings.h" + +#include "llvm-c/BitReader.h" +#include "llvm-c/Core.h" +#include "llvm-c/Object.h" +#include "llvm/ADT/ArrayRef.h" +#include "llvm/ADT/DenseSet.h" +#include "llvm/ADT/SmallVector.h" +#include "llvm/Analysis/Lint.h" +#include "llvm/Analysis/Passes.h" +#include "llvm/IR/IRBuilder.h" +#include "llvm/IR/InlineAsm.h" +#include "llvm/IR/LLVMContext.h" +#include "llvm/IR/Module.h" +#include "llvm/Support/CommandLine.h" +#include "llvm/Support/Debug.h" +#include "llvm/Support/DynamicLibrary.h" +#include "llvm/Support/FormattedStream.h" +#include "llvm/Support/JSON.h" +#include "llvm/Support/Memory.h" +#include "llvm/Support/SourceMgr.h" +#include "llvm/Support/TargetSelect.h" +#include "llvm/Support/Timer.h" +#include "llvm/Support/raw_ostream.h" +#include "llvm/Target/TargetMachine.h" +#include "llvm/Target/TargetOptions.h" +#include "llvm/Transforms/IPO.h" +#include "llvm/Transforms/Instrumentation.h" +#include "llvm/Transforms/Scalar.h" + +#define LLVM_VERSION_GE(major, minor) \ + (LLVM_VERSION_MAJOR > (major) || \ + LLVM_VERSION_MAJOR == (major) && LLVM_VERSION_MINOR >= (minor)) + +#define LLVM_VERSION_LT(major, minor) (!LLVM_VERSION_GE((major), (minor))) + +#include "llvm/IR/LegacyPassManager.h" + +#include "llvm/Bitcode/BitcodeReader.h" +#include "llvm/Bitcode/BitcodeWriter.h" + +#include "llvm/IR/DIBuilder.h" +#include "llvm/IR/DebugInfo.h" +#include "llvm/IR/IRPrintingPasses.h" +#include "llvm/Linker/Linker.h" + +#include "llvm/TargetParser/Triple.h" + +extern "C" void LLVMRustSetLastError(const char *); + +enum class LLVMRustResult { Success, Failure }; + +enum LLVMRustAttribute { + AlwaysInline = 0, + ByVal = 1, + Cold = 2, + InlineHint = 3, + MinSize = 4, + Naked = 5, + NoAlias = 6, + NoCapture = 7, + NoInline = 8, + NonNull = 9, + NoRedZone = 10, + NoReturn = 11, + NoUnwind = 12, + OptimizeForSize = 13, + ReadOnly = 14, + SExt = 15, + StructRet = 16, + UWTable = 17, + ZExt = 18, + InReg = 19, + SanitizeThread = 20, + SanitizeAddress = 21, + SanitizeMemory = 22, + NonLazyBind = 23, + OptimizeNone = 24, + ReadNone = 26, + SanitizeHWAddress = 28, + WillReturn = 29, + StackProtectReq = 30, + StackProtectStrong = 31, + StackProtect = 32, + NoUndef = 33, + SanitizeMemTag = 34, + NoCfCheck = 35, + ShadowCallStack = 36, + AllocSize = 37, + AllocatedPointer = 38, + AllocAlign = 39, + SanitizeSafeStack = 40, + FnRetThunkExtern = 41, + Writable = 42, + DeadOnUnwind = 43, +}; + +typedef struct OpaqueRustString *RustStringRef; +typedef struct LLVMOpaqueTwine *LLVMTwineRef; +typedef struct LLVMOpaqueSMDiagnostic *LLVMSMDiagnosticRef; + +extern "C" void LLVMRustStringWriteImpl(RustStringRef Str, const char *Ptr, + size_t Size); + +class RawRustStringOstream : public llvm::raw_ostream { + RustStringRef Str; + uint64_t Pos; + + void write_impl(const char *Ptr, size_t Size) override { + LLVMRustStringWriteImpl(Str, Ptr, Size); + Pos += Size; + } + + uint64_t current_pos() const override { return Pos; } + +public: + explicit RawRustStringOstream(RustStringRef Str) : Str(Str), Pos(0) {} + + ~RawRustStringOstream() { + // LLVM requires this. + flush(); + } +}; diff --git a/compiler/rustc_llvm/llvm-wrapper/Linker.cpp b/compiler/rustc_llvm/llvm-wrapper/Linker.cpp new file mode 100644 index 00000000000..f43128ed550 --- /dev/null +++ b/compiler/rustc_llvm/llvm-wrapper/Linker.cpp @@ -0,0 +1,41 @@ +#include "llvm/Linker/Linker.h" +#include "SuppressLLVMWarnings.h" + +#include "LLVMWrapper.h" + +using namespace llvm; + +struct RustLinker { + Linker L; + LLVMContext &Ctx; + + RustLinker(Module &M) : L(M), Ctx(M.getContext()) {} +}; + +extern "C" RustLinker *LLVMRustLinkerNew(LLVMModuleRef DstRef) { + Module *Dst = unwrap(DstRef); + + return new RustLinker(*Dst); +} + +extern "C" void LLVMRustLinkerFree(RustLinker *L) { delete L; } + +extern "C" bool LLVMRustLinkerAdd(RustLinker *L, char *BC, size_t Len) { + std::unique_ptr<MemoryBuffer> Buf = + MemoryBuffer::getMemBufferCopy(StringRef(BC, Len)); + + Expected<std::unique_ptr<Module>> SrcOrError = + llvm::getLazyBitcodeModule(Buf->getMemBufferRef(), L->Ctx); + if (!SrcOrError) { + LLVMRustSetLastError(toString(SrcOrError.takeError()).c_str()); + return false; + } + + auto Src = std::move(*SrcOrError); + + if (L->L.linkInModule(std::move(Src))) { + LLVMRustSetLastError(""); + return false; + } + return true; +} diff --git a/compiler/rustc_llvm/llvm-wrapper/PassWrapper.cpp b/compiler/rustc_llvm/llvm-wrapper/PassWrapper.cpp new file mode 100644 index 00000000000..283c4fbbb7c --- /dev/null +++ b/compiler/rustc_llvm/llvm-wrapper/PassWrapper.cpp @@ -0,0 +1,1626 @@ +#include <stdio.h> + +#include <cstddef> +#include <iomanip> +#include <set> +#include <vector> + +#include "LLVMWrapper.h" + +#include "llvm/Analysis/AliasAnalysis.h" +#include "llvm/Analysis/TargetLibraryInfo.h" +#include "llvm/Analysis/TargetTransformInfo.h" +#include "llvm/Bitcode/BitcodeWriter.h" +#include "llvm/CodeGen/CommandFlags.h" +#include "llvm/CodeGen/TargetSubtargetInfo.h" +#include "llvm/IR/AssemblyAnnotationWriter.h" +#include "llvm/IR/AutoUpgrade.h" +#include "llvm/IR/IntrinsicInst.h" +#include "llvm/IR/Verifier.h" +#include "llvm/LTO/LTO.h" +#include "llvm/MC/TargetRegistry.h" +#include "llvm/Object/IRObjectFile.h" +#include "llvm/Object/ObjectFile.h" +#include "llvm/Passes/PassBuilder.h" +#include "llvm/Passes/PassPlugin.h" +#include "llvm/Passes/StandardInstrumentations.h" +#include "llvm/Support/CBindingWrapping.h" +#include "llvm/Support/FileSystem.h" +#include "llvm/Support/VirtualFileSystem.h" +#include "llvm/Target/TargetMachine.h" +#include "llvm/Transforms/IPO/AlwaysInliner.h" +#include "llvm/Transforms/IPO/FunctionImport.h" +#include "llvm/Transforms/IPO/Internalize.h" +#include "llvm/Transforms/IPO/LowerTypeTests.h" +#include "llvm/Transforms/IPO/ThinLTOBitcodeWriter.h" +#include "llvm/Transforms/Utils/AddDiscriminators.h" +#include "llvm/Transforms/Utils/FunctionImportUtils.h" +#if LLVM_VERSION_GE(18, 0) +#include "llvm/TargetParser/Host.h" +#endif +#include "llvm/Support/TimeProfiler.h" +#include "llvm/Transforms/Instrumentation.h" +#include "llvm/Transforms/Instrumentation/AddressSanitizer.h" +#include "llvm/Transforms/Instrumentation/DataFlowSanitizer.h" +#if LLVM_VERSION_GE(19, 0) +#include "llvm/Support/PGOOptions.h" +#endif +#include "llvm/Transforms/Instrumentation/GCOVProfiler.h" +#include "llvm/Transforms/Instrumentation/HWAddressSanitizer.h" +#include "llvm/Transforms/Instrumentation/InstrProfiling.h" +#include "llvm/Transforms/Instrumentation/MemorySanitizer.h" +#include "llvm/Transforms/Instrumentation/ThreadSanitizer.h" +#include "llvm/Transforms/Utils.h" +#include "llvm/Transforms/Utils/CanonicalizeAliases.h" +#include "llvm/Transforms/Utils/NameAnonGlobals.h" + +using namespace llvm; + +static codegen::RegisterCodeGenFlags CGF; + +typedef struct LLVMOpaquePass *LLVMPassRef; +typedef struct LLVMOpaqueTargetMachine *LLVMTargetMachineRef; + +DEFINE_STDCXX_CONVERSION_FUNCTIONS(Pass, LLVMPassRef) +DEFINE_STDCXX_CONVERSION_FUNCTIONS(TargetMachine, LLVMTargetMachineRef) + +extern "C" void LLVMRustTimeTraceProfilerInitialize() { + timeTraceProfilerInitialize( + /* TimeTraceGranularity */ 0, + /* ProcName */ "rustc"); +} + +extern "C" void LLVMRustTimeTraceProfilerFinishThread() { + timeTraceProfilerFinishThread(); +} + +extern "C" void LLVMRustTimeTraceProfilerFinish(const char *FileName) { + auto FN = StringRef(FileName); + std::error_code EC; + auto OS = raw_fd_ostream(FN, EC, sys::fs::CD_CreateAlways); + + timeTraceProfilerWrite(OS); + timeTraceProfilerCleanup(); +} + +#ifdef LLVM_COMPONENT_X86 +#define SUBTARGET_X86 SUBTARGET(X86) +#else +#define SUBTARGET_X86 +#endif + +#ifdef LLVM_COMPONENT_ARM +#define SUBTARGET_ARM SUBTARGET(ARM) +#else +#define SUBTARGET_ARM +#endif + +#ifdef LLVM_COMPONENT_AARCH64 +#define SUBTARGET_AARCH64 SUBTARGET(AArch64) +#else +#define SUBTARGET_AARCH64 +#endif + +#ifdef LLVM_COMPONENT_AVR +#define SUBTARGET_AVR SUBTARGET(AVR) +#else +#define SUBTARGET_AVR +#endif + +#ifdef LLVM_COMPONENT_M68k +#define SUBTARGET_M68K SUBTARGET(M68k) +#else +#define SUBTARGET_M68K +#endif + +#ifdef LLVM_COMPONENT_CSKY +#define SUBTARGET_CSKY SUBTARGET(CSKY) +#else +#define SUBTARGET_CSKY +#endif + +#ifdef LLVM_COMPONENT_MIPS +#define SUBTARGET_MIPS SUBTARGET(Mips) +#else +#define SUBTARGET_MIPS +#endif + +#ifdef LLVM_COMPONENT_POWERPC +#define SUBTARGET_PPC SUBTARGET(PPC) +#else +#define SUBTARGET_PPC +#endif + +#ifdef LLVM_COMPONENT_SYSTEMZ +#define SUBTARGET_SYSTEMZ SUBTARGET(SystemZ) +#else +#define SUBTARGET_SYSTEMZ +#endif + +#ifdef LLVM_COMPONENT_MSP430 +#define SUBTARGET_MSP430 SUBTARGET(MSP430) +#else +#define SUBTARGET_MSP430 +#endif + +#ifdef LLVM_COMPONENT_RISCV +#define SUBTARGET_RISCV SUBTARGET(RISCV) +#else +#define SUBTARGET_RISCV +#endif + +#ifdef LLVM_COMPONENT_SPARC +#define SUBTARGET_SPARC SUBTARGET(Sparc) +#else +#define SUBTARGET_SPARC +#endif + +#ifdef LLVM_COMPONENT_XTENSA +#define SUBTARGET_XTENSA SUBTARGET(XTENSA) +#else +#define SUBTARGET_XTENSA +#endif + +#ifdef LLVM_COMPONENT_HEXAGON +#define SUBTARGET_HEXAGON SUBTARGET(Hexagon) +#else +#define SUBTARGET_HEXAGON +#endif + +#ifdef LLVM_COMPONENT_LOONGARCH +#define SUBTARGET_LOONGARCH SUBTARGET(LoongArch) +#else +#define SUBTARGET_LOONGARCH +#endif + +#define GEN_SUBTARGETS \ + SUBTARGET_X86 \ + SUBTARGET_ARM \ + SUBTARGET_AARCH64 \ + SUBTARGET_AVR \ + SUBTARGET_M68K \ + SUBTARGET_CSKY \ + SUBTARGET_MIPS \ + SUBTARGET_PPC \ + SUBTARGET_SYSTEMZ \ + SUBTARGET_MSP430 \ + SUBTARGET_SPARC \ + SUBTARGET_HEXAGON \ + SUBTARGET_XTENSA \ + SUBTARGET_RISCV \ + SUBTARGET_LOONGARCH + +#define SUBTARGET(x) \ + namespace llvm { \ + extern const SubtargetFeatureKV x##FeatureKV[]; \ + extern const SubtargetFeatureKV x##SubTypeKV[]; \ + } + +GEN_SUBTARGETS +#undef SUBTARGET + +extern "C" bool LLVMRustHasFeature(LLVMTargetMachineRef TM, + const char *Feature) { + TargetMachine *Target = unwrap(TM); + const MCSubtargetInfo *MCInfo = Target->getMCSubtargetInfo(); + return MCInfo->checkFeatures(std::string("+") + Feature); +} + +enum class LLVMRustCodeModel { + Tiny, + Small, + Kernel, + Medium, + Large, + None, +}; + +static std::optional<CodeModel::Model> fromRust(LLVMRustCodeModel Model) { + switch (Model) { + case LLVMRustCodeModel::Tiny: + return CodeModel::Tiny; + case LLVMRustCodeModel::Small: + return CodeModel::Small; + case LLVMRustCodeModel::Kernel: + return CodeModel::Kernel; + case LLVMRustCodeModel::Medium: + return CodeModel::Medium; + case LLVMRustCodeModel::Large: + return CodeModel::Large; + case LLVMRustCodeModel::None: + return std::nullopt; + default: + report_fatal_error("Bad CodeModel."); + } +} + +enum class LLVMRustCodeGenOptLevel { + None, + Less, + Default, + Aggressive, +}; + +#if LLVM_VERSION_GE(18, 0) +using CodeGenOptLevelEnum = llvm::CodeGenOptLevel; +#else +using CodeGenOptLevelEnum = llvm::CodeGenOpt::Level; +#endif + +static CodeGenOptLevelEnum fromRust(LLVMRustCodeGenOptLevel Level) { + switch (Level) { + case LLVMRustCodeGenOptLevel::None: + return CodeGenOptLevelEnum::None; + case LLVMRustCodeGenOptLevel::Less: + return CodeGenOptLevelEnum::Less; + case LLVMRustCodeGenOptLevel::Default: + return CodeGenOptLevelEnum::Default; + case LLVMRustCodeGenOptLevel::Aggressive: + return CodeGenOptLevelEnum::Aggressive; + default: + report_fatal_error("Bad CodeGenOptLevel."); + } +} + +enum class LLVMRustPassBuilderOptLevel { + O0, + O1, + O2, + O3, + Os, + Oz, +}; + +static OptimizationLevel fromRust(LLVMRustPassBuilderOptLevel Level) { + switch (Level) { + case LLVMRustPassBuilderOptLevel::O0: + return OptimizationLevel::O0; + case LLVMRustPassBuilderOptLevel::O1: + return OptimizationLevel::O1; + case LLVMRustPassBuilderOptLevel::O2: + return OptimizationLevel::O2; + case LLVMRustPassBuilderOptLevel::O3: + return OptimizationLevel::O3; + case LLVMRustPassBuilderOptLevel::Os: + return OptimizationLevel::Os; + case LLVMRustPassBuilderOptLevel::Oz: + return OptimizationLevel::Oz; + default: + report_fatal_error("Bad PassBuilderOptLevel."); + } +} + +enum class LLVMRustRelocModel { + Static, + PIC, + DynamicNoPic, + ROPI, + RWPI, + ROPIRWPI, +}; + +static Reloc::Model fromRust(LLVMRustRelocModel RustReloc) { + switch (RustReloc) { + case LLVMRustRelocModel::Static: + return Reloc::Static; + case LLVMRustRelocModel::PIC: + return Reloc::PIC_; + case LLVMRustRelocModel::DynamicNoPic: + return Reloc::DynamicNoPIC; + case LLVMRustRelocModel::ROPI: + return Reloc::ROPI; + case LLVMRustRelocModel::RWPI: + return Reloc::RWPI; + case LLVMRustRelocModel::ROPIRWPI: + return Reloc::ROPI_RWPI; + } + report_fatal_error("Bad RelocModel."); +} + +/// getLongestEntryLength - Return the length of the longest entry in the table. +template <typename KV> static size_t getLongestEntryLength(ArrayRef<KV> Table) { + size_t MaxLen = 0; + for (auto &I : Table) + MaxLen = std::max(MaxLen, std::strlen(I.Key)); + return MaxLen; +} + +using PrintBackendInfo = void(void *, const char *Data, size_t Len); + +extern "C" void LLVMRustPrintTargetCPUs(LLVMTargetMachineRef TM, + const char *TargetCPU, + PrintBackendInfo Print, void *Out) { + const TargetMachine *Target = unwrap(TM); + const Triple::ArchType HostArch = + Triple(sys::getDefaultTargetTriple()).getArch(); + const Triple::ArchType TargetArch = Target->getTargetTriple().getArch(); + + std::ostringstream Buf; + + const MCSubtargetInfo *MCInfo = Target->getMCSubtargetInfo(); + const ArrayRef<SubtargetSubTypeKV> CPUTable = + MCInfo->getAllProcessorDescriptions(); + unsigned MaxCPULen = getLongestEntryLength(CPUTable); + + Buf << "Available CPUs for this target:\n"; + // Don't print the "native" entry when the user specifies --target with a + // different arch since that could be wrong or misleading. + if (HostArch == TargetArch) { + MaxCPULen = std::max(MaxCPULen, (unsigned)std::strlen("native")); + const StringRef HostCPU = sys::getHostCPUName(); + Buf << " " << std::left << std::setw(MaxCPULen) << "native" + << " - Select the CPU of the current host " + "(currently " + << HostCPU.str() << ").\n"; + } + for (auto &CPU : CPUTable) { + // Compare cpu against current target to label the default + if (strcmp(CPU.Key, TargetCPU) == 0) { + Buf << " " << std::left << std::setw(MaxCPULen) << CPU.Key + << " - This is the default target CPU for the current build target " + "(currently " + << Target->getTargetTriple().str() << ")."; + } else { + Buf << " " << CPU.Key; + } + Buf << "\n"; + } + + const auto &BufString = Buf.str(); + Print(Out, BufString.data(), BufString.size()); +} + +extern "C" size_t LLVMRustGetTargetFeaturesCount(LLVMTargetMachineRef TM) { +#if LLVM_VERSION_GE(18, 0) + const TargetMachine *Target = unwrap(TM); + const MCSubtargetInfo *MCInfo = Target->getMCSubtargetInfo(); + const ArrayRef<SubtargetFeatureKV> FeatTable = + MCInfo->getAllProcessorFeatures(); + return FeatTable.size(); +#else + return 0; +#endif +} + +extern "C" void LLVMRustGetTargetFeature(LLVMTargetMachineRef TM, size_t Index, + const char **Feature, + const char **Desc) { +#if LLVM_VERSION_GE(18, 0) + const TargetMachine *Target = unwrap(TM); + const MCSubtargetInfo *MCInfo = Target->getMCSubtargetInfo(); + const ArrayRef<SubtargetFeatureKV> FeatTable = + MCInfo->getAllProcessorFeatures(); + const SubtargetFeatureKV Feat = FeatTable[Index]; + *Feature = Feat.Key; + *Desc = Feat.Desc; +#endif +} + +extern "C" const char *LLVMRustGetHostCPUName(size_t *len) { + StringRef Name = sys::getHostCPUName(); + *len = Name.size(); + return Name.data(); +} + +extern "C" LLVMTargetMachineRef LLVMRustCreateTargetMachine( + const char *TripleStr, const char *CPU, const char *Feature, + const char *ABIStr, LLVMRustCodeModel RustCM, LLVMRustRelocModel RustReloc, + LLVMRustCodeGenOptLevel RustOptLevel, bool UseSoftFloat, + bool FunctionSections, bool DataSections, bool UniqueSectionNames, + bool TrapUnreachable, bool Singlethread, bool VerboseAsm, + bool EmitStackSizeSection, bool RelaxELFRelocations, bool UseInitArray, + const char *SplitDwarfFile, const char *OutputObjFile, + const char *DebugInfoCompression, bool UseEmulatedTls, + const char *ArgsCstrBuff, size_t ArgsCstrBuffLen) { + + auto OptLevel = fromRust(RustOptLevel); + auto RM = fromRust(RustReloc); + auto CM = fromRust(RustCM); + + std::string Error; + auto Trip = Triple(Triple::normalize(TripleStr)); + const llvm::Target *TheTarget = + TargetRegistry::lookupTarget(Trip.getTriple(), Error); + if (TheTarget == nullptr) { + LLVMRustSetLastError(Error.c_str()); + return nullptr; + } + + TargetOptions Options = codegen::InitTargetOptionsFromCodeGenFlags(Trip); + + Options.FloatABIType = FloatABI::Default; + if (UseSoftFloat) { + Options.FloatABIType = FloatABI::Soft; + } + Options.DataSections = DataSections; + Options.FunctionSections = FunctionSections; + Options.UniqueSectionNames = UniqueSectionNames; + Options.MCOptions.AsmVerbose = VerboseAsm; + // Always preserve comments that were written by the user + Options.MCOptions.PreserveAsmComments = true; + Options.MCOptions.ABIName = ABIStr; + if (SplitDwarfFile) { + Options.MCOptions.SplitDwarfFile = SplitDwarfFile; + } + if (OutputObjFile) { + Options.ObjectFilenameForDebug = OutputObjFile; + } + if (!strcmp("zlib", DebugInfoCompression) && + llvm::compression::zlib::isAvailable()) { +#if LLVM_VERSION_GE(19, 0) + Options.MCOptions.CompressDebugSections = DebugCompressionType::Zlib; +#else + Options.CompressDebugSections = DebugCompressionType::Zlib; +#endif + } else if (!strcmp("zstd", DebugInfoCompression) && + llvm::compression::zstd::isAvailable()) { +#if LLVM_VERSION_GE(19, 0) + Options.MCOptions.CompressDebugSections = DebugCompressionType::Zstd; +#else + Options.CompressDebugSections = DebugCompressionType::Zstd; +#endif + } else if (!strcmp("none", DebugInfoCompression)) { +#if LLVM_VERSION_GE(19, 0) + Options.MCOptions.CompressDebugSections = DebugCompressionType::None; +#else + Options.CompressDebugSections = DebugCompressionType::None; +#endif + } + +#if LLVM_VERSION_GE(19, 0) + Options.MCOptions.X86RelaxRelocations = RelaxELFRelocations; +#else + Options.RelaxELFRelocations = RelaxELFRelocations; +#endif + Options.UseInitArray = UseInitArray; + Options.EmulatedTLS = UseEmulatedTls; + + if (TrapUnreachable) { + // Tell LLVM to codegen `unreachable` into an explicit trap instruction. + // This limits the extent of possible undefined behavior in some cases, as + // it prevents control flow from "falling through" into whatever code + // happens to be laid out next in memory. + Options.TrapUnreachable = true; + // But don't emit traps after other traps or no-returns unnecessarily. + // ...except for when targeting WebAssembly, because the NoTrapAfterNoreturn + // option causes bugs in the LLVM WebAssembly backend. You should be able to + // remove this check when Rust's minimum supported LLVM version is >= 18 + // https://github.com/llvm/llvm-project/pull/65876 + if (!Trip.isWasm()) { + Options.NoTrapAfterNoreturn = true; + } + } + + if (Singlethread) { + Options.ThreadModel = ThreadModel::Single; + } + + Options.EmitStackSizeSection = EmitStackSizeSection; + + if (ArgsCstrBuff != nullptr) { + int buffer_offset = 0; + assert(ArgsCstrBuff[ArgsCstrBuffLen - 1] == '\0'); + + const size_t arg0_len = std::strlen(ArgsCstrBuff); + char *arg0 = new char[arg0_len + 1]; + memcpy(arg0, ArgsCstrBuff, arg0_len); + arg0[arg0_len] = '\0'; + buffer_offset += arg0_len + 1; + + const int num_cmd_arg_strings = std::count( + &ArgsCstrBuff[buffer_offset], &ArgsCstrBuff[ArgsCstrBuffLen], '\0'); + + std::string *cmd_arg_strings = new std::string[num_cmd_arg_strings]; + for (int i = 0; i < num_cmd_arg_strings; ++i) { + assert(buffer_offset < ArgsCstrBuffLen); + const int len = std::strlen(ArgsCstrBuff + buffer_offset); + cmd_arg_strings[i] = std::string(&ArgsCstrBuff[buffer_offset], len); + buffer_offset += len + 1; + } + + assert(buffer_offset == ArgsCstrBuffLen); + + Options.MCOptions.Argv0 = arg0; + Options.MCOptions.CommandLineArgs = + llvm::ArrayRef<std::string>(cmd_arg_strings, num_cmd_arg_strings); + } + + TargetMachine *TM = TheTarget->createTargetMachine( + Trip.getTriple(), CPU, Feature, Options, RM, CM, OptLevel); + return wrap(TM); +} + +extern "C" void LLVMRustDisposeTargetMachine(LLVMTargetMachineRef TM) { + + MCTargetOptions &MCOptions = unwrap(TM)->Options.MCOptions; + delete[] MCOptions.Argv0; + delete[] MCOptions.CommandLineArgs.data(); + + delete unwrap(TM); +} + +// Unfortunately, the LLVM C API doesn't provide a way to create the +// TargetLibraryInfo pass, so we use this method to do so. +extern "C" void LLVMRustAddLibraryInfo(LLVMPassManagerRef PMR, LLVMModuleRef M, + bool DisableSimplifyLibCalls) { + auto TargetTriple = Triple(unwrap(M)->getTargetTriple()); + auto TLII = TargetLibraryInfoImpl(TargetTriple); + if (DisableSimplifyLibCalls) + TLII.disableAllFunctions(); + unwrap(PMR)->add(new TargetLibraryInfoWrapperPass(TLII)); +} + +extern "C" void LLVMRustSetLLVMOptions(int Argc, char **Argv) { + // Initializing the command-line options more than once is not allowed. So, + // check if they've already been initialized. (This could happen if we're + // being called from rustpkg, for example). If the arguments change, then + // that's just kinda unfortunate. + static bool Initialized = false; + if (Initialized) + return; + Initialized = true; + cl::ParseCommandLineOptions(Argc, Argv); +} + +enum class LLVMRustFileType { + AssemblyFile, + ObjectFile, +}; + +static CodeGenFileType fromRust(LLVMRustFileType Type) { + switch (Type) { + case LLVMRustFileType::AssemblyFile: +#if LLVM_VERSION_GE(18, 0) + return CodeGenFileType::AssemblyFile; +#else + return CGFT_AssemblyFile; +#endif + case LLVMRustFileType::ObjectFile: +#if LLVM_VERSION_GE(18, 0) + return CodeGenFileType::ObjectFile; +#else + return CGFT_ObjectFile; +#endif + default: + report_fatal_error("Bad FileType."); + } +} + +extern "C" LLVMRustResult +LLVMRustWriteOutputFile(LLVMTargetMachineRef Target, LLVMPassManagerRef PMR, + LLVMModuleRef M, const char *Path, const char *DwoPath, + LLVMRustFileType RustFileType) { + llvm::legacy::PassManager *PM = unwrap<llvm::legacy::PassManager>(PMR); + auto FileType = fromRust(RustFileType); + + std::string ErrorInfo; + std::error_code EC; + auto OS = raw_fd_ostream(Path, EC, sys::fs::OF_None); + if (EC) + ErrorInfo = EC.message(); + if (ErrorInfo != "") { + LLVMRustSetLastError(ErrorInfo.c_str()); + return LLVMRustResult::Failure; + } + + auto BOS = buffer_ostream(OS); + if (DwoPath) { + auto DOS = raw_fd_ostream(DwoPath, EC, sys::fs::OF_None); + EC.clear(); + if (EC) + ErrorInfo = EC.message(); + if (ErrorInfo != "") { + LLVMRustSetLastError(ErrorInfo.c_str()); + return LLVMRustResult::Failure; + } + auto DBOS = buffer_ostream(DOS); + unwrap(Target)->addPassesToEmitFile(*PM, BOS, &DBOS, FileType, false); + PM->run(*unwrap(M)); + } else { + unwrap(Target)->addPassesToEmitFile(*PM, BOS, nullptr, FileType, false); + PM->run(*unwrap(M)); + } + + // Apparently `addPassesToEmitFile` adds a pointer to our on-the-stack output + // stream (OS), so the only real safe place to delete this is here? Don't we + // wish this was written in Rust? + LLVMDisposePassManager(PMR); + return LLVMRustResult::Success; +} + +extern "C" typedef void (*LLVMRustSelfProfileBeforePassCallback)( + void *, // LlvmSelfProfiler + const char *, // pass name + const char *); // IR name +extern "C" typedef void (*LLVMRustSelfProfileAfterPassCallback)( + void *); // LlvmSelfProfiler + +std::string LLVMRustwrappedIrGetName(const llvm::Any &WrappedIr) { + if (const auto *Cast = any_cast<const Module *>(&WrappedIr)) + return (*Cast)->getName().str(); + if (const auto *Cast = any_cast<const Function *>(&WrappedIr)) + return (*Cast)->getName().str(); + if (const auto *Cast = any_cast<const Loop *>(&WrappedIr)) + return (*Cast)->getName().str(); + if (const auto *Cast = any_cast<const LazyCallGraph::SCC *>(&WrappedIr)) + return (*Cast)->getName(); + return "<UNKNOWN>"; +} + +void LLVMSelfProfileInitializeCallbacks( + PassInstrumentationCallbacks &PIC, void *LlvmSelfProfiler, + LLVMRustSelfProfileBeforePassCallback BeforePassCallback, + LLVMRustSelfProfileAfterPassCallback AfterPassCallback) { + PIC.registerBeforeNonSkippedPassCallback( + [LlvmSelfProfiler, BeforePassCallback](StringRef Pass, llvm::Any Ir) { + std::string PassName = Pass.str(); + std::string IrName = LLVMRustwrappedIrGetName(Ir); + BeforePassCallback(LlvmSelfProfiler, PassName.c_str(), IrName.c_str()); + }); + + PIC.registerAfterPassCallback( + [LlvmSelfProfiler, AfterPassCallback]( + StringRef Pass, llvm::Any IR, const PreservedAnalyses &Preserved) { + AfterPassCallback(LlvmSelfProfiler); + }); + + PIC.registerAfterPassInvalidatedCallback( + [LlvmSelfProfiler, + AfterPassCallback](StringRef Pass, const PreservedAnalyses &Preserved) { + AfterPassCallback(LlvmSelfProfiler); + }); + + PIC.registerBeforeAnalysisCallback( + [LlvmSelfProfiler, BeforePassCallback](StringRef Pass, llvm::Any Ir) { + std::string PassName = Pass.str(); + std::string IrName = LLVMRustwrappedIrGetName(Ir); + BeforePassCallback(LlvmSelfProfiler, PassName.c_str(), IrName.c_str()); + }); + + PIC.registerAfterAnalysisCallback( + [LlvmSelfProfiler, AfterPassCallback](StringRef Pass, llvm::Any Ir) { + AfterPassCallback(LlvmSelfProfiler); + }); +} + +enum class LLVMRustOptStage { + PreLinkNoLTO, + PreLinkThinLTO, + PreLinkFatLTO, + ThinLTO, + FatLTO, +}; + +struct LLVMRustSanitizerOptions { + bool SanitizeAddress; + bool SanitizeAddressRecover; + bool SanitizeCFI; + bool SanitizeDataFlow; + char **SanitizeDataFlowABIList; + size_t SanitizeDataFlowABIListLen; + bool SanitizeKCFI; + bool SanitizeMemory; + bool SanitizeMemoryRecover; + int SanitizeMemoryTrackOrigins; + bool SanitizeThread; + bool SanitizeHWAddress; + bool SanitizeHWAddressRecover; + bool SanitizeKernelAddress; + bool SanitizeKernelAddressRecover; +}; + +extern "C" LLVMRustResult LLVMRustOptimize( + LLVMModuleRef ModuleRef, LLVMTargetMachineRef TMRef, + LLVMRustPassBuilderOptLevel OptLevelRust, LLVMRustOptStage OptStage, + bool IsLinkerPluginLTO, bool NoPrepopulatePasses, bool VerifyIR, + bool UseThinLTOBuffers, bool MergeFunctions, bool UnrollLoops, + bool SLPVectorize, bool LoopVectorize, bool DisableSimplifyLibCalls, + bool EmitLifetimeMarkers, LLVMRustSanitizerOptions *SanitizerOptions, + const char *PGOGenPath, const char *PGOUsePath, bool InstrumentCoverage, + const char *InstrProfileOutput, bool InstrumentGCOV, + const char *PGOSampleUsePath, bool DebugInfoForProfiling, + void *LlvmSelfProfiler, + LLVMRustSelfProfileBeforePassCallback BeforePassCallback, + LLVMRustSelfProfileAfterPassCallback AfterPassCallback, + const char *ExtraPasses, size_t ExtraPassesLen, const char *LLVMPlugins, + size_t LLVMPluginsLen) { + Module *TheModule = unwrap(ModuleRef); + TargetMachine *TM = unwrap(TMRef); + OptimizationLevel OptLevel = fromRust(OptLevelRust); + + PipelineTuningOptions PTO; + PTO.LoopUnrolling = UnrollLoops; + PTO.LoopInterleaving = UnrollLoops; + PTO.LoopVectorization = LoopVectorize; + PTO.SLPVectorization = SLPVectorize; + PTO.MergeFunctions = MergeFunctions; + + // FIXME: We may want to expose this as an option. + bool DebugPassManager = false; + + PassInstrumentationCallbacks PIC; + StandardInstrumentations SI(TheModule->getContext(), DebugPassManager); + SI.registerCallbacks(PIC); + + if (LlvmSelfProfiler) { + LLVMSelfProfileInitializeCallbacks(PIC, LlvmSelfProfiler, + BeforePassCallback, AfterPassCallback); + } + + std::optional<PGOOptions> PGOOpt; + auto FS = vfs::getRealFileSystem(); + if (PGOGenPath) { + assert(!PGOUsePath && !PGOSampleUsePath); + PGOOpt = PGOOptions(PGOGenPath, "", "", "", FS, PGOOptions::IRInstr, + PGOOptions::NoCSAction, +#if LLVM_VERSION_GE(19, 0) + PGOOptions::ColdFuncOpt::Default, +#endif + DebugInfoForProfiling); + } else if (PGOUsePath) { + assert(!PGOSampleUsePath); + PGOOpt = PGOOptions(PGOUsePath, "", "", "", FS, PGOOptions::IRUse, + PGOOptions::NoCSAction, +#if LLVM_VERSION_GE(19, 0) + PGOOptions::ColdFuncOpt::Default, +#endif + DebugInfoForProfiling); + } else if (PGOSampleUsePath) { + PGOOpt = PGOOptions(PGOSampleUsePath, "", "", "", FS, PGOOptions::SampleUse, + PGOOptions::NoCSAction, +#if LLVM_VERSION_GE(19, 0) + PGOOptions::ColdFuncOpt::Default, +#endif + DebugInfoForProfiling); + } else if (DebugInfoForProfiling) { + PGOOpt = PGOOptions("", "", "", "", FS, PGOOptions::NoAction, + PGOOptions::NoCSAction, +#if LLVM_VERSION_GE(19, 0) + PGOOptions::ColdFuncOpt::Default, +#endif + DebugInfoForProfiling); + } + + auto PB = PassBuilder(TM, PTO, PGOOpt, &PIC); + LoopAnalysisManager LAM; + FunctionAnalysisManager FAM; + CGSCCAnalysisManager CGAM; + ModuleAnalysisManager MAM; + + if (LLVMPluginsLen) { + auto PluginsStr = StringRef(LLVMPlugins, LLVMPluginsLen); + SmallVector<StringRef> Plugins; + PluginsStr.split(Plugins, ',', -1, false); + for (auto PluginPath : Plugins) { + auto Plugin = PassPlugin::Load(PluginPath.str()); + if (!Plugin) { + auto Err = Plugin.takeError(); + auto ErrMsg = llvm::toString(std::move(Err)); + LLVMRustSetLastError(ErrMsg.c_str()); + return LLVMRustResult::Failure; + } + Plugin->registerPassBuilderCallbacks(PB); + } + } + + FAM.registerPass([&] { return PB.buildDefaultAAPipeline(); }); + + Triple TargetTriple(TheModule->getTargetTriple()); + std::unique_ptr<TargetLibraryInfoImpl> TLII( + new TargetLibraryInfoImpl(TargetTriple)); + if (DisableSimplifyLibCalls) + TLII->disableAllFunctions(); + FAM.registerPass([&] { return TargetLibraryAnalysis(*TLII); }); + + PB.registerModuleAnalyses(MAM); + PB.registerCGSCCAnalyses(CGAM); + PB.registerFunctionAnalyses(FAM); + PB.registerLoopAnalyses(LAM); + PB.crossRegisterProxies(LAM, FAM, CGAM, MAM); + + // We manually collect pipeline callbacks so we can apply them at O0, where + // the PassBuilder does not create a pipeline. + std::vector<std::function<void(ModulePassManager &, OptimizationLevel)>> + PipelineStartEPCallbacks; + std::vector<std::function<void(ModulePassManager &, OptimizationLevel)>> + OptimizerLastEPCallbacks; + + if (!IsLinkerPluginLTO && SanitizerOptions && SanitizerOptions->SanitizeCFI && + !NoPrepopulatePasses) { + PipelineStartEPCallbacks.push_back( + [](ModulePassManager &MPM, OptimizationLevel Level) { + MPM.addPass(LowerTypeTestsPass(/*ExportSummary=*/nullptr, + /*ImportSummary=*/nullptr, + /*DropTypeTests=*/false)); + }); + } + + if (VerifyIR) { + PipelineStartEPCallbacks.push_back( + [VerifyIR](ModulePassManager &MPM, OptimizationLevel Level) { + MPM.addPass(VerifierPass()); + }); + } + + if (InstrumentGCOV) { + PipelineStartEPCallbacks.push_back( + [](ModulePassManager &MPM, OptimizationLevel Level) { + MPM.addPass(GCOVProfilerPass(GCOVOptions::getDefault())); + }); + } + + if (InstrumentCoverage) { + PipelineStartEPCallbacks.push_back( + [InstrProfileOutput](ModulePassManager &MPM, OptimizationLevel Level) { + InstrProfOptions Options; + if (InstrProfileOutput) { + Options.InstrProfileOutput = InstrProfileOutput; + } + // cargo run tests in multhreading mode by default + // so use atomics for coverage counters + Options.Atomic = true; +#if LLVM_VERSION_GE(18, 0) + MPM.addPass(InstrProfilingLoweringPass(Options, false)); +#else + MPM.addPass(InstrProfiling(Options, false)); +#endif + }); + } + + if (SanitizerOptions) { + if (SanitizerOptions->SanitizeDataFlow) { + std::vector<std::string> ABIListFiles( + SanitizerOptions->SanitizeDataFlowABIList, + SanitizerOptions->SanitizeDataFlowABIList + + SanitizerOptions->SanitizeDataFlowABIListLen); + OptimizerLastEPCallbacks.push_back( + [ABIListFiles](ModulePassManager &MPM, OptimizationLevel Level) { + MPM.addPass(DataFlowSanitizerPass(ABIListFiles)); + }); + } + + if (SanitizerOptions->SanitizeMemory) { + MemorySanitizerOptions Options( + SanitizerOptions->SanitizeMemoryTrackOrigins, + SanitizerOptions->SanitizeMemoryRecover, + /*CompileKernel=*/false, + /*EagerChecks=*/true); + OptimizerLastEPCallbacks.push_back( + [Options](ModulePassManager &MPM, OptimizationLevel Level) { + MPM.addPass(MemorySanitizerPass(Options)); + }); + } + + if (SanitizerOptions->SanitizeThread) { + OptimizerLastEPCallbacks.push_back([](ModulePassManager &MPM, + OptimizationLevel Level) { + MPM.addPass(ModuleThreadSanitizerPass()); + MPM.addPass(createModuleToFunctionPassAdaptor(ThreadSanitizerPass())); + }); + } + + if (SanitizerOptions->SanitizeAddress || + SanitizerOptions->SanitizeKernelAddress) { + OptimizerLastEPCallbacks.push_back( + [SanitizerOptions](ModulePassManager &MPM, OptimizationLevel Level) { + auto CompileKernel = SanitizerOptions->SanitizeKernelAddress; + AddressSanitizerOptions opts = AddressSanitizerOptions{ + CompileKernel, + SanitizerOptions->SanitizeAddressRecover || + SanitizerOptions->SanitizeKernelAddressRecover, + /*UseAfterScope=*/true, + AsanDetectStackUseAfterReturnMode::Runtime, + }; + MPM.addPass(AddressSanitizerPass(opts)); + }); + } + if (SanitizerOptions->SanitizeHWAddress) { + OptimizerLastEPCallbacks.push_back( + [SanitizerOptions](ModulePassManager &MPM, OptimizationLevel Level) { + HWAddressSanitizerOptions opts( + /*CompileKernel=*/false, + SanitizerOptions->SanitizeHWAddressRecover, + /*DisableOptimization=*/false); + MPM.addPass(HWAddressSanitizerPass(opts)); + }); + } + } + + ModulePassManager MPM; + bool NeedThinLTOBufferPasses = UseThinLTOBuffers; + if (!NoPrepopulatePasses) { + // The pre-link pipelines don't support O0 and require using + // buildO0DefaultPipeline() instead. At the same time, the LTO pipelines do + // support O0 and using them is required. + bool IsLTO = OptStage == LLVMRustOptStage::ThinLTO || + OptStage == LLVMRustOptStage::FatLTO; + if (OptLevel == OptimizationLevel::O0 && !IsLTO) { + for (const auto &C : PipelineStartEPCallbacks) + PB.registerPipelineStartEPCallback(C); + for (const auto &C : OptimizerLastEPCallbacks) + PB.registerOptimizerLastEPCallback(C); + + // Pass false as we manually schedule ThinLTOBufferPasses below. + MPM = PB.buildO0DefaultPipeline(OptLevel, /* PreLinkLTO */ false); + } else { + for (const auto &C : PipelineStartEPCallbacks) + PB.registerPipelineStartEPCallback(C); + for (const auto &C : OptimizerLastEPCallbacks) + PB.registerOptimizerLastEPCallback(C); + + switch (OptStage) { + case LLVMRustOptStage::PreLinkNoLTO: + MPM = PB.buildPerModuleDefaultPipeline(OptLevel, DebugPassManager); + break; + case LLVMRustOptStage::PreLinkThinLTO: + MPM = PB.buildThinLTOPreLinkDefaultPipeline(OptLevel); + NeedThinLTOBufferPasses = false; + break; + case LLVMRustOptStage::PreLinkFatLTO: + MPM = PB.buildLTOPreLinkDefaultPipeline(OptLevel); + NeedThinLTOBufferPasses = false; + break; + case LLVMRustOptStage::ThinLTO: + // FIXME: Does it make sense to pass the ModuleSummaryIndex? + // It only seems to be needed for C++ specific optimizations. + MPM = PB.buildThinLTODefaultPipeline(OptLevel, nullptr); + break; + case LLVMRustOptStage::FatLTO: + MPM = PB.buildLTODefaultPipeline(OptLevel, nullptr); + break; + } + } + } else { + // We're not building any of the default pipelines but we still want to + // add the verifier, instrumentation, etc passes if they were requested + for (const auto &C : PipelineStartEPCallbacks) + C(MPM, OptLevel); + for (const auto &C : OptimizerLastEPCallbacks) + C(MPM, OptLevel); + } + + if (ExtraPassesLen) { + if (auto Err = + PB.parsePassPipeline(MPM, StringRef(ExtraPasses, ExtraPassesLen))) { + std::string ErrMsg = toString(std::move(Err)); + LLVMRustSetLastError(ErrMsg.c_str()); + return LLVMRustResult::Failure; + } + } + + if (NeedThinLTOBufferPasses) { + MPM.addPass(CanonicalizeAliasesPass()); + MPM.addPass(NameAnonGlobalPass()); + } + + // Upgrade all calls to old intrinsics first. + for (Module::iterator I = TheModule->begin(), E = TheModule->end(); I != E;) + UpgradeCallsToIntrinsic(&*I++); // must be post-increment, as we remove + + MPM.run(*TheModule, MAM); + return LLVMRustResult::Success; +} + +// Callback to demangle function name +// Parameters: +// * name to be demangled +// * name len +// * output buffer +// * output buffer len +// Returns len of demangled string, or 0 if demangle failed. +typedef size_t (*DemangleFn)(const char *, size_t, char *, size_t); + +namespace { + +class RustAssemblyAnnotationWriter : public AssemblyAnnotationWriter { + DemangleFn Demangle; + std::vector<char> Buf; + +public: + RustAssemblyAnnotationWriter(DemangleFn Demangle) : Demangle(Demangle) {} + + // Return empty string if demangle failed + // or if name does not need to be demangled + StringRef CallDemangle(StringRef name) { + if (!Demangle) { + return StringRef(); + } + + if (Buf.size() < name.size() * 2) { + // Semangled name usually shorter than mangled, + // but allocate twice as much memory just in case + Buf.resize(name.size() * 2); + } + + auto R = Demangle(name.data(), name.size(), Buf.data(), Buf.size()); + if (!R) { + // Demangle failed. + return StringRef(); + } + + auto Demangled = StringRef(Buf.data(), R); + if (Demangled == name) { + // Do not print anything if demangled name is equal to mangled. + return StringRef(); + } + + return Demangled; + } + + void emitFunctionAnnot(const Function *F, + formatted_raw_ostream &OS) override { + StringRef Demangled = CallDemangle(F->getName()); + if (Demangled.empty()) { + return; + } + + OS << "; " << Demangled << "\n"; + } + + void emitInstructionAnnot(const Instruction *I, + formatted_raw_ostream &OS) override { + const char *Name; + const Value *Value; + if (const CallInst *CI = dyn_cast<CallInst>(I)) { + Name = "call"; + Value = CI->getCalledOperand(); + } else if (const InvokeInst *II = dyn_cast<InvokeInst>(I)) { + Name = "invoke"; + Value = II->getCalledOperand(); + } else { + // Could demangle more operations, e. g. + // `store %place, @function`. + return; + } + + if (!Value->hasName()) { + return; + } + + StringRef Demangled = CallDemangle(Value->getName()); + if (Demangled.empty()) { + return; + } + + OS << "; " << Name << " " << Demangled << "\n"; + } +}; + +} // namespace + +extern "C" LLVMRustResult LLVMRustPrintModule(LLVMModuleRef M, const char *Path, + DemangleFn Demangle) { + std::string ErrorInfo; + std::error_code EC; + auto OS = raw_fd_ostream(Path, EC, sys::fs::OF_None); + if (EC) + ErrorInfo = EC.message(); + if (ErrorInfo != "") { + LLVMRustSetLastError(ErrorInfo.c_str()); + return LLVMRustResult::Failure; + } + + auto AAW = RustAssemblyAnnotationWriter(Demangle); + auto FOS = formatted_raw_ostream(OS); + unwrap(M)->print(FOS, &AAW); + + return LLVMRustResult::Success; +} + +extern "C" void LLVMRustPrintPasses() { + PassBuilder PB; + PB.printPassNames(outs()); +} + +extern "C" void LLVMRustRunRestrictionPass(LLVMModuleRef M, char **Symbols, + size_t Len) { + auto PreserveFunctions = [=](const GlobalValue &GV) { + // Preserve LLVM-injected, ASAN-related symbols. + // See also https://github.com/rust-lang/rust/issues/113404. + if (GV.getName() == "___asan_globals_registered") { + return true; + } + + // Preserve symbols exported from Rust modules. + for (size_t I = 0; I < Len; I++) { + if (GV.getName() == Symbols[I]) { + return true; + } + } + return false; + }; + + internalizeModule(*unwrap(M), PreserveFunctions); +} + +extern "C" void +LLVMRustSetDataLayoutFromTargetMachine(LLVMModuleRef Module, + LLVMTargetMachineRef TMR) { + TargetMachine *Target = unwrap(TMR); + unwrap(Module)->setDataLayout(Target->createDataLayout()); +} + +extern "C" void LLVMRustSetModulePICLevel(LLVMModuleRef M) { + unwrap(M)->setPICLevel(PICLevel::Level::BigPIC); +} + +extern "C" void LLVMRustSetModulePIELevel(LLVMModuleRef M) { + unwrap(M)->setPIELevel(PIELevel::Level::Large); +} + +extern "C" void LLVMRustSetModuleCodeModel(LLVMModuleRef M, + LLVMRustCodeModel Model) { + auto CM = fromRust(Model); + if (!CM) + return; + unwrap(M)->setCodeModel(*CM); +} + +// Here you'll find an implementation of ThinLTO as used by the Rust compiler +// right now. This ThinLTO support is only enabled on "recent ish" versions of +// LLVM, and otherwise it's just blanket rejected from other compilers. +// +// Most of this implementation is straight copied from LLVM. At the time of +// this writing it wasn't *quite* suitable to reuse more code from upstream +// for our purposes, but we should strive to upstream this support once it's +// ready to go! I figure we may want a bit of testing locally first before +// sending this upstream to LLVM. I hear though they're quite eager to receive +// feedback like this! +// +// If you're reading this code and wondering "what in the world" or you're +// working "good lord by LLVM upgrade is *still* failing due to these bindings" +// then fear not! (ok maybe fear a little). All code here is mostly based +// on `lib/LTO/ThinLTOCodeGenerator.cpp` in LLVM. +// +// You'll find that the general layout here roughly corresponds to the `run` +// method in that file as well as `ProcessThinLTOModule`. Functions are +// specifically commented below as well, but if you're updating this code +// or otherwise trying to understand it, the LLVM source will be useful in +// interpreting the mysteries within. +// +// Otherwise I'll apologize in advance, it probably requires a relatively +// significant investment on your part to "truly understand" what's going on +// here. Not saying I do myself, but it took me awhile staring at LLVM's source +// and various online resources about ThinLTO to make heads or tails of all +// this. + +// This is a shared data structure which *must* be threadsafe to share +// read-only amongst threads. This also corresponds basically to the arguments +// of the `ProcessThinLTOModule` function in the LLVM source. +struct LLVMRustThinLTOData { + // The combined index that is the global analysis over all modules we're + // performing ThinLTO for. This is mostly managed by LLVM. + ModuleSummaryIndex Index; + + // All modules we may look at, stored as in-memory serialized versions. This + // is later used when inlining to ensure we can extract any module to inline + // from. + StringMap<MemoryBufferRef> ModuleMap; + + // A set that we manage of everything we *don't* want internalized. Note that + // this includes all transitive references right now as well, but it may not + // always! + DenseSet<GlobalValue::GUID> GUIDPreservedSymbols; + + // Not 100% sure what these are, but they impact what's internalized and + // what's inlined across modules, I believe. +#if LLVM_VERSION_GE(18, 0) + DenseMap<StringRef, FunctionImporter::ImportMapTy> ImportLists; + DenseMap<StringRef, FunctionImporter::ExportSetTy> ExportLists; + DenseMap<StringRef, GVSummaryMapTy> ModuleToDefinedGVSummaries; +#else + StringMap<FunctionImporter::ImportMapTy> ImportLists; + StringMap<FunctionImporter::ExportSetTy> ExportLists; + StringMap<GVSummaryMapTy> ModuleToDefinedGVSummaries; +#endif + StringMap<std::map<GlobalValue::GUID, GlobalValue::LinkageTypes>> ResolvedODR; + + LLVMRustThinLTOData() : Index(/* HaveGVs = */ false) {} +}; + +// Just an argument to the `LLVMRustCreateThinLTOData` function below. +struct LLVMRustThinLTOModule { + const char *identifier; + const char *data; + size_t len; +}; + +// This is copied from `lib/LTO/ThinLTOCodeGenerator.cpp`, not sure what it +// does. +static const GlobalValueSummary * +getFirstDefinitionForLinker(const GlobalValueSummaryList &GVSummaryList) { + auto StrongDefForLinker = llvm::find_if( + GVSummaryList, [](const std::unique_ptr<GlobalValueSummary> &Summary) { + auto Linkage = Summary->linkage(); + return !GlobalValue::isAvailableExternallyLinkage(Linkage) && + !GlobalValue::isWeakForLinker(Linkage); + }); + if (StrongDefForLinker != GVSummaryList.end()) + return StrongDefForLinker->get(); + + auto FirstDefForLinker = llvm::find_if( + GVSummaryList, [](const std::unique_ptr<GlobalValueSummary> &Summary) { + auto Linkage = Summary->linkage(); + return !GlobalValue::isAvailableExternallyLinkage(Linkage); + }); + if (FirstDefForLinker == GVSummaryList.end()) + return nullptr; + return FirstDefForLinker->get(); +} + +// The main entry point for creating the global ThinLTO analysis. The structure +// here is basically the same as before threads are spawned in the `run` +// function of `lib/LTO/ThinLTOCodeGenerator.cpp`. +extern "C" LLVMRustThinLTOData * +LLVMRustCreateThinLTOData(LLVMRustThinLTOModule *modules, int num_modules, + const char **preserved_symbols, int num_symbols) { + auto Ret = std::make_unique<LLVMRustThinLTOData>(); + + // Load each module's summary and merge it into one combined index + for (int i = 0; i < num_modules; i++) { + auto module = &modules[i]; + auto buffer = StringRef(module->data, module->len); + auto mem_buffer = MemoryBufferRef(buffer, module->identifier); + + Ret->ModuleMap[module->identifier] = mem_buffer; + +#if LLVM_VERSION_GE(18, 0) + if (Error Err = readModuleSummaryIndex(mem_buffer, Ret->Index)) { +#else + if (Error Err = readModuleSummaryIndex(mem_buffer, Ret->Index, i)) { +#endif + LLVMRustSetLastError(toString(std::move(Err)).c_str()); + return nullptr; + } + } + + // Collect for each module the list of function it defines (GUID -> Summary) + Ret->Index.collectDefinedGVSummariesPerModule( + Ret->ModuleToDefinedGVSummaries); + + // Convert the preserved symbols set from string to GUID, this is then needed + // for internalization. + for (int i = 0; i < num_symbols; i++) { + auto GUID = GlobalValue::getGUID(preserved_symbols[i]); + Ret->GUIDPreservedSymbols.insert(GUID); + } + + // Collect the import/export lists for all modules from the call-graph in the + // combined index + // + // This is copied from `lib/LTO/ThinLTOCodeGenerator.cpp` + auto deadIsPrevailing = [&](GlobalValue::GUID G) { + return PrevailingType::Unknown; + }; + // We don't have a complete picture in our use of ThinLTO, just our immediate + // crate, so we need `ImportEnabled = false` to limit internalization. + // Otherwise, we sometimes lose `static` values -- see #60184. + computeDeadSymbolsWithConstProp(Ret->Index, Ret->GUIDPreservedSymbols, + deadIsPrevailing, + /* ImportEnabled = */ false); + // Resolve LinkOnce/Weak symbols, this has to be computed early be cause it + // impacts the caching. + // + // This is copied from `lib/LTO/ThinLTOCodeGenerator.cpp` with some of this + // being lifted from `lib/LTO/LTO.cpp` as well + DenseMap<GlobalValue::GUID, const GlobalValueSummary *> PrevailingCopy; + for (auto &I : Ret->Index) { + if (I.second.SummaryList.size() > 1) + PrevailingCopy[I.first] = + getFirstDefinitionForLinker(I.second.SummaryList); + } + auto isPrevailing = [&](GlobalValue::GUID GUID, const GlobalValueSummary *S) { + const auto &Prevailing = PrevailingCopy.find(GUID); + if (Prevailing == PrevailingCopy.end()) + return true; + return Prevailing->second == S; + }; + ComputeCrossModuleImport(Ret->Index, Ret->ModuleToDefinedGVSummaries, + isPrevailing, Ret->ImportLists, Ret->ExportLists); + + auto recordNewLinkage = [&](StringRef ModuleIdentifier, + GlobalValue::GUID GUID, + GlobalValue::LinkageTypes NewLinkage) { + Ret->ResolvedODR[ModuleIdentifier][GUID] = NewLinkage; + }; + + // Uses FromPrevailing visibility scheme which works for many binary + // formats. We probably could and should use ELF visibility scheme for many of + // our targets, however. + lto::Config conf; + thinLTOResolvePrevailingInIndex(conf, Ret->Index, isPrevailing, + recordNewLinkage, Ret->GUIDPreservedSymbols); + + // Here we calculate an `ExportedGUIDs` set for use in the `isExported` + // callback below. This callback below will dictate the linkage for all + // summaries in the index, and we basically just only want to ensure that dead + // symbols are internalized. Otherwise everything that's already external + // linkage will stay as external, and internal will stay as internal. + std::set<GlobalValue::GUID> ExportedGUIDs; + for (auto &List : Ret->Index) { + for (auto &GVS : List.second.SummaryList) { + if (GlobalValue::isLocalLinkage(GVS->linkage())) + continue; + auto GUID = GVS->getOriginalName(); + if (GVS->flags().Live) + ExportedGUIDs.insert(GUID); + } + } + auto isExported = [&](StringRef ModuleIdentifier, ValueInfo VI) { + const auto &ExportList = Ret->ExportLists.find(ModuleIdentifier); + return (ExportList != Ret->ExportLists.end() && + ExportList->second.count(VI)) || + ExportedGUIDs.count(VI.getGUID()); + }; + thinLTOInternalizeAndPromoteInIndex(Ret->Index, isExported, isPrevailing); + + return Ret.release(); +} + +extern "C" void LLVMRustFreeThinLTOData(LLVMRustThinLTOData *Data) { + delete Data; +} + +// Below are the various passes that happen *per module* when doing ThinLTO. +// +// In other words, these are the functions that are all run concurrently +// with one another, one per module. The passes here correspond to the analysis +// passes in `lib/LTO/ThinLTOCodeGenerator.cpp`, currently found in the +// `ProcessThinLTOModule` function. Here they're split up into separate steps +// so rustc can save off the intermediate bytecode between each step. + +static bool clearDSOLocalOnDeclarations(Module &Mod, TargetMachine &TM) { + // When linking an ELF shared object, dso_local should be dropped. We + // conservatively do this for -fpic. + bool ClearDSOLocalOnDeclarations = TM.getTargetTriple().isOSBinFormatELF() && + TM.getRelocationModel() != Reloc::Static && + Mod.getPIELevel() == PIELevel::Default; + return ClearDSOLocalOnDeclarations; +} + +extern "C" bool LLVMRustPrepareThinLTORename(const LLVMRustThinLTOData *Data, + LLVMModuleRef M, + LLVMTargetMachineRef TM) { + Module &Mod = *unwrap(M); + TargetMachine &Target = *unwrap(TM); + + bool ClearDSOLocal = clearDSOLocalOnDeclarations(Mod, Target); + bool error = renameModuleForThinLTO(Mod, Data->Index, ClearDSOLocal); + + if (error) { + LLVMRustSetLastError("renameModuleForThinLTO failed"); + return false; + } + return true; +} + +extern "C" bool +LLVMRustPrepareThinLTOResolveWeak(const LLVMRustThinLTOData *Data, + LLVMModuleRef M) { + Module &Mod = *unwrap(M); + const auto &DefinedGlobals = + Data->ModuleToDefinedGVSummaries.lookup(Mod.getModuleIdentifier()); + thinLTOFinalizeInModule(Mod, DefinedGlobals, /*PropagateAttrs=*/true); + return true; +} + +extern "C" bool +LLVMRustPrepareThinLTOInternalize(const LLVMRustThinLTOData *Data, + LLVMModuleRef M) { + Module &Mod = *unwrap(M); + const auto &DefinedGlobals = + Data->ModuleToDefinedGVSummaries.lookup(Mod.getModuleIdentifier()); + thinLTOInternalizeModule(Mod, DefinedGlobals); + return true; +} + +extern "C" bool LLVMRustPrepareThinLTOImport(const LLVMRustThinLTOData *Data, + LLVMModuleRef M, + LLVMTargetMachineRef TM) { + Module &Mod = *unwrap(M); + TargetMachine &Target = *unwrap(TM); + + const auto &ImportList = Data->ImportLists.lookup(Mod.getModuleIdentifier()); + auto Loader = [&](StringRef Identifier) { + const auto &Memory = Data->ModuleMap.lookup(Identifier); + auto &Context = Mod.getContext(); + auto MOrErr = getLazyBitcodeModule(Memory, Context, true, true); + + if (!MOrErr) + return MOrErr; + + // The rest of this closure is a workaround for + // https://bugs.llvm.org/show_bug.cgi?id=38184 where during ThinLTO imports + // we accidentally import wasm custom sections into different modules, + // duplicating them by in the final output artifact. + // + // The issue is worked around here by manually removing the + // `wasm.custom_sections` named metadata node from any imported module. This + // we know isn't used by any optimization pass so there's no need for it to + // be imported. + // + // Note that the metadata is currently lazily loaded, so we materialize it + // here before looking up if there's metadata inside. The `FunctionImporter` + // will immediately materialize metadata anyway after an import, so this + // shouldn't be a perf hit. + if (Error Err = (*MOrErr)->materializeMetadata()) { + Expected<std::unique_ptr<Module>> Ret(std::move(Err)); + return Ret; + } + + auto *WasmCustomSections = + (*MOrErr)->getNamedMetadata("wasm.custom_sections"); + if (WasmCustomSections) + WasmCustomSections->eraseFromParent(); + + // `llvm.ident` named metadata also gets duplicated. + auto *llvmIdent = (*MOrErr)->getNamedMetadata("llvm.ident"); + if (llvmIdent) + llvmIdent->eraseFromParent(); + + return MOrErr; + }; + bool ClearDSOLocal = clearDSOLocalOnDeclarations(Mod, Target); + auto Importer = FunctionImporter(Data->Index, Loader, ClearDSOLocal); + Expected<bool> Result = Importer.importFunctions(Mod, ImportList); + if (!Result) { + LLVMRustSetLastError(toString(Result.takeError()).c_str()); + return false; + } + return true; +} + +// This struct and various functions are sort of a hack right now, but the +// problem is that we've got in-memory LLVM modules after we generate and +// optimize all codegen-units for one compilation in rustc. To be compatible +// with the LTO support above we need to serialize the modules plus their +// ThinLTO summary into memory. +// +// This structure is basically an owned version of a serialize module, with +// a ThinLTO summary attached. +struct LLVMRustThinLTOBuffer { + std::string data; + std::string thin_link_data; +}; + +extern "C" LLVMRustThinLTOBuffer * +LLVMRustThinLTOBufferCreate(LLVMModuleRef M, bool is_thin, bool emit_summary) { + auto Ret = std::make_unique<LLVMRustThinLTOBuffer>(); + { + auto OS = raw_string_ostream(Ret->data); + auto ThinLinkOS = raw_string_ostream(Ret->thin_link_data); + { + if (is_thin) { + PassBuilder PB; + LoopAnalysisManager LAM; + FunctionAnalysisManager FAM; + CGSCCAnalysisManager CGAM; + ModuleAnalysisManager MAM; + PB.registerModuleAnalyses(MAM); + PB.registerCGSCCAnalyses(CGAM); + PB.registerFunctionAnalyses(FAM); + PB.registerLoopAnalyses(LAM); + PB.crossRegisterProxies(LAM, FAM, CGAM, MAM); + ModulePassManager MPM; + // We only pass ThinLinkOS to be filled in if we want the summary, + // because otherwise LLVM does extra work and may double-emit some + // errors or warnings. + MPM.addPass( + ThinLTOBitcodeWriterPass(OS, emit_summary ? &ThinLinkOS : nullptr)); + MPM.run(*unwrap(M), MAM); + } else { + WriteBitcodeToFile(*unwrap(M), OS); + } + } + } + return Ret.release(); +} + +extern "C" void LLVMRustThinLTOBufferFree(LLVMRustThinLTOBuffer *Buffer) { + delete Buffer; +} + +extern "C" const void * +LLVMRustThinLTOBufferPtr(const LLVMRustThinLTOBuffer *Buffer) { + return Buffer->data.data(); +} + +extern "C" size_t +LLVMRustThinLTOBufferLen(const LLVMRustThinLTOBuffer *Buffer) { + return Buffer->data.length(); +} + +extern "C" const void * +LLVMRustThinLTOBufferThinLinkDataPtr(const LLVMRustThinLTOBuffer *Buffer) { + return Buffer->thin_link_data.data(); +} + +extern "C" size_t +LLVMRustThinLTOBufferThinLinkDataLen(const LLVMRustThinLTOBuffer *Buffer) { + return Buffer->thin_link_data.length(); +} + +// This is what we used to parse upstream bitcode for actual ThinLTO +// processing. We'll call this once per module optimized through ThinLTO, and +// it'll be called concurrently on many threads. +extern "C" LLVMModuleRef LLVMRustParseBitcodeForLTO(LLVMContextRef Context, + const char *data, + size_t len, + const char *identifier) { + auto Data = StringRef(data, len); + auto Buffer = MemoryBufferRef(Data, identifier); + unwrap(Context)->enableDebugTypeODRUniquing(); + Expected<std::unique_ptr<Module>> SrcOrError = + parseBitcodeFile(Buffer, *unwrap(Context)); + if (!SrcOrError) { + LLVMRustSetLastError(toString(SrcOrError.takeError()).c_str()); + return nullptr; + } + return wrap(std::move(*SrcOrError).release()); +} + +// Find a section of an object file by name. Fail if the section is missing or +// empty. +extern "C" const char *LLVMRustGetSliceFromObjectDataByName(const char *data, + size_t len, + const char *name, + size_t *out_len) { + *out_len = 0; + auto Data = StringRef(data, len); + auto Buffer = MemoryBufferRef(Data, ""); // The id is unused. + file_magic Type = identify_magic(Buffer.getBuffer()); + Expected<std::unique_ptr<object::ObjectFile>> ObjFileOrError = + object::ObjectFile::createObjectFile(Buffer, Type); + if (!ObjFileOrError) { + LLVMRustSetLastError(toString(ObjFileOrError.takeError()).c_str()); + return nullptr; + } + for (const object::SectionRef &Sec : (*ObjFileOrError)->sections()) { + Expected<StringRef> Name = Sec.getName(); + if (Name && *Name == name) { + Expected<StringRef> SectionOrError = Sec.getContents(); + if (!SectionOrError) { + LLVMRustSetLastError(toString(SectionOrError.takeError()).c_str()); + return nullptr; + } + *out_len = SectionOrError->size(); + return SectionOrError->data(); + } + } + LLVMRustSetLastError("could not find requested section"); + return nullptr; +} + +// Computes the LTO cache key for the provided 'ModId' in the given 'Data', +// storing the result in 'KeyOut'. +// Currently, this cache key is a SHA-1 hash of anything that could affect +// the result of optimizing this module (e.g. module imports, exports, liveness +// of access globals, etc). +// The precise details are determined by LLVM in `computeLTOCacheKey`, which is +// used during the normal linker-plugin incremental thin-LTO process. +extern "C" void LLVMRustComputeLTOCacheKey(RustStringRef KeyOut, + const char *ModId, + LLVMRustThinLTOData *Data) { + SmallString<40> Key; + llvm::lto::Config conf; + const auto &ImportList = Data->ImportLists.lookup(ModId); + const auto &ExportList = Data->ExportLists.lookup(ModId); + const auto &ResolvedODR = Data->ResolvedODR.lookup(ModId); + const auto &DefinedGlobals = Data->ModuleToDefinedGVSummaries.lookup(ModId); + std::set<GlobalValue::GUID> CfiFunctionDefs; + std::set<GlobalValue::GUID> CfiFunctionDecls; + + // Based on the 'InProcessThinBackend' constructor in LLVM + for (auto &Name : Data->Index.cfiFunctionDefs()) + CfiFunctionDefs.insert( + GlobalValue::getGUID(GlobalValue::dropLLVMManglingEscape(Name))); + for (auto &Name : Data->Index.cfiFunctionDecls()) + CfiFunctionDecls.insert( + GlobalValue::getGUID(GlobalValue::dropLLVMManglingEscape(Name))); + + llvm::computeLTOCacheKey(Key, conf, Data->Index, ModId, ImportList, + ExportList, ResolvedODR, DefinedGlobals, + CfiFunctionDefs, CfiFunctionDecls); + + LLVMRustStringWriteImpl(KeyOut, Key.c_str(), Key.size()); +} diff --git a/compiler/rustc_llvm/llvm-wrapper/README b/compiler/rustc_llvm/llvm-wrapper/README new file mode 100644 index 00000000000..e1c6dd07d2b --- /dev/null +++ b/compiler/rustc_llvm/llvm-wrapper/README @@ -0,0 +1,16 @@ +This directory currently contains some LLVM support code. This will generally +be sent upstream to LLVM in time; for now it lives here. + +NOTE: the LLVM C++ ABI is subject to between-version breakage and must *never* +be exposed to Rust. To allow for easy auditing of that, all Rust-exposed types +must be typedef-ed as "LLVMXyz", or "LLVMRustXyz" if they were defined here. + +Functions that return a failure status and leave the error in +the LLVM last error should return an LLVMRustResult rather than an +int or anything to avoid confusion. + +When translating enums, add a single `Other` variant as the first +one to allow for new variants to be added. It should abort when used +as an input. + +All other types must not be typedef-ed as such. diff --git a/compiler/rustc_llvm/llvm-wrapper/RustWrapper.cpp b/compiler/rustc_llvm/llvm-wrapper/RustWrapper.cpp new file mode 100644 index 00000000000..14757b27a37 --- /dev/null +++ b/compiler/rustc_llvm/llvm-wrapper/RustWrapper.cpp @@ -0,0 +1,2193 @@ +#include "LLVMWrapper.h" +#include "llvm/ADT/Statistic.h" +#include "llvm/Bitcode/BitcodeWriter.h" +#include "llvm/IR/DebugInfoMetadata.h" +#include "llvm/IR/DiagnosticHandler.h" +#include "llvm/IR/DiagnosticInfo.h" +#include "llvm/IR/DiagnosticPrinter.h" +#include "llvm/IR/GlobalVariable.h" +#include "llvm/IR/Instructions.h" +#include "llvm/IR/Intrinsics.h" +#include "llvm/IR/IntrinsicsARM.h" +#include "llvm/IR/LLVMRemarkStreamer.h" +#include "llvm/IR/Mangler.h" +#include "llvm/IR/Value.h" +#include "llvm/Object/Archive.h" +#include "llvm/Object/COFFImportFile.h" +#include "llvm/Object/ObjectFile.h" +#include "llvm/Pass.h" +#include "llvm/Remarks/RemarkFormat.h" +#include "llvm/Remarks/RemarkSerializer.h" +#include "llvm/Remarks/RemarkStreamer.h" +#include "llvm/Support/ModRef.h" +#include "llvm/Support/Signals.h" +#include "llvm/Support/ToolOutputFile.h" + +#include <iostream> + +// for raw `write` in the bad-alloc handler +#ifdef _MSC_VER +#include <io.h> +#else +#include <unistd.h> +#endif + +//===----------------------------------------------------------------------=== +// +// This file defines alternate interfaces to core functions that are more +// readily callable by Rust's FFI. +// +//===----------------------------------------------------------------------=== + +using namespace llvm; +using namespace llvm::sys; +using namespace llvm::object; + +// LLVMAtomicOrdering is already an enum - don't create another +// one. +static AtomicOrdering fromRust(LLVMAtomicOrdering Ordering) { + switch (Ordering) { + case LLVMAtomicOrderingNotAtomic: + return AtomicOrdering::NotAtomic; + case LLVMAtomicOrderingUnordered: + return AtomicOrdering::Unordered; + case LLVMAtomicOrderingMonotonic: + return AtomicOrdering::Monotonic; + case LLVMAtomicOrderingAcquire: + return AtomicOrdering::Acquire; + case LLVMAtomicOrderingRelease: + return AtomicOrdering::Release; + case LLVMAtomicOrderingAcquireRelease: + return AtomicOrdering::AcquireRelease; + case LLVMAtomicOrderingSequentiallyConsistent: + return AtomicOrdering::SequentiallyConsistent; + } + + report_fatal_error("Invalid LLVMAtomicOrdering value!"); +} + +static LLVM_THREAD_LOCAL char *LastError; + +// Custom error handler for fatal LLVM errors. +// +// Notably it exits the process with code 101, unlike LLVM's default of 1. +static void FatalErrorHandler(void *UserData, const char *Reason, + bool GenCrashDiag) { + // Once upon a time we emitted "LLVM ERROR:" specifically to mimic LLVM. Then, + // we developed crater and other tools which only expose logs, not error + // codes. Use a more greppable prefix that will still match the "LLVM ERROR:" + // prefix. + std::cerr << "rustc-LLVM ERROR: " << Reason << std::endl; + + // Since this error handler exits the process, we have to run any cleanup that + // LLVM would run after handling the error. This might change with an LLVM + // upgrade. + // + // In practice, this will do nothing, because the only cleanup LLVM does is + // to remove all files that were registered with it via a frontend calling + // one of the `createOutputFile` family of functions in LLVM and passing true + // to RemoveFileOnSignal, something that rustc does not do. However, it would + // be... inadvisable to suddenly stop running these handlers, if LLVM gets + // "interesting" ideas in the future about what cleanup should be done. + // We might even find it useful for generating less artifacts. + sys::RunInterruptHandlers(); + + exit(101); +} + +// Custom error handler for bad-alloc LLVM errors. +// +// It aborts the process without any further allocations, similar to LLVM's +// default except that may be configured to `throw std::bad_alloc()` instead. +static void BadAllocErrorHandler(void *UserData, const char *Reason, + bool GenCrashDiag) { + const char *OOM = "rustc-LLVM ERROR: out of memory\n"; + (void)!::write(2, OOM, strlen(OOM)); + (void)!::write(2, Reason, strlen(Reason)); + (void)!::write(2, "\n", 1); + abort(); +} + +extern "C" void LLVMRustInstallErrorHandlers() { + install_bad_alloc_error_handler(BadAllocErrorHandler); + install_fatal_error_handler(FatalErrorHandler); + install_out_of_memory_new_handler(); +} + +extern "C" void LLVMRustDisableSystemDialogsOnCrash() { + sys::DisableSystemDialogsOnCrash(); +} + +extern "C" char *LLVMRustGetLastError(void) { + char *Ret = LastError; + LastError = nullptr; + return Ret; +} + +extern "C" void LLVMRustSetLastError(const char *Err) { + free((void *)LastError); + LastError = strdup(Err); +} + +extern "C" LLVMContextRef LLVMRustContextCreate(bool shouldDiscardNames) { + auto ctx = new LLVMContext(); + ctx->setDiscardValueNames(shouldDiscardNames); + return wrap(ctx); +} + +extern "C" void LLVMRustSetNormalizedTarget(LLVMModuleRef M, + const char *Triple) { + unwrap(M)->setTargetTriple(Triple::normalize(Triple)); +} + +extern "C" const char *LLVMRustPrintPassTimings(size_t *Len) { + std::string buf; + auto SS = raw_string_ostream(buf); + TimerGroup::printAll(SS); + SS.flush(); + *Len = buf.length(); + char *CStr = (char *)malloc(*Len); + memcpy(CStr, buf.c_str(), *Len); + return CStr; +} + +extern "C" const char *LLVMRustPrintStatistics(size_t *Len) { + std::string buf; + auto SS = raw_string_ostream(buf); + llvm::PrintStatistics(SS); + SS.flush(); + *Len = buf.length(); + char *CStr = (char *)malloc(*Len); + memcpy(CStr, buf.c_str(), *Len); + return CStr; +} + +extern "C" LLVMValueRef LLVMRustGetNamedValue(LLVMModuleRef M, const char *Name, + size_t NameLen) { + return wrap(unwrap(M)->getNamedValue(StringRef(Name, NameLen))); +} + +enum class LLVMRustTailCallKind { + None, + Tail, + MustTail, + NoTail, +}; + +static CallInst::TailCallKind fromRust(LLVMRustTailCallKind Kind) { + switch (Kind) { + case LLVMRustTailCallKind::None: + return CallInst::TailCallKind::TCK_None; + case LLVMRustTailCallKind::Tail: + return CallInst::TailCallKind::TCK_Tail; + case LLVMRustTailCallKind::MustTail: + return CallInst::TailCallKind::TCK_MustTail; + case LLVMRustTailCallKind::NoTail: + return CallInst::TailCallKind::TCK_NoTail; + default: + report_fatal_error("bad CallInst::TailCallKind."); + } +} + +extern "C" void LLVMRustSetTailCallKind(LLVMValueRef Call, + LLVMRustTailCallKind TCK) { + unwrap<CallInst>(Call)->setTailCallKind(fromRust(TCK)); +} + +extern "C" LLVMValueRef LLVMRustGetOrInsertFunction(LLVMModuleRef M, + const char *Name, + size_t NameLen, + LLVMTypeRef FunctionTy) { + return wrap(unwrap(M) + ->getOrInsertFunction(StringRef(Name, NameLen), + unwrap<FunctionType>(FunctionTy)) + .getCallee()); +} + +extern "C" LLVMValueRef LLVMRustGetOrInsertGlobal(LLVMModuleRef M, + const char *Name, + size_t NameLen, + LLVMTypeRef Ty) { + Module *Mod = unwrap(M); + auto NameRef = StringRef(Name, NameLen); + + // We don't use Module::getOrInsertGlobal because that returns a Constant*, + // which may either be the real GlobalVariable*, or a constant bitcast of it + // if our type doesn't match the original declaration. We always want the + // GlobalVariable* so we can access linkage, visibility, etc. + GlobalVariable *GV = Mod->getGlobalVariable(NameRef, true); + if (!GV) + GV = new GlobalVariable(*Mod, unwrap(Ty), false, + GlobalValue::ExternalLinkage, nullptr, NameRef); + return wrap(GV); +} + +extern "C" LLVMValueRef LLVMRustInsertPrivateGlobal(LLVMModuleRef M, + LLVMTypeRef Ty) { + return wrap(new GlobalVariable(*unwrap(M), unwrap(Ty), false, + GlobalValue::PrivateLinkage, nullptr)); +} + +static Attribute::AttrKind fromRust(LLVMRustAttribute Kind) { + switch (Kind) { + case AlwaysInline: + return Attribute::AlwaysInline; + case ByVal: + return Attribute::ByVal; + case Cold: + return Attribute::Cold; + case InlineHint: + return Attribute::InlineHint; + case MinSize: + return Attribute::MinSize; + case Naked: + return Attribute::Naked; + case NoAlias: + return Attribute::NoAlias; + case NoCapture: + return Attribute::NoCapture; + case NoCfCheck: + return Attribute::NoCfCheck; + case NoInline: + return Attribute::NoInline; + case NonNull: + return Attribute::NonNull; + case NoRedZone: + return Attribute::NoRedZone; + case NoReturn: + return Attribute::NoReturn; + case NoUnwind: + return Attribute::NoUnwind; + case OptimizeForSize: + return Attribute::OptimizeForSize; + case ReadOnly: + return Attribute::ReadOnly; + case SExt: + return Attribute::SExt; + case StructRet: + return Attribute::StructRet; + case UWTable: + return Attribute::UWTable; + case ZExt: + return Attribute::ZExt; + case InReg: + return Attribute::InReg; + case SanitizeThread: + return Attribute::SanitizeThread; + case SanitizeAddress: + return Attribute::SanitizeAddress; + case SanitizeMemory: + return Attribute::SanitizeMemory; + case NonLazyBind: + return Attribute::NonLazyBind; + case OptimizeNone: + return Attribute::OptimizeNone; + case ReadNone: + return Attribute::ReadNone; + case SanitizeHWAddress: + return Attribute::SanitizeHWAddress; + case WillReturn: + return Attribute::WillReturn; + case StackProtectReq: + return Attribute::StackProtectReq; + case StackProtectStrong: + return Attribute::StackProtectStrong; + case StackProtect: + return Attribute::StackProtect; + case NoUndef: + return Attribute::NoUndef; + case SanitizeMemTag: + return Attribute::SanitizeMemTag; + case ShadowCallStack: + return Attribute::ShadowCallStack; + case AllocSize: + return Attribute::AllocSize; + case AllocatedPointer: + return Attribute::AllocatedPointer; + case AllocAlign: + return Attribute::AllocAlign; + case SanitizeSafeStack: + return Attribute::SafeStack; + case FnRetThunkExtern: + return Attribute::FnRetThunkExtern; +#if LLVM_VERSION_GE(18, 0) + case Writable: + return Attribute::Writable; + case DeadOnUnwind: + return Attribute::DeadOnUnwind; +#else + case Writable: + case DeadOnUnwind: + report_fatal_error("Not supported on this LLVM version"); +#endif + } + report_fatal_error("bad AttributeKind"); +} + +template <typename T> +static inline void AddAttributes(T *t, unsigned Index, LLVMAttributeRef *Attrs, + size_t AttrsLen) { + AttributeList PAL = t->getAttributes(); + auto B = AttrBuilder(t->getContext()); + for (LLVMAttributeRef Attr : ArrayRef<LLVMAttributeRef>(Attrs, AttrsLen)) + B.addAttribute(unwrap(Attr)); + AttributeList PALNew = PAL.addAttributesAtIndex(t->getContext(), Index, B); + t->setAttributes(PALNew); +} + +extern "C" void LLVMRustAddFunctionAttributes(LLVMValueRef Fn, unsigned Index, + LLVMAttributeRef *Attrs, + size_t AttrsLen) { + Function *F = unwrap<Function>(Fn); + AddAttributes(F, Index, Attrs, AttrsLen); +} + +extern "C" void LLVMRustAddCallSiteAttributes(LLVMValueRef Instr, + unsigned Index, + LLVMAttributeRef *Attrs, + size_t AttrsLen) { + CallBase *Call = unwrap<CallBase>(Instr); + AddAttributes(Call, Index, Attrs, AttrsLen); +} + +extern "C" LLVMAttributeRef +LLVMRustCreateAttrNoValue(LLVMContextRef C, LLVMRustAttribute RustAttr) { + return wrap(Attribute::get(*unwrap(C), fromRust(RustAttr))); +} + +extern "C" LLVMAttributeRef LLVMRustCreateAlignmentAttr(LLVMContextRef C, + uint64_t Bytes) { + return wrap(Attribute::getWithAlignment(*unwrap(C), llvm::Align(Bytes))); +} + +extern "C" LLVMAttributeRef LLVMRustCreateDereferenceableAttr(LLVMContextRef C, + uint64_t Bytes) { + return wrap(Attribute::getWithDereferenceableBytes(*unwrap(C), Bytes)); +} + +extern "C" LLVMAttributeRef +LLVMRustCreateDereferenceableOrNullAttr(LLVMContextRef C, uint64_t Bytes) { + return wrap(Attribute::getWithDereferenceableOrNullBytes(*unwrap(C), Bytes)); +} + +extern "C" LLVMAttributeRef LLVMRustCreateByValAttr(LLVMContextRef C, + LLVMTypeRef Ty) { + return wrap(Attribute::getWithByValType(*unwrap(C), unwrap(Ty))); +} + +extern "C" LLVMAttributeRef LLVMRustCreateStructRetAttr(LLVMContextRef C, + LLVMTypeRef Ty) { + return wrap(Attribute::getWithStructRetType(*unwrap(C), unwrap(Ty))); +} + +extern "C" LLVMAttributeRef LLVMRustCreateElementTypeAttr(LLVMContextRef C, + LLVMTypeRef Ty) { + return wrap(Attribute::get(*unwrap(C), Attribute::ElementType, unwrap(Ty))); +} + +extern "C" LLVMAttributeRef LLVMRustCreateUWTableAttr(LLVMContextRef C, + bool Async) { + return wrap(Attribute::getWithUWTableKind( + *unwrap(C), Async ? UWTableKind::Async : UWTableKind::Sync)); +} + +extern "C" LLVMAttributeRef +LLVMRustCreateAllocSizeAttr(LLVMContextRef C, uint32_t ElementSizeArg) { + return wrap(Attribute::getWithAllocSizeArgs(*unwrap(C), ElementSizeArg, + std::nullopt)); +} + +// These values **must** match ffi::AllocKindFlags. +// It _happens_ to match the LLVM values of llvm::AllocFnKind, +// but that's happenstance and we do explicit conversions before +// passing them to LLVM. +enum class LLVMRustAllocKindFlags : uint64_t { + Unknown = 0, + Alloc = 1, + Realloc = 1 << 1, + Free = 1 << 2, + Uninitialized = 1 << 3, + Zeroed = 1 << 4, + Aligned = 1 << 5, +}; + +static LLVMRustAllocKindFlags operator&(LLVMRustAllocKindFlags A, + LLVMRustAllocKindFlags B) { + return static_cast<LLVMRustAllocKindFlags>(static_cast<uint64_t>(A) & + static_cast<uint64_t>(B)); +} + +static bool isSet(LLVMRustAllocKindFlags F) { + return F != LLVMRustAllocKindFlags::Unknown; +} + +static llvm::AllocFnKind allocKindFromRust(LLVMRustAllocKindFlags F) { + llvm::AllocFnKind AFK = llvm::AllocFnKind::Unknown; + if (isSet(F & LLVMRustAllocKindFlags::Alloc)) { + AFK |= llvm::AllocFnKind::Alloc; + } + if (isSet(F & LLVMRustAllocKindFlags::Realloc)) { + AFK |= llvm::AllocFnKind::Realloc; + } + if (isSet(F & LLVMRustAllocKindFlags::Free)) { + AFK |= llvm::AllocFnKind::Free; + } + if (isSet(F & LLVMRustAllocKindFlags::Uninitialized)) { + AFK |= llvm::AllocFnKind::Uninitialized; + } + if (isSet(F & LLVMRustAllocKindFlags::Zeroed)) { + AFK |= llvm::AllocFnKind::Zeroed; + } + if (isSet(F & LLVMRustAllocKindFlags::Aligned)) { + AFK |= llvm::AllocFnKind::Aligned; + } + return AFK; +} + +extern "C" LLVMAttributeRef LLVMRustCreateAllocKindAttr(LLVMContextRef C, + uint64_t AllocKindArg) { + return wrap( + Attribute::get(*unwrap(C), Attribute::AllocKind, + static_cast<uint64_t>(allocKindFromRust( + static_cast<LLVMRustAllocKindFlags>(AllocKindArg))))); +} + +// Simplified representation of `MemoryEffects` across the FFI boundary. +// +// Each variant corresponds to one of the static factory methods on +// `MemoryEffects`. +enum class LLVMRustMemoryEffects { + None, + ReadOnly, + InaccessibleMemOnly, +}; + +extern "C" LLVMAttributeRef +LLVMRustCreateMemoryEffectsAttr(LLVMContextRef C, + LLVMRustMemoryEffects Effects) { + switch (Effects) { + case LLVMRustMemoryEffects::None: + return wrap( + Attribute::getWithMemoryEffects(*unwrap(C), MemoryEffects::none())); + case LLVMRustMemoryEffects::ReadOnly: + return wrap( + Attribute::getWithMemoryEffects(*unwrap(C), MemoryEffects::readOnly())); + case LLVMRustMemoryEffects::InaccessibleMemOnly: + return wrap(Attribute::getWithMemoryEffects( + *unwrap(C), MemoryEffects::inaccessibleMemOnly())); + default: + report_fatal_error("bad MemoryEffects."); + } +} + +// Enable all fast-math flags, including those which will cause floating-point +// operations to return poison for some well-defined inputs. This function can +// only be used to build unsafe Rust intrinsics. That unsafety does permit +// additional optimizations, but at the time of writing, their value is not +// well-understood relative to those enabled by LLVMRustSetAlgebraicMath. +// +// https://llvm.org/docs/LangRef.html#fast-math-flags +extern "C" void LLVMRustSetFastMath(LLVMValueRef V) { + if (auto I = dyn_cast<Instruction>(unwrap<Value>(V))) { + I->setFast(true); + } +} + +// Enable fast-math flags which permit algebraic transformations that are not +// allowed by IEEE floating point. For example: a + (b + c) = (a + b) + c and a +// / b = a * (1 / b) Note that this does NOT enable any flags which can cause a +// floating-point operation on well-defined inputs to return poison, and +// therefore this function can be used to build safe Rust intrinsics (such as +// fadd_algebraic). +// +// https://llvm.org/docs/LangRef.html#fast-math-flags +extern "C" void LLVMRustSetAlgebraicMath(LLVMValueRef V) { + if (auto I = dyn_cast<Instruction>(unwrap<Value>(V))) { + I->setHasAllowReassoc(true); + I->setHasAllowContract(true); + I->setHasAllowReciprocal(true); + I->setHasNoSignedZeros(true); + } +} + +// Enable the reassoc fast-math flag, allowing transformations that pretend +// floating-point addition and multiplication are associative. +// +// Note that this does NOT enable any flags which can cause a floating-point +// operation on well-defined inputs to return poison, and therefore this +// function can be used to build safe Rust intrinsics (such as fadd_algebraic). +// +// https://llvm.org/docs/LangRef.html#fast-math-flags +extern "C" void LLVMRustSetAllowReassoc(LLVMValueRef V) { + if (auto I = dyn_cast<Instruction>(unwrap<Value>(V))) { + I->setHasAllowReassoc(true); + } +} + +extern "C" LLVMValueRef +LLVMRustBuildAtomicLoad(LLVMBuilderRef B, LLVMTypeRef Ty, LLVMValueRef Source, + const char *Name, LLVMAtomicOrdering Order) { + Value *Ptr = unwrap(Source); + LoadInst *LI = unwrap(B)->CreateLoad(unwrap(Ty), Ptr, Name); + LI->setAtomic(fromRust(Order)); + return wrap(LI); +} + +extern "C" LLVMValueRef LLVMRustBuildAtomicStore(LLVMBuilderRef B, + LLVMValueRef V, + LLVMValueRef Target, + LLVMAtomicOrdering Order) { + StoreInst *SI = unwrap(B)->CreateStore(unwrap(V), unwrap(Target)); + SI->setAtomic(fromRust(Order)); + return wrap(SI); +} + +enum class LLVMRustAsmDialect { + Att, + Intel, +}; + +static InlineAsm::AsmDialect fromRust(LLVMRustAsmDialect Dialect) { + switch (Dialect) { + case LLVMRustAsmDialect::Att: + return InlineAsm::AD_ATT; + case LLVMRustAsmDialect::Intel: + return InlineAsm::AD_Intel; + default: + report_fatal_error("bad AsmDialect."); + } +} + +extern "C" LLVMValueRef +LLVMRustInlineAsm(LLVMTypeRef Ty, char *AsmString, size_t AsmStringLen, + char *Constraints, size_t ConstraintsLen, + LLVMBool HasSideEffects, LLVMBool IsAlignStack, + LLVMRustAsmDialect Dialect, LLVMBool CanThrow) { + return wrap(InlineAsm::get( + unwrap<FunctionType>(Ty), StringRef(AsmString, AsmStringLen), + StringRef(Constraints, ConstraintsLen), HasSideEffects, IsAlignStack, + fromRust(Dialect), CanThrow)); +} + +extern "C" bool LLVMRustInlineAsmVerify(LLVMTypeRef Ty, char *Constraints, + size_t ConstraintsLen) { + // llvm::Error converts to true if it is an error. + return !llvm::errorToBool(InlineAsm::verify( + unwrap<FunctionType>(Ty), StringRef(Constraints, ConstraintsLen))); +} + +typedef DIBuilder *LLVMRustDIBuilderRef; + +template <typename DIT> DIT *unwrapDIPtr(LLVMMetadataRef Ref) { + return (DIT *)(Ref ? unwrap<MDNode>(Ref) : nullptr); +} + +#define DIDescriptor DIScope +#define DIArray DINodeArray +#define unwrapDI unwrapDIPtr + +// These values **must** match debuginfo::DIFlags! They also *happen* +// to match LLVM, but that isn't required as we do giant sets of +// matching below. The value shouldn't be directly passed to LLVM. +enum class LLVMRustDIFlags : uint32_t { + FlagZero = 0, + FlagPrivate = 1, + FlagProtected = 2, + FlagPublic = 3, + FlagFwdDecl = (1 << 2), + FlagAppleBlock = (1 << 3), + FlagBlockByrefStruct = (1 << 4), + FlagVirtual = (1 << 5), + FlagArtificial = (1 << 6), + FlagExplicit = (1 << 7), + FlagPrototyped = (1 << 8), + FlagObjcClassComplete = (1 << 9), + FlagObjectPointer = (1 << 10), + FlagVector = (1 << 11), + FlagStaticMember = (1 << 12), + FlagLValueReference = (1 << 13), + FlagRValueReference = (1 << 14), + FlagExternalTypeRef = (1 << 15), + FlagIntroducedVirtual = (1 << 18), + FlagBitField = (1 << 19), + FlagNoReturn = (1 << 20), + // Do not add values that are not supported by the minimum LLVM + // version we support! see llvm/include/llvm/IR/DebugInfoFlags.def +}; + +inline LLVMRustDIFlags operator&(LLVMRustDIFlags A, LLVMRustDIFlags B) { + return static_cast<LLVMRustDIFlags>(static_cast<uint32_t>(A) & + static_cast<uint32_t>(B)); +} + +inline LLVMRustDIFlags operator|(LLVMRustDIFlags A, LLVMRustDIFlags B) { + return static_cast<LLVMRustDIFlags>(static_cast<uint32_t>(A) | + static_cast<uint32_t>(B)); +} + +inline LLVMRustDIFlags &operator|=(LLVMRustDIFlags &A, LLVMRustDIFlags B) { + return A = A | B; +} + +inline bool isSet(LLVMRustDIFlags F) { return F != LLVMRustDIFlags::FlagZero; } + +inline LLVMRustDIFlags visibility(LLVMRustDIFlags F) { + return static_cast<LLVMRustDIFlags>(static_cast<uint32_t>(F) & 0x3); +} + +static DINode::DIFlags fromRust(LLVMRustDIFlags Flags) { + DINode::DIFlags Result = DINode::DIFlags::FlagZero; + + switch (visibility(Flags)) { + case LLVMRustDIFlags::FlagPrivate: + Result |= DINode::DIFlags::FlagPrivate; + break; + case LLVMRustDIFlags::FlagProtected: + Result |= DINode::DIFlags::FlagProtected; + break; + case LLVMRustDIFlags::FlagPublic: + Result |= DINode::DIFlags::FlagPublic; + break; + default: + // The rest are handled below + break; + } + + if (isSet(Flags & LLVMRustDIFlags::FlagFwdDecl)) { + Result |= DINode::DIFlags::FlagFwdDecl; + } + if (isSet(Flags & LLVMRustDIFlags::FlagAppleBlock)) { + Result |= DINode::DIFlags::FlagAppleBlock; + } + if (isSet(Flags & LLVMRustDIFlags::FlagVirtual)) { + Result |= DINode::DIFlags::FlagVirtual; + } + if (isSet(Flags & LLVMRustDIFlags::FlagArtificial)) { + Result |= DINode::DIFlags::FlagArtificial; + } + if (isSet(Flags & LLVMRustDIFlags::FlagExplicit)) { + Result |= DINode::DIFlags::FlagExplicit; + } + if (isSet(Flags & LLVMRustDIFlags::FlagPrototyped)) { + Result |= DINode::DIFlags::FlagPrototyped; + } + if (isSet(Flags & LLVMRustDIFlags::FlagObjcClassComplete)) { + Result |= DINode::DIFlags::FlagObjcClassComplete; + } + if (isSet(Flags & LLVMRustDIFlags::FlagObjectPointer)) { + Result |= DINode::DIFlags::FlagObjectPointer; + } + if (isSet(Flags & LLVMRustDIFlags::FlagVector)) { + Result |= DINode::DIFlags::FlagVector; + } + if (isSet(Flags & LLVMRustDIFlags::FlagStaticMember)) { + Result |= DINode::DIFlags::FlagStaticMember; + } + if (isSet(Flags & LLVMRustDIFlags::FlagLValueReference)) { + Result |= DINode::DIFlags::FlagLValueReference; + } + if (isSet(Flags & LLVMRustDIFlags::FlagRValueReference)) { + Result |= DINode::DIFlags::FlagRValueReference; + } + if (isSet(Flags & LLVMRustDIFlags::FlagIntroducedVirtual)) { + Result |= DINode::DIFlags::FlagIntroducedVirtual; + } + if (isSet(Flags & LLVMRustDIFlags::FlagBitField)) { + Result |= DINode::DIFlags::FlagBitField; + } + if (isSet(Flags & LLVMRustDIFlags::FlagNoReturn)) { + Result |= DINode::DIFlags::FlagNoReturn; + } + + return Result; +} + +// These values **must** match debuginfo::DISPFlags! They also *happen* +// to match LLVM, but that isn't required as we do giant sets of +// matching below. The value shouldn't be directly passed to LLVM. +enum class LLVMRustDISPFlags : uint32_t { + SPFlagZero = 0, + SPFlagVirtual = 1, + SPFlagPureVirtual = 2, + SPFlagLocalToUnit = (1 << 2), + SPFlagDefinition = (1 << 3), + SPFlagOptimized = (1 << 4), + SPFlagMainSubprogram = (1 << 5), + // Do not add values that are not supported by the minimum LLVM + // version we support! see llvm/include/llvm/IR/DebugInfoFlags.def + // (In LLVM < 8, createFunction supported these as separate bool arguments.) +}; + +inline LLVMRustDISPFlags operator&(LLVMRustDISPFlags A, LLVMRustDISPFlags B) { + return static_cast<LLVMRustDISPFlags>(static_cast<uint32_t>(A) & + static_cast<uint32_t>(B)); +} + +inline LLVMRustDISPFlags operator|(LLVMRustDISPFlags A, LLVMRustDISPFlags B) { + return static_cast<LLVMRustDISPFlags>(static_cast<uint32_t>(A) | + static_cast<uint32_t>(B)); +} + +inline LLVMRustDISPFlags &operator|=(LLVMRustDISPFlags &A, + LLVMRustDISPFlags B) { + return A = A | B; +} + +inline bool isSet(LLVMRustDISPFlags F) { + return F != LLVMRustDISPFlags::SPFlagZero; +} + +inline LLVMRustDISPFlags virtuality(LLVMRustDISPFlags F) { + return static_cast<LLVMRustDISPFlags>(static_cast<uint32_t>(F) & 0x3); +} + +static DISubprogram::DISPFlags fromRust(LLVMRustDISPFlags SPFlags) { + DISubprogram::DISPFlags Result = DISubprogram::DISPFlags::SPFlagZero; + + switch (virtuality(SPFlags)) { + case LLVMRustDISPFlags::SPFlagVirtual: + Result |= DISubprogram::DISPFlags::SPFlagVirtual; + break; + case LLVMRustDISPFlags::SPFlagPureVirtual: + Result |= DISubprogram::DISPFlags::SPFlagPureVirtual; + break; + default: + // The rest are handled below + break; + } + + if (isSet(SPFlags & LLVMRustDISPFlags::SPFlagLocalToUnit)) { + Result |= DISubprogram::DISPFlags::SPFlagLocalToUnit; + } + if (isSet(SPFlags & LLVMRustDISPFlags::SPFlagDefinition)) { + Result |= DISubprogram::DISPFlags::SPFlagDefinition; + } + if (isSet(SPFlags & LLVMRustDISPFlags::SPFlagOptimized)) { + Result |= DISubprogram::DISPFlags::SPFlagOptimized; + } + if (isSet(SPFlags & LLVMRustDISPFlags::SPFlagMainSubprogram)) { + Result |= DISubprogram::DISPFlags::SPFlagMainSubprogram; + } + + return Result; +} + +enum class LLVMRustDebugEmissionKind { + NoDebug, + FullDebug, + LineTablesOnly, + DebugDirectivesOnly, +}; + +static DICompileUnit::DebugEmissionKind +fromRust(LLVMRustDebugEmissionKind Kind) { + switch (Kind) { + case LLVMRustDebugEmissionKind::NoDebug: + return DICompileUnit::DebugEmissionKind::NoDebug; + case LLVMRustDebugEmissionKind::FullDebug: + return DICompileUnit::DebugEmissionKind::FullDebug; + case LLVMRustDebugEmissionKind::LineTablesOnly: + return DICompileUnit::DebugEmissionKind::LineTablesOnly; + case LLVMRustDebugEmissionKind::DebugDirectivesOnly: + return DICompileUnit::DebugEmissionKind::DebugDirectivesOnly; + default: + report_fatal_error("bad DebugEmissionKind."); + } +} + +enum class LLVMRustDebugNameTableKind { + Default, + GNU, + None, +}; + +static DICompileUnit::DebugNameTableKind +fromRust(LLVMRustDebugNameTableKind Kind) { + switch (Kind) { + case LLVMRustDebugNameTableKind::Default: + return DICompileUnit::DebugNameTableKind::Default; + case LLVMRustDebugNameTableKind::GNU: + return DICompileUnit::DebugNameTableKind::GNU; + case LLVMRustDebugNameTableKind::None: + return DICompileUnit::DebugNameTableKind::None; + default: + report_fatal_error("bad DebugNameTableKind."); + } +} + +enum class LLVMRustChecksumKind { + None, + MD5, + SHA1, + SHA256, +}; + +static std::optional<DIFile::ChecksumKind> fromRust(LLVMRustChecksumKind Kind) { + switch (Kind) { + case LLVMRustChecksumKind::None: + return std::nullopt; + case LLVMRustChecksumKind::MD5: + return DIFile::ChecksumKind::CSK_MD5; + case LLVMRustChecksumKind::SHA1: + return DIFile::ChecksumKind::CSK_SHA1; + case LLVMRustChecksumKind::SHA256: + return DIFile::ChecksumKind::CSK_SHA256; + default: + report_fatal_error("bad ChecksumKind."); + } +} + +extern "C" uint32_t LLVMRustDebugMetadataVersion() { + return DEBUG_METADATA_VERSION; +} + +extern "C" uint32_t LLVMRustVersionPatch() { return LLVM_VERSION_PATCH; } + +extern "C" uint32_t LLVMRustVersionMinor() { return LLVM_VERSION_MINOR; } + +extern "C" uint32_t LLVMRustVersionMajor() { return LLVM_VERSION_MAJOR; } + +extern "C" void LLVMRustAddModuleFlagU32(LLVMModuleRef M, + Module::ModFlagBehavior MergeBehavior, + const char *Name, uint32_t Value) { + unwrap(M)->addModuleFlag(MergeBehavior, Name, Value); +} + +extern "C" void LLVMRustAddModuleFlagString( + LLVMModuleRef M, Module::ModFlagBehavior MergeBehavior, const char *Name, + const char *Value, size_t ValueLen) { + unwrap(M)->addModuleFlag( + MergeBehavior, Name, + MDString::get(unwrap(M)->getContext(), StringRef(Value, ValueLen))); +} + +extern "C" bool LLVMRustHasModuleFlag(LLVMModuleRef M, const char *Name, + size_t Len) { + return unwrap(M)->getModuleFlag(StringRef(Name, Len)) != nullptr; +} + +extern "C" void LLVMRustGlobalAddMetadata(LLVMValueRef Global, unsigned Kind, + LLVMMetadataRef MD) { + unwrap<GlobalObject>(Global)->addMetadata(Kind, *unwrap<MDNode>(MD)); +} + +extern "C" LLVMRustDIBuilderRef LLVMRustDIBuilderCreate(LLVMModuleRef M) { + return new DIBuilder(*unwrap(M)); +} + +extern "C" void LLVMRustDIBuilderDispose(LLVMRustDIBuilderRef Builder) { + delete Builder; +} + +extern "C" void LLVMRustDIBuilderFinalize(LLVMRustDIBuilderRef Builder) { + Builder->finalize(); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateCompileUnit( + LLVMRustDIBuilderRef Builder, unsigned Lang, LLVMMetadataRef FileRef, + const char *Producer, size_t ProducerLen, bool isOptimized, + const char *Flags, unsigned RuntimeVer, const char *SplitName, + size_t SplitNameLen, LLVMRustDebugEmissionKind Kind, uint64_t DWOId, + bool SplitDebugInlining, LLVMRustDebugNameTableKind TableKind) { + auto *File = unwrapDI<DIFile>(FileRef); + + return wrap(Builder->createCompileUnit( + Lang, File, StringRef(Producer, ProducerLen), isOptimized, Flags, + RuntimeVer, StringRef(SplitName, SplitNameLen), fromRust(Kind), DWOId, + SplitDebugInlining, false, fromRust(TableKind))); +} + +extern "C" LLVMMetadataRef +LLVMRustDIBuilderCreateFile(LLVMRustDIBuilderRef Builder, const char *Filename, + size_t FilenameLen, const char *Directory, + size_t DirectoryLen, LLVMRustChecksumKind CSKind, + const char *Checksum, size_t ChecksumLen) { + + std::optional<DIFile::ChecksumKind> llvmCSKind = fromRust(CSKind); + std::optional<DIFile::ChecksumInfo<StringRef>> CSInfo{}; + if (llvmCSKind) + CSInfo.emplace(*llvmCSKind, StringRef{Checksum, ChecksumLen}); + return wrap(Builder->createFile(StringRef(Filename, FilenameLen), + StringRef(Directory, DirectoryLen), CSInfo)); +} + +extern "C" LLVMMetadataRef +LLVMRustDIBuilderCreateSubroutineType(LLVMRustDIBuilderRef Builder, + LLVMMetadataRef ParameterTypes) { + return wrap(Builder->createSubroutineType( + DITypeRefArray(unwrap<MDTuple>(ParameterTypes)))); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateFunction( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Scope, const char *Name, + size_t NameLen, const char *LinkageName, size_t LinkageNameLen, + LLVMMetadataRef File, unsigned LineNo, LLVMMetadataRef Ty, + unsigned ScopeLine, LLVMRustDIFlags Flags, LLVMRustDISPFlags SPFlags, + LLVMValueRef MaybeFn, LLVMMetadataRef TParam, LLVMMetadataRef Decl) { + DITemplateParameterArray TParams = + DITemplateParameterArray(unwrap<MDTuple>(TParam)); + DISubprogram::DISPFlags llvmSPFlags = fromRust(SPFlags); + DINode::DIFlags llvmFlags = fromRust(Flags); + DISubprogram *Sub = Builder->createFunction( + unwrapDI<DIScope>(Scope), StringRef(Name, NameLen), + StringRef(LinkageName, LinkageNameLen), unwrapDI<DIFile>(File), LineNo, + unwrapDI<DISubroutineType>(Ty), ScopeLine, llvmFlags, llvmSPFlags, + TParams, unwrapDIPtr<DISubprogram>(Decl)); + if (MaybeFn) + unwrap<Function>(MaybeFn)->setSubprogram(Sub); + return wrap(Sub); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateMethod( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Scope, const char *Name, + size_t NameLen, const char *LinkageName, size_t LinkageNameLen, + LLVMMetadataRef File, unsigned LineNo, LLVMMetadataRef Ty, + LLVMRustDIFlags Flags, LLVMRustDISPFlags SPFlags, LLVMMetadataRef TParam) { + DITemplateParameterArray TParams = + DITemplateParameterArray(unwrap<MDTuple>(TParam)); + DISubprogram::DISPFlags llvmSPFlags = fromRust(SPFlags); + DINode::DIFlags llvmFlags = fromRust(Flags); + DISubprogram *Sub = Builder->createMethod( + unwrapDI<DIScope>(Scope), StringRef(Name, NameLen), + StringRef(LinkageName, LinkageNameLen), unwrapDI<DIFile>(File), LineNo, + unwrapDI<DISubroutineType>(Ty), 0, 0, + nullptr, // VTable params aren't used + llvmFlags, llvmSPFlags, TParams); + return wrap(Sub); +} + +extern "C" LLVMMetadataRef +LLVMRustDIBuilderCreateBasicType(LLVMRustDIBuilderRef Builder, const char *Name, + size_t NameLen, uint64_t SizeInBits, + unsigned Encoding) { + return wrap( + Builder->createBasicType(StringRef(Name, NameLen), SizeInBits, Encoding)); +} + +extern "C" LLVMMetadataRef +LLVMRustDIBuilderCreateTypedef(LLVMRustDIBuilderRef Builder, + LLVMMetadataRef Type, const char *Name, + size_t NameLen, LLVMMetadataRef File, + unsigned LineNo, LLVMMetadataRef Scope) { + return wrap(Builder->createTypedef( + unwrap<DIType>(Type), StringRef(Name, NameLen), unwrap<DIFile>(File), + LineNo, unwrapDIPtr<DIScope>(Scope))); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreatePointerType( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef PointeeTy, + uint64_t SizeInBits, uint32_t AlignInBits, unsigned AddressSpace, + const char *Name, size_t NameLen) { + return wrap(Builder->createPointerType(unwrapDI<DIType>(PointeeTy), + SizeInBits, AlignInBits, AddressSpace, + StringRef(Name, NameLen))); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateStructType( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Scope, const char *Name, + size_t NameLen, LLVMMetadataRef File, unsigned LineNumber, + uint64_t SizeInBits, uint32_t AlignInBits, LLVMRustDIFlags Flags, + LLVMMetadataRef DerivedFrom, LLVMMetadataRef Elements, unsigned RunTimeLang, + LLVMMetadataRef VTableHolder, const char *UniqueId, size_t UniqueIdLen) { + return wrap(Builder->createStructType( + unwrapDI<DIDescriptor>(Scope), StringRef(Name, NameLen), + unwrapDI<DIFile>(File), LineNumber, SizeInBits, AlignInBits, + fromRust(Flags), unwrapDI<DIType>(DerivedFrom), + DINodeArray(unwrapDI<MDTuple>(Elements)), RunTimeLang, + unwrapDI<DIType>(VTableHolder), StringRef(UniqueId, UniqueIdLen))); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateVariantPart( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Scope, const char *Name, + size_t NameLen, LLVMMetadataRef File, unsigned LineNumber, + uint64_t SizeInBits, uint32_t AlignInBits, LLVMRustDIFlags Flags, + LLVMMetadataRef Discriminator, LLVMMetadataRef Elements, + const char *UniqueId, size_t UniqueIdLen) { + return wrap(Builder->createVariantPart( + unwrapDI<DIDescriptor>(Scope), StringRef(Name, NameLen), + unwrapDI<DIFile>(File), LineNumber, SizeInBits, AlignInBits, + fromRust(Flags), unwrapDI<DIDerivedType>(Discriminator), + DINodeArray(unwrapDI<MDTuple>(Elements)), + StringRef(UniqueId, UniqueIdLen))); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateMemberType( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Scope, const char *Name, + size_t NameLen, LLVMMetadataRef File, unsigned LineNo, uint64_t SizeInBits, + uint32_t AlignInBits, uint64_t OffsetInBits, LLVMRustDIFlags Flags, + LLVMMetadataRef Ty) { + return wrap(Builder->createMemberType( + unwrapDI<DIDescriptor>(Scope), StringRef(Name, NameLen), + unwrapDI<DIFile>(File), LineNo, SizeInBits, AlignInBits, OffsetInBits, + fromRust(Flags), unwrapDI<DIType>(Ty))); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateVariantMemberType( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Scope, const char *Name, + size_t NameLen, LLVMMetadataRef File, unsigned LineNo, uint64_t SizeInBits, + uint32_t AlignInBits, uint64_t OffsetInBits, LLVMValueRef Discriminant, + LLVMRustDIFlags Flags, LLVMMetadataRef Ty) { + llvm::ConstantInt *D = nullptr; + if (Discriminant) { + D = unwrap<llvm::ConstantInt>(Discriminant); + } + return wrap(Builder->createVariantMemberType( + unwrapDI<DIDescriptor>(Scope), StringRef(Name, NameLen), + unwrapDI<DIFile>(File), LineNo, SizeInBits, AlignInBits, OffsetInBits, D, + fromRust(Flags), unwrapDI<DIType>(Ty))); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateStaticMemberType( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Scope, const char *Name, + size_t NameLen, LLVMMetadataRef File, unsigned LineNo, LLVMMetadataRef Ty, + LLVMRustDIFlags Flags, LLVMValueRef val, uint32_t AlignInBits) { + return wrap(Builder->createStaticMemberType( + unwrapDI<DIDescriptor>(Scope), StringRef(Name, NameLen), + unwrapDI<DIFile>(File), LineNo, unwrapDI<DIType>(Ty), fromRust(Flags), + unwrap<llvm::ConstantInt>(val), +#if LLVM_VERSION_GE(18, 0) + llvm::dwarf::DW_TAG_member, +#endif + AlignInBits)); +} + +extern "C" LLVMMetadataRef +LLVMRustDIBuilderCreateLexicalBlock(LLVMRustDIBuilderRef Builder, + LLVMMetadataRef Scope, LLVMMetadataRef File, + unsigned Line, unsigned Col) { + return wrap(Builder->createLexicalBlock(unwrapDI<DIDescriptor>(Scope), + unwrapDI<DIFile>(File), Line, Col)); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateLexicalBlockFile( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Scope, LLVMMetadataRef File) { + return wrap(Builder->createLexicalBlockFile(unwrapDI<DIDescriptor>(Scope), + unwrapDI<DIFile>(File))); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateStaticVariable( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Context, const char *Name, + size_t NameLen, const char *LinkageName, size_t LinkageNameLen, + LLVMMetadataRef File, unsigned LineNo, LLVMMetadataRef Ty, + bool IsLocalToUnit, LLVMValueRef V, LLVMMetadataRef Decl = nullptr, + uint32_t AlignInBits = 0) { + llvm::GlobalVariable *InitVal = cast<llvm::GlobalVariable>(unwrap(V)); + + llvm::DIExpression *InitExpr = nullptr; + if (llvm::ConstantInt *IntVal = llvm::dyn_cast<llvm::ConstantInt>(InitVal)) { + InitExpr = Builder->createConstantValueExpression( + IntVal->getValue().getSExtValue()); + } else if (llvm::ConstantFP *FPVal = + llvm::dyn_cast<llvm::ConstantFP>(InitVal)) { + InitExpr = Builder->createConstantValueExpression( + FPVal->getValueAPF().bitcastToAPInt().getZExtValue()); + } + + llvm::DIGlobalVariableExpression *VarExpr = + Builder->createGlobalVariableExpression( + unwrapDI<DIDescriptor>(Context), StringRef(Name, NameLen), + StringRef(LinkageName, LinkageNameLen), unwrapDI<DIFile>(File), + LineNo, unwrapDI<DIType>(Ty), IsLocalToUnit, + /* isDefined */ true, InitExpr, unwrapDIPtr<MDNode>(Decl), + /* templateParams */ nullptr, AlignInBits); + + InitVal->setMetadata("dbg", VarExpr); + + return wrap(VarExpr); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateVariable( + LLVMRustDIBuilderRef Builder, unsigned Tag, LLVMMetadataRef Scope, + const char *Name, size_t NameLen, LLVMMetadataRef File, unsigned LineNo, + LLVMMetadataRef Ty, bool AlwaysPreserve, LLVMRustDIFlags Flags, + unsigned ArgNo, uint32_t AlignInBits) { + if (Tag == 0x100) { // DW_TAG_auto_variable + return wrap(Builder->createAutoVariable( + unwrapDI<DIDescriptor>(Scope), StringRef(Name, NameLen), + unwrapDI<DIFile>(File), LineNo, unwrapDI<DIType>(Ty), AlwaysPreserve, + fromRust(Flags), AlignInBits)); + } else { + return wrap(Builder->createParameterVariable( + unwrapDI<DIDescriptor>(Scope), StringRef(Name, NameLen), ArgNo, + unwrapDI<DIFile>(File), LineNo, unwrapDI<DIType>(Ty), AlwaysPreserve, + fromRust(Flags))); + } +} + +extern "C" LLVMMetadataRef +LLVMRustDIBuilderCreateArrayType(LLVMRustDIBuilderRef Builder, uint64_t Size, + uint32_t AlignInBits, LLVMMetadataRef Ty, + LLVMMetadataRef Subscripts) { + return wrap( + Builder->createArrayType(Size, AlignInBits, unwrapDI<DIType>(Ty), + DINodeArray(unwrapDI<MDTuple>(Subscripts)))); +} + +extern "C" LLVMMetadataRef +LLVMRustDIBuilderGetOrCreateSubrange(LLVMRustDIBuilderRef Builder, int64_t Lo, + int64_t Count) { + return wrap(Builder->getOrCreateSubrange(Lo, Count)); +} + +extern "C" LLVMMetadataRef +LLVMRustDIBuilderGetOrCreateArray(LLVMRustDIBuilderRef Builder, + LLVMMetadataRef *Ptr, unsigned Count) { + Metadata **DataValue = unwrap(Ptr); + return wrap( + Builder->getOrCreateArray(ArrayRef<Metadata *>(DataValue, Count)).get()); +} + +extern "C" void LLVMRustDIBuilderInsertDeclareAtEnd( + LLVMRustDIBuilderRef Builder, LLVMValueRef V, LLVMMetadataRef VarInfo, + uint64_t *AddrOps, unsigned AddrOpsCount, LLVMMetadataRef DL, + LLVMBasicBlockRef InsertAtEnd) { + Builder->insertDeclare(unwrap(V), unwrap<DILocalVariable>(VarInfo), + Builder->createExpression( + llvm::ArrayRef<uint64_t>(AddrOps, AddrOpsCount)), + DebugLoc(cast<MDNode>(unwrap(DL))), + unwrap(InsertAtEnd)); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateEnumerator( + LLVMRustDIBuilderRef Builder, const char *Name, size_t NameLen, + const uint64_t Value[2], unsigned SizeInBits, bool IsUnsigned) { + return wrap(Builder->createEnumerator( + StringRef(Name, NameLen), + APSInt(APInt(SizeInBits, ArrayRef<uint64_t>(Value, 2)), IsUnsigned))); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateEnumerationType( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Scope, const char *Name, + size_t NameLen, LLVMMetadataRef File, unsigned LineNumber, + uint64_t SizeInBits, uint32_t AlignInBits, LLVMMetadataRef Elements, + LLVMMetadataRef ClassTy, bool IsScoped) { + return wrap(Builder->createEnumerationType( + unwrapDI<DIDescriptor>(Scope), StringRef(Name, NameLen), + unwrapDI<DIFile>(File), LineNumber, SizeInBits, AlignInBits, + DINodeArray(unwrapDI<MDTuple>(Elements)), unwrapDI<DIType>(ClassTy), +#if LLVM_VERSION_GE(18, 0) + /* RunTimeLang */ 0, +#endif + "", IsScoped)); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateUnionType( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Scope, const char *Name, + size_t NameLen, LLVMMetadataRef File, unsigned LineNumber, + uint64_t SizeInBits, uint32_t AlignInBits, LLVMRustDIFlags Flags, + LLVMMetadataRef Elements, unsigned RunTimeLang, const char *UniqueId, + size_t UniqueIdLen) { + return wrap(Builder->createUnionType( + unwrapDI<DIDescriptor>(Scope), StringRef(Name, NameLen), + unwrapDI<DIFile>(File), LineNumber, SizeInBits, AlignInBits, + fromRust(Flags), DINodeArray(unwrapDI<MDTuple>(Elements)), RunTimeLang, + StringRef(UniqueId, UniqueIdLen))); +} + +extern "C" LLVMMetadataRef LLVMRustDIBuilderCreateTemplateTypeParameter( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef Scope, const char *Name, + size_t NameLen, LLVMMetadataRef Ty) { + bool IsDefault = false; // FIXME: should we ever set this true? + return wrap(Builder->createTemplateTypeParameter( + unwrapDI<DIDescriptor>(Scope), StringRef(Name, NameLen), + unwrapDI<DIType>(Ty), IsDefault)); +} + +extern "C" LLVMMetadataRef +LLVMRustDIBuilderCreateNameSpace(LLVMRustDIBuilderRef Builder, + LLVMMetadataRef Scope, const char *Name, + size_t NameLen, bool ExportSymbols) { + return wrap(Builder->createNameSpace( + unwrapDI<DIDescriptor>(Scope), StringRef(Name, NameLen), ExportSymbols)); +} + +extern "C" void LLVMRustDICompositeTypeReplaceArrays( + LLVMRustDIBuilderRef Builder, LLVMMetadataRef CompositeTy, + LLVMMetadataRef Elements, LLVMMetadataRef Params) { + DICompositeType *Tmp = unwrapDI<DICompositeType>(CompositeTy); + Builder->replaceArrays(Tmp, DINodeArray(unwrap<MDTuple>(Elements)), + DINodeArray(unwrap<MDTuple>(Params))); +} + +extern "C" LLVMMetadataRef +LLVMRustDIBuilderCreateDebugLocation(unsigned Line, unsigned Column, + LLVMMetadataRef ScopeRef, + LLVMMetadataRef InlinedAt) { + MDNode *Scope = unwrapDIPtr<MDNode>(ScopeRef); + DILocation *Loc = DILocation::get(Scope->getContext(), Line, Column, Scope, + unwrapDIPtr<MDNode>(InlinedAt)); + return wrap(Loc); +} + +extern "C" uint64_t LLVMRustDIBuilderCreateOpDeref() { + return dwarf::DW_OP_deref; +} + +extern "C" uint64_t LLVMRustDIBuilderCreateOpPlusUconst() { + return dwarf::DW_OP_plus_uconst; +} + +extern "C" int64_t LLVMRustDIBuilderCreateOpLLVMFragment() { + return dwarf::DW_OP_LLVM_fragment; +} + +extern "C" void LLVMRustWriteTypeToString(LLVMTypeRef Ty, RustStringRef Str) { + auto OS = RawRustStringOstream(Str); + unwrap<llvm::Type>(Ty)->print(OS); +} + +extern "C" void LLVMRustWriteValueToString(LLVMValueRef V, RustStringRef Str) { + auto OS = RawRustStringOstream(Str); + if (!V) { + OS << "(null)"; + } else { + OS << "("; + unwrap<llvm::Value>(V)->getType()->print(OS); + OS << ":"; + unwrap<llvm::Value>(V)->print(OS); + OS << ")"; + } +} + +DEFINE_SIMPLE_CONVERSION_FUNCTIONS(Twine, LLVMTwineRef) + +extern "C" void LLVMRustWriteTwineToString(LLVMTwineRef T, RustStringRef Str) { + auto OS = RawRustStringOstream(Str); + unwrap(T)->print(OS); +} + +extern "C" void LLVMRustUnpackOptimizationDiagnostic( + LLVMDiagnosticInfoRef DI, RustStringRef PassNameOut, + LLVMValueRef *FunctionOut, unsigned *Line, unsigned *Column, + RustStringRef FilenameOut, RustStringRef MessageOut) { + // Undefined to call this not on an optimization diagnostic! + llvm::DiagnosticInfoOptimizationBase *Opt = + static_cast<llvm::DiagnosticInfoOptimizationBase *>(unwrap(DI)); + + auto PassNameOS = RawRustStringOstream(PassNameOut); + PassNameOS << Opt->getPassName(); + *FunctionOut = wrap(&Opt->getFunction()); + + auto FilenameOS = RawRustStringOstream(FilenameOut); + DiagnosticLocation loc = Opt->getLocation(); + if (loc.isValid()) { + *Line = loc.getLine(); + *Column = loc.getColumn(); + FilenameOS << loc.getAbsolutePath(); + } + + auto MessageOS = RawRustStringOstream(MessageOut); + MessageOS << Opt->getMsg(); +} + +enum class LLVMRustDiagnosticLevel { + Error, + Warning, + Note, + Remark, +}; + +extern "C" void LLVMRustUnpackInlineAsmDiagnostic( + LLVMDiagnosticInfoRef DI, LLVMRustDiagnosticLevel *LevelOut, + uint64_t *CookieOut, LLVMTwineRef *MessageOut) { + // Undefined to call this not on an inline assembly diagnostic! + llvm::DiagnosticInfoInlineAsm *IA = + static_cast<llvm::DiagnosticInfoInlineAsm *>(unwrap(DI)); + + *CookieOut = IA->getLocCookie(); + *MessageOut = wrap(&IA->getMsgStr()); + + switch (IA->getSeverity()) { + case DS_Error: + *LevelOut = LLVMRustDiagnosticLevel::Error; + break; + case DS_Warning: + *LevelOut = LLVMRustDiagnosticLevel::Warning; + break; + case DS_Note: + *LevelOut = LLVMRustDiagnosticLevel::Note; + break; + case DS_Remark: + *LevelOut = LLVMRustDiagnosticLevel::Remark; + break; + default: + report_fatal_error("Invalid LLVMRustDiagnosticLevel value!"); + } +} + +extern "C" void LLVMRustWriteDiagnosticInfoToString(LLVMDiagnosticInfoRef DI, + RustStringRef Str) { + auto OS = RawRustStringOstream(Str); + auto DP = DiagnosticPrinterRawOStream(OS); + unwrap(DI)->print(DP); +} + +enum class LLVMRustDiagnosticKind { + Other, + InlineAsm, + StackSize, + DebugMetadataVersion, + SampleProfile, + OptimizationRemark, + OptimizationRemarkMissed, + OptimizationRemarkAnalysis, + OptimizationRemarkAnalysisFPCommute, + OptimizationRemarkAnalysisAliasing, + OptimizationRemarkOther, + OptimizationFailure, + PGOProfile, + Linker, + Unsupported, + SrcMgr, +}; + +static LLVMRustDiagnosticKind toRust(DiagnosticKind Kind) { + switch (Kind) { + case DK_InlineAsm: + return LLVMRustDiagnosticKind::InlineAsm; + case DK_StackSize: + return LLVMRustDiagnosticKind::StackSize; + case DK_DebugMetadataVersion: + return LLVMRustDiagnosticKind::DebugMetadataVersion; + case DK_SampleProfile: + return LLVMRustDiagnosticKind::SampleProfile; + case DK_OptimizationRemark: + case DK_MachineOptimizationRemark: + return LLVMRustDiagnosticKind::OptimizationRemark; + case DK_OptimizationRemarkMissed: + case DK_MachineOptimizationRemarkMissed: + return LLVMRustDiagnosticKind::OptimizationRemarkMissed; + case DK_OptimizationRemarkAnalysis: + case DK_MachineOptimizationRemarkAnalysis: + return LLVMRustDiagnosticKind::OptimizationRemarkAnalysis; + case DK_OptimizationRemarkAnalysisFPCommute: + return LLVMRustDiagnosticKind::OptimizationRemarkAnalysisFPCommute; + case DK_OptimizationRemarkAnalysisAliasing: + return LLVMRustDiagnosticKind::OptimizationRemarkAnalysisAliasing; + case DK_PGOProfile: + return LLVMRustDiagnosticKind::PGOProfile; + case DK_Linker: + return LLVMRustDiagnosticKind::Linker; + case DK_Unsupported: + return LLVMRustDiagnosticKind::Unsupported; + case DK_SrcMgr: + return LLVMRustDiagnosticKind::SrcMgr; + default: + return (Kind >= DK_FirstRemark && Kind <= DK_LastRemark) + ? LLVMRustDiagnosticKind::OptimizationRemarkOther + : LLVMRustDiagnosticKind::Other; + } +} + +extern "C" LLVMRustDiagnosticKind +LLVMRustGetDiagInfoKind(LLVMDiagnosticInfoRef DI) { + return toRust((DiagnosticKind)unwrap(DI)->getKind()); +} + +// This is kept distinct from LLVMGetTypeKind, because when +// a new type kind is added, the Rust-side enum must be +// updated or UB will result. +extern "C" LLVMTypeKind LLVMRustGetTypeKind(LLVMTypeRef Ty) { + switch (unwrap(Ty)->getTypeID()) { + case Type::VoidTyID: + return LLVMVoidTypeKind; + case Type::HalfTyID: + return LLVMHalfTypeKind; + case Type::FloatTyID: + return LLVMFloatTypeKind; + case Type::DoubleTyID: + return LLVMDoubleTypeKind; + case Type::X86_FP80TyID: + return LLVMX86_FP80TypeKind; + case Type::FP128TyID: + return LLVMFP128TypeKind; + case Type::PPC_FP128TyID: + return LLVMPPC_FP128TypeKind; + case Type::LabelTyID: + return LLVMLabelTypeKind; + case Type::MetadataTyID: + return LLVMMetadataTypeKind; + case Type::IntegerTyID: + return LLVMIntegerTypeKind; + case Type::FunctionTyID: + return LLVMFunctionTypeKind; + case Type::StructTyID: + return LLVMStructTypeKind; + case Type::ArrayTyID: + return LLVMArrayTypeKind; + case Type::PointerTyID: + return LLVMPointerTypeKind; + case Type::FixedVectorTyID: + return LLVMVectorTypeKind; + case Type::X86_MMXTyID: + return LLVMX86_MMXTypeKind; + case Type::TokenTyID: + return LLVMTokenTypeKind; + case Type::ScalableVectorTyID: + return LLVMScalableVectorTypeKind; + case Type::BFloatTyID: + return LLVMBFloatTypeKind; + case Type::X86_AMXTyID: + return LLVMX86_AMXTypeKind; + default: { + std::string error; + auto stream = llvm::raw_string_ostream(error); + stream << "Rust does not support the TypeID: " << unwrap(Ty)->getTypeID() + << " for the type: " << *unwrap(Ty); + stream.flush(); + report_fatal_error(error.c_str()); + } + } +} + +DEFINE_SIMPLE_CONVERSION_FUNCTIONS(SMDiagnostic, LLVMSMDiagnosticRef) + +extern "C" LLVMSMDiagnosticRef LLVMRustGetSMDiagnostic(LLVMDiagnosticInfoRef DI, + unsigned *Cookie) { + llvm::DiagnosticInfoSrcMgr *SM = + static_cast<llvm::DiagnosticInfoSrcMgr *>(unwrap(DI)); + *Cookie = SM->getLocCookie(); + return wrap(&SM->getSMDiag()); +} + +extern "C" bool +LLVMRustUnpackSMDiagnostic(LLVMSMDiagnosticRef DRef, RustStringRef MessageOut, + RustStringRef BufferOut, + LLVMRustDiagnosticLevel *LevelOut, unsigned *LocOut, + unsigned *RangesOut, size_t *NumRanges) { + SMDiagnostic &D = *unwrap(DRef); + auto MessageOS = RawRustStringOstream(MessageOut); + MessageOS << D.getMessage(); + + switch (D.getKind()) { + case SourceMgr::DK_Error: + *LevelOut = LLVMRustDiagnosticLevel::Error; + break; + case SourceMgr::DK_Warning: + *LevelOut = LLVMRustDiagnosticLevel::Warning; + break; + case SourceMgr::DK_Note: + *LevelOut = LLVMRustDiagnosticLevel::Note; + break; + case SourceMgr::DK_Remark: + *LevelOut = LLVMRustDiagnosticLevel::Remark; + break; + default: + report_fatal_error("Invalid LLVMRustDiagnosticLevel value!"); + } + + if (D.getLoc() == SMLoc()) + return false; + + const SourceMgr &LSM = *D.getSourceMgr(); + const MemoryBuffer *LBuf = + LSM.getMemoryBuffer(LSM.FindBufferContainingLoc(D.getLoc())); + LLVMRustStringWriteImpl(BufferOut, LBuf->getBufferStart(), + LBuf->getBufferSize()); + + *LocOut = D.getLoc().getPointer() - LBuf->getBufferStart(); + + *NumRanges = std::min(*NumRanges, D.getRanges().size()); + size_t LineStart = *LocOut - (size_t)D.getColumnNo(); + for (size_t i = 0; i < *NumRanges; i++) { + RangesOut[i * 2] = LineStart + D.getRanges()[i].first; + RangesOut[i * 2 + 1] = LineStart + D.getRanges()[i].second; + } + + return true; +} + +extern "C" OperandBundleDef *LLVMRustBuildOperandBundleDef(const char *Name, + LLVMValueRef *Inputs, + unsigned NumInputs) { + return new OperandBundleDef(Name, + ArrayRef<Value *>(unwrap(Inputs), NumInputs)); +} + +extern "C" void LLVMRustFreeOperandBundleDef(OperandBundleDef *Bundle) { + delete Bundle; +} + +// OpBundlesIndirect is an array of pointers (*not* a pointer to an array). +extern "C" LLVMValueRef LLVMRustBuildCall(LLVMBuilderRef B, LLVMTypeRef Ty, + LLVMValueRef Fn, LLVMValueRef *Args, + unsigned NumArgs, + OperandBundleDef **OpBundlesIndirect, + unsigned NumOpBundles) { + Value *Callee = unwrap(Fn); + FunctionType *FTy = unwrap<FunctionType>(Ty); + + // FIXME: Is there a way around this? + SmallVector<OperandBundleDef> OpBundles; + OpBundles.reserve(NumOpBundles); + for (unsigned i = 0; i < NumOpBundles; ++i) { + OpBundles.push_back(*OpBundlesIndirect[i]); + } + + return wrap(unwrap(B)->CreateCall(FTy, Callee, + ArrayRef<Value *>(unwrap(Args), NumArgs), + ArrayRef<OperandBundleDef>(OpBundles))); +} + +extern "C" LLVMValueRef +LLVMRustGetInstrProfIncrementIntrinsic(LLVMModuleRef M) { + return wrap(llvm::Intrinsic::getDeclaration( + unwrap(M), llvm::Intrinsic::instrprof_increment)); +} + +extern "C" LLVMValueRef +LLVMRustGetInstrProfMCDCParametersIntrinsic(LLVMModuleRef M) { +#if LLVM_VERSION_GE(18, 0) + return wrap(llvm::Intrinsic::getDeclaration( + unwrap(M), llvm::Intrinsic::instrprof_mcdc_parameters)); +#else + report_fatal_error("LLVM 18.0 is required for mcdc intrinsic functions"); +#endif +} + +extern "C" LLVMValueRef +LLVMRustGetInstrProfMCDCTVBitmapUpdateIntrinsic(LLVMModuleRef M) { +#if LLVM_VERSION_GE(18, 0) + return wrap(llvm::Intrinsic::getDeclaration( + unwrap(M), llvm::Intrinsic::instrprof_mcdc_tvbitmap_update)); +#else + report_fatal_error("LLVM 18.0 is required for mcdc intrinsic functions"); +#endif +} + +extern "C" LLVMValueRef +LLVMRustGetInstrProfMCDCCondBitmapIntrinsic(LLVMModuleRef M) { +#if LLVM_VERSION_GE(18, 0) + return wrap(llvm::Intrinsic::getDeclaration( + unwrap(M), llvm::Intrinsic::instrprof_mcdc_condbitmap_update)); +#else + report_fatal_error("LLVM 18.0 is required for mcdc intrinsic functions"); +#endif +} + +extern "C" LLVMValueRef LLVMRustBuildMemCpy(LLVMBuilderRef B, LLVMValueRef Dst, + unsigned DstAlign, LLVMValueRef Src, + unsigned SrcAlign, + LLVMValueRef Size, + bool IsVolatile) { + return wrap(unwrap(B)->CreateMemCpy(unwrap(Dst), MaybeAlign(DstAlign), + unwrap(Src), MaybeAlign(SrcAlign), + unwrap(Size), IsVolatile)); +} + +extern "C" LLVMValueRef +LLVMRustBuildMemMove(LLVMBuilderRef B, LLVMValueRef Dst, unsigned DstAlign, + LLVMValueRef Src, unsigned SrcAlign, LLVMValueRef Size, + bool IsVolatile) { + return wrap(unwrap(B)->CreateMemMove(unwrap(Dst), MaybeAlign(DstAlign), + unwrap(Src), MaybeAlign(SrcAlign), + unwrap(Size), IsVolatile)); +} + +extern "C" LLVMValueRef LLVMRustBuildMemSet(LLVMBuilderRef B, LLVMValueRef Dst, + unsigned DstAlign, LLVMValueRef Val, + LLVMValueRef Size, + bool IsVolatile) { + return wrap(unwrap(B)->CreateMemSet(unwrap(Dst), unwrap(Val), unwrap(Size), + MaybeAlign(DstAlign), IsVolatile)); +} + +// OpBundlesIndirect is an array of pointers (*not* a pointer to an array). +extern "C" LLVMValueRef +LLVMRustBuildInvoke(LLVMBuilderRef B, LLVMTypeRef Ty, LLVMValueRef Fn, + LLVMValueRef *Args, unsigned NumArgs, + LLVMBasicBlockRef Then, LLVMBasicBlockRef Catch, + OperandBundleDef **OpBundlesIndirect, unsigned NumOpBundles, + const char *Name) { + Value *Callee = unwrap(Fn); + FunctionType *FTy = unwrap<FunctionType>(Ty); + + // FIXME: Is there a way around this? + SmallVector<OperandBundleDef> OpBundles; + OpBundles.reserve(NumOpBundles); + for (unsigned i = 0; i < NumOpBundles; ++i) { + OpBundles.push_back(*OpBundlesIndirect[i]); + } + + return wrap(unwrap(B)->CreateInvoke(FTy, Callee, unwrap(Then), unwrap(Catch), + ArrayRef<Value *>(unwrap(Args), NumArgs), + ArrayRef<OperandBundleDef>(OpBundles), + Name)); +} + +// OpBundlesIndirect is an array of pointers (*not* a pointer to an array). +extern "C" LLVMValueRef +LLVMRustBuildCallBr(LLVMBuilderRef B, LLVMTypeRef Ty, LLVMValueRef Fn, + LLVMBasicBlockRef DefaultDest, + LLVMBasicBlockRef *IndirectDests, unsigned NumIndirectDests, + LLVMValueRef *Args, unsigned NumArgs, + OperandBundleDef **OpBundlesIndirect, unsigned NumOpBundles, + const char *Name) { + Value *Callee = unwrap(Fn); + FunctionType *FTy = unwrap<FunctionType>(Ty); + + // FIXME: Is there a way around this? + std::vector<BasicBlock *> IndirectDestsUnwrapped; + IndirectDestsUnwrapped.reserve(NumIndirectDests); + for (unsigned i = 0; i < NumIndirectDests; ++i) { + IndirectDestsUnwrapped.push_back(unwrap(IndirectDests[i])); + } + + // FIXME: Is there a way around this? + SmallVector<OperandBundleDef> OpBundles; + OpBundles.reserve(NumOpBundles); + for (unsigned i = 0; i < NumOpBundles; ++i) { + OpBundles.push_back(*OpBundlesIndirect[i]); + } + + return wrap( + unwrap(B)->CreateCallBr(FTy, Callee, unwrap(DefaultDest), + ArrayRef<BasicBlock *>(IndirectDestsUnwrapped), + ArrayRef<Value *>(unwrap(Args), NumArgs), + ArrayRef<OperandBundleDef>(OpBundles), Name)); +} + +extern "C" void LLVMRustPositionBuilderAtStart(LLVMBuilderRef B, + LLVMBasicBlockRef BB) { + auto Point = unwrap(BB)->getFirstInsertionPt(); + unwrap(B)->SetInsertPoint(unwrap(BB), Point); +} + +extern "C" void LLVMRustSetComdat(LLVMModuleRef M, LLVMValueRef V, + const char *Name, size_t NameLen) { + Triple TargetTriple = Triple(unwrap(M)->getTargetTriple()); + GlobalObject *GV = unwrap<GlobalObject>(V); + if (TargetTriple.supportsCOMDAT()) { + StringRef NameRef(Name, NameLen); + GV->setComdat(unwrap(M)->getOrInsertComdat(NameRef)); + } +} + +enum class LLVMRustLinkage { + ExternalLinkage = 0, + AvailableExternallyLinkage = 1, + LinkOnceAnyLinkage = 2, + LinkOnceODRLinkage = 3, + WeakAnyLinkage = 4, + WeakODRLinkage = 5, + AppendingLinkage = 6, + InternalLinkage = 7, + PrivateLinkage = 8, + ExternalWeakLinkage = 9, + CommonLinkage = 10, +}; + +static LLVMRustLinkage toRust(LLVMLinkage Linkage) { + switch (Linkage) { + case LLVMExternalLinkage: + return LLVMRustLinkage::ExternalLinkage; + case LLVMAvailableExternallyLinkage: + return LLVMRustLinkage::AvailableExternallyLinkage; + case LLVMLinkOnceAnyLinkage: + return LLVMRustLinkage::LinkOnceAnyLinkage; + case LLVMLinkOnceODRLinkage: + return LLVMRustLinkage::LinkOnceODRLinkage; + case LLVMWeakAnyLinkage: + return LLVMRustLinkage::WeakAnyLinkage; + case LLVMWeakODRLinkage: + return LLVMRustLinkage::WeakODRLinkage; + case LLVMAppendingLinkage: + return LLVMRustLinkage::AppendingLinkage; + case LLVMInternalLinkage: + return LLVMRustLinkage::InternalLinkage; + case LLVMPrivateLinkage: + return LLVMRustLinkage::PrivateLinkage; + case LLVMExternalWeakLinkage: + return LLVMRustLinkage::ExternalWeakLinkage; + case LLVMCommonLinkage: + return LLVMRustLinkage::CommonLinkage; + default: + report_fatal_error("Invalid LLVMRustLinkage value!"); + } +} + +static LLVMLinkage fromRust(LLVMRustLinkage Linkage) { + switch (Linkage) { + case LLVMRustLinkage::ExternalLinkage: + return LLVMExternalLinkage; + case LLVMRustLinkage::AvailableExternallyLinkage: + return LLVMAvailableExternallyLinkage; + case LLVMRustLinkage::LinkOnceAnyLinkage: + return LLVMLinkOnceAnyLinkage; + case LLVMRustLinkage::LinkOnceODRLinkage: + return LLVMLinkOnceODRLinkage; + case LLVMRustLinkage::WeakAnyLinkage: + return LLVMWeakAnyLinkage; + case LLVMRustLinkage::WeakODRLinkage: + return LLVMWeakODRLinkage; + case LLVMRustLinkage::AppendingLinkage: + return LLVMAppendingLinkage; + case LLVMRustLinkage::InternalLinkage: + return LLVMInternalLinkage; + case LLVMRustLinkage::PrivateLinkage: + return LLVMPrivateLinkage; + case LLVMRustLinkage::ExternalWeakLinkage: + return LLVMExternalWeakLinkage; + case LLVMRustLinkage::CommonLinkage: + return LLVMCommonLinkage; + } + report_fatal_error("Invalid LLVMRustLinkage value!"); +} + +extern "C" LLVMRustLinkage LLVMRustGetLinkage(LLVMValueRef V) { + return toRust(LLVMGetLinkage(V)); +} + +extern "C" void LLVMRustSetLinkage(LLVMValueRef V, + LLVMRustLinkage RustLinkage) { + LLVMSetLinkage(V, fromRust(RustLinkage)); +} + +extern "C" bool LLVMRustConstIntGetZExtValue(LLVMValueRef CV, uint64_t *value) { + auto C = unwrap<llvm::ConstantInt>(CV); + if (C->getBitWidth() > 64) + return false; + *value = C->getZExtValue(); + return true; +} + +// Returns true if both high and low were successfully set. Fails in case +// constant wasn’t any of the common sizes (1, 8, 16, 32, 64, 128 bits) +extern "C" bool LLVMRustConstInt128Get(LLVMValueRef CV, bool sext, + uint64_t *high, uint64_t *low) { + auto C = unwrap<llvm::ConstantInt>(CV); + if (C->getBitWidth() > 128) { + return false; + } + APInt AP; + if (sext) { + AP = C->getValue().sext(128); + } else { + AP = C->getValue().zext(128); + } + *low = AP.getLoBits(64).getZExtValue(); + *high = AP.getHiBits(64).getZExtValue(); + return true; +} + +enum class LLVMRustVisibility { + Default = 0, + Hidden = 1, + Protected = 2, +}; + +static LLVMRustVisibility toRust(LLVMVisibility Vis) { + switch (Vis) { + case LLVMDefaultVisibility: + return LLVMRustVisibility::Default; + case LLVMHiddenVisibility: + return LLVMRustVisibility::Hidden; + case LLVMProtectedVisibility: + return LLVMRustVisibility::Protected; + } + report_fatal_error("Invalid LLVMRustVisibility value!"); +} + +static LLVMVisibility fromRust(LLVMRustVisibility Vis) { + switch (Vis) { + case LLVMRustVisibility::Default: + return LLVMDefaultVisibility; + case LLVMRustVisibility::Hidden: + return LLVMHiddenVisibility; + case LLVMRustVisibility::Protected: + return LLVMProtectedVisibility; + } + report_fatal_error("Invalid LLVMRustVisibility value!"); +} + +extern "C" LLVMRustVisibility LLVMRustGetVisibility(LLVMValueRef V) { + return toRust(LLVMGetVisibility(V)); +} + +extern "C" void LLVMRustSetVisibility(LLVMValueRef V, + LLVMRustVisibility RustVisibility) { + LLVMSetVisibility(V, fromRust(RustVisibility)); +} + +extern "C" void LLVMRustSetDSOLocal(LLVMValueRef Global, bool is_dso_local) { + unwrap<GlobalValue>(Global)->setDSOLocal(is_dso_local); +} + +struct LLVMRustModuleBuffer { + std::string data; +}; + +extern "C" LLVMRustModuleBuffer *LLVMRustModuleBufferCreate(LLVMModuleRef M) { + auto Ret = std::make_unique<LLVMRustModuleBuffer>(); + { + auto OS = raw_string_ostream(Ret->data); + WriteBitcodeToFile(*unwrap(M), OS); + } + return Ret.release(); +} + +extern "C" void LLVMRustModuleBufferFree(LLVMRustModuleBuffer *Buffer) { + delete Buffer; +} + +extern "C" const void * +LLVMRustModuleBufferPtr(const LLVMRustModuleBuffer *Buffer) { + return Buffer->data.data(); +} + +extern "C" size_t LLVMRustModuleBufferLen(const LLVMRustModuleBuffer *Buffer) { + return Buffer->data.length(); +} + +extern "C" uint64_t LLVMRustModuleCost(LLVMModuleRef M) { + auto f = unwrap(M)->functions(); + return std::distance(std::begin(f), std::end(f)); +} + +extern "C" void LLVMRustModuleInstructionStats(LLVMModuleRef M, + RustStringRef Str) { + auto OS = RawRustStringOstream(Str); + auto JOS = llvm::json::OStream(OS); + auto Module = unwrap(M); + + JOS.object([&] { + JOS.attribute("module", Module->getName()); + JOS.attribute("total", Module->getInstructionCount()); + }); +} + +// Vector reductions: +extern "C" LLVMValueRef LLVMRustBuildVectorReduceFAdd(LLVMBuilderRef B, + LLVMValueRef Acc, + LLVMValueRef Src) { + return wrap(unwrap(B)->CreateFAddReduce(unwrap(Acc), unwrap(Src))); +} +extern "C" LLVMValueRef LLVMRustBuildVectorReduceFMul(LLVMBuilderRef B, + LLVMValueRef Acc, + LLVMValueRef Src) { + return wrap(unwrap(B)->CreateFMulReduce(unwrap(Acc), unwrap(Src))); +} +extern "C" LLVMValueRef LLVMRustBuildVectorReduceAdd(LLVMBuilderRef B, + LLVMValueRef Src) { + return wrap(unwrap(B)->CreateAddReduce(unwrap(Src))); +} +extern "C" LLVMValueRef LLVMRustBuildVectorReduceMul(LLVMBuilderRef B, + LLVMValueRef Src) { + return wrap(unwrap(B)->CreateMulReduce(unwrap(Src))); +} +extern "C" LLVMValueRef LLVMRustBuildVectorReduceAnd(LLVMBuilderRef B, + LLVMValueRef Src) { + return wrap(unwrap(B)->CreateAndReduce(unwrap(Src))); +} +extern "C" LLVMValueRef LLVMRustBuildVectorReduceOr(LLVMBuilderRef B, + LLVMValueRef Src) { + return wrap(unwrap(B)->CreateOrReduce(unwrap(Src))); +} +extern "C" LLVMValueRef LLVMRustBuildVectorReduceXor(LLVMBuilderRef B, + LLVMValueRef Src) { + return wrap(unwrap(B)->CreateXorReduce(unwrap(Src))); +} +extern "C" LLVMValueRef LLVMRustBuildVectorReduceMin(LLVMBuilderRef B, + LLVMValueRef Src, + bool IsSigned) { + return wrap(unwrap(B)->CreateIntMinReduce(unwrap(Src), IsSigned)); +} +extern "C" LLVMValueRef LLVMRustBuildVectorReduceMax(LLVMBuilderRef B, + LLVMValueRef Src, + bool IsSigned) { + return wrap(unwrap(B)->CreateIntMaxReduce(unwrap(Src), IsSigned)); +} +extern "C" LLVMValueRef +LLVMRustBuildVectorReduceFMin(LLVMBuilderRef B, LLVMValueRef Src, bool NoNaN) { + Instruction *I = unwrap(B)->CreateFPMinReduce(unwrap(Src)); + I->setHasNoNaNs(NoNaN); + return wrap(I); +} +extern "C" LLVMValueRef +LLVMRustBuildVectorReduceFMax(LLVMBuilderRef B, LLVMValueRef Src, bool NoNaN) { + Instruction *I = unwrap(B)->CreateFPMaxReduce(unwrap(Src)); + I->setHasNoNaNs(NoNaN); + return wrap(I); +} + +extern "C" LLVMValueRef LLVMRustBuildMinNum(LLVMBuilderRef B, LLVMValueRef LHS, + LLVMValueRef RHS) { + return wrap(unwrap(B)->CreateMinNum(unwrap(LHS), unwrap(RHS))); +} +extern "C" LLVMValueRef LLVMRustBuildMaxNum(LLVMBuilderRef B, LLVMValueRef LHS, + LLVMValueRef RHS) { + return wrap(unwrap(B)->CreateMaxNum(unwrap(LHS), unwrap(RHS))); +} + +// This struct contains all necessary info about a symbol exported from a DLL. +struct LLVMRustCOFFShortExport { + const char *name; + bool ordinal_present; + // The value of `ordinal` is only meaningful if `ordinal_present` is true. + uint16_t ordinal; +}; + +// Machine must be a COFF machine type, as defined in PE specs. +extern "C" LLVMRustResult +LLVMRustWriteImportLibrary(const char *ImportName, const char *Path, + const LLVMRustCOFFShortExport *Exports, + size_t NumExports, uint16_t Machine, bool MinGW) { + std::vector<llvm::object::COFFShortExport> ConvertedExports; + ConvertedExports.reserve(NumExports); + + for (size_t i = 0; i < NumExports; ++i) { + bool ordinal_present = Exports[i].ordinal_present; + uint16_t ordinal = ordinal_present ? Exports[i].ordinal : 0; + ConvertedExports.push_back(llvm::object::COFFShortExport{ + Exports[i].name, // Name + std::string{}, // ExtName + std::string{}, // SymbolName + std::string{}, // AliasTarget +#if LLVM_VERSION_GE(19, 0) + std::string{}, // ExportAs +#endif + ordinal, // Ordinal + ordinal_present, // Noname + false, // Data + false, // Private + false // Constant + }); + } + + auto Error = llvm::object::writeImportLibrary( + ImportName, Path, ConvertedExports, + static_cast<llvm::COFF::MachineTypes>(Machine), MinGW); + if (Error) { + std::string errorString; + auto stream = llvm::raw_string_ostream(errorString); + stream << Error; + stream.flush(); + LLVMRustSetLastError(errorString.c_str()); + return LLVMRustResult::Failure; + } else { + return LLVMRustResult::Success; + } +} + +// Transfers ownership of DiagnosticHandler unique_ptr to the caller. +extern "C" DiagnosticHandler * +LLVMRustContextGetDiagnosticHandler(LLVMContextRef C) { + std::unique_ptr<DiagnosticHandler> DH = unwrap(C)->getDiagnosticHandler(); + return DH.release(); +} + +// Sets unique_ptr to object of DiagnosticHandler to provide custom diagnostic +// handling. Ownership of the handler is moved to the LLVMContext. +extern "C" void LLVMRustContextSetDiagnosticHandler(LLVMContextRef C, + DiagnosticHandler *DH) { + unwrap(C)->setDiagnosticHandler(std::unique_ptr<DiagnosticHandler>(DH)); +} + +using LLVMDiagnosticHandlerTy = DiagnosticHandler::DiagnosticHandlerTy; + +// Configures a diagnostic handler that invokes provided callback when a +// backend needs to emit a diagnostic. +// +// When RemarkAllPasses is true, remarks are enabled for all passes. Otherwise +// the RemarkPasses array specifies individual passes for which remarks will be +// enabled. +// +// If RemarkFilePath is not NULL, optimization remarks will be streamed directly +// into this file, bypassing the diagnostics handler. +extern "C" void LLVMRustContextConfigureDiagnosticHandler( + LLVMContextRef C, LLVMDiagnosticHandlerTy DiagnosticHandlerCallback, + void *DiagnosticHandlerContext, bool RemarkAllPasses, + const char *const *RemarkPasses, size_t RemarkPassesLen, + const char *RemarkFilePath, bool PGOAvailable) { + + class RustDiagnosticHandler final : public DiagnosticHandler { + public: + RustDiagnosticHandler( + LLVMDiagnosticHandlerTy DiagnosticHandlerCallback, + void *DiagnosticHandlerContext, bool RemarkAllPasses, + std::vector<std::string> RemarkPasses, + std::unique_ptr<ToolOutputFile> RemarksFile, + std::unique_ptr<llvm::remarks::RemarkStreamer> RemarkStreamer, + std::unique_ptr<LLVMRemarkStreamer> LlvmRemarkStreamer) + : DiagnosticHandlerCallback(DiagnosticHandlerCallback), + DiagnosticHandlerContext(DiagnosticHandlerContext), + RemarkAllPasses(RemarkAllPasses), + RemarkPasses(std::move(RemarkPasses)), + RemarksFile(std::move(RemarksFile)), + RemarkStreamer(std::move(RemarkStreamer)), + LlvmRemarkStreamer(std::move(LlvmRemarkStreamer)) {} + + virtual bool handleDiagnostics(const DiagnosticInfo &DI) override { + // If this diagnostic is one of the optimization remark kinds, we can + // check if it's enabled before emitting it. This can avoid many + // short-lived allocations when unpacking the diagnostic and converting + // its various C++ strings into rust strings. + // FIXME: some diagnostic infos still allocate before we get here, and + // avoiding that would be good in the future. That will require changing a + // few call sites in LLVM. + if (auto *OptDiagBase = dyn_cast<DiagnosticInfoOptimizationBase>(&DI)) { + if (OptDiagBase->isEnabled()) { + if (this->LlvmRemarkStreamer) { + this->LlvmRemarkStreamer->emit(*OptDiagBase); + return true; + } + } else { + return true; + } + } + if (DiagnosticHandlerCallback) { +#if LLVM_VERSION_GE(19, 0) + DiagnosticHandlerCallback(&DI, DiagnosticHandlerContext); +#else + DiagnosticHandlerCallback(DI, DiagnosticHandlerContext); +#endif + return true; + } + return false; + } + + bool isAnalysisRemarkEnabled(StringRef PassName) const override { + return isRemarkEnabled(PassName); + } + + bool isMissedOptRemarkEnabled(StringRef PassName) const override { + return isRemarkEnabled(PassName); + } + + bool isPassedOptRemarkEnabled(StringRef PassName) const override { + return isRemarkEnabled(PassName); + } + + bool isAnyRemarkEnabled() const override { + return RemarkAllPasses || !RemarkPasses.empty(); + } + + private: + bool isRemarkEnabled(StringRef PassName) const { + if (RemarkAllPasses) + return true; + + for (auto &Pass : RemarkPasses) + if (Pass == PassName) + return true; + + return false; + } + + LLVMDiagnosticHandlerTy DiagnosticHandlerCallback = nullptr; + void *DiagnosticHandlerContext = nullptr; + + bool RemarkAllPasses = false; + std::vector<std::string> RemarkPasses; + + // Since LlvmRemarkStreamer contains a pointer to RemarkStreamer, the + // ordering of the three members below is important. + std::unique_ptr<ToolOutputFile> RemarksFile; + std::unique_ptr<llvm::remarks::RemarkStreamer> RemarkStreamer; + std::unique_ptr<LLVMRemarkStreamer> LlvmRemarkStreamer; + }; + + std::vector<std::string> Passes; + for (size_t I = 0; I != RemarkPassesLen; ++I) { + Passes.push_back(RemarkPasses[I]); + } + + // We need to hold onto both the streamers and the opened file + std::unique_ptr<ToolOutputFile> RemarkFile; + std::unique_ptr<llvm::remarks::RemarkStreamer> RemarkStreamer; + std::unique_ptr<LLVMRemarkStreamer> LlvmRemarkStreamer; + + if (RemarkFilePath != nullptr) { + if (PGOAvailable) { + // Enable PGO hotness data for remarks, if available + unwrap(C)->setDiagnosticsHotnessRequested(true); + } + + std::error_code EC; + RemarkFile = std::make_unique<ToolOutputFile>( + RemarkFilePath, EC, llvm::sys::fs::OF_TextWithCRLF); + if (EC) { + std::string Error = std::string("Cannot create remark file: ") + + toString(errorCodeToError(EC)); + report_fatal_error(Twine(Error)); + } + + // Do not delete the file after we gather remarks + RemarkFile->keep(); + + auto RemarkSerializer = remarks::createRemarkSerializer( + llvm::remarks::Format::YAML, remarks::SerializerMode::Separate, + RemarkFile->os()); + if (Error E = RemarkSerializer.takeError()) { + std::string Error = std::string("Cannot create remark serializer: ") + + toString(std::move(E)); + report_fatal_error(Twine(Error)); + } + RemarkStreamer = std::make_unique<llvm::remarks::RemarkStreamer>( + std::move(*RemarkSerializer)); + LlvmRemarkStreamer = std::make_unique<LLVMRemarkStreamer>(*RemarkStreamer); + } + + unwrap(C)->setDiagnosticHandler(std::make_unique<RustDiagnosticHandler>( + DiagnosticHandlerCallback, DiagnosticHandlerContext, RemarkAllPasses, + Passes, std::move(RemarkFile), std::move(RemarkStreamer), + std::move(LlvmRemarkStreamer))); +} + +extern "C" void LLVMRustGetMangledName(LLVMValueRef V, RustStringRef Str) { + auto OS = RawRustStringOstream(Str); + GlobalValue *GV = unwrap<GlobalValue>(V); + Mangler().getNameWithPrefix(OS, GV, true); +} + +extern "C" int32_t LLVMRustGetElementTypeArgIndex(LLVMValueRef CallSite) { + auto *CB = unwrap<CallBase>(CallSite); + switch (CB->getIntrinsicID()) { + case Intrinsic::arm_ldrex: + return 0; + case Intrinsic::arm_strex: + return 1; + } + return -1; +} + +extern "C" bool LLVMRustIsBitcode(char *ptr, size_t len) { + return identify_magic(StringRef(ptr, len)) == file_magic::bitcode; +} + +extern "C" bool LLVMRustIsNonGVFunctionPointerTy(LLVMValueRef V) { + if (unwrap<Value>(V)->getType()->isPointerTy()) { + if (auto *GV = dyn_cast<GlobalValue>(unwrap<Value>(V))) { + if (GV->getValueType()->isFunctionTy()) + return false; + } + return true; + } + return false; +} + +extern "C" bool LLVMRustLLVMHasZlibCompressionForDebugSymbols() { + return llvm::compression::zlib::isAvailable(); +} + +extern "C" bool LLVMRustLLVMHasZstdCompressionForDebugSymbols() { + return llvm::compression::zstd::isAvailable(); +} + +// Operations on composite constants. +// These are clones of LLVM api functions that will become available in future +// releases. They can be removed once Rust's minimum supported LLVM version +// supports them. See https://github.com/rust-lang/rust/issues/121868 See +// https://llvm.org/doxygen/group__LLVMCCoreValueConstantComposite.html + +// FIXME: Remove when Rust's minimum supported LLVM version reaches 19. +// https://github.com/llvm/llvm-project/commit/e1405e4f71c899420ebf8262d5e9745598419df8 +#if LLVM_VERSION_LT(19, 0) +extern "C" LLVMValueRef LLVMConstStringInContext2(LLVMContextRef C, + const char *Str, + size_t Length, + bool DontNullTerminate) { + return wrap(ConstantDataArray::getString(*unwrap(C), StringRef(Str, Length), + !DontNullTerminate)); +} +#endif diff --git a/compiler/rustc_llvm/llvm-wrapper/SuppressLLVMWarnings.h b/compiler/rustc_llvm/llvm-wrapper/SuppressLLVMWarnings.h new file mode 100644 index 00000000000..09ecfd54cd8 --- /dev/null +++ b/compiler/rustc_llvm/llvm-wrapper/SuppressLLVMWarnings.h @@ -0,0 +1,17 @@ +#ifndef _rustc_llvm_SuppressLLVMWarnings_h +#define _rustc_llvm_SuppressLLVMWarnings_h + +// LLVM currently generates many warnings when compiled using MSVC. These +// warnings make it difficult to diagnose real problems when working on C++ +// code, so we suppress them. + +#ifdef _MSC_VER +#pragma warning(disable : 4530) // C++ exception handler used, but unwind + // semantics are not enabled. +#pragma warning( \ + disable : 4624) // 'xxx': destructor was implicitly defined as deleted +#pragma warning( \ + disable : 4244) // conversion from 'xxx' to 'yyy', possible loss of data +#endif + +#endif // _rustc_llvm_SuppressLLVMWarnings_h diff --git a/compiler/rustc_llvm/llvm-wrapper/SymbolWrapper.cpp b/compiler/rustc_llvm/llvm-wrapper/SymbolWrapper.cpp new file mode 100644 index 00000000000..a9d1362a338 --- /dev/null +++ b/compiler/rustc_llvm/llvm-wrapper/SymbolWrapper.cpp @@ -0,0 +1,99 @@ +// Derived from code in LLVM, which is: +// 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 + +// Derived from: +// * https://github.com/llvm/llvm-project/blob/8ef3e895ad8ab1724e2b87cabad1dacdc7a397a3/llvm/include/llvm/Object/ArchiveWriter.h +// * https://github.com/llvm/llvm-project/blob/8ef3e895ad8ab1724e2b87cabad1dacdc7a397a3/llvm/lib/Object/ArchiveWriter.cpp + +#include "SuppressLLVMWarnings.h" +#include "llvm/ADT/SmallString.h" +#include "llvm/IR/LLVMContext.h" +#include "llvm/Object/ObjectFile.h" +#include <llvm/Support/raw_ostream.h> + +using namespace llvm; +using namespace llvm::sys; +using namespace llvm::object; + +static bool isArchiveSymbol(const object::BasicSymbolRef &S) { + Expected<uint32_t> SymFlagsOrErr = S.getFlags(); + if (!SymFlagsOrErr) + // FIXME: Actually report errors helpfully. + report_fatal_error(SymFlagsOrErr.takeError()); + if (*SymFlagsOrErr & object::SymbolRef::SF_FormatSpecific) + return false; + if (!(*SymFlagsOrErr & object::SymbolRef::SF_Global)) + return false; + if (*SymFlagsOrErr & object::SymbolRef::SF_Undefined) + return false; + return true; +} + +typedef void *(*LLVMRustGetSymbolsCallback)(void *, const char *); +typedef void *(*LLVMRustGetSymbolsErrorCallback)(const char *); + +// Note: This is implemented in C++ instead of using the C api from Rust as +// IRObjectFile doesn't implement getSymbolName, only printSymbolName, which is +// inaccessible from the C api. +extern "C" void * +LLVMRustGetSymbols(char *BufPtr, size_t BufLen, void *State, + LLVMRustGetSymbolsCallback Callback, + LLVMRustGetSymbolsErrorCallback ErrorCallback) { + std::unique_ptr<MemoryBuffer> Buf = MemoryBuffer::getMemBuffer( + StringRef(BufPtr, BufLen), StringRef("LLVMRustGetSymbolsObject"), false); + SmallString<0> SymNameBuf; + auto SymName = raw_svector_ostream(SymNameBuf); + + // In the scenario when LLVMContext is populated SymbolicFile will contain a + // reference to it, thus SymbolicFile should be destroyed first. + LLVMContext Context; + std::unique_ptr<object::SymbolicFile> Obj; + + const file_magic Type = identify_magic(Buf->getBuffer()); + if (!object::SymbolicFile::isSymbolicFile(Type, &Context)) { + return 0; + } + + if (Type == file_magic::bitcode) { + auto ObjOrErr = object::SymbolicFile::createSymbolicFile( + Buf->getMemBufferRef(), file_magic::bitcode, &Context); + if (!ObjOrErr) { + Error E = ObjOrErr.takeError(); + SmallString<0> ErrorBuf; + auto Error = raw_svector_ostream(ErrorBuf); + Error << E << '\0'; + return ErrorCallback(Error.str().data()); + } + Obj = std::move(*ObjOrErr); + } else { + auto ObjOrErr = + object::SymbolicFile::createSymbolicFile(Buf->getMemBufferRef()); + if (!ObjOrErr) { + Error E = ObjOrErr.takeError(); + SmallString<0> ErrorBuf; + auto Error = raw_svector_ostream(ErrorBuf); + Error << E << '\0'; + return ErrorCallback(Error.str().data()); + } + Obj = std::move(*ObjOrErr); + } + + for (const object::BasicSymbolRef &S : Obj->symbols()) { + if (!isArchiveSymbol(S)) + continue; + if (Error E = S.printName(SymName)) { + SmallString<0> ErrorBuf; + auto Error = raw_svector_ostream(ErrorBuf); + Error << E << '\0'; + return ErrorCallback(Error.str().data()); + } + SymName << '\0'; + if (void *E = Callback(State, SymNameBuf.str().data())) { + return E; + } + SymNameBuf.clear(); + } + return 0; +} |