diff options
Diffstat (limited to 'lib/Bitcode/Writer')
| -rw-r--r-- | lib/Bitcode/Writer/BitWriter.cpp | 58 | ||||
| -rw-r--r-- | lib/Bitcode/Writer/BitcodeWriter.cpp | 1449 | ||||
| -rw-r--r-- | lib/Bitcode/Writer/BitcodeWriterPass.cpp | 56 | ||||
| -rw-r--r-- | lib/Bitcode/Writer/CMakeLists.txt | 9 | ||||
| -rw-r--r-- | lib/Bitcode/Writer/Makefile | 15 | ||||
| -rw-r--r-- | lib/Bitcode/Writer/Serialize.cpp | 118 | ||||
| -rw-r--r-- | lib/Bitcode/Writer/SerializeAPFloat.cpp | 21 | ||||
| -rw-r--r-- | lib/Bitcode/Writer/SerializeAPInt.cpp | 31 | ||||
| -rw-r--r-- | lib/Bitcode/Writer/ValueEnumerator.cpp | 347 | ||||
| -rw-r--r-- | lib/Bitcode/Writer/ValueEnumerator.h | 127 |
10 files changed, 2231 insertions, 0 deletions
diff --git a/lib/Bitcode/Writer/BitWriter.cpp b/lib/Bitcode/Writer/BitWriter.cpp new file mode 100644 index 0000000..8834964 --- /dev/null +++ b/lib/Bitcode/Writer/BitWriter.cpp @@ -0,0 +1,58 @@ +//===-- BitWriter.cpp -----------------------------------------------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is distributed under the University of Illinois Open Source +// License. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// + +#include "llvm-c/BitWriter.h" +#include "llvm/Bitcode/ReaderWriter.h" +#include <fstream> + +using namespace llvm; + + +/*===-- Operations on modules ---------------------------------------------===*/ + +int LLVMWriteBitcodeToFile(LLVMModuleRef M, const char *Path) { + std::ofstream OS(Path, std::ios_base::out|std::ios::trunc|std::ios::binary); + + if (!OS.fail()) + WriteBitcodeToFile(unwrap(M), OS); + + if (OS.fail()) + return -1; + + return 0; +} + +#if defined(__GNUC__) && (__GNUC__ > 3 || __GNUC__ == 3 && __GNUC_MINOR >= 4) +#include <ext/stdio_filebuf.h> + +// FIXME: Control this with configure? Provide some portable abstraction in +// libSystem? As is, the user will just get a linker error if they use this on +// non-GCC. Some C++ stdlibs even have ofstream::ofstream(int fd). +int LLVMWriteBitcodeToFileHandle(LLVMModuleRef M, int FileHandle) { + __gnu_cxx::stdio_filebuf<char> Buffer(FileHandle, std::ios_base::out | + std::ios::trunc | + std::ios::binary); + std::ostream OS(&Buffer); + + if (!OS.fail()) + WriteBitcodeToFile(unwrap(M), OS); + + if (OS.fail()) + return -1; + + return 0; +} + +#else + +int LLVMWriteBitcodeToFileHandle(LLVMModuleRef M, int FileHandle) { + return -1; // Not supported. +} + +#endif diff --git a/lib/Bitcode/Writer/BitcodeWriter.cpp b/lib/Bitcode/Writer/BitcodeWriter.cpp new file mode 100644 index 0000000..bfc029c --- /dev/null +++ b/lib/Bitcode/Writer/BitcodeWriter.cpp @@ -0,0 +1,1449 @@ +//===--- Bitcode/Writer/BitcodeWriter.cpp - Bitcode Writer ----------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is distributed under the University of Illinois Open Source +// License. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// +// +// Bitcode writer implementation. +// +//===----------------------------------------------------------------------===// + +#include "llvm/Bitcode/ReaderWriter.h" +#include "llvm/Bitcode/BitstreamWriter.h" +#include "llvm/Bitcode/LLVMBitCodes.h" +#include "ValueEnumerator.h" +#include "llvm/Constants.h" +#include "llvm/DerivedTypes.h" +#include "llvm/InlineAsm.h" +#include "llvm/Instructions.h" +#include "llvm/MDNode.h" +#include "llvm/Module.h" +#include "llvm/TypeSymbolTable.h" +#include "llvm/ValueSymbolTable.h" +#include "llvm/Support/MathExtras.h" +#include "llvm/Support/Streams.h" +#include "llvm/Support/raw_ostream.h" +#include "llvm/System/Program.h" +using namespace llvm; + +/// These are manifest constants used by the bitcode writer. They do not need to +/// be kept in sync with the reader, but need to be consistent within this file. +enum { + CurVersion = 0, + + // VALUE_SYMTAB_BLOCK abbrev id's. + VST_ENTRY_8_ABBREV = bitc::FIRST_APPLICATION_ABBREV, + VST_ENTRY_7_ABBREV, + VST_ENTRY_6_ABBREV, + VST_BBENTRY_6_ABBREV, + + // CONSTANTS_BLOCK abbrev id's. + CONSTANTS_SETTYPE_ABBREV = bitc::FIRST_APPLICATION_ABBREV, + CONSTANTS_INTEGER_ABBREV, + CONSTANTS_CE_CAST_Abbrev, + CONSTANTS_NULL_Abbrev, + + // FUNCTION_BLOCK abbrev id's. + FUNCTION_INST_LOAD_ABBREV = bitc::FIRST_APPLICATION_ABBREV, + FUNCTION_INST_BINOP_ABBREV, + FUNCTION_INST_CAST_ABBREV, + FUNCTION_INST_RET_VOID_ABBREV, + FUNCTION_INST_RET_VAL_ABBREV, + FUNCTION_INST_UNREACHABLE_ABBREV +}; + + +static unsigned GetEncodedCastOpcode(unsigned Opcode) { + switch (Opcode) { + default: assert(0 && "Unknown cast instruction!"); + case Instruction::Trunc : return bitc::CAST_TRUNC; + case Instruction::ZExt : return bitc::CAST_ZEXT; + case Instruction::SExt : return bitc::CAST_SEXT; + case Instruction::FPToUI : return bitc::CAST_FPTOUI; + case Instruction::FPToSI : return bitc::CAST_FPTOSI; + case Instruction::UIToFP : return bitc::CAST_UITOFP; + case Instruction::SIToFP : return bitc::CAST_SITOFP; + case Instruction::FPTrunc : return bitc::CAST_FPTRUNC; + case Instruction::FPExt : return bitc::CAST_FPEXT; + case Instruction::PtrToInt: return bitc::CAST_PTRTOINT; + case Instruction::IntToPtr: return bitc::CAST_INTTOPTR; + case Instruction::BitCast : return bitc::CAST_BITCAST; + } +} + +static unsigned GetEncodedBinaryOpcode(unsigned Opcode) { + switch (Opcode) { + default: assert(0 && "Unknown binary instruction!"); + case Instruction::Add: return bitc::BINOP_ADD; + case Instruction::Sub: return bitc::BINOP_SUB; + case Instruction::Mul: return bitc::BINOP_MUL; + case Instruction::UDiv: return bitc::BINOP_UDIV; + case Instruction::FDiv: + case Instruction::SDiv: return bitc::BINOP_SDIV; + case Instruction::URem: return bitc::BINOP_UREM; + case Instruction::FRem: + case Instruction::SRem: return bitc::BINOP_SREM; + case Instruction::Shl: return bitc::BINOP_SHL; + case Instruction::LShr: return bitc::BINOP_LSHR; + case Instruction::AShr: return bitc::BINOP_ASHR; + case Instruction::And: return bitc::BINOP_AND; + case Instruction::Or: return bitc::BINOP_OR; + case Instruction::Xor: return bitc::BINOP_XOR; + } +} + + + +static void WriteStringRecord(unsigned Code, const std::string &Str, + unsigned AbbrevToUse, BitstreamWriter &Stream) { + SmallVector<unsigned, 64> Vals; + + // Code: [strchar x N] + for (unsigned i = 0, e = Str.size(); i != e; ++i) + Vals.push_back(Str[i]); + + // Emit the finished record. + Stream.EmitRecord(Code, Vals, AbbrevToUse); +} + +// Emit information about parameter attributes. +static void WriteAttributeTable(const ValueEnumerator &VE, + BitstreamWriter &Stream) { + const std::vector<AttrListPtr> &Attrs = VE.getAttributes(); + if (Attrs.empty()) return; + + Stream.EnterSubblock(bitc::PARAMATTR_BLOCK_ID, 3); + + SmallVector<uint64_t, 64> Record; + for (unsigned i = 0, e = Attrs.size(); i != e; ++i) { + const AttrListPtr &A = Attrs[i]; + for (unsigned i = 0, e = A.getNumSlots(); i != e; ++i) { + const AttributeWithIndex &PAWI = A.getSlot(i); + Record.push_back(PAWI.Index); + + // FIXME: remove in LLVM 3.0 + // Store the alignment in the bitcode as a 16-bit raw value instead of a + // 5-bit log2 encoded value. Shift the bits above the alignment up by + // 11 bits. + uint64_t FauxAttr = PAWI.Attrs & 0xffff; + if (PAWI.Attrs & Attribute::Alignment) + FauxAttr |= (1ull<<16)<<(((PAWI.Attrs & Attribute::Alignment)-1) >> 16); + FauxAttr |= (PAWI.Attrs & (0x3FFull << 21)) << 11; + + Record.push_back(FauxAttr); + } + + Stream.EmitRecord(bitc::PARAMATTR_CODE_ENTRY, Record); + Record.clear(); + } + + Stream.ExitBlock(); +} + +/// WriteTypeTable - Write out the type table for a module. +static void WriteTypeTable(const ValueEnumerator &VE, BitstreamWriter &Stream) { + const ValueEnumerator::TypeList &TypeList = VE.getTypes(); + + Stream.EnterSubblock(bitc::TYPE_BLOCK_ID, 4 /*count from # abbrevs */); + SmallVector<uint64_t, 64> TypeVals; + + // Abbrev for TYPE_CODE_POINTER. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::TYPE_CODE_POINTER)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, + Log2_32_Ceil(VE.getTypes().size()+1))); + Abbv->Add(BitCodeAbbrevOp(0)); // Addrspace = 0 + unsigned PtrAbbrev = Stream.EmitAbbrev(Abbv); + + // Abbrev for TYPE_CODE_FUNCTION. + Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::TYPE_CODE_FUNCTION)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 1)); // isvararg + Abbv->Add(BitCodeAbbrevOp(0)); // FIXME: DEAD value, remove in LLVM 3.0 + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, + Log2_32_Ceil(VE.getTypes().size()+1))); + unsigned FunctionAbbrev = Stream.EmitAbbrev(Abbv); + + // Abbrev for TYPE_CODE_STRUCT. + Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::TYPE_CODE_STRUCT)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 1)); // ispacked + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, + Log2_32_Ceil(VE.getTypes().size()+1))); + unsigned StructAbbrev = Stream.EmitAbbrev(Abbv); + + // Abbrev for TYPE_CODE_ARRAY. + Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::TYPE_CODE_ARRAY)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 8)); // size + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, + Log2_32_Ceil(VE.getTypes().size()+1))); + unsigned ArrayAbbrev = Stream.EmitAbbrev(Abbv); + + // Emit an entry count so the reader can reserve space. + TypeVals.push_back(TypeList.size()); + Stream.EmitRecord(bitc::TYPE_CODE_NUMENTRY, TypeVals); + TypeVals.clear(); + + // Loop over all of the types, emitting each in turn. + for (unsigned i = 0, e = TypeList.size(); i != e; ++i) { + const Type *T = TypeList[i].first; + int AbbrevToUse = 0; + unsigned Code = 0; + + switch (T->getTypeID()) { + default: assert(0 && "Unknown type!"); + case Type::VoidTyID: Code = bitc::TYPE_CODE_VOID; break; + case Type::FloatTyID: Code = bitc::TYPE_CODE_FLOAT; break; + case Type::DoubleTyID: Code = bitc::TYPE_CODE_DOUBLE; break; + case Type::X86_FP80TyID: Code = bitc::TYPE_CODE_X86_FP80; break; + case Type::FP128TyID: Code = bitc::TYPE_CODE_FP128; break; + case Type::PPC_FP128TyID: Code = bitc::TYPE_CODE_PPC_FP128; break; + case Type::LabelTyID: Code = bitc::TYPE_CODE_LABEL; break; + case Type::OpaqueTyID: Code = bitc::TYPE_CODE_OPAQUE; break; + case Type::MetadataTyID: Code = bitc::TYPE_CODE_METADATA; break; + case Type::IntegerTyID: + // INTEGER: [width] + Code = bitc::TYPE_CODE_INTEGER; + TypeVals.push_back(cast<IntegerType>(T)->getBitWidth()); + break; + case Type::PointerTyID: { + const PointerType *PTy = cast<PointerType>(T); + // POINTER: [pointee type, address space] + Code = bitc::TYPE_CODE_POINTER; + TypeVals.push_back(VE.getTypeID(PTy->getElementType())); + unsigned AddressSpace = PTy->getAddressSpace(); + TypeVals.push_back(AddressSpace); + if (AddressSpace == 0) AbbrevToUse = PtrAbbrev; + break; + } + case Type::FunctionTyID: { + const FunctionType *FT = cast<FunctionType>(T); + // FUNCTION: [isvararg, attrid, retty, paramty x N] + Code = bitc::TYPE_CODE_FUNCTION; + TypeVals.push_back(FT->isVarArg()); + TypeVals.push_back(0); // FIXME: DEAD: remove in llvm 3.0 + TypeVals.push_back(VE.getTypeID(FT->getReturnType())); + for (unsigned i = 0, e = FT->getNumParams(); i != e; ++i) + TypeVals.push_back(VE.getTypeID(FT->getParamType(i))); + AbbrevToUse = FunctionAbbrev; + break; + } + case Type::StructTyID: { + const StructType *ST = cast<StructType>(T); + // STRUCT: [ispacked, eltty x N] + Code = bitc::TYPE_CODE_STRUCT; + TypeVals.push_back(ST->isPacked()); + // Output all of the element types. + for (StructType::element_iterator I = ST->element_begin(), + E = ST->element_end(); I != E; ++I) + TypeVals.push_back(VE.getTypeID(*I)); + AbbrevToUse = StructAbbrev; + break; + } + case Type::ArrayTyID: { + const ArrayType *AT = cast<ArrayType>(T); + // ARRAY: [numelts, eltty] + Code = bitc::TYPE_CODE_ARRAY; + TypeVals.push_back(AT->getNumElements()); + TypeVals.push_back(VE.getTypeID(AT->getElementType())); + AbbrevToUse = ArrayAbbrev; + break; + } + case Type::VectorTyID: { + const VectorType *VT = cast<VectorType>(T); + // VECTOR [numelts, eltty] + Code = bitc::TYPE_CODE_VECTOR; + TypeVals.push_back(VT->getNumElements()); + TypeVals.push_back(VE.getTypeID(VT->getElementType())); + break; + } + } + + // Emit the finished record. + Stream.EmitRecord(Code, TypeVals, AbbrevToUse); + TypeVals.clear(); + } + + Stream.ExitBlock(); +} + +static unsigned getEncodedLinkage(const GlobalValue *GV) { + switch (GV->getLinkage()) { + default: assert(0 && "Invalid linkage!"); + case GlobalValue::GhostLinkage: // Map ghost linkage onto external. + case GlobalValue::ExternalLinkage: return 0; + case GlobalValue::WeakAnyLinkage: return 1; + case GlobalValue::AppendingLinkage: return 2; + case GlobalValue::InternalLinkage: return 3; + case GlobalValue::LinkOnceAnyLinkage: return 4; + case GlobalValue::DLLImportLinkage: return 5; + case GlobalValue::DLLExportLinkage: return 6; + case GlobalValue::ExternalWeakLinkage: return 7; + case GlobalValue::CommonLinkage: return 8; + case GlobalValue::PrivateLinkage: return 9; + case GlobalValue::WeakODRLinkage: return 10; + case GlobalValue::LinkOnceODRLinkage: return 11; + case GlobalValue::AvailableExternallyLinkage: return 12; + } +} + +static unsigned getEncodedVisibility(const GlobalValue *GV) { + switch (GV->getVisibility()) { + default: assert(0 && "Invalid visibility!"); + case GlobalValue::DefaultVisibility: return 0; + case GlobalValue::HiddenVisibility: return 1; + case GlobalValue::ProtectedVisibility: return 2; + } +} + +// Emit top-level description of module, including target triple, inline asm, +// descriptors for global variables, and function prototype info. +static void WriteModuleInfo(const Module *M, const ValueEnumerator &VE, + BitstreamWriter &Stream) { + // Emit the list of dependent libraries for the Module. + for (Module::lib_iterator I = M->lib_begin(), E = M->lib_end(); I != E; ++I) + WriteStringRecord(bitc::MODULE_CODE_DEPLIB, *I, 0/*TODO*/, Stream); + + // Emit various pieces of data attached to a module. + if (!M->getTargetTriple().empty()) + WriteStringRecord(bitc::MODULE_CODE_TRIPLE, M->getTargetTriple(), + 0/*TODO*/, Stream); + if (!M->getDataLayout().empty()) + WriteStringRecord(bitc::MODULE_CODE_DATALAYOUT, M->getDataLayout(), + 0/*TODO*/, Stream); + if (!M->getModuleInlineAsm().empty()) + WriteStringRecord(bitc::MODULE_CODE_ASM, M->getModuleInlineAsm(), + 0/*TODO*/, Stream); + + // Emit information about sections and GC, computing how many there are. Also + // compute the maximum alignment value. + std::map<std::string, unsigned> SectionMap; + std::map<std::string, unsigned> GCMap; + unsigned MaxAlignment = 0; + unsigned MaxGlobalType = 0; + for (Module::const_global_iterator GV = M->global_begin(),E = M->global_end(); + GV != E; ++GV) { + MaxAlignment = std::max(MaxAlignment, GV->getAlignment()); + MaxGlobalType = std::max(MaxGlobalType, VE.getTypeID(GV->getType())); + + if (!GV->hasSection()) continue; + // Give section names unique ID's. + unsigned &Entry = SectionMap[GV->getSection()]; + if (Entry != 0) continue; + WriteStringRecord(bitc::MODULE_CODE_SECTIONNAME, GV->getSection(), + 0/*TODO*/, Stream); + Entry = SectionMap.size(); + } + for (Module::const_iterator F = M->begin(), E = M->end(); F != E; ++F) { + MaxAlignment = std::max(MaxAlignment, F->getAlignment()); + if (F->hasSection()) { + // Give section names unique ID's. + unsigned &Entry = SectionMap[F->getSection()]; + if (!Entry) { + WriteStringRecord(bitc::MODULE_CODE_SECTIONNAME, F->getSection(), + 0/*TODO*/, Stream); + Entry = SectionMap.size(); + } + } + if (F->hasGC()) { + // Same for GC names. + unsigned &Entry = GCMap[F->getGC()]; + if (!Entry) { + WriteStringRecord(bitc::MODULE_CODE_GCNAME, F->getGC(), + 0/*TODO*/, Stream); + Entry = GCMap.size(); + } + } + } + + // Emit abbrev for globals, now that we know # sections and max alignment. + unsigned SimpleGVarAbbrev = 0; + if (!M->global_empty()) { + // Add an abbrev for common globals with no visibility or thread localness. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::MODULE_CODE_GLOBALVAR)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, + Log2_32_Ceil(MaxGlobalType+1))); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 1)); // Constant. + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // Initializer. + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 4)); // Linkage. + if (MaxAlignment == 0) // Alignment. + Abbv->Add(BitCodeAbbrevOp(0)); + else { + unsigned MaxEncAlignment = Log2_32(MaxAlignment)+1; + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, + Log2_32_Ceil(MaxEncAlignment+1))); + } + if (SectionMap.empty()) // Section. + Abbv->Add(BitCodeAbbrevOp(0)); + else + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, + Log2_32_Ceil(SectionMap.size()+1))); + // Don't bother emitting vis + thread local. + SimpleGVarAbbrev = Stream.EmitAbbrev(Abbv); + } + + // Emit the global variable information. + SmallVector<unsigned, 64> Vals; + for (Module::const_global_iterator GV = M->global_begin(),E = M->global_end(); + GV != E; ++GV) { + unsigned AbbrevToUse = 0; + + // GLOBALVAR: [type, isconst, initid, + // linkage, alignment, section, visibility, threadlocal] + Vals.push_back(VE.getTypeID(GV->getType())); + Vals.push_back(GV->isConstant()); + Vals.push_back(GV->isDeclaration() ? 0 : + (VE.getValueID(GV->getInitializer()) + 1)); + Vals.push_back(getEncodedLinkage(GV)); + Vals.push_back(Log2_32(GV->getAlignment())+1); + Vals.push_back(GV->hasSection() ? SectionMap[GV->getSection()] : 0); + if (GV->isThreadLocal() || + GV->getVisibility() != GlobalValue::DefaultVisibility) { + Vals.push_back(getEncodedVisibility(GV)); + Vals.push_back(GV->isThreadLocal()); + } else { + AbbrevToUse = SimpleGVarAbbrev; + } + + Stream.EmitRecord(bitc::MODULE_CODE_GLOBALVAR, Vals, AbbrevToUse); + Vals.clear(); + } + + // Emit the function proto information. + for (Module::const_iterator F = M->begin(), E = M->end(); F != E; ++F) { + // FUNCTION: [type, callingconv, isproto, paramattr, + // linkage, alignment, section, visibility, gc] + Vals.push_back(VE.getTypeID(F->getType())); + Vals.push_back(F->getCallingConv()); + Vals.push_back(F->isDeclaration()); + Vals.push_back(getEncodedLinkage(F)); + Vals.push_back(VE.getAttributeID(F->getAttributes())); + Vals.push_back(Log2_32(F->getAlignment())+1); + Vals.push_back(F->hasSection() ? SectionMap[F->getSection()] : 0); + Vals.push_back(getEncodedVisibility(F)); + Vals.push_back(F->hasGC() ? GCMap[F->getGC()] : 0); + + unsigned AbbrevToUse = 0; + Stream.EmitRecord(bitc::MODULE_CODE_FUNCTION, Vals, AbbrevToUse); + Vals.clear(); + } + + + // Emit the alias information. + for (Module::const_alias_iterator AI = M->alias_begin(), E = M->alias_end(); + AI != E; ++AI) { + Vals.push_back(VE.getTypeID(AI->getType())); + Vals.push_back(VE.getValueID(AI->getAliasee())); + Vals.push_back(getEncodedLinkage(AI)); + Vals.push_back(getEncodedVisibility(AI)); + unsigned AbbrevToUse = 0; + Stream.EmitRecord(bitc::MODULE_CODE_ALIAS, Vals, AbbrevToUse); + Vals.clear(); + } +} + + +static void WriteConstants(unsigned FirstVal, unsigned LastVal, + const ValueEnumerator &VE, + BitstreamWriter &Stream, bool isGlobal) { + if (FirstVal == LastVal) return; + + Stream.EnterSubblock(bitc::CONSTANTS_BLOCK_ID, 4); + + unsigned AggregateAbbrev = 0; + unsigned String8Abbrev = 0; + unsigned CString7Abbrev = 0; + unsigned CString6Abbrev = 0; + unsigned MDString8Abbrev = 0; + unsigned MDString6Abbrev = 0; + // If this is a constant pool for the module, emit module-specific abbrevs. + if (isGlobal) { + // Abbrev for CST_CODE_AGGREGATE. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::CST_CODE_AGGREGATE)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, Log2_32_Ceil(LastVal+1))); + AggregateAbbrev = Stream.EmitAbbrev(Abbv); + + // Abbrev for CST_CODE_STRING. + Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::CST_CODE_STRING)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 8)); + String8Abbrev = Stream.EmitAbbrev(Abbv); + // Abbrev for CST_CODE_CSTRING. + Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::CST_CODE_CSTRING)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 7)); + CString7Abbrev = Stream.EmitAbbrev(Abbv); + // Abbrev for CST_CODE_CSTRING. + Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::CST_CODE_CSTRING)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Char6)); + CString6Abbrev = Stream.EmitAbbrev(Abbv); + + // Abbrev for CST_CODE_MDSTRING. + Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::CST_CODE_MDSTRING)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 8)); + MDString8Abbrev = Stream.EmitAbbrev(Abbv); + // Abbrev for CST_CODE_MDSTRING. + Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::CST_CODE_MDSTRING)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Char6)); + MDString6Abbrev = Stream.EmitAbbrev(Abbv); + } + + SmallVector<uint64_t, 64> Record; + + const ValueEnumerator::ValueList &Vals = VE.getValues(); + const Type *LastTy = 0; + for (unsigned i = FirstVal; i != LastVal; ++i) { + const Value *V = Vals[i].first; + // If we need to switch types, do so now. + if (V->getType() != LastTy) { + LastTy = V->getType(); + Record.push_back(VE.getTypeID(LastTy)); + Stream.EmitRecord(bitc::CST_CODE_SETTYPE, Record, + CONSTANTS_SETTYPE_ABBREV); + Record.clear(); + } + + if (const InlineAsm *IA = dyn_cast<InlineAsm>(V)) { + Record.push_back(unsigned(IA->hasSideEffects())); + + // Add the asm string. + const std::string &AsmStr = IA->getAsmString(); + Record.push_back(AsmStr.size()); + for (unsigned i = 0, e = AsmStr.size(); i != e; ++i) + Record.push_back(AsmStr[i]); + + // Add the constraint string. + const std::string &ConstraintStr = IA->getConstraintString(); + Record.push_back(ConstraintStr.size()); + for (unsigned i = 0, e = ConstraintStr.size(); i != e; ++i) + Record.push_back(ConstraintStr[i]); + Stream.EmitRecord(bitc::CST_CODE_INLINEASM, Record); + Record.clear(); + continue; + } + const Constant *C = cast<Constant>(V); + unsigned Code = -1U; + unsigned AbbrevToUse = 0; + if (C->isNullValue()) { + Code = bitc::CST_CODE_NULL; + } else if (isa<UndefValue>(C)) { + Code = bitc::CST_CODE_UNDEF; + } else if (const ConstantInt *IV = dyn_cast<ConstantInt>(C)) { + if (IV->getBitWidth() <= 64) { + int64_t V = IV->getSExtValue(); + if (V >= 0) + Record.push_back(V << 1); + else + Record.push_back((-V << 1) | 1); + Code = bitc::CST_CODE_INTEGER; + AbbrevToUse = CONSTANTS_INTEGER_ABBREV; + } else { // Wide integers, > 64 bits in size. + // We have an arbitrary precision integer value to write whose + // bit width is > 64. However, in canonical unsigned integer + // format it is likely that the high bits are going to be zero. + // So, we only write the number of active words. + unsigned NWords = IV->getValue().getActiveWords(); + const uint64_t *RawWords = IV->getValue().getRawData(); + for (unsigned i = 0; i != NWords; ++i) { + int64_t V = RawWords[i]; + if (V >= 0) + Record.push_back(V << 1); + else + Record.push_back((-V << 1) | 1); + } + Code = bitc::CST_CODE_WIDE_INTEGER; + } + } else if (const ConstantFP *CFP = dyn_cast<ConstantFP>(C)) { + Code = bitc::CST_CODE_FLOAT; + const Type *Ty = CFP->getType(); + if (Ty == Type::FloatTy || Ty == Type::DoubleTy) { + Record.push_back(CFP->getValueAPF().bitcastToAPInt().getZExtValue()); + } else if (Ty == Type::X86_FP80Ty) { + // api needed to prevent premature destruction + // bits are not in the same order as a normal i80 APInt, compensate. + APInt api = CFP->getValueAPF().bitcastToAPInt(); + const uint64_t *p = api.getRawData(); + Record.push_back((p[1] << 48) | (p[0] >> 16)); + Record.push_back(p[0] & 0xffffLL); + } else if (Ty == Type::FP128Ty || Ty == Type::PPC_FP128Ty) { + APInt api = CFP->getValueAPF().bitcastToAPInt(); + const uint64_t *p = api.getRawData(); + Record.push_back(p[0]); + Record.push_back(p[1]); + } else { + assert (0 && "Unknown FP type!"); + } + } else if (isa<ConstantArray>(C) && cast<ConstantArray>(C)->isString()) { + // Emit constant strings specially. + unsigned NumOps = C->getNumOperands(); + // If this is a null-terminated string, use the denser CSTRING encoding. + if (C->getOperand(NumOps-1)->isNullValue()) { + Code = bitc::CST_CODE_CSTRING; + --NumOps; // Don't encode the null, which isn't allowed by char6. + } else { + Code = bitc::CST_CODE_STRING; + AbbrevToUse = String8Abbrev; + } + bool isCStr7 = Code == bitc::CST_CODE_CSTRING; + bool isCStrChar6 = Code == bitc::CST_CODE_CSTRING; + for (unsigned i = 0; i != NumOps; ++i) { + unsigned char V = cast<ConstantInt>(C->getOperand(i))->getZExtValue(); + Record.push_back(V); + isCStr7 &= (V & 128) == 0; + if (isCStrChar6) + isCStrChar6 = BitCodeAbbrevOp::isChar6(V); + } + + if (isCStrChar6) + AbbrevToUse = CString6Abbrev; + else if (isCStr7) + AbbrevToUse = CString7Abbrev; + } else if (isa<ConstantArray>(C) || isa<ConstantStruct>(V) || + isa<ConstantVector>(V)) { + Code = bitc::CST_CODE_AGGREGATE; + for (unsigned i = 0, e = C->getNumOperands(); i != e; ++i) + Record.push_back(VE.getValueID(C->getOperand(i))); + AbbrevToUse = AggregateAbbrev; + } else if (const ConstantExpr *CE = dyn_cast<ConstantExpr>(C)) { + switch (CE->getOpcode()) { + default: + if (Instruction::isCast(CE->getOpcode())) { + Code = bitc::CST_CODE_CE_CAST; + Record.push_back(GetEncodedCastOpcode(CE->getOpcode())); + Record.push_back(VE.getTypeID(C->getOperand(0)->getType())); + Record.push_back(VE.getValueID(C->getOperand(0))); + AbbrevToUse = CONSTANTS_CE_CAST_Abbrev; + } else { + assert(CE->getNumOperands() == 2 && "Unknown constant expr!"); + Code = bitc::CST_CODE_CE_BINOP; + Record.push_back(GetEncodedBinaryOpcode(CE->getOpcode())); + Record.push_back(VE.getValueID(C->getOperand(0))); + Record.push_back(VE.getValueID(C->getOperand(1))); + } + break; + case Instruction::GetElementPtr: + Code = bitc::CST_CODE_CE_GEP; + for (unsigned i = 0, e = CE->getNumOperands(); i != e; ++i) { + Record.push_back(VE.getTypeID(C->getOperand(i)->getType())); + Record.push_back(VE.getValueID(C->getOperand(i))); + } + break; + case Instruction::Select: + Code = bitc::CST_CODE_CE_SELECT; + Record.push_back(VE.getValueID(C->getOperand(0))); + Record.push_back(VE.getValueID(C->getOperand(1))); + Record.push_back(VE.getValueID(C->getOperand(2))); + break; + case Instruction::ExtractElement: + Code = bitc::CST_CODE_CE_EXTRACTELT; + Record.push_back(VE.getTypeID(C->getOperand(0)->getType())); + Record.push_back(VE.getValueID(C->getOperand(0))); + Record.push_back(VE.getValueID(C->getOperand(1))); + break; + case Instruction::InsertElement: + Code = bitc::CST_CODE_CE_INSERTELT; + Record.push_back(VE.getValueID(C->getOperand(0))); + Record.push_back(VE.getValueID(C->getOperand(1))); + Record.push_back(VE.getValueID(C->getOperand(2))); + break; + case Instruction::ShuffleVector: + // If the return type and argument types are the same, this is a + // standard shufflevector instruction. If the types are different, + // then the shuffle is widening or truncating the input vectors, and + // the argument type must also be encoded. + if (C->getType() == C->getOperand(0)->getType()) { + Code = bitc::CST_CODE_CE_SHUFFLEVEC; + } else { + Code = bitc::CST_CODE_CE_SHUFVEC_EX; + Record.push_back(VE.getTypeID(C->getOperand(0)->getType())); + } + Record.push_back(VE.getValueID(C->getOperand(0))); + Record.push_back(VE.getValueID(C->getOperand(1))); + Record.push_back(VE.getValueID(C->getOperand(2))); + break; + case Instruction::ICmp: + case Instruction::FCmp: + case Instruction::VICmp: + case Instruction::VFCmp: + if (isa<VectorType>(C->getOperand(0)->getType()) + && (CE->getOpcode() == Instruction::ICmp + || CE->getOpcode() == Instruction::FCmp)) { + // compare returning vector of Int1Ty + assert(0 && "Unsupported constant!"); + } else { + Code = bitc::CST_CODE_CE_CMP; + } + Record.push_back(VE.getTypeID(C->getOperand(0)->getType())); + Record.push_back(VE.getValueID(C->getOperand(0))); + Record.push_back(VE.getValueID(C->getOperand(1))); + Record.push_back(CE->getPredicate()); + break; + } + } else if (const MDString *S = dyn_cast<MDString>(C)) { + Code = bitc::CST_CODE_MDSTRING; + AbbrevToUse = MDString6Abbrev; + for (unsigned i = 0, e = S->size(); i != e; ++i) { + char V = S->begin()[i]; + Record.push_back(V); + + if (!BitCodeAbbrevOp::isChar6(V)) + AbbrevToUse = MDString8Abbrev; + } + } else if (const MDNode *N = dyn_cast<MDNode>(C)) { + Code = bitc::CST_CODE_MDNODE; + for (unsigned i = 0, e = N->getNumElements(); i != e; ++i) { + if (N->getElement(i)) { + Record.push_back(VE.getTypeID(N->getElement(i)->getType())); + Record.push_back(VE.getValueID(N->getElement(i))); + } else { + Record.push_back(VE.getTypeID(Type::VoidTy)); + Record.push_back(0); + } + } + } else { + assert(0 && "Unknown constant!"); + } + Stream.EmitRecord(Code, Record, AbbrevToUse); + Record.clear(); + } + + Stream.ExitBlock(); +} + +static void WriteModuleConstants(const ValueEnumerator &VE, + BitstreamWriter &Stream) { + const ValueEnumerator::ValueList &Vals = VE.getValues(); + + // Find the first constant to emit, which is the first non-globalvalue value. + // We know globalvalues have been emitted by WriteModuleInfo. + for (unsigned i = 0, e = Vals.size(); i != e; ++i) { + if (!isa<GlobalValue>(Vals[i].first)) { + WriteConstants(i, Vals.size(), VE, Stream, true); + return; + } + } +} + +/// PushValueAndType - The file has to encode both the value and type id for +/// many values, because we need to know what type to create for forward +/// references. However, most operands are not forward references, so this type +/// field is not needed. +/// +/// This function adds V's value ID to Vals. If the value ID is higher than the +/// instruction ID, then it is a forward reference, and it also includes the +/// type ID. +static bool PushValueAndType(const Value *V, unsigned InstID, + SmallVector<unsigned, 64> &Vals, + ValueEnumerator &VE) { + unsigned ValID = VE.getValueID(V); + Vals.push_back(ValID); + if (ValID >= InstID) { + Vals.push_back(VE.getTypeID(V->getType())); + return true; + } + return false; +} + +/// WriteInstruction - Emit an instruction to the specified stream. +static void WriteInstruction(const Instruction &I, unsigned InstID, + ValueEnumerator &VE, BitstreamWriter &Stream, + SmallVector<unsigned, 64> &Vals) { + unsigned Code = 0; + unsigned AbbrevToUse = 0; + switch (I.getOpcode()) { + default: + if (Instruction::isCast(I.getOpcode())) { + Code = bitc::FUNC_CODE_INST_CAST; + if (!PushValueAndType(I.getOperand(0), InstID, Vals, VE)) + AbbrevToUse = FUNCTION_INST_CAST_ABBREV; + Vals.push_back(VE.getTypeID(I.getType())); + Vals.push_back(GetEncodedCastOpcode(I.getOpcode())); + } else { + assert(isa<BinaryOperator>(I) && "Unknown instruction!"); + Code = bitc::FUNC_CODE_INST_BINOP; + if (!PushValueAndType(I.getOperand(0), InstID, Vals, VE)) + AbbrevToUse = FUNCTION_INST_BINOP_ABBREV; + Vals.push_back(VE.getValueID(I.getOperand(1))); + Vals.push_back(GetEncodedBinaryOpcode(I.getOpcode())); + } + break; + + case Instruction::GetElementPtr: + Code = bitc::FUNC_CODE_INST_GEP; + for (unsigned i = 0, e = I.getNumOperands(); i != e; ++i) + PushValueAndType(I.getOperand(i), InstID, Vals, VE); + break; + case Instruction::ExtractValue: { + Code = bitc::FUNC_CODE_INST_EXTRACTVAL; + PushValueAndType(I.getOperand(0), InstID, Vals, VE); + const ExtractValueInst *EVI = cast<ExtractValueInst>(&I); + for (const unsigned *i = EVI->idx_begin(), *e = EVI->idx_end(); i != e; ++i) + Vals.push_back(*i); + break; + } + case Instruction::InsertValue: { + Code = bitc::FUNC_CODE_INST_INSERTVAL; + PushValueAndType(I.getOperand(0), InstID, Vals, VE); + PushValueAndType(I.getOperand(1), InstID, Vals, VE); + const InsertValueInst *IVI = cast<InsertValueInst>(&I); + for (const unsigned *i = IVI->idx_begin(), *e = IVI->idx_end(); i != e; ++i) + Vals.push_back(*i); + break; + } + case Instruction::Select: + Code = bitc::FUNC_CODE_INST_VSELECT; + PushValueAndType(I.getOperand(1), InstID, Vals, VE); + Vals.push_back(VE.getValueID(I.getOperand(2))); + PushValueAndType(I.getOperand(0), InstID, Vals, VE); + break; + case Instruction::ExtractElement: + Code = bitc::FUNC_CODE_INST_EXTRACTELT; + PushValueAndType(I.getOperand(0), InstID, Vals, VE); + Vals.push_back(VE.getValueID(I.getOperand(1))); + break; + case Instruction::InsertElement: + Code = bitc::FUNC_CODE_INST_INSERTELT; + PushValueAndType(I.getOperand(0), InstID, Vals, VE); + Vals.push_back(VE.getValueID(I.getOperand(1))); + Vals.push_back(VE.getValueID(I.getOperand(2))); + break; + case Instruction::ShuffleVector: + Code = bitc::FUNC_CODE_INST_SHUFFLEVEC; + PushValueAndType(I.getOperand(0), InstID, Vals, VE); + Vals.push_back(VE.getValueID(I.getOperand(1))); + Vals.push_back(VE.getValueID(I.getOperand(2))); + break; + case Instruction::ICmp: + case Instruction::FCmp: + case Instruction::VICmp: + case Instruction::VFCmp: + if (I.getOpcode() == Instruction::ICmp + || I.getOpcode() == Instruction::FCmp) { + // compare returning Int1Ty or vector of Int1Ty + Code = bitc::FUNC_CODE_INST_CMP2; + } else { + Code = bitc::FUNC_CODE_INST_CMP; + } + PushValueAndType(I.getOperand(0), InstID, Vals, VE); + Vals.push_back(VE.getValueID(I.getOperand(1))); + Vals.push_back(cast<CmpInst>(I).getPredicate()); + break; + + case Instruction::Ret: + { + Code = bitc::FUNC_CODE_INST_RET; + unsigned NumOperands = I.getNumOperands(); + if (NumOperands == 0) + AbbrevToUse = FUNCTION_INST_RET_VOID_ABBREV; + else if (NumOperands == 1) { + if (!PushValueAndType(I.getOperand(0), InstID, Vals, VE)) + AbbrevToUse = FUNCTION_INST_RET_VAL_ABBREV; + } else { + for (unsigned i = 0, e = NumOperands; i != e; ++i) + PushValueAndType(I.getOperand(i), InstID, Vals, VE); + } + } + break; + case Instruction::Br: + { + Code = bitc::FUNC_CODE_INST_BR; + BranchInst &II(cast<BranchInst>(I)); + Vals.push_back(VE.getValueID(II.getSuccessor(0))); + if (II.isConditional()) { + Vals.push_back(VE.getValueID(II.getSuccessor(1))); + Vals.push_back(VE.getValueID(II.getCondition())); + } + } + break; + case Instruction::Switch: + Code = bitc::FUNC_CODE_INST_SWITCH; + Vals.push_back(VE.getTypeID(I.getOperand(0)->getType())); + for (unsigned i = 0, e = I.getNumOperands(); i != e; ++i) + Vals.push_back(VE.getValueID(I.getOperand(i))); + break; + case Instruction::Invoke: { + const InvokeInst *II = cast<InvokeInst>(&I); + const Value *Callee(II->getCalledValue()); + const PointerType *PTy = cast<PointerType>(Callee->getType()); + const FunctionType *FTy = cast<FunctionType>(PTy->getElementType()); + Code = bitc::FUNC_CODE_INST_INVOKE; + + Vals.push_back(VE.getAttributeID(II->getAttributes())); + Vals.push_back(II->getCallingConv()); + Vals.push_back(VE.getValueID(II->getNormalDest())); + Vals.push_back(VE.getValueID(II->getUnwindDest())); + PushValueAndType(Callee, InstID, Vals, VE); + + // Emit value #'s for the fixed parameters. + for (unsigned i = 0, e = FTy->getNumParams(); i != e; ++i) + Vals.push_back(VE.getValueID(I.getOperand(i+3))); // fixed param. + + // Emit type/value pairs for varargs params. + if (FTy->isVarArg()) { + for (unsigned i = 3+FTy->getNumParams(), e = I.getNumOperands(); + i != e; ++i) + PushValueAndType(I.getOperand(i), InstID, Vals, VE); // vararg + } + break; + } + case Instruction::Unwind: + Code = bitc::FUNC_CODE_INST_UNWIND; + break; + case Instruction::Unreachable: + Code = bitc::FUNC_CODE_INST_UNREACHABLE; + AbbrevToUse = FUNCTION_INST_UNREACHABLE_ABBREV; + break; + + case Instruction::PHI: + Code = bitc::FUNC_CODE_INST_PHI; + Vals.push_back(VE.getTypeID(I.getType())); + for (unsigned i = 0, e = I.getNumOperands(); i != e; ++i) + Vals.push_back(VE.getValueID(I.getOperand(i))); + break; + + case Instruction::Malloc: + Code = bitc::FUNC_CODE_INST_MALLOC; + Vals.push_back(VE.getTypeID(I.getType())); + Vals.push_back(VE.getValueID(I.getOperand(0))); // size. + Vals.push_back(Log2_32(cast<MallocInst>(I).getAlignment())+1); + break; + + case Instruction::Free: + Code = bitc::FUNC_CODE_INST_FREE; + PushValueAndType(I.getOperand(0), InstID, Vals, VE); + break; + + case Instruction::Alloca: + Code = bitc::FUNC_CODE_INST_ALLOCA; + Vals.push_back(VE.getTypeID(I.getType())); + Vals.push_back(VE.getValueID(I.getOperand(0))); // size. + Vals.push_back(Log2_32(cast<AllocaInst>(I).getAlignment())+1); + break; + + case Instruction::Load: + Code = bitc::FUNC_CODE_INST_LOAD; + if (!PushValueAndType(I.getOperand(0), InstID, Vals, VE)) // ptr + AbbrevToUse = FUNCTION_INST_LOAD_ABBREV; + + Vals.push_back(Log2_32(cast<LoadInst>(I).getAlignment())+1); + Vals.push_back(cast<LoadInst>(I).isVolatile()); + break; + case Instruction::Store: + Code = bitc::FUNC_CODE_INST_STORE2; + PushValueAndType(I.getOperand(1), InstID, Vals, VE); // ptrty + ptr + Vals.push_back(VE.getValueID(I.getOperand(0))); // val. + Vals.push_back(Log2_32(cast<StoreInst>(I).getAlignment())+1); + Vals.push_back(cast<StoreInst>(I).isVolatile()); + break; + case Instruction::Call: { + const PointerType *PTy = cast<PointerType>(I.getOperand(0)->getType()); + const FunctionType *FTy = cast<FunctionType>(PTy->getElementType()); + + Code = bitc::FUNC_CODE_INST_CALL; + + const CallInst *CI = cast<CallInst>(&I); + Vals.push_back(VE.getAttributeID(CI->getAttributes())); + Vals.push_back((CI->getCallingConv() << 1) | unsigned(CI->isTailCall())); + PushValueAndType(CI->getOperand(0), InstID, Vals, VE); // Callee + + // Emit value #'s for the fixed parameters. + for (unsigned i = 0, e = FTy->getNumParams(); i != e; ++i) + Vals.push_back(VE.getValueID(I.getOperand(i+1))); // fixed param. + + // Emit type/value pairs for varargs params. + if (FTy->isVarArg()) { + unsigned NumVarargs = I.getNumOperands()-1-FTy->getNumParams(); + for (unsigned i = I.getNumOperands()-NumVarargs, e = I.getNumOperands(); + i != e; ++i) + PushValueAndType(I.getOperand(i), InstID, Vals, VE); // varargs + } + break; + } + case Instruction::VAArg: + Code = bitc::FUNC_CODE_INST_VAARG; + Vals.push_back(VE.getTypeID(I.getOperand(0)->getType())); // valistty + Vals.push_back(VE.getValueID(I.getOperand(0))); // valist. + Vals.push_back(VE.getTypeID(I.getType())); // restype. + break; + } + + Stream.EmitRecord(Code, Vals, AbbrevToUse); + Vals.clear(); +} + +// Emit names for globals/functions etc. +static void WriteValueSymbolTable(const ValueSymbolTable &VST, + const ValueEnumerator &VE, + BitstreamWriter &Stream) { + if (VST.empty()) return; + Stream.EnterSubblock(bitc::VALUE_SYMTAB_BLOCK_ID, 4); + + // FIXME: Set up the abbrev, we know how many values there are! + // FIXME: We know if the type names can use 7-bit ascii. + SmallVector<unsigned, 64> NameVals; + + for (ValueSymbolTable::const_iterator SI = VST.begin(), SE = VST.end(); + SI != SE; ++SI) { + + const ValueName &Name = *SI; + + // Figure out the encoding to use for the name. + bool is7Bit = true; + bool isChar6 = true; + for (const char *C = Name.getKeyData(), *E = C+Name.getKeyLength(); + C != E; ++C) { + if (isChar6) + isChar6 = BitCodeAbbrevOp::isChar6(*C); + if ((unsigned char)*C & 128) { + is7Bit = false; + break; // don't bother scanning the rest. + } + } + + unsigned AbbrevToUse = VST_ENTRY_8_ABBREV; + + // VST_ENTRY: [valueid, namechar x N] + // VST_BBENTRY: [bbid, namechar x N] + unsigned Code; + if (isa<BasicBlock>(SI->getValue())) { + Code = bitc::VST_CODE_BBENTRY; + if (isChar6) + AbbrevToUse = VST_BBENTRY_6_ABBREV; + } else { + Code = bitc::VST_CODE_ENTRY; + if (isChar6) + AbbrevToUse = VST_ENTRY_6_ABBREV; + else if (is7Bit) + AbbrevToUse = VST_ENTRY_7_ABBREV; + } + + NameVals.push_back(VE.getValueID(SI->getValue())); + for (const char *P = Name.getKeyData(), + *E = Name.getKeyData()+Name.getKeyLength(); P != E; ++P) + NameVals.push_back((unsigned char)*P); + + // Emit the finished record. + Stream.EmitRecord(Code, NameVals, AbbrevToUse); + NameVals.clear(); + } + Stream.ExitBlock(); +} + +/// WriteFunction - Emit a function body to the module stream. +static void WriteFunction(const Function &F, ValueEnumerator &VE, + BitstreamWriter &Stream) { + Stream.EnterSubblock(bitc::FUNCTION_BLOCK_ID, 4); + VE.incorporateFunction(F); + + SmallVector<unsigned, 64> Vals; + + // Emit the number of basic blocks, so the reader can create them ahead of + // time. + Vals.push_back(VE.getBasicBlocks().size()); + Stream.EmitRecord(bitc::FUNC_CODE_DECLAREBLOCKS, Vals); + Vals.clear(); + + // If there are function-local constants, emit them now. + unsigned CstStart, CstEnd; + VE.getFunctionConstantRange(CstStart, CstEnd); + WriteConstants(CstStart, CstEnd, VE, Stream, false); + + // Keep a running idea of what the instruction ID is. + unsigned InstID = CstEnd; + + // Finally, emit all the instructions, in order. + for (Function::const_iterator BB = F.begin(), E = F.end(); BB != E; ++BB) + for (BasicBlock::const_iterator I = BB->begin(), E = BB->end(); + I != E; ++I) { + WriteInstruction(*I, InstID, VE, Stream, Vals); + if (I->getType() != Type::VoidTy) + ++InstID; + } + + // Emit names for all the instructions etc. + WriteValueSymbolTable(F.getValueSymbolTable(), VE, Stream); + + VE.purgeFunction(); + Stream.ExitBlock(); +} + +/// WriteTypeSymbolTable - Emit a block for the specified type symtab. +static void WriteTypeSymbolTable(const TypeSymbolTable &TST, + const ValueEnumerator &VE, + BitstreamWriter &Stream) { + if (TST.empty()) return; + + Stream.EnterSubblock(bitc::TYPE_SYMTAB_BLOCK_ID, 3); + + // 7-bit fixed width VST_CODE_ENTRY strings. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::VST_CODE_ENTRY)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, + Log2_32_Ceil(VE.getTypes().size()+1))); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 7)); + unsigned V7Abbrev = Stream.EmitAbbrev(Abbv); + + SmallVector<unsigned, 64> NameVals; + + for (TypeSymbolTable::const_iterator TI = TST.begin(), TE = TST.end(); + TI != TE; ++TI) { + // TST_ENTRY: [typeid, namechar x N] + NameVals.push_back(VE.getTypeID(TI->second)); + + const std::string &Str = TI->first; + bool is7Bit = true; + for (unsigned i = 0, e = Str.size(); i != e; ++i) { + NameVals.push_back((unsigned char)Str[i]); + if (Str[i] & 128) + is7Bit = false; + } + + // Emit the finished record. + Stream.EmitRecord(bitc::VST_CODE_ENTRY, NameVals, is7Bit ? V7Abbrev : 0); + NameVals.clear(); + } + + Stream.ExitBlock(); +} + +// Emit blockinfo, which defines the standard abbreviations etc. +static void WriteBlockInfo(const ValueEnumerator &VE, BitstreamWriter &Stream) { + // We only want to emit block info records for blocks that have multiple + // instances: CONSTANTS_BLOCK, FUNCTION_BLOCK and VALUE_SYMTAB_BLOCK. Other + // blocks can defined their abbrevs inline. + Stream.EnterBlockInfoBlock(2); + + { // 8-bit fixed-width VST_ENTRY/VST_BBENTRY strings. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 3)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 8)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 8)); + if (Stream.EmitBlockInfoAbbrev(bitc::VALUE_SYMTAB_BLOCK_ID, + Abbv) != VST_ENTRY_8_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + + { // 7-bit fixed width VST_ENTRY strings. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::VST_CODE_ENTRY)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 8)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 7)); + if (Stream.EmitBlockInfoAbbrev(bitc::VALUE_SYMTAB_BLOCK_ID, + Abbv) != VST_ENTRY_7_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + { // 6-bit char6 VST_ENTRY strings. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::VST_CODE_ENTRY)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 8)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Char6)); + if (Stream.EmitBlockInfoAbbrev(bitc::VALUE_SYMTAB_BLOCK_ID, + Abbv) != VST_ENTRY_6_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + { // 6-bit char6 VST_BBENTRY strings. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::VST_CODE_BBENTRY)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 8)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Array)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Char6)); + if (Stream.EmitBlockInfoAbbrev(bitc::VALUE_SYMTAB_BLOCK_ID, + Abbv) != VST_BBENTRY_6_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + + + + { // SETTYPE abbrev for CONSTANTS_BLOCK. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::CST_CODE_SETTYPE)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, + Log2_32_Ceil(VE.getTypes().size()+1))); + if (Stream.EmitBlockInfoAbbrev(bitc::CONSTANTS_BLOCK_ID, + Abbv) != CONSTANTS_SETTYPE_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + + { // INTEGER abbrev for CONSTANTS_BLOCK. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::CST_CODE_INTEGER)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 8)); + if (Stream.EmitBlockInfoAbbrev(bitc::CONSTANTS_BLOCK_ID, + Abbv) != CONSTANTS_INTEGER_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + + { // CE_CAST abbrev for CONSTANTS_BLOCK. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::CST_CODE_CE_CAST)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 4)); // cast opc + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, // typeid + Log2_32_Ceil(VE.getTypes().size()+1))); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 8)); // value id + + if (Stream.EmitBlockInfoAbbrev(bitc::CONSTANTS_BLOCK_ID, + Abbv) != CONSTANTS_CE_CAST_Abbrev) + assert(0 && "Unexpected abbrev ordering!"); + } + { // NULL abbrev for CONSTANTS_BLOCK. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::CST_CODE_NULL)); + if (Stream.EmitBlockInfoAbbrev(bitc::CONSTANTS_BLOCK_ID, + Abbv) != CONSTANTS_NULL_Abbrev) + assert(0 && "Unexpected abbrev ordering!"); + } + + // FIXME: This should only use space for first class types! + + { // INST_LOAD abbrev for FUNCTION_BLOCK. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::FUNC_CODE_INST_LOAD)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // Ptr + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 4)); // Align + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 1)); // volatile + if (Stream.EmitBlockInfoAbbrev(bitc::FUNCTION_BLOCK_ID, + Abbv) != FUNCTION_INST_LOAD_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + { // INST_BINOP abbrev for FUNCTION_BLOCK. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::FUNC_CODE_INST_BINOP)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // LHS + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // RHS + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 4)); // opc + if (Stream.EmitBlockInfoAbbrev(bitc::FUNCTION_BLOCK_ID, + Abbv) != FUNCTION_INST_BINOP_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + { // INST_CAST abbrev for FUNCTION_BLOCK. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::FUNC_CODE_INST_CAST)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // OpVal + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, // dest ty + Log2_32_Ceil(VE.getTypes().size()+1))); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::Fixed, 4)); // opc + if (Stream.EmitBlockInfoAbbrev(bitc::FUNCTION_BLOCK_ID, + Abbv) != FUNCTION_INST_CAST_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + + { // INST_RET abbrev for FUNCTION_BLOCK. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::FUNC_CODE_INST_RET)); + if (Stream.EmitBlockInfoAbbrev(bitc::FUNCTION_BLOCK_ID, + Abbv) != FUNCTION_INST_RET_VOID_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + { // INST_RET abbrev for FUNCTION_BLOCK. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::FUNC_CODE_INST_RET)); + Abbv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // ValID + if (Stream.EmitBlockInfoAbbrev(bitc::FUNCTION_BLOCK_ID, + Abbv) != FUNCTION_INST_RET_VAL_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + { // INST_UNREACHABLE abbrev for FUNCTION_BLOCK. + BitCodeAbbrev *Abbv = new BitCodeAbbrev(); + Abbv->Add(BitCodeAbbrevOp(bitc::FUNC_CODE_INST_UNREACHABLE)); + if (Stream.EmitBlockInfoAbbrev(bitc::FUNCTION_BLOCK_ID, + Abbv) != FUNCTION_INST_UNREACHABLE_ABBREV) + assert(0 && "Unexpected abbrev ordering!"); + } + + Stream.ExitBlock(); +} + + +/// WriteModule - Emit the specified module to the bitstream. +static void WriteModule(const Module *M, BitstreamWriter &Stream) { + Stream.EnterSubblock(bitc::MODULE_BLOCK_ID, 3); + + // Emit the version number if it is non-zero. + if (CurVersion) { + SmallVector<unsigned, 1> Vals; + Vals.push_back(CurVersion); + Stream.EmitRecord(bitc::MODULE_CODE_VERSION, Vals); + } + + // Analyze the module, enumerating globals, functions, etc. + ValueEnumerator VE(M); + + // Emit blockinfo, which defines the standard abbreviations etc. + WriteBlockInfo(VE, Stream); + + // Emit information about parameter attributes. + WriteAttributeTable(VE, Stream); + + // Emit information describing all of the types in the module. + WriteTypeTable(VE, Stream); + + // Emit top-level description of module, including target triple, inline asm, + // descriptors for global variables, and function prototype info. + WriteModuleInfo(M, VE, Stream); + + // Emit constants. + WriteModuleConstants(VE, Stream); + + // If we have any aggregate values in the value table, purge them - these can + // only be used to initialize global variables. Doing so makes the value + // namespace smaller for code in functions. + int NumNonAggregates = VE.PurgeAggregateValues(); + if (NumNonAggregates != -1) { + SmallVector<unsigned, 1> Vals; + Vals.push_back(NumNonAggregates); + Stream.EmitRecord(bitc::MODULE_CODE_PURGEVALS, Vals); + } + + // Emit function bodies. + for (Module::const_iterator I = M->begin(), E = M->end(); I != E; ++I) + if (!I->isDeclaration()) + WriteFunction(*I, VE, Stream); + + // Emit the type symbol table information. + WriteTypeSymbolTable(M->getTypeSymbolTable(), VE, Stream); + + // Emit names for globals/functions etc. + WriteValueSymbolTable(M->getValueSymbolTable(), VE, Stream); + + Stream.ExitBlock(); +} + +/// EmitDarwinBCHeader - If generating a bc file on darwin, we have to emit a +/// header and trailer to make it compatible with the system archiver. To do +/// this we emit the following header, and then emit a trailer that pads the +/// file out to be a multiple of 16 bytes. +/// +/// struct bc_header { +/// uint32_t Magic; // 0x0B17C0DE +/// uint32_t Version; // Version, currently always 0. +/// uint32_t BitcodeOffset; // Offset to traditional bitcode file. +/// uint32_t BitcodeSize; // Size of traditional bitcode file. +/// uint32_t CPUType; // CPU specifier. +/// ... potentially more later ... +/// }; +enum { + DarwinBCSizeFieldOffset = 3*4, // Offset to bitcode_size. + DarwinBCHeaderSize = 5*4 +}; + +static void EmitDarwinBCHeader(BitstreamWriter &Stream, + const std::string &TT) { + unsigned CPUType = ~0U; + + // Match x86_64-*, i[3-9]86-*, powerpc-*, powerpc64-*. The CPUType is a + // magic number from /usr/include/mach/machine.h. It is ok to reproduce the + // specific constants here because they are implicitly part of the Darwin ABI. + enum { + DARWIN_CPU_ARCH_ABI64 = 0x01000000, + DARWIN_CPU_TYPE_X86 = 7, + DARWIN_CPU_TYPE_POWERPC = 18 + }; + + if (TT.find("x86_64-") == 0) + CPUType = DARWIN_CPU_TYPE_X86 | DARWIN_CPU_ARCH_ABI64; + else if (TT.size() >= 5 && TT[0] == 'i' && TT[2] == '8' && TT[3] == '6' && + TT[4] == '-' && TT[1] - '3' < 6) + CPUType = DARWIN_CPU_TYPE_X86; + else if (TT.find("powerpc-") == 0) + CPUType = DARWIN_CPU_TYPE_POWERPC; + else if (TT.find("powerpc64-") == 0) + CPUType = DARWIN_CPU_TYPE_POWERPC | DARWIN_CPU_ARCH_ABI64; + + // Traditional Bitcode starts after header. + unsigned BCOffset = DarwinBCHeaderSize; + + Stream.Emit(0x0B17C0DE, 32); + Stream.Emit(0 , 32); // Version. + Stream.Emit(BCOffset , 32); + Stream.Emit(0 , 32); // Filled in later. + Stream.Emit(CPUType , 32); +} + +/// EmitDarwinBCTrailer - Emit the darwin epilog after the bitcode file and +/// finalize the header. +static void EmitDarwinBCTrailer(BitstreamWriter &Stream, unsigned BufferSize) { + // Update the size field in the header. + Stream.BackpatchWord(DarwinBCSizeFieldOffset, BufferSize-DarwinBCHeaderSize); + + // If the file is not a multiple of 16 bytes, insert dummy padding. + while (BufferSize & 15) { + Stream.Emit(0, 8); + ++BufferSize; + } +} + + +/// WriteBitcodeToFile - Write the specified module to the specified output +/// stream. +void llvm::WriteBitcodeToFile(const Module *M, std::ostream &Out) { + raw_os_ostream RawOut(Out); + // If writing to stdout, set binary mode. + if (llvm::cout == Out) + sys::Program::ChangeStdoutToBinary(); + WriteBitcodeToFile(M, RawOut); +} + +/// WriteBitcodeToFile - Write the specified module to the specified output +/// stream. +void llvm::WriteBitcodeToFile(const Module *M, raw_ostream &Out) { + std::vector<unsigned char> Buffer; + BitstreamWriter Stream(Buffer); + + Buffer.reserve(256*1024); + + WriteBitcodeToStream( M, Stream ); + + // If writing to stdout, set binary mode. + if (&llvm::outs() == &Out) + sys::Program::ChangeStdoutToBinary(); + + // Write the generated bitstream to "Out". + Out.write((char*)&Buffer.front(), Buffer.size()); + + // Make sure it hits disk now. + Out.flush(); +} + +/// WriteBitcodeToStream - Write the specified module to the specified output +/// stream. +void llvm::WriteBitcodeToStream(const Module *M, BitstreamWriter &Stream) { + // If this is darwin, emit a file header and trailer if needed. + bool isDarwin = M->getTargetTriple().find("-darwin") != std::string::npos; + if (isDarwin) + EmitDarwinBCHeader(Stream, M->getTargetTriple()); + + // Emit the file header. + Stream.Emit((unsigned)'B', 8); + Stream.Emit((unsigned)'C', 8); + Stream.Emit(0x0, 4); + Stream.Emit(0xC, 4); + Stream.Emit(0xE, 4); + Stream.Emit(0xD, 4); + + // Emit the module. + WriteModule(M, Stream); + + if (isDarwin) + EmitDarwinBCTrailer(Stream, Stream.getBuffer().size()); +} diff --git a/lib/Bitcode/Writer/BitcodeWriterPass.cpp b/lib/Bitcode/Writer/BitcodeWriterPass.cpp new file mode 100644 index 0000000..209cf09 --- /dev/null +++ b/lib/Bitcode/Writer/BitcodeWriterPass.cpp @@ -0,0 +1,56 @@ +//===--- Bitcode/Writer/BitcodeWriterPass.cpp - Bitcode Writer ------------===// +// +// The LLVM Compiler Infrastructure +// +// This file is distributed under the University of Illinois Open Source +// License. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// +// +// BitcodeWriterPass implementation. +// +//===----------------------------------------------------------------------===// + +#include "llvm/Bitcode/ReaderWriter.h" +#include "llvm/Pass.h" +using namespace llvm; + +namespace { + class WriteBitcodePass : public ModulePass { + // FIXME: Kill off std::ostream + std::ostream *Out; + raw_ostream *RawOut; // raw_ostream to print on + public: + static char ID; // Pass identification, replacement for typeid + explicit WriteBitcodePass(std::ostream &o) + : ModulePass(&ID), Out(&o), RawOut(0) {} + explicit WriteBitcodePass(raw_ostream &o) + : ModulePass(&ID), Out(0), RawOut(&o) {} + + const char *getPassName() const { return "Bitcode Writer"; } + + bool runOnModule(Module &M) { + if (Out) { + WriteBitcodeToFile(&M, *Out); + } else { + WriteBitcodeToFile(&M, *RawOut); + } + return false; + } + }; +} + +char WriteBitcodePass::ID = 0; + +/// CreateBitcodeWriterPass - Create and return a pass that writes the module +/// to the specified ostream. +ModulePass *llvm::CreateBitcodeWriterPass(std::ostream &Str) { + return new WriteBitcodePass(Str); +} + + +/// createBitcodeWriterPass - Create and return a pass that writes the module +/// to the specified ostream. +ModulePass *llvm::createBitcodeWriterPass(raw_ostream &Str) { + return new WriteBitcodePass(Str); +} diff --git a/lib/Bitcode/Writer/CMakeLists.txt b/lib/Bitcode/Writer/CMakeLists.txt new file mode 100644 index 0000000..ac5bb99 --- /dev/null +++ b/lib/Bitcode/Writer/CMakeLists.txt @@ -0,0 +1,9 @@ +add_llvm_library(LLVMBitWriter + BitWriter.cpp + BitcodeWriter.cpp + BitcodeWriterPass.cpp + Serialize.cpp + SerializeAPFloat.cpp + SerializeAPInt.cpp + ValueEnumerator.cpp + ) diff --git a/lib/Bitcode/Writer/Makefile b/lib/Bitcode/Writer/Makefile new file mode 100644 index 0000000..7b0bd72 --- /dev/null +++ b/lib/Bitcode/Writer/Makefile @@ -0,0 +1,15 @@ +##===- lib/Bitcode/Reader/Makefile -------------------------*- Makefile -*-===## +# +# The LLVM Compiler Infrastructure +# +# This file is distributed under the University of Illinois Open Source +# License. See LICENSE.TXT for details. +# +##===----------------------------------------------------------------------===## + +LEVEL = ../../.. +LIBRARYNAME = LLVMBitWriter +BUILD_ARCHIVE = 1 + +include $(LEVEL)/Makefile.common + diff --git a/lib/Bitcode/Writer/Serialize.cpp b/lib/Bitcode/Writer/Serialize.cpp new file mode 100644 index 0000000..79464a6 --- /dev/null +++ b/lib/Bitcode/Writer/Serialize.cpp @@ -0,0 +1,118 @@ +//==- Serialize.cpp - Generic Object Serialization to Bitcode ----*- C++ -*-==// +// +// The LLVM Compiler Infrastructure +// +// This file is distributed under the University of Illinois Open Source +// License. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// +// +// This file defines the internal methods used for object serialization. +// +//===----------------------------------------------------------------------===// + +#include "llvm/Bitcode/Serialize.h" +#include "string.h" + +#ifdef DEBUG_BACKPATCH +#include "llvm/Support/Streams.h" +#endif + +using namespace llvm; + +Serializer::Serializer(BitstreamWriter& stream) + : Stream(stream), BlockLevel(0) {} + +Serializer::~Serializer() { + if (inRecord()) + EmitRecord(); + + while (BlockLevel > 0) + Stream.ExitBlock(); + + Stream.FlushToWord(); +} + +void Serializer::EmitRecord() { + assert(Record.size() > 0 && "Cannot emit empty record."); + Stream.EmitRecord(8,Record); + Record.clear(); +} + +void Serializer::EnterBlock(unsigned BlockID,unsigned CodeLen) { + FlushRecord(); + Stream.EnterSubblock(BlockID,CodeLen); + ++BlockLevel; +} + +void Serializer::ExitBlock() { + assert (BlockLevel > 0); + --BlockLevel; + FlushRecord(); + Stream.ExitBlock(); +} + +void Serializer::EmitInt(uint64_t X) { + assert (BlockLevel > 0); + Record.push_back(X); +} + +void Serializer::EmitSInt(int64_t X) { + if (X >= 0) + EmitInt(X << 1); + else + EmitInt((-X << 1) | 1); +} + +void Serializer::EmitCStr(const char* s, const char* end) { + Record.push_back(end - s); + + while(s != end) { + Record.push_back(*s); + ++s; + } +} + +void Serializer::EmitCStr(const char* s) { + EmitCStr(s,s+strlen(s)); +} + +SerializedPtrID Serializer::getPtrId(const void* ptr) { + if (!ptr) + return 0; + + MapTy::iterator I = PtrMap.find(ptr); + + if (I == PtrMap.end()) { + unsigned id = PtrMap.size()+1; +#ifdef DEBUG_BACKPATCH + llvm::cerr << "Registered PTR: " << ptr << " => " << id << "\n"; +#endif + PtrMap[ptr] = id; + return id; + } + else return I->second; +} + +bool Serializer::isRegistered(const void* ptr) const { + MapTy::const_iterator I = PtrMap.find(ptr); + return I != PtrMap.end(); +} + + +#define INT_EMIT(TYPE)\ +void SerializeTrait<TYPE>::Emit(Serializer&S, TYPE X) { S.EmitInt(X); } + +INT_EMIT(bool) +INT_EMIT(unsigned char) +INT_EMIT(unsigned short) +INT_EMIT(unsigned int) +INT_EMIT(unsigned long) + +#define SINT_EMIT(TYPE)\ +void SerializeTrait<TYPE>::Emit(Serializer&S, TYPE X) { S.EmitSInt(X); } + +SINT_EMIT(signed char) +SINT_EMIT(signed short) +SINT_EMIT(signed int) +SINT_EMIT(signed long) diff --git a/lib/Bitcode/Writer/SerializeAPFloat.cpp b/lib/Bitcode/Writer/SerializeAPFloat.cpp new file mode 100644 index 0000000..25d954f --- /dev/null +++ b/lib/Bitcode/Writer/SerializeAPFloat.cpp @@ -0,0 +1,21 @@ +//===-- SerializeAPInt.cpp - Serialization for APFloat ---------*- C++ -*--===// +// +// The LLVM Compiler Infrastructure +// +// This file is distributed under the University of Illinois Open Source +// License. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// +// +// This file implements serialization of APFloat. +// +//===----------------------------------------------------------------------===// + +#include "llvm/ADT/APFloat.h" +#include "llvm/Bitcode/Serialize.h" + +using namespace llvm; + +void APFloat::Emit(Serializer& S) const { + S.Emit(bitcastToAPInt()); +} diff --git a/lib/Bitcode/Writer/SerializeAPInt.cpp b/lib/Bitcode/Writer/SerializeAPInt.cpp new file mode 100644 index 0000000..47792c7 --- /dev/null +++ b/lib/Bitcode/Writer/SerializeAPInt.cpp @@ -0,0 +1,31 @@ +//===-- SerializeAPInt.cpp - Serialization for APInts ----------*- C++ -*--===// +// +// The LLVM Compiler Infrastructure +// +// This file is distributed under the University of Illinois Open Source +// License. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// +// +// This file implements serialization of APInts. +// +//===----------------------------------------------------------------------===// + +#include "llvm/ADT/APInt.h" +#include "llvm/Bitcode/Serialize.h" +#include <cassert> + +using namespace llvm; + +void APInt::Emit(Serializer& S) const { + S.EmitInt(BitWidth); + + if (isSingleWord()) + S.EmitInt(VAL); + else { + uint32_t NumWords = getNumWords(); + S.EmitInt(NumWords); + for (unsigned i = 0; i < NumWords; ++i) + S.EmitInt(pVal[i]); + } +} diff --git a/lib/Bitcode/Writer/ValueEnumerator.cpp b/lib/Bitcode/Writer/ValueEnumerator.cpp new file mode 100644 index 0000000..8002a36 --- /dev/null +++ b/lib/Bitcode/Writer/ValueEnumerator.cpp @@ -0,0 +1,347 @@ +//===-- ValueEnumerator.cpp - Number values and types for bitcode writer --===// +// +// The LLVM Compiler Infrastructure +// +// This file is distributed under the University of Illinois Open Source +// License. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// +// +// This file implements the ValueEnumerator class. +// +//===----------------------------------------------------------------------===// + +#include "ValueEnumerator.h" +#include "llvm/Constants.h" +#include "llvm/DerivedTypes.h" +#include "llvm/MDNode.h" +#include "llvm/Module.h" +#include "llvm/TypeSymbolTable.h" +#include "llvm/ValueSymbolTable.h" +#include "llvm/Instructions.h" +#include <algorithm> +using namespace llvm; + +static bool isSingleValueType(const std::pair<const llvm::Type*, + unsigned int> &P) { + return P.first->isSingleValueType(); +} + +static bool isIntegerValue(const std::pair<const Value*, unsigned> &V) { + return isa<IntegerType>(V.first->getType()); +} + +static bool CompareByFrequency(const std::pair<const llvm::Type*, + unsigned int> &P1, + const std::pair<const llvm::Type*, + unsigned int> &P2) { + return P1.second > P2.second; +} + +/// ValueEnumerator - Enumerate module-level information. +ValueEnumerator::ValueEnumerator(const Module *M) { + // Enumerate the global variables. + for (Module::const_global_iterator I = M->global_begin(), + E = M->global_end(); I != E; ++I) + EnumerateValue(I); + + // Enumerate the functions. + for (Module::const_iterator I = M->begin(), E = M->end(); I != E; ++I) { + EnumerateValue(I); + EnumerateAttributes(cast<Function>(I)->getAttributes()); + } + + // Enumerate the aliases. + for (Module::const_alias_iterator I = M->alias_begin(), E = M->alias_end(); + I != E; ++I) + EnumerateValue(I); + + // Remember what is the cutoff between globalvalue's and other constants. + unsigned FirstConstant = Values.size(); + + // Enumerate the global variable initializers. + for (Module::const_global_iterator I = M->global_begin(), + E = M->global_end(); I != E; ++I) + if (I->hasInitializer()) + EnumerateValue(I->getInitializer()); + + // Enumerate the aliasees. + for (Module::const_alias_iterator I = M->alias_begin(), E = M->alias_end(); + I != E; ++I) + EnumerateValue(I->getAliasee()); + + // Enumerate types used by the type symbol table. + EnumerateTypeSymbolTable(M->getTypeSymbolTable()); + + // Insert constants that are named at module level into the slot pool so that + // the module symbol table can refer to them... + EnumerateValueSymbolTable(M->getValueSymbolTable()); + + // Enumerate types used by function bodies and argument lists. + for (Module::const_iterator F = M->begin(), E = M->end(); F != E; ++F) { + + for (Function::const_arg_iterator I = F->arg_begin(), E = F->arg_end(); + I != E; ++I) + EnumerateType(I->getType()); + + for (Function::const_iterator BB = F->begin(), E = F->end(); BB != E; ++BB) + for (BasicBlock::const_iterator I = BB->begin(), E = BB->end(); I!=E;++I){ + for (User::const_op_iterator OI = I->op_begin(), E = I->op_end(); + OI != E; ++OI) + EnumerateOperandType(*OI); + EnumerateType(I->getType()); + if (const CallInst *CI = dyn_cast<CallInst>(I)) + EnumerateAttributes(CI->getAttributes()); + else if (const InvokeInst *II = dyn_cast<InvokeInst>(I)) + EnumerateAttributes(II->getAttributes()); + } + } + + // Optimize constant ordering. + OptimizeConstants(FirstConstant, Values.size()); + + // Sort the type table by frequency so that most commonly used types are early + // in the table (have low bit-width). + std::stable_sort(Types.begin(), Types.end(), CompareByFrequency); + + // Partition the Type ID's so that the single-value types occur before the + // aggregate types. This allows the aggregate types to be dropped from the + // type table after parsing the global variable initializers. + std::partition(Types.begin(), Types.end(), isSingleValueType); + + // Now that we rearranged the type table, rebuild TypeMap. + for (unsigned i = 0, e = Types.size(); i != e; ++i) + TypeMap[Types[i].first] = i+1; +} + +// Optimize constant ordering. +namespace { + struct CstSortPredicate { + ValueEnumerator &VE; + explicit CstSortPredicate(ValueEnumerator &ve) : VE(ve) {} + bool operator()(const std::pair<const Value*, unsigned> &LHS, + const std::pair<const Value*, unsigned> &RHS) { + // Sort by plane. + if (LHS.first->getType() != RHS.first->getType()) + return VE.getTypeID(LHS.first->getType()) < + VE.getTypeID(RHS.first->getType()); + // Then by frequency. + return LHS.second > RHS.second; + } + }; +} + +/// OptimizeConstants - Reorder constant pool for denser encoding. +void ValueEnumerator::OptimizeConstants(unsigned CstStart, unsigned CstEnd) { + if (CstStart == CstEnd || CstStart+1 == CstEnd) return; + + CstSortPredicate P(*this); + std::stable_sort(Values.begin()+CstStart, Values.begin()+CstEnd, P); + + // Ensure that integer constants are at the start of the constant pool. This + // is important so that GEP structure indices come before gep constant exprs. + std::partition(Values.begin()+CstStart, Values.begin()+CstEnd, + isIntegerValue); + + // Rebuild the modified portion of ValueMap. + for (; CstStart != CstEnd; ++CstStart) + ValueMap[Values[CstStart].first] = CstStart+1; +} + + +/// EnumerateTypeSymbolTable - Insert all of the types in the specified symbol +/// table. +void ValueEnumerator::EnumerateTypeSymbolTable(const TypeSymbolTable &TST) { + for (TypeSymbolTable::const_iterator TI = TST.begin(), TE = TST.end(); + TI != TE; ++TI) + EnumerateType(TI->second); +} + +/// EnumerateValueSymbolTable - Insert all of the values in the specified symbol +/// table into the values table. +void ValueEnumerator::EnumerateValueSymbolTable(const ValueSymbolTable &VST) { + for (ValueSymbolTable::const_iterator VI = VST.begin(), VE = VST.end(); + VI != VE; ++VI) + EnumerateValue(VI->getValue()); +} + +void ValueEnumerator::EnumerateValue(const Value *V) { + assert(V->getType() != Type::VoidTy && "Can't insert void values!"); + + // Check to see if it's already in! + unsigned &ValueID = ValueMap[V]; + if (ValueID) { + // Increment use count. + Values[ValueID-1].second++; + return; + } + + // Enumerate the type of this value. + EnumerateType(V->getType()); + + if (const Constant *C = dyn_cast<Constant>(V)) { + if (isa<GlobalValue>(C)) { + // Initializers for globals are handled explicitly elsewhere. + } else if (isa<ConstantArray>(C) && cast<ConstantArray>(C)->isString()) { + // Do not enumerate the initializers for an array of simple characters. + // The initializers just polute the value table, and we emit the strings + // specially. + } else if (C->getNumOperands()) { + // If a constant has operands, enumerate them. This makes sure that if a + // constant has uses (for example an array of const ints), that they are + // inserted also. + + // We prefer to enumerate them with values before we enumerate the user + // itself. This makes it more likely that we can avoid forward references + // in the reader. We know that there can be no cycles in the constants + // graph that don't go through a global variable. + for (User::const_op_iterator I = C->op_begin(), E = C->op_end(); + I != E; ++I) + EnumerateValue(*I); + + // Finally, add the value. Doing this could make the ValueID reference be + // dangling, don't reuse it. + Values.push_back(std::make_pair(V, 1U)); + ValueMap[V] = Values.size(); + return; + } else if (const MDNode *N = dyn_cast<MDNode>(C)) { + for (MDNode::const_elem_iterator I = N->elem_begin(), E = N->elem_end(); + I != E; ++I) { + if (*I) + EnumerateValue(*I); + else + EnumerateType(Type::VoidTy); + } + + Values.push_back(std::make_pair(V, 1U)); + ValueMap[V] = Values.size(); + return; + } + } + + // Add the value. + Values.push_back(std::make_pair(V, 1U)); + ValueID = Values.size(); +} + + +void ValueEnumerator::EnumerateType(const Type *Ty) { + unsigned &TypeID = TypeMap[Ty]; + + if (TypeID) { + // If we've already seen this type, just increase its occurrence count. + Types[TypeID-1].second++; + return; + } + + // First time we saw this type, add it. + Types.push_back(std::make_pair(Ty, 1U)); + TypeID = Types.size(); + + // Enumerate subtypes. + for (Type::subtype_iterator I = Ty->subtype_begin(), E = Ty->subtype_end(); + I != E; ++I) + EnumerateType(*I); +} + +// Enumerate the types for the specified value. If the value is a constant, +// walk through it, enumerating the types of the constant. +void ValueEnumerator::EnumerateOperandType(const Value *V) { + EnumerateType(V->getType()); + if (const Constant *C = dyn_cast<Constant>(V)) { + // If this constant is already enumerated, ignore it, we know its type must + // be enumerated. + if (ValueMap.count(V)) return; + + // This constant may have operands, make sure to enumerate the types in + // them. + for (unsigned i = 0, e = C->getNumOperands(); i != e; ++i) + EnumerateOperandType(C->getOperand(i)); + + if (const MDNode *N = dyn_cast<MDNode>(V)) { + for (unsigned i = 0, e = N->getNumElements(); i != e; ++i) + EnumerateOperandType(N->getElement(i)); + } + } +} + +void ValueEnumerator::EnumerateAttributes(const AttrListPtr &PAL) { + if (PAL.isEmpty()) return; // null is always 0. + // Do a lookup. + unsigned &Entry = AttributeMap[PAL.getRawPointer()]; + if (Entry == 0) { + // Never saw this before, add it. + Attributes.push_back(PAL); + Entry = Attributes.size(); + } +} + + +/// PurgeAggregateValues - If there are any aggregate values at the end of the +/// value list, remove them and return the count of the remaining values. If +/// there are none, return -1. +int ValueEnumerator::PurgeAggregateValues() { + // If there are no aggregate values at the end of the list, return -1. + if (Values.empty() || Values.back().first->getType()->isSingleValueType()) + return -1; + + // Otherwise, remove aggregate values... + while (!Values.empty() && !Values.back().first->getType()->isSingleValueType()) + Values.pop_back(); + + // ... and return the new size. + return Values.size(); +} + +void ValueEnumerator::incorporateFunction(const Function &F) { + NumModuleValues = Values.size(); + + // Adding function arguments to the value table. + for(Function::const_arg_iterator I = F.arg_begin(), E = F.arg_end(); + I != E; ++I) + EnumerateValue(I); + + FirstFuncConstantID = Values.size(); + + // Add all function-level constants to the value table. + for (Function::const_iterator BB = F.begin(), E = F.end(); BB != E; ++BB) { + for (BasicBlock::const_iterator I = BB->begin(), E = BB->end(); I!=E; ++I) + for (User::const_op_iterator OI = I->op_begin(), E = I->op_end(); + OI != E; ++OI) { + if ((isa<Constant>(*OI) && !isa<GlobalValue>(*OI)) || + isa<InlineAsm>(*OI)) + EnumerateValue(*OI); + } + BasicBlocks.push_back(BB); + ValueMap[BB] = BasicBlocks.size(); + } + + // Optimize the constant layout. + OptimizeConstants(FirstFuncConstantID, Values.size()); + + // Add the function's parameter attributes so they are available for use in + // the function's instruction. + EnumerateAttributes(F.getAttributes()); + + FirstInstID = Values.size(); + + // Add all of the instructions. + for (Function::const_iterator BB = F.begin(), E = F.end(); BB != E; ++BB) { + for (BasicBlock::const_iterator I = BB->begin(), E = BB->end(); I!=E; ++I) { + if (I->getType() != Type::VoidTy) + EnumerateValue(I); + } + } +} + +void ValueEnumerator::purgeFunction() { + /// Remove purged values from the ValueMap. + for (unsigned i = NumModuleValues, e = Values.size(); i != e; ++i) + ValueMap.erase(Values[i].first); + for (unsigned i = 0, e = BasicBlocks.size(); i != e; ++i) + ValueMap.erase(BasicBlocks[i]); + + Values.resize(NumModuleValues); + BasicBlocks.clear(); +} + diff --git a/lib/Bitcode/Writer/ValueEnumerator.h b/lib/Bitcode/Writer/ValueEnumerator.h new file mode 100644 index 0000000..bb0324b --- /dev/null +++ b/lib/Bitcode/Writer/ValueEnumerator.h @@ -0,0 +1,127 @@ +//===-- Bitcode/Writer/ValueEnumerator.h - Number values --------*- C++ -*-===// +// +// The LLVM Compiler Infrastructure +// +// This file is distributed under the University of Illinois Open Source +// License. See LICENSE.TXT for details. +// +//===----------------------------------------------------------------------===// +// +// This class gives values and types Unique ID's. +// +//===----------------------------------------------------------------------===// + +#ifndef VALUE_ENUMERATOR_H +#define VALUE_ENUMERATOR_H + +#include "llvm/ADT/DenseMap.h" +#include "llvm/Attributes.h" +#include <vector> + +namespace llvm { + +class Type; +class Value; +class BasicBlock; +class Function; +class Module; +class AttrListPtr; +class TypeSymbolTable; +class ValueSymbolTable; + +class ValueEnumerator { +public: + // For each type, we remember its Type* and occurrence frequency. + typedef std::vector<std::pair<const Type*, unsigned> > TypeList; + + // For each value, we remember its Value* and occurrence frequency. + typedef std::vector<std::pair<const Value*, unsigned> > ValueList; +private: + typedef DenseMap<const Type*, unsigned> TypeMapType; + TypeMapType TypeMap; + TypeList Types; + + typedef DenseMap<const Value*, unsigned> ValueMapType; + ValueMapType ValueMap; + ValueList Values; + + typedef DenseMap<void*, unsigned> AttributeMapType; + AttributeMapType AttributeMap; + std::vector<AttrListPtr> Attributes; + + /// BasicBlocks - This contains all the basic blocks for the currently + /// incorporated function. Their reverse mapping is stored in ValueMap. + std::vector<const BasicBlock*> BasicBlocks; + + /// When a function is incorporated, this is the size of the Values list + /// before incorporation. + unsigned NumModuleValues; + unsigned FirstFuncConstantID; + unsigned FirstInstID; + + ValueEnumerator(const ValueEnumerator &); // DO NOT IMPLEMENT + void operator=(const ValueEnumerator &); // DO NOT IMPLEMENT +public: + ValueEnumerator(const Module *M); + + unsigned getValueID(const Value *V) const { + ValueMapType::const_iterator I = ValueMap.find(V); + assert(I != ValueMap.end() && "Value not in slotcalculator!"); + return I->second-1; + } + + unsigned getTypeID(const Type *T) const { + TypeMapType::const_iterator I = TypeMap.find(T); + assert(I != TypeMap.end() && "Type not in ValueEnumerator!"); + return I->second-1; + } + + unsigned getAttributeID(const AttrListPtr &PAL) const { + if (PAL.isEmpty()) return 0; // Null maps to zero. + AttributeMapType::const_iterator I = AttributeMap.find(PAL.getRawPointer()); + assert(I != AttributeMap.end() && "Attribute not in ValueEnumerator!"); + return I->second; + } + + /// getFunctionConstantRange - Return the range of values that corresponds to + /// function-local constants. + void getFunctionConstantRange(unsigned &Start, unsigned &End) const { + Start = FirstFuncConstantID; + End = FirstInstID; + } + + const ValueList &getValues() const { return Values; } + const TypeList &getTypes() const { return Types; } + const std::vector<const BasicBlock*> &getBasicBlocks() const { + return BasicBlocks; + } + const std::vector<AttrListPtr> &getAttributes() const { + return Attributes; + } + + /// PurgeAggregateValues - If there are any aggregate values at the end of the + /// value list, remove them and return the count of the remaining values. If + /// there are none, return -1. + int PurgeAggregateValues(); + + /// incorporateFunction/purgeFunction - If you'd like to deal with a function, + /// use these two methods to get its data into the ValueEnumerator! + /// + void incorporateFunction(const Function &F); + void purgeFunction(); + +private: + void OptimizeConstants(unsigned CstStart, unsigned CstEnd); + + void EnumerateValue(const Value *V); + void EnumerateType(const Type *T); + void EnumerateOperandType(const Value *V); + void EnumerateAttributes(const AttrListPtr &PAL); + + void EnumerateTypeSymbolTable(const TypeSymbolTable &ST); + void EnumerateValueSymbolTable(const ValueSymbolTable &ST); +}; + +} // End llvm namespace + +#endif |
