1 //===- BitCodes.h - Enum values for the bitstream format --------*- C++ -*-===// 2 // 3 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. 4 // See https://llvm.org/LICENSE.txt for license information. 5 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception 6 // 7 //===----------------------------------------------------------------------===// 8 // 9 // This header defines bitstream enum values. 10 // 11 // The enum values defined in this file should be considered permanent. If 12 // new features are added, they should have values added at the end of the 13 // respective lists. 14 // 15 //===----------------------------------------------------------------------===// 16 17 #ifndef LLVM_BITSTREAM_BITCODES_H 18 #define LLVM_BITSTREAM_BITCODES_H 19 20 #include "llvm/ADT/SmallVector.h" 21 #include "llvm/ADT/StringExtras.h" 22 #include "llvm/Support/DataTypes.h" 23 #include "llvm/Support/ErrorHandling.h" 24 #include <cassert> 25 26 namespace llvm { 27 /// Offsets of the 32-bit fields of bitstream wrapper header. 28 enum BitstreamWrapperHeader : unsigned { 29 BWH_MagicField = 0 * 4, 30 BWH_VersionField = 1 * 4, 31 BWH_OffsetField = 2 * 4, 32 BWH_SizeField = 3 * 4, 33 BWH_CPUTypeField = 4 * 4, 34 BWH_HeaderSize = 5 * 4 35 }; 36 37 namespace bitc { 38 enum StandardWidths { 39 BlockIDWidth = 8, // We use VBR-8 for block IDs. 40 CodeLenWidth = 4, // Codelen are VBR-4. 41 BlockSizeWidth = 32 // BlockSize up to 2^32 32-bit words = 16GB per block. 42 }; 43 44 // The standard abbrev namespace always has a way to exit a block, enter a 45 // nested block, define abbrevs, and define an unabbreviated record. 46 enum FixedAbbrevIDs { 47 END_BLOCK = 0, // Must be zero to guarantee termination for broken bitcode. 48 ENTER_SUBBLOCK = 1, 49 50 /// DEFINE_ABBREV - Defines an abbrev for the current block. It consists 51 /// of a vbr5 for # operand infos. Each operand info is emitted with a 52 /// single bit to indicate if it is a literal encoding. If so, the value is 53 /// emitted with a vbr8. If not, the encoding is emitted as 3 bits followed 54 /// by the info value as a vbr5 if needed. 55 DEFINE_ABBREV = 2, 56 57 // UNABBREV_RECORDs are emitted with a vbr6 for the record code, followed by 58 // a vbr6 for the # operands, followed by vbr6's for each operand. 59 UNABBREV_RECORD = 3, 60 61 // This is not a code, this is a marker for the first abbrev assignment. 62 FIRST_APPLICATION_ABBREV = 4 63 }; 64 65 /// StandardBlockIDs - All bitcode files can optionally include a BLOCKINFO 66 /// block, which contains metadata about other blocks in the file. 67 enum StandardBlockIDs { 68 /// BLOCKINFO_BLOCK is used to define metadata about blocks, for example, 69 /// standard abbrevs that should be available to all blocks of a specified 70 /// ID. 71 BLOCKINFO_BLOCK_ID = 0, 72 73 // Block IDs 1-7 are reserved for future expansion. 74 FIRST_APPLICATION_BLOCKID = 8 75 }; 76 77 /// BlockInfoCodes - The blockinfo block contains metadata about user-defined 78 /// blocks. 79 enum BlockInfoCodes { 80 // DEFINE_ABBREV has magic semantics here, applying to the current SETBID'd 81 // block, instead of the BlockInfo block. 82 83 BLOCKINFO_CODE_SETBID = 1, // SETBID: [blockid#] 84 BLOCKINFO_CODE_BLOCKNAME = 2, // BLOCKNAME: [name] 85 BLOCKINFO_CODE_SETRECORDNAME = 3 // BLOCKINFO_CODE_SETRECORDNAME: 86 // [id, name] 87 }; 88 89 } // End bitc namespace 90 91 /// BitCodeAbbrevOp - This describes one or more operands in an abbreviation. 92 /// This is actually a union of two different things: 93 /// 1. It could be a literal integer value ("the operand is always 17"). 94 /// 2. It could be an encoding specification ("this operand encoded like so"). 95 /// 96 class BitCodeAbbrevOp { 97 uint64_t Val; // A literal value or data for an encoding. 98 bool IsLiteral : 1; // Indicate whether this is a literal value or not. 99 unsigned Enc : 3; // The encoding to use. 100 public: 101 enum Encoding { 102 Fixed = 1, // A fixed width field, Val specifies number of bits. 103 VBR = 2, // A VBR field where Val specifies the width of each chunk. 104 Array = 3, // A sequence of fields, next field species elt encoding. 105 Char6 = 4, // A 6-bit fixed field which maps to [a-zA-Z0-9._]. 106 Blob = 5 // 32-bit aligned array of 8-bit characters. 107 }; 108 109 static bool isValidEncoding(uint64_t E) { 110 return E >= 1 && E <= 5; 111 } 112 113 explicit BitCodeAbbrevOp(uint64_t V) : Val(V), IsLiteral(true) {} 114 explicit BitCodeAbbrevOp(Encoding E, uint64_t Data = 0) 115 : Val(Data), IsLiteral(false), Enc(E) {} 116 117 bool isLiteral() const { return IsLiteral; } 118 bool isEncoding() const { return !IsLiteral; } 119 120 // Accessors for literals. 121 uint64_t getLiteralValue() const { assert(isLiteral()); return Val; } 122 123 // Accessors for encoding info. 124 Encoding getEncoding() const { assert(isEncoding()); return (Encoding)Enc; } 125 uint64_t getEncodingData() const { 126 assert(isEncoding() && hasEncodingData()); 127 return Val; 128 } 129 130 bool hasEncodingData() const { return hasEncodingData(getEncoding()); } 131 static bool hasEncodingData(Encoding E) { 132 switch (E) { 133 case Fixed: 134 case VBR: 135 return true; 136 case Array: 137 case Char6: 138 case Blob: 139 return false; 140 } 141 report_fatal_error("Invalid encoding"); 142 } 143 144 /// isChar6 - Return true if this character is legal in the Char6 encoding. 145 static bool isChar6(char C) { return isAlnum(C) || C == '.' || C == '_'; } 146 static unsigned EncodeChar6(char C) { 147 if (C >= 'a' && C <= 'z') return C-'a'; 148 if (C >= 'A' && C <= 'Z') return C-'A'+26; 149 if (C >= '0' && C <= '9') return C-'0'+26+26; 150 if (C == '.') return 62; 151 if (C == '_') return 63; 152 llvm_unreachable("Not a value Char6 character!"); 153 } 154 155 static char DecodeChar6(unsigned V) { 156 assert((V & ~63) == 0 && "Not a Char6 encoded character!"); 157 return "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789._" 158 [V]; 159 } 160 161 }; 162 163 /// BitCodeAbbrev - This class represents an abbreviation record. An 164 /// abbreviation allows a complex record that has redundancy to be stored in a 165 /// specialized format instead of the fully-general, fully-vbr, format. 166 class BitCodeAbbrev { 167 SmallVector<BitCodeAbbrevOp, 32> OperandList; 168 169 public: 170 BitCodeAbbrev() = default; 171 172 explicit BitCodeAbbrev(std::initializer_list<BitCodeAbbrevOp> OperandList) 173 : OperandList(OperandList) {} 174 175 unsigned getNumOperandInfos() const { 176 return static_cast<unsigned>(OperandList.size()); 177 } 178 const BitCodeAbbrevOp &getOperandInfo(unsigned N) const { 179 return OperandList[N]; 180 } 181 182 void Add(const BitCodeAbbrevOp &OpInfo) { 183 OperandList.push_back(OpInfo); 184 } 185 }; 186 } // End llvm namespace 187 188 #endif 189