-
-
[原创]入门编译原理之前端对接LLVM IR
-
2023-9-5 21:30
-
接上篇[原创]入门编译原理之前端体验，本篇将前端解析到的AST对接到LLVM IR。本文目标明确，就直贴代码了。
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 | import llvmlite.ir as irimport llvmlite.binding as llvm# 初始化LLVMllvm.initialize()llvm.initialize_native_target()llvm.initialize_native_asmprinter()# 创建LLVM模块module = ir.Module(name="my_module")# 创建一个LLVM上下文和函数context = ir.Context()function = None# 创建LLVM IR构建器builder = None# 创建一个LLVM上下文和函数# context = llvm.getGlobalContext()# function = None# Token 类用于表示词法分析器生成的令牌class Token: def __init__(self, type, value=None): self.type = type self.value = value# Lexer 类负责将输入文本解析成令牌流class Lexer: def __init__(self, text): self.text = text # 要解析的文本 self.pos = 0 # 当前解析位置 # 获取下一个令牌 def get_next_token(self): if self.pos >= len(self.text): return Token("EOF") # 如果已经到达文本末尾，返回一个表示结束的令牌 current_char = self.text[self.pos] # 如果当前字符是字母，则解析标识符 if current_char.isalpha(): identifier = "" while self.pos < len(self.text) and self.text[self.pos].isalnum(): identifier += self.text[self.pos] self.pos += 1 return Token("IDENTIFIER", identifier) # 如果当前字符是数字，则解析数字 if current_char.isdigit(): self.pos += 1 return Token("NUMBER", int(current_char)) # 如果当前字符是运算符，则解析运算符 if current_char in "+-*/": self.pos += 1 return Token("OPERATOR", current_char) # 如果当前字符是等号，则解析为赋值符号 if current_char == "=": self.pos += 1 return Token("ASSIGN", "=") # 如果当前字符是分号，则解析为分号 if current_char == ";": self.pos += 1 return Token("SEMICOLON", ";") # 如果当前字符是左括号，则解析为左括号 if current_char == "(": self.pos += 1 return Token("LPAREN", "(") # 如果当前字符是右括号，则解析为右括号 if current_char == ")": self.pos += 1 return Token("RPAREN", ")") # 如果当前字符是空格或制表符，则忽略并获取下一个令牌 if current_char in " \t": self.pos += 1 return self.get_next_token() raise ValueError("Invalid character") # 如果遇到无法识别的字符，引发异常# Parser 类负责解析令牌流并计算结果class Parser: def __init__(self, lexer): self.lexer = lexer # 词法分析器 self.current_token = self.lexer.get_next_token() # 当前令牌 self.variables = {} # 存储变量名和值的字典 # 解析整个表达式并生成LLVM IR def parse(self): global builder, function results = [] # 创建主函数 main_function_type = ir.FunctionType(ir.IntType(32), ()) function = ir.Function(module, main_function_type, name="main") block = function.append_basic_block(name="entry") builder = ir.IRBuilder(block) while self.current_token.type != "EOF": result = self.parse_statement() results.append(result) if self.current_token.type == "SEMICOLON": self.eat("SEMICOLON") builder.ret(results[-1]) return results # 解析表达式并生成LLVM IR def parse_expression(self, min_precedence=0): left = self.parse_atom() while self.current_token.type == "OPERATOR" and self.precedence(self.current_token.value) >= min_precedence: operator = self.current_token.value self.eat("OPERATOR") right = self.parse_expression(self.precedence(operator) + 1) if operator == "+": result = builder.add(left, right, name="addtmp") elif operator == "-": result = builder.sub(left, right, name="subtmp") elif operator == "*": result = builder.mul(left, right, name="multmp") elif operator == "/": result = builder.sdiv(left, right, name="divtmp") left = result return left # 解析原子表达式并生成LLVM IR def parse_atom(self): if self.current_token.type == "NUMBER": value = self.current_token.value self.eat("NUMBER") return ir.Constant(ir.IntType(32), value) elif self.current_token.type == "IDENTIFIER": variable_name = self.current_token.value self.eat("IDENTIFIER") if variable_name in self.variables: return self.variables[variable_name] else: raise ValueError(f"Undefined variable: {variable_name}") elif self.current_token.type == "LPAREN": self.eat("LPAREN") expression = self.parse_expression() self.eat("RPAREN") return expression else: raise ValueError("Invalid syntax") def parse_statement(self): if self.current_token.type == "IDENTIFIER": variable_name = self.current_token.value self.eat("IDENTIFIER") self.eat("ASSIGN") expression_value = self.parse_expression() self.eat("SEMICOLON") self.variables[variable_name] = expression_value return expression_value else: return self.parse_expression() def eat(self, token_type): if self.current_token.type == token_type: self.current_token = self.lexer.get_next_token() else: raise ValueError("Unexpected token") def precedence(self, operator): precedence = {"+": 1, "-": 1, "*": 2, "/": 2} return precedence.get(operator, 0) def apply_operator(self, left, operator, right): if operator == "+": return left + right elif operator == "-": return left - right elif operator == "*": return left * right elif operator == "/": return left / right# 计算函数，接受一个表达式并返回计算结果def calculate(expression): lexer = Lexer(expression) parser = Parser(lexer) results = parser.parse() # 使用 str 方法将 LLVM IR 内部表示转换为格式化的LLVM IR 汇编代码 formatted_ir = str(module) # 打印格式化的 LLVM IR 汇编代码 # print(formatted_ir) return results, formatted_ir'''# 测试代码expression = "x = 3 * (4-1); y = 2*x + 2;"# expression = "7"results = calculate(expression)print(results) # 输出结果为 [12, 14, None]'''# 测试代码expression = "x = 3 * (4-1); y = 2*x + 2;"llvm_ir = calculate(expression)print("\n------------------------------------------")print("LLVM IR的内部表示:\n", llvm_ir[0])print("\n------------------------------------------")print("LLVM IR的内部表示格式化和排版的LLVM IR汇编代码:\n", llvm_ir[1]) |
输出结果:
下篇就是将IR转成汇编语言了... ...
[CTF入门培训]顶尖高校博士及硕士团队亲授《30小时教你玩转CTF》，视频+靶场+题目！助力进入CTF世界
最后于 2023-9-5 21:37
被_THINCT编辑
，原因：
