編譯器設計概論

This course is an introduction to development tools for embedded systems. Our focus will be on the compiler construction, from the most basic to some introduction to optimization, object code format, and some topics related to embedded systems. In the introduction to the design and implementation of compilers for an imperative language, we will study various components in a compiler, including scanners, parsers, symbol tables, semantic analyzers, and code generators, and the overall organization of a compiler. In the scanners, we will study regular expressions and the transformation of a regular expression into a deterministic finite state machine. In the parsers, we will study LL and LR parsing methods and the construction of various parsing tables. There are many tools that can generate scanners and parsers from formal specifications. We will also study these tools and use them to generate the scanner and parser for our project. We will study and implement the code generation components of a compiler. The symbol tables are a data structure shared by most components in the compiler. We will study how to process variables, types, expressions, records, arrays, pointers, and simple control structures. Procedures and packages will also be studied. Finally, we will study techniques for code generation and local optimization. If time permits, we will study global optimization techniques in compilers. This course will include an integrated programming project. We will give you a detailed specification of a simplified language that is very similar to Ada. It will be our target language. We will also supply the documents of a scanner generator and of a parser generator. A target language and its simulator will also be provided. You will implement a scanner, a parser, and other components used to generate code for a hypothetical machine. To help you make progress, we will set three check points during the semester and a final demonstration at the end of the semester. You will need to finish a scanner, a parser, and a primitive code generator at the checkpoints, respectively. At the end of the semester, you will need to show (off) a complete compiler that generate code for the hypothetical machine. We will provide a simulator for the machine to run your object code. In the project, you will need to implement a minimum set of features, which will be announced later in the class. If you implement extra features, you will earn extra points.

programming, data structure

無備註

http://www.cs.nctu.edu.tw/~wuuyang/Lecture/lecture.compiler.html

期中考(30%) 期末考(30%) 程式作業 (40%)

• introduction

  introduction to compiler organization

  講授：

  3

• a simple compiler

  Show a simple compiler.

  講授：

  3

• scanning and its implementation

  regular expressions

  講授：

  3

  實作：

  6

• parsing

  grammars and parsing LL(1) grammars and parsers LR parsing

  講授：

  9

  實作：

  9

• basic code generation

  semantic processing symbol table run-time storage organization processing declarations processing expressions and data struct references translating control structures Translating procedures and functions

  講授：

  33

  實作：

  40

• Java and JVM

  Java code generation

  講授：

  3

C.N. Fischer, R.K. Cytron, and R.J. LeBlanc, Jr., Crafting a Compiler, Addison-Wesley, 2010.

地點

EC332B

時間

星期二 1:20-3:10 PM.

聯絡方式

office phone: 56614, email: wuuyang@cs.nctu.edu.tw