Compilers: Principles, Techniques, and Tools
Chapter 1 Introduction
Dongjie He
University of New South Wales
https://dongjiehe.github.io/teaching/compiler/
29 Jun 2023
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 1 / 21
Outline
1 Programming Languages
Evolution
Basics
2 Language Processors
Compilers, Interpreters, and Other Language Processors
A Language-Processing System
3 Compiler Structure
Lexical Analysis
Syntax Analysis
Semantic Analysis
Intermediate Code Generation and Optimization
Code Generation
Symbol-Table Management
4 Applications of Compiler Technology
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 2 / 21
Programming Languages Evolution
Evolution of Programming Languages
Machine Languages: 1940’s, sequences of 0’s and 1’s
programming was slow, tedious, and error-prone.
programs were hard to understand and modify.
Assembly Languages: early 1950’s
mnemonics of machine instructions
macros: parameterized shorthands for frequently used instructions
High-Level Languages: 1955 to present
programming is easier, more natural, and more robust.
General Purpose Languages: Fortran, Cobol, Lisp, C, C++, Java, ...
Domain-Specific Languages: NOMAD, SQL, Postscript, ...
Logic- and constraint-based Languages: Prolog, OPS5, ...
Other Classification: 0
Imperative languages: C, C++, Java, Rust, ...
Functional languages: ML, Haskell, OCaml, ...
Constraint logic languages: Prolog, Datalog, ...
Von Neumann language/Object-oriented language/Scripting
.
language
. . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 3 / 21
Programming Languages Basics
Programming Language Basics I
A language policy allows the compiler to decide an issue.
Static policy: the issue is decided at compile time.
Dynamic policy: the issue is decided at run time.
e.g., “static int x;”, “static” in Java enables the compiler to
determine the location of ‘x’ in memory.
The scope of a declaration of x is the region in which uses of x refer
to this declaration.
Static scope: can be determined by looking only at the program.
Dynamic scope: determined by the program runs.
int i; // global i (defvar x 1) ; global variable 'x' with value 1
void f(...) { (defun foo () (message "The value of 'x' in foo: %s" x))
int i; // local i (defun bar ()
i = 3; // use of local i (let ((x 2)) ; local variable 'x' with value 2
} (foo))) ; Call foo from within bar
x = i + 1; // use of global i (bar) ; Call bar from the global scope
A C/Java Program An Emacs-Lisp Program
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 4 / 21
Programming Languages Basics
Programming Language Basics II
A block is a grouping of declarations and statements.
C family languages use braces { and } to delimit a block.
Algol and Pascal use begin and end.
Syntax of blocks in C
stmt := block | · · · block := declarations stmts
Block structure: blocks nested inside each other
Static-scope of a declaration D (which belongs to block B) of name x
B is the most closely nested block containing D
the scope of D is all of B, except for any blocks B′ nested to any depth
within B, in which x is redeclared.
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 5 / 21
Programming Languages Basics
Programming Language Basics III
Member Scope in Classes/Structures
the scope of a member declaration x in a class C extends to any
subclass C′ , except if C′ has a local declaration of the same name x
Explicit Access Control
public/protected/private
Parameter Passing Mechanisms
Actual Parameters/Formal Parameters
e.g., “A id(A p) {return p;} r = id(a); ”
#include <stdio.h> #include <stdio.h> procedure swap(a,b); Algol-60
void incrementByValue(int num) { void incrementByReference(int* num) { integer a, b;
num = num + 1; (*num) = (*num) + 1; begin
printf("Inside function: %d\n", num); printf("Inside function: %d\n", *num); integer tmp;
} } tmp := a; a := b; b := tmp;
int main() { int main() { end;
int x = 5; int x = 5; integer i, x;
printf("Before function call: %d\n", x); printf("Before function call: %d\n", x); integer array arr[0:9];
incrementByValue(x); incrementByReference(&x); x := 2;
printf("After function call: %d\n", x); printf("After function call: %d\n", x); for i := 0 step 1 until 9 do arr[i] := 10 - i;
return 0; return 0; swap(x, arr[x]);
} } print(x); printArray(arr);
(a) Call by Value (b) Call by Reference . . . . . . (c)
. Call
. . by. Name
. . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 6 / 21
Programming Languages Basics
Programming Language Basics III
Environments and States
environment: a mapping from names to locations
state: a mapping from locations to their values
e.g., “x = y + 1” changes the value in the location denoted by name x
Aliasing
x and y are aliases of each other if they can refer to the same location
#include <stdio.h>
int main() {
int x[3] = {1, 2, 3};
int* y = x;
y[1] = 4;
printf("%d", x[1]);
return 0;
} . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 7 / 21
Language Processors Compilers, Interpreters, and Other Language Processors
Compilers and Interpreters
Compilers
source program Compiler target program
e.g., hello.c e.g., gcc e.g., hello.o
Interpreters
source program
e.g., hello.py Interpreter output
input e.g., python3
Hybrid Compilers
Translator source program
e.g., javac e.g., hello.java
intermediate program
e.g., hello.class Virtual Machine output
input e.g., JVM
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 8 / 21
Language Processors Compilers, Interpreters, and Other Language Processors
Other Language Processors
Preprocessors: C Preprocessor, “/usr/bin/cpp”
modify source code before compilation
allow to write code in a more convenient and expressive manner
macro expansion, conditional compilation, file inclusion, constant or
variable substitution
Assemblers
translate assembly language code into relocatable machine code
specific to the target architecture or processor
Linkers: GNU Linker, “/usr/bin/ld”
combine object files to create an executable program or a library
static linking/dynamic linking
Loaders: put all executable object files into memory for execution
Debuggers: GNU Debugger, “/usr/bin/gdb”
Breakpoints, Stepping, Variable or Memory inspection, Call stack
analysis, ...
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 9 / 21
Language Processors A Language-Processing System
A Language-Processing System
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 10 / 21
Compiler Structure
Phases of a Compiler
front end or analysis part
character stream
intermediate representation
Lexical Analyzer Machine-Independent
Code Optimizer
token stream
intermediate representation
Syntax Analyzer
Code Generator
syntax tree
Semantic Analyzer target-machine code
syntax tree Machine-Dependent
Code Optimizer
Intermediate Code Generator target-machine code
back end or synthesis part
Symbol Table
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 11 / 21
Compiler Structure Lexical Analysis
Lexical Analysis
lexical analysis/scanning
grouping characters into meaningful lexemes sequences
lexeme = ⟨token-name, attribute-value⟩
token-name: abstract symbol used during syntax analysis
attribute-value: stored in symbol-table and used in semantic analysis
and code generation
An example
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 12 / 21
Compiler Structure Syntax Analysis
Syntax Analysis
syntax: describes the proper form of programs
syntax analysis/parsing
transform tokens into an Intermediate Representation, e.g., syntax tree
IR depicts the grammatical structure of the token stream
An example
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 13 / 21
Compiler Structure Semantic Analysis
Semantic Analysis
semantics: define what programs mean
semantic analysis
check semantic consistency with the language definition
gather and save type information in syntax tree or symbol table
type checking and type conversions (coercions)
An example
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 14 / 21
Compiler Structure Intermediate Code Generation and Optimization
Intermediate Code Generation
generate low-level/machine-like intermediate representation
a program for an abstract machine
easy to produce and easy to translate into target machine
An example: three-address code
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 15 / 21
Compiler Structure Intermediate Code Generation and Optimization
Code Optimization
machine-independent/machine-dependent code-optimization
objectives: produce better target code
faster running time
shorter target code
consuming less power
...
An example
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 16 / 21
Compiler Structure Code Generation
Code Generation
map an intermediate representation into target language
instruction selection
register allocation
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 17 / 21
Compiler Structure Symbol-Table Management
Symbol-Table Management
Symbol table:
a data structure recording each variable name and attributes, including
storage, type, scope, ...
should be efficient for finding records and storing or retrieving data
from the records.
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 18 / 21
Applications of Compiler Technology
Applications of Compiler Technology
Compiler design is not only about compilers
Applications
Implementation of High-Level Programming Languages
support increased levels of programming abstractions
e.g., data abstraction and inheritance in Java, type system in Rust
Optimizations for Computer Architectures
parallelism: instruction level and processor level
memory hierarchy: closest to the processor being fastest but smallest
Design of New Computer Architectures
RISC architecture
Specialized architecture, e.g., VLIW, SIMD, vector machine, ...
Program Translations
Binary translation: translate binary code for one machine to another
Hardware synthesis: translate RTL descriptions into gates
Database Query Interpreters, Compiled Simulation
Software Productivity Tools
e.g., bug detection, type checking, bounds. checking,
. . . . . . ....
. . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 19 / 21
End-of-Chapter
Compilers: Principles, Techniques, and Tools
Chapter 1 Introduction
Dongjie He
University of New South Wales
https://dongjiehe.github.io/teaching/compiler/
29 Jun 2023
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 20 / 21
End-of-Chapter
Lab 1: Get Familiar with LLVM’s Code Structure
LLVM Project, https://llvm.org/, is a collection of modular and
reusable compiler and toolchain technologies.
LLVM Core: a modern source- and target-independent optimizer
Clang: an LLVM native C/C++/Objective-C compiler
LLD: a linker
...
Task 1: install LLVM on your machine
https://llvm.org/docs/GettingStarted.html#getting-the-source-code-and-building-llvm
Task 2: getting familiar with LLVM’s code structure
. . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . .
UNSW, Sydney Compilers 29 Jun 2023 21 / 21
