1. COMPILER DESIGN
UNIT-I

2. A typical compilation process Preprocessor
Source program with macros
A typical compilation process
Preprocessor
Source program
Compiler
Target assembly program
Try g++ with –v, -E, -S flags
on linprog.
assembler
Relocatable machine code
linker
Absolute machine code

3. What is a compiler?
A program that reads a program written in one language (source language) and translates it into an equivalent program in another language (target language).
Two components
Understand the program (make sure it is correct)
Rewrite the program in the target language.
Traditionally, the source language is a high level language and the target language is a low level language (machine code).
Source
program
Target
program
compiler
Error message

4. Compilation Phases and Passes
Compilation of a program proceeds through a fixed series of phases
Each phase use an (intermediate) form of the program produced by an earlier phase
Subsequent phases operate on lower-level code representations
Each phase may consist of a number of passes over the program representation
Pascal, FORTRAN, C languages designed for one-pass compilation, which explains the need for function prototypes

5. Compiler Front- and Back-end
Abstract syntax tree or other intermediate form
Source program (character stream)
Scanner (lexical analysis)
Machine-Independent Code Improvement
Tokens
Parser (syntax analysis)
Modified intermediate form
Front end analysis
Back end synthesis
Target Code Generation
Parse tree
Semantic Analysis and Intermediate Code Generation
Assembly or object code
Machine-Specific Code Improvement
Abstract syntax tree or other intermediate form
Modified assembly or object code

6. Scanner: Lexical Analysis
Lexical analysis breaks up a program into tokens
Grouping characters into non-separatable units (tokens)
Changing a stream to characters to a stream of tokens
program gcd (input, output); var i, j : integer; begin   read (i, j);   while i <> j do     if i > j then i := i - j else j := j - i;   writeln (i) end.
program gcd ( input , output ) ; var i , j : integer ; begin read ( i , j ) ; while i <> j do if i > j then i := i j else j := i i ;  writeln ( i ) end .

7. Scanner: Lexical Analysis
What kind of errors can be reported by lexical analyzer?
A = b

8. Parser: Syntax Analysis
Checks whether the token stream meets the grammatical specification of the language and generates the syntax tree.
A syntax error is produced by the compiler when the program does not meet the grammatical specification.
For grammatically correct program, this phase generates an internal representation that is easy to manipulate in later phases
Typically a syntax tree (also called a parse tree).
A grammar of a programming language is typically described by a context free grammer, which also defines the structure of the parse tree.

9. Context-Free Grammars
A context-free grammar defines the syntax of a programming language
The syntax defines the syntactic categories for language constructs
Statements
Expressions
Declarations
Categories are subdivided into more detailed categories
A Statement is a
For-statement
If-statement
Assignment
<statement> ::= <for-statement> | <if-statement> | <assignment> <for-statement> ::= for ( <expression> ; <expression> ; <expression> ) <statement> <assignment> ::= <identifier> := <expression>

10. Example: Micro Pascal
<Program> ::= program <id> ( <id> <More_ids> ) ; <Block> . <Block> ::= <Variables> begin <Stmt> <More_Stmts> end <More_ids> ::= , <id> <More_ids> |  <Variables> ::= var <id> <More_ids> : <Type> ; <More_Variables> |  <More_Variables> ::= <id> <More_ids> : <Type> ; <More_Variables> |  <Stmt> ::= <id> := <Exp> | if <Exp> then <Stmt> else <Stmt> | while <Exp> do <Stmt> | begin <Stmt> <More_Stmts> end <Exp> ::= <num> | <id> | <Exp> + <Exp> | <Exp> - <Exp>

11. Parsing examples
Pos = init + / rate * 60  id1 = id2 + / id3 * const  syntax error (exp ::= exp + exp cannot be reduced).
Pos = init + rate * 60  id1 = id2 + id3 * const  :=
id1 +
id2 *
id3 60

12. Semantic Analysis
Semantic analysis is applied by a compiler to discover the meaning of a program by analyzing its parse tree or abstract syntax tree.
A program without grammatical errors may not always be correct program.
pos = init + rate * 60
What if pos is a class while init and rate are integers?
This kind of errors cannot be found by the parser
Semantic analysis finds this type of error and ensure that the program has a meaning.

13. Semantic Analysis
Static semantic checks (done by the compiler) are performed at compile time
Type checking
Every variable is declared before used
Identifiers are used in appropriate contexts
Check subroutine call arguments
Check labels
Dynamic semantic checks are performed at run time, and the compiler produces code that performs these checks
Array subscript values are within bounds
Arithmetic errors, e.g. division by zero
Pointers are not dereferenced unless pointing to valid object
A variable is used but hasn't been initialized
When a check fails at run time, an exception is raised

14. Semantic Analysis and Strong Typing
A language is strongly typed "if (type) errors are always detected"
Errors are either detected at compile time or at run time
Examples of such errors are listed on previous slide
Languages that are strongly typed are Ada, Java, ML, Haskell
Languages that are not strongly typed are Fortran, Pascal, C/C++, Lisp
Strong typing makes language safe and easier to use, but potentially slower because of dynamic semantic checks

15. Code Generation and Intermediate Code Forms
A typical intermediate form of code produced by the semantic analyzer is an abstract syntax tree (AST)
The AST is annotated with useful information such as pointers to the symbol table entry of identifiers
Example AST for the gcd program in Pascal

16. Code Generation and Intermediate Code Forms
Other intermediate code forms
intermediate code is something that is both close to the final machine code and easy to manipulate (for optimization). One example is the three-address code:
dst = op1 op op2
The three-address code for the assignment statement:
temp1 = 60
temp2 = id3 + temp1
temp3 = id2 + temp2
id1 = temp3
Machine-independent Intermediate code improvement
temp1 = id3 * 60.0
id1 = id2 + temp1

17. Target Code Generation and Optimization
From the machine-independent form assembly or object code is generated by the compiler
MOVF id3, R2
MULF #60.0, R2
MOVF id2, R1
ADDF R2, R1
MOVF R1, id1
This machine-specific code is optimized to exploit specific hardware features
COP4020 Fall 2006

18. The role of lexical analyzer
token
Lexical Analyzer
Parser
Source
program
To semantic
analysis
getNextToken
Symbol
table

19. Why to separate Lexical analysis and parsing
Simplicity of design
Improving compiler efficiency
Enhancing compiler portability

20. Tokens, Patterns and Lexemes
A token is a pair a token name and an optional token value
A pattern is a description of the form that the lexemes of a token may take
A lexeme is a sequence of characters in the source program that matches the pattern for a token

21. Example Token Informal description Sample lexemesif
Characters i, f if
else
Characters e, l, s, e
else
comparison
<=, !=
< or > or <= or >= or == or != id
Letter followed by letter and digits
pi, score, D2
number
Any numeric constant
, 0, 6.02e23
literal
Anything but “ sorrounded by “
“core dumped”
printf(“total = %d\n”, score);

22. Attributes for tokens E = M * C ** 2
<id, pointer to symbol table entry for E>
<assign-op>
<id, pointer to symbol table entry for M>
<mult-op>
<id, pointer to symbol table entry for C>
<exp-op>
<number, integer value 2>

23. Lexical errors
Some errors are out of power of lexical analyzer to recognize:
fi (a == f(x)) …
However it may be able to recognize errors like:
d = 2r
Such errors are recognized when no pattern for tokens matches a character sequence

24. Error recovery
Panic mode: successive characters are ignored until we reach to a well formed token
Delete one character from the remaining input
Insert a missing character into the remaining input
Replace a character by another character
Transpose two adjacent characters

25. Input buffering
Sometimes lexical analyzer needs to look ahead some symbols to decide about the token to return
In C language: we need to look after -, = or < to decide what token to return
In Fortran: DO 5 I = 1.25
We need to introduce a two buffer scheme to handle large look-aheads safely
E = M * C * * 2 eof

26. Sentinels E = M eof * C * * 2 eof eof
Switch (*forward++) { case eof: if (forward is at end of first buffer) { reload second buffer; forward = beginning of second buffer; } else if {forward is at end of second buffer) { reload first buffer;\ forward = beginning of first buffer; else /* eof within a buffer marks the end of input */ terminate lexical analysis; break; cases for the other characters;

27. Specification of tokens
In theory of compilation regular expressions are used to formalize the specification of tokens
Regular expressions are means for specifying regular languages
Example:
Letter_(letter_ | digit)*
Each regular expression is a pattern specifying the form of strings

28. Regular expressions
Ɛ is a regular expression, L(Ɛ) = {Ɛ}
If a is a symbol in ∑then a is a regular expression, L(a) = {a}
(r) | (s) is a regular expression denoting the language L(r) ∪ L(s)
(r)(s) is a regular expression denoting the language L(r)L(s)
(r)* is a regular expression denoting (L9r))*
(r) is a regular expression denting L(r)

29. Regular definitions d1 -> r1 d2 -> r2 … dn -> rn Example:
letter_ -> A | B | … | Z | a | b | … | Z | _
digit -> 0 | 1 | … | 9
id > letter_ (letter_ | digit)*

30. Extensions One or more instances: (r)+ Zero of one instances: r?
Character classes: [abc]
Example:
letter_ -> [A-Za-z_]
digit -> [0-9]
id > letter_(letter|digit)*

31. Recognition of tokens
Starting point is the language grammar to understand the tokens:
stmt -> if expr then stmt
| if expr then stmt else stmt
| Ɛ
expr -> term relop term
| term
term -> id
| number

32. Recognition of tokens (cont.)
The next step is to formalize the patterns:
digit -> [0-9]
Digits -> digit+
number -> digit(.digits)? (E[+-]? Digit)?
letter -> [A-Za-z_]
id > letter (letter|digit)*
If > if
Then -> then
Else > else
Relop -> < | > | <= | >= | = | <>
We also need to handle whitespaces:
ws -> (blank | tab | newline)+

33. Transition diagrams
Transition diagram for relop

34. Transition diagrams (cont.)
Transition diagram for reserved words and identifiers

35. Transition diagrams (cont.)
Transition diagram for unsigned numbers

36. Transition diagrams (cont.)
Transition diagram for whitespace

37. Architecture of a transition-diagram-based lexical analyzer
TOKEN getRelop() { TOKEN retToken = new (RELOP) while (1) { /* repeat character processing until a return or failure occurs */ switch(state) { case 0: c= nextchar(); if (c == ‘<‘) state = 1; else if (c == ‘=‘) state = 5; else if (c == ‘>’) state = 6; else fail(); /* lexeme is not a relop */ break; case 1: … … case 8: retract(); retToken.attribute = GT; return(retToken); }

38. Lexical Analyzer Generator - Lex
Lex Source program
lex.l
Lexical Compiler
lex.yy.c C
compiler
lex.yy.c
a.out
a.out
Sequence of tokens
Input stream

39. Structure of Lex programs
declarations %%
translation rules
auxiliary functions
Pattern {Action}

40. Example
%{ /* definitions of manifest constants LT, LE, EQ, NE, GT, GE, IF, THEN, ELSE, ID, NUMBER, RELOP */ %} /* regular definitions delim [ \t\n] ws {delim}+ letter [A-Za-z] digit [0-9] id {letter}({letter}|{digit})* number {digit}+(\.{digit}+)?(E[+-]?{digit}+)? %% {ws} {/* no action and no return */} if {return(IF);} then {return(THEN);} else {return(ELSE);} {id} {yylval = (int) installID(); return(ID); } {number} {yylval = (int) installNum(); return(NUMBER);} …
Int installID() {/* funtion to install the lexeme, whose first character is pointed to by yytext, and whose length is yyleng, into the symbol table and return a pointer thereto */ }
Int installNum() { /* similar to installID, but puts numerical constants into a separate table */

41. Finite Automata Regular expressions = specification
Finite automata = implementation
A finite automaton consists of
An input alphabet 
A set of states S
A start state n
A set of accepting states F  S
A set of transitions state input state

42. In state s1 on input “a” go to state s2
Finite Automata
Transition
s1 a s2
Is read
In state s1 on input “a” go to state s2
If end of input
If in accepting state => accept, othewise => reject
If no transition possible => reject

43. Finite Automata State Graphs
The start state
An accepting state a
A transition

44. A Simple Example A finite automaton that accepts only “1”
A finite automaton accepts a string if we can follow transitions labeled with the characters in the string from the start to some accepting state 1

45. Another Simple Example
A finite automaton accepting any number of 1’s followed by a single 0
Alphabet: {0,1}
Check that “1110” is accepted but “110…” is not 1

46. And Another Example Alphabet {0,1} What language does this recognize?1 1

47. And Another Example Alphabet still { 0, 1 }
The operation of the automaton is not completely defined by the input
On input “11” the automaton could be in either state 1

48. Epsilon Moves
Another kind of transition: -moves  A B
Machine can move from state A to state B without reading input

49. Deterministic and Nondeterministic Automata
Deterministic Finite Automata (DFA)
One transition per input per state
No -moves
Nondeterministic Finite Automata (NFA)
Can have multiple transitions for one input in a given state
Can have -moves
Finite automata have finite memory
Need only to encode the current state

50. Execution of Finite Automata
A DFA can take only one path through the state graph
Completely determined by input
NFAs can choose
Whether to make -moves
Which of multiple transitions for a single input to take

51. Acceptance of NFAs Input:
An NFA can get into multiple states 1
Input: 1 1
Rule: NFA accepts if it can get in a final state

52. NFA vs. DFA (1)
NFAs and DFAs recognize the same set of languages (regular languages)
DFAs are easier to implement
There are no choices to consider

53. NFA vs. DFA (2) DFA can be exponentially larger than NFA
For a given language the NFA can be simpler than the DFA 1
NFA 1
DFA
DFA can be exponentially larger than NFA

54. Regular Expressions to Finite Automata
High-level sketch
NFA
Regular
expressions
DFA
Lexical
Specification
Table-driven
Implementation of DFA

55. Regular Expressions to NFA (1)
For each kind of rexp, define an NFA
Notation: NFA for rexp A A
For  
For input a a

56. Regular Expressions to NFA (2)
For AB A B 
For A | B A B 

57. Regular Expressions to NFA (3)
For A*  A  

58. Example of RegExp -> NFA conversion
Consider the regular expression
(1 | 0)*1
The NFA is  A H 1 C E  B G 1 I J D F 

59. Next NFA Regular expressions DFA Lexical Table-driven Specification
Implementation of DFA

60. NFA to DFA. The Trick Simulate the NFA Each state of resulting DFA
= a non-empty subset of states of the NFA
Start state
= the set of NFA states reachable through -moves from NFA start state
Add a transition S a S’ to DFA iff
S’ is the set of NFA states reachable from the states in S after seeing the input a
considering -moves as well

61. NFA -> DFA Example  1   1      1 1 1 C E A B G H I J D FG H I J   D F  
FGABCDHI 1
ABCDHI 1 1
EJGABCDHI

62. NFA to DFA. Remark An NFA may be in many states at any time
How many different states ?
If there are N states, the NFA must be in some subset of those N states
How many non-empty subsets are there?
2N - 1 = finitely many, but exponentially many

63. Implementation A DFA can be implemented by a 2D table T
One dimension is “states”
Other dimension is “input symbols”
For every transition Si a Sk define T[i,a] = k
DFA “execution”
If in state Si and input a, read T[i,a] = k and skip to state Sk
Very efficient

64. Table Implementation of a DFAT 1 S 1 1 U 1 S T U

65. Implementation (Cont.)
NFA -> DFA conversion is at the heart of tools such as flex or jflex
But, DFAs can be huge
In practice, flex-like tools trade off speed for space in the choice of NFA and DFA representations